Freely Available British and Irish Public Legal Information
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
Irish Court of Appeal
You are here: BAILII >> Databases >> Irish Court of Appeal >> Used Cars Importers Ireland Ltd v Minister for Finance & Anor (Unapproved) [2020] IECA 298 (06 November 2020)
URL: http://www.bailii.org/ie/cases/IECA/2020/2020IECA298.html
Cite as: [2020] IECA 298
[New search] [Printable PDF version] [Help]
UNAPPROVED
THE COURT OF APPEAL
CIVIL
Neutral Citation Number [2020] IECA 298
Court of Appeal Record No. 2014/1170
High Court Record No. 1995/1988P
Donnelly J.
Faherty J.
Murray J.
Between:
USED CAR IMPORTERS OF IRELAND LIMITED
APPELLANT
- AND -
THE MINISTER FOR FINANCE, THE REVENUE COMMISSIONERS, IRELAND AND THE ATTORNEY GENERAL
RESPONDENTS
JUDGMENT of Mr. Justice Murray delivered on the 6th of November 2020
Introduction - 1 -
Motor Vehicle Excise Duty . - 5 -
The 1992 Acts ... - 12 -
II THE DETERMINATION OF THE OMSP FOR USED IMPORTS
The development of the process for the valuation of used imports .. - 17 -
The introduction of VRT ... - 37 -
The 1994 correspondence - 50 -
III THE PROCEEDINGS AND THE EVOLUTION OF THE SYSTEM FOR DETERMINING THE OMSP
The proceedings ... - 61 -
The controversy around publication of the valuation database .. - 74 -
The Report of the Working Group - 81 -
Improvements to Revenues system .. - 88 -
IV REVENUES METHODOLOGY FOR THE DETERMINATION OF THE OMSP OF USED VEHICLES
UCIIs claim and Revenues methodology .. - 97 -
The documents - 100 -
Mr. Butlers evidence - 103 -
The evidence of Mr. Campion and Mr. Fogarty - 110 -
V THE CASE .. - 117 -
The Appeal and its operation . - 127 -
The High Court judgment ... - 133 -
The relevance of the appeal - 143 -
VII THE OBLIGATION TO TAKE ACCOUNT OF THE ACTUAL SELLING PRICE
The argument . - 155 -
The High Court judgment .. - 157 -
The legal test .. - 158 -
Application . - 164 -
VIII THE TEST FOR DETERMINING THE LEGALITY OF REVENUES METHODOLOGY ... - 171 -
IX ALLEGED INCONSISTENCIES IN REVENUES EXPLANATION OF ITS SYSTEM
The argument .. - 183 -
The evidence ... - 186 -
Analysis .. - 188 -
The High Court Judgment .. - 201 -
The High Court judgment -204 -
UCIIs case . - 215 -
The correct legal analysis .. - 222 -
Refining the grounds of challenge and the depreciation tables . - 226 -
Reliance upon the Car Sales Guide - 231 -
Systematic overvaluation - 238 -
Were the Courts conclusions as to the reasonableness of Revenues methodology sufficiently reasoned?............................................................................................................................... - 252 -
XI PROPORTIONALITY .. - 258 -
XII THE CLAIM BASED ON BREACH OF CONSTITUTIONAL RIGHTS
Findings of the trial Judge . - 262 -
The claim - 264 -
Relevant principles of constitutional law ... - 268 -
XIII TRANSPARENCY, LEGAL CERTAINTY, STATUTORY DISCRETIONS AND FIXED POLICIES
The Argument .. -270 -
The High Court judgment ... - 271 -
Relevant legal principles - 277 -
Revenues arguments .. - 291 -
Conclusion .. - 298 -
XIV WAS UCII ENTITLED TO MAINTAIN A CLAIM UNDER EU LAW ?
The High Court judgment .. - 306 -
Analysis ... -310 -
The principles . - 315 -
Obligation of transparency in EU law ... - 331 -
The issue around the residual value: the arguments of the parties and findings of the High Court .- 342 -
The case law - 351 -
Analysis .. - 365 -
The compatibility of Revenues valuation methodology with EU law: the High Court judgment . - 371 -
The VAT Directive .. - 380 -
XVI DAMAGES AND RECOVERY OF TAX
The causes of action for damages and the High Court judgment .. - 386 -
Assessment . - 392 -
The claim for recovery of VRT paid ... - 399 -
The findings of the High Court as to loss ... - 401 -
The claim for declaratory relief . - 408 -
XVIII CONCLUSIONS AND ORDERS ... - 415 -
I THE CONTEXT
Introduction
1. The plaintiff (UCII) was incorporated in 1989 and began trading that year. As of its establishment Niall ODowling and Fintan Riordan were equal shareholders in and the only directors of the company. Both had been involved in different aspects of the motor trade for many years. Their intention was that UCII would develop a business of sourcing used cars abroad (including in other European Union member states) and selling the vehicles thus imported to other dealers and to the public. In particular, Mr. ODowling and Mr. Riordan had come to the view that there was a market for, and shortage of, good quality second hand Japanese cars in Ireland. Mr. ODowling identified contacts in Japan and eventually put in place a system for purchasing vehicles at auction there. He secured shipping lines and necessary credit facilities.
2. A disused factory premises was acquired near the port in Cork city. From that premises UCII sold the cars it imported to a network of dealers it had established, as well as to the general public. In the early years UCIIs business grew rapidly so that by 1992 it was selling in the region of 670 units per annum. Mr. Riordan and Mr. ODowling believed that with the introduction of the single European market in 1993 the opportunities for their car importation business would further expand.
3. On 1 January 1993 (and coincident with the opening of the single market) the legal provisions governing Vehicle Registration Tax (VRT) took effect. VRT was payable on the registration of all motor vehicles including those imported into the State. It replaced a tax known as motor vehicle excise duty (MVED) previously levied on the importation or delivery of such vehicles. VRT was initially introduced and regulated by the provisions of ss. 132 and 133 of the Finance Act 1992. These have since been amended from time to time. The effect of the statutory provisions governing VRT throughout the period with which this action was concerned was that the tax was to be imposed by reference to the chargeable value of the motor vehicle in question, defined in turn as its open market selling price (OMSP). For second hand vehicles, the OMSP was dependent upon the opinion of the Revenue Commissioners (Revenue) as to the price the vehicle might reasonably be expected to fetch on a sale in the open market.
4. UCII said that from the introduction of VRT it encountered difficulties in ascertaining the methodology by which Revenue determined the OMSP of used cars it imported. That methodology, UCII claimed, had resulted in it paying more VRT than was properly due in respect of vehicles it sold. In this action it complained that the manner in which VRT was operated and administered by Revenue over the period from 1993 to 2010 was contrary to the governing legislation, to principles of legitimate expectation, proportionality, legality, equality and legal certainty and that it was unreasonable. It further contended that Revenue had breached EU law by discriminating against imports. All of this, UCII said, resulted in VRT being imposed on used car imports at a rate that was too high, thereby increasing the price of those cars and damaging its business. It sought various declaratory reliefs arising from these claims, including declarations that the provisions pursuant to which VRT was calculated and levied were invalid having regard to the Constitution. It also claimed for the return of taxes paid by it and/or damages in a total amount of 131M. Following a thirty-three-day trial Murphy J. ([2013] IEHC 128) refused UCII all of the relief claimed by it.
Motor Vehicle Excise Duty
5. Until 1985 there were significant restrictions on the importation of motor vehicles into Ireland. Originally, imports of fully built cars were restricted by a system of quotas and tariffs, the former being made available only to companies prepared to engage in motor assembly in the State. The Motor Vehicles (Registration of Importers) Act 1968 modified this system and provided for a system of registered importers of motor vehicles, with the requirements attached to the grant of such registration being operated in such a way that by the beginning of the 1980s the majority of cars supplied to the Irish market were imports controlled by manufacturers previously engaged in domestic motor assembly. In effect, private citizens were prohibited from importing motor vehicles into the State and the undertakings that could do so were limited in number - by 1985 there were only 33 firms licensed to engage in vehicle importation. This regime, which would otherwise have been contrary to those provisions of the Treaty of Rome governing the free movement of goods, was permitted by a derogation provided for in the Treaty of Accession (Protocol No. 7). This derogation expired on 31 December 1984.
6. In anticipation of the changes to the system of importation that would be thus necessary, a restructuring of the taxation regime for motor cars occurred in 1979. The Imposition of Duties (Excise Duties on Motor Vehicles, Televisions and Gramophones Records) Order S.I. 57/1979 provided for inter alia MVED to be charged, levied and paid on motor vehicles delivered in or imported into, the State. MVED was based on a chargeable value, in turn referable to an OMSP. For motor vehicles this was defined as the price (exclusive of taxes) that, in the opinion of Revenue, a motor vehicle might reasonably be expected to fetch on a first arms-length sale after manufacture or importation in the open market in the State to a dealer. As explained in UCIIs expert evidence to the High Court, this reflected the fact that the tax on both assembled cars and fully built units was based on the wholesale landed price in Ireland.
7. It became necessary to relax the prohibition on importation from January 1985. In that connection, the Imposition of Duties (No.272) (Excise Duties on Motor Vehicles) Order 1984, S.I. 353/1984 (the 1984 Regulations) amended the 1979 Order by changing the definition of OMSP. For new vehicles delivered or imported for re-sale the chargeable basis for the duty was referenced to the retail price at the time of the charging of the duty. Where not delivered or imported for re-sale, the basis for the duty was the retail price of a motor vehicle of a similar or corresponding type to the vehicle on sale by retail. For these purposes, retail price did not mean the price actually charged on any particular sale : it was defined as the recommended retail price (RRP), inclusive of taxes and duties standing declared to Revenue by the relevant manufacturer or importer (or where more than one, the wholesale importer) in respect of the vehicle. Revenue had the power to determine the retail price where none was declared or where that declared was unreasonably low.
8. While a system of ad valorem duty charged by reference to a recommended retail price operates relatively smoothly for new vehicles, the identification of an appropriate reference point for the charging of such a tax on used cars is more complex and, if it is to be accurate, should take account of the wide variety of factors that affect the value of such a vehicle. The process of determining that value was described during the trial as noisy, the price obtainable for used cars being affected by their condition, their kilometrage, their age and by whether any additional features (or extras) have been incorporated into them. Indeed, the evidence before the High Court was that that price even varied - in some cases significantly - according to the county in which the vehicle had been registered.
9. The 1984 Regulations addressed this via a clear, if crude, mechanism. They provided for the chargeable value of imported used vehicles to be determined by reference to the retail price of a new motor vehicle of a similar or corresponding type reduced by reference to a depreciation scale detailed in a Table to the instrument. This Table sought to reflect depreciation at specified percentages depending on the age of the vehicle. This single statutory scale was almost linear for the first six years thereafter flattening out. The evidence to the High Court was that the rate of depreciation envisaged in the order was particularly low, resulting in a high chargeable value of the purposes of excise duty, and underestimating - in particular - the depreciation of one, two, and three-year-old cars. The depreciation scale was itself both artificial and acknowledged not to generally reflect the market value of the vehicle. The system further failed to take account of differing depreciation rates in different models, making no allowances for the mileage or condition of an individual vehicle.
10. That said, the parameters within which the duty was imposed were readily identifiable - the RRP of vehicles was publicly available, and the applicable rate of depreciation was provided for by law and capable of ascertainment. When UCIIs business started and as it acquired vehicles in Japan, it was able to determine the tax liability for any particular vehicle and to work backwards from that to fix a ceiling of what it should pay for each vehicle.
11. The advent of the Single Market meant that this system of import duties had to be removed and the extent of the revenue generated by the duty meant that it had, in some way, to be replaced. That replacement took the form of VRT, a tax to be imposed on all vehicles at the point of first registration in the State. By fixing the event giving rise to the liability to pay the tax in this way the Oireachtas both enabled the revenue previously generated by MVED to continue while at the same time seeking to avoid discriminating to that end against imports from other Members States of the European Community. As the provisions to govern VRT were being formulated and, the evidence was, in anticipation of a significant increase in business with the abolition and replacement of MVED, Mr. ODowling and Mr. Riordan purchased a larger site in the vicinity of the port at Cork City which included a capacity to hold up to 2,100 cars in bond. They leased this to UCII. In July 1992 they retained a former Revenue official, Mr. OHerlihy, to advise them in relation to the changeover to the new tax.
The 1992 Acts
12. Although not taking effect until January 1 1993, ss. 132 and 133 of the Finance Act 1992 (FA92) were enacted in May 1992. In their original form these sections did not make specific provision for used vehicles instead providing generally that VRT was to be charged by reference to the value of a vehicle, that that value was to be the OMSP, and that in respect of new vehicles the supplying manufacturer or distributor would declare to Revenue its opinion as to the price that vehicle might reasonably be expected to fetch on a first arms length sale. In respect of such new vehicles, the price so declared would be deemed to be the OMSP.
13. Section 9 of the Finance (No.2) Act 1992 (2FA92) was enacted in December and amended s. 133 so as to make specific provision for used cars (or, as the distinction was described in the Act, vehicles other than new vehicles). What differentiated vehicles that were new and those that were not for this purpose was whether they were less than three months old or had a mileage of less than 3,000 km.
14. These sections combined to make three fundamental changes to the system for taxation imposed on the sale of imported motor vehicles. First, the point at which the charge was to be applied was changed from that of importation or, in some cases, delivery, to that of the registration of the vehicle. This, it might be noted, inevitably entailed a significant increase in the administrative burden placed on the system of tax collection. Moving the event that triggered the liability from the control of a small number of importers (there were less than 30 at the time of the introduction of the tax) into the general process for the registration of motor vehicles necessarily affected the scale of the operation required to determine the chargeable basis for individual vehicles. However Revenue went about determining the OMSP of used cars for these purposes, it had to accommodate that increased scale in its process. Second, the new system - at least formally - removed RRP as the basis for the tax, and for used vehicles the depreciation table which had been a feature of the 1984 Regulations was also removed. Third, FA92 made specific provision for the taking into account of specified features of the car and, as will be explained, was understood as requiring regard to be had to the condition and mileage of the vehicle.
15. However, two features of the MVED regime were retained. The approach whereby chargeable value was not stated to be related to the transaction price remained a feature of the system introduced by FA92 as amended. A distinction between the method of calculation of the charge for new and for used vehicles was also maintained. The OMSP for a new car was defined (s.133(2)(b) FA92) as the price, inclusive of all taxes and duties, which in the opinion of the sole wholesale distributor, a model of that make and specification might reasonably be expected to fetch on a first arms length sale thereof in the State by retail, this being fixed by reference to the declaration of the manufacturer or distributor. As matters transpired, when declared this did not include any charge for delivery or preparation but it did include VRT. In respect of a used vehicle, the opinion of the manufacturer or distributor was replaced with the opinion of Revenue. Thus, the OMSP for all vehicles was mandated by the legislation as (as it was described by the economist Mr. McDowell in giving expert evidence for UCII) an artificial price. It was thus defined in respect of a used vehicle by s. 133(3)(c) FA92:
the price, inclusive of all taxes and duties, which in the opinion of the Revenue Commissioners, a model of that make and specification might reasonably be expected to fetch on a first arms length sale thereof in the State by retail.
16. As I explain later, s. 138 of the FA 1992 provided for an appeal against a decision of the Commissioners as to the amount of tax charged and s. 141 provided for the power to adopt regulations inter alia for the purposes of prescribing the method of charging, securing and collecting VRT. No such Regulations were promulgated during the period with which this action was concerned. Finally, the effect of s.136 of the Act was that an authorised dealer in motor vehicles (and UCII was such a dealer) could not deliver, send out or otherwise make available for use a vehicle which required to be registered, without registering it - and thus paying the tax due on registration.
II THE DETERMINATION OF THE OMSP FOR USED IMPORTS
The development of the process for the valuation of used imports
17. Both at trial and in the course of this appeal UCII placed some emphasis on the temporal gap between FA92 and 2FA92. Once the former was enacted Revenue had to address how it was going to go about determining the OMSP of used vehicles in accordance with those provisions. FA92 provided no guidance as to how this was to be done. 2FA92 provided some more definition by expressly relating the value of used vehicles to the opinion of Revenue. In fact, by the time that legislation was introduced the process of determining the methodology by which the OMSP would be decided had concluded.
18. Mr. Niall Butler, Surveyor, Custom and Excise was the person tasked with the establishment of the Central Vehicle Office (CVO) (to be Revenues central administrative office and operational hub for the implementation and management of VRT). The CVO was located in Rosslare, Co. Wexford. He was assigned as manager in May 1992 and remained in that position until July 1995. Part of his function involved the development of the technical and business aspects of a valuation system for new and used vehicles. He commenced the design of the valuation system for used vehicles in June of that year.
19. Mr. Butlers evidence was that he was not provided with any written instructions as to how he should go about designing that system and it was accepted that there was no protocol identifying how Revenue would implement the 1992 Act. He set about preparing a report which was completed and issued on August 18 1992. He referred to this report throughout his evidence in the High Court as a framework proposal. He prepared it without the benefit of legal advice. The system was not revisited after 2FA92 was promulgated and indeed Mr. Butler accepted under cross examination that he could not recall paying any particular attention to the amended definition in 2FA92 (Day 23 p. 77 Q.219).
20. Mr. Butlers report outlines a system which was similar in many respects to that in place under the MVED regime. It proposed that the OMSP of a used vehicle would be calculated in the first instance by identifying the OMSP of a vehicle of the same make and model as new. In that regard it was necessary to distinguish between different categories of vehicle. The prices of those vehicles available in the Irish market could be checked by reference to the price declared by the manufacturer or distributor. Where the exact same model was not available on the Irish market but had at one point been sold in the State the last available value would be identified and updated by reference to the Consumer Price Index. Where the vehicle had never been available on the retail market in the State the nearest model would be established by reference to the guides available on that date, by consultation with the trade if necessary and by reference to any established precedents. Where the vehicle did not fall within one of these categories and was unusual, a range of similar models would be examined and a mean value determined. Importantly, the object - as explained by Mr. Butler in the course of his evidence - was to ascertain the price of the vehicle when sold as new as of the date of the valuation (Day 23 p. 87 Q. 259).
21. The OMSP of the new vehicle thus ascertained would be reduced in accordance with a depreciation table with further reductions then made for the condition and mileage. Inspections of the vehicles were to be directed to assessing the vehicle for the purpose of these deductions. As explained in the report the proposal was that the officer examining the vehicle would note the model, any extras, and the mileage on the clock and make an informed judgment on the condition. He would access Revenues own system which generated a list of models and statistical codes from which the officer would identify the exact model. That would provide a value for a model of the age of the vehicle in question. The officer was then to key in data regarding condition, extras, and mileage following which the system should make appropriate adjustments and thus identify the OMSP.
22. In the course of his evidence Mr. Butler explained that Revenue initially decided to adopt five tables for the purposes of addressing depreciation. He said:
An examination was conducted on a wide variety of models as detailed in the Car Salesman Guide. The guide prices for year one and as many subsequent years as were relevant were examined. The percentage depreciation as against the current retail selling price was calculated. Where possible, a cross check was made with advertised selling prices.
23. The Car Salesmans Guide referred to here (entitled in the papers with which this Court has been furnished and referred to throughout the evidence in the High Court as The Car Sales Guide) featured prominently in the High Court and occupied the focus of a number of different aspects of this appeal. The evidence of Mr. Butler was to the effect that it was the primary - but not the only - source used to determine prices for the purposes of constructing (and thereafter periodically reviewing) the depreciation tables. It would appear that at the time of the introduction of VRT the Guide was produced twice yearly by a very small company in north Dublin for the purposes of sale within the trade - although as one of the revenue documents observed it was theoretically available to the general public. The evidence was that by the time of the hearing in the High Court it was being updated monthly. It provided prices paid in relation to domestic cars over an eight-year period. It thus did not provide information regarding models that were more than eight years old, did not cover Japanese models, published only the prices obtained by main dealers (therefore excluded other car dealers, and prices paid for sales at auction) and quoted prices that were margin inclusive, incorporating the cost of servicing and some degree of warranty.
24. While the introductory text to the Car Sales Guide changed from year to year, for much of the period to which the proceedings related, it stated:
Guide prices quoted in this book are those which a main dealer would normally receive for a particular car in this range.
25. However, Mr. Butler stressed in his evidence, the Car Sales Guide was not the only source used for this purpose. Regard was also had to newspaper advertisements and trade magazines and this was supplemented by information received from individual dealers. The evidence from Revenue was clear as to the purposes of this exercise. Mr. Butler explained it as follows under cross examination (Day 23 p. 54 Q.114):
We were not valuing individual cars. What we were doing was building a system which took into account most of the variables that contribute to establishing a second-hand vehicle and putting that on a computer system and making sure it stood up and did reflect actual values.
26. As I have noted, Mr. Butlers report acknowledged that Revenue had to do more than simply develop a method for determining the value of the car when new and a system by which that value could be adjusted to reflect the age of the vehicle. It was also necessary to take account of condition and of mileage. The former was addressed by identifying three categories of condition - good, fair and poor. Good would be the standard and would involve no reduction. In respect of vehicles categorised as presenting a fair condition the OMSP would be reduced by 5% and for cars in poor condition it was reduced by 10%. A mileage adjustment was also to be made loosely based on mileage allowances given by the motor trade. For diesel vehicles the proposal involved a reduction of £20,000 per year and other vehicles £15,000 per year. Then, for every 1000 miles that a vehicle exceeded the standard allowance an amount of £50 would be credited with a limit of £500.00. Mileages of less than 1000 would not be counted. These, Mr. Butler explained in evidence, were reflective of the market reductions for mileage at that time and they were put in place in consultation with the trade (Day 23 p. 108 Q. 317).
27. Shortly after it was completed Mr. Butlers report was sent to external consultants - Edge Anderson - for review. They reported their opinion on August 24. They described the process for the ascertainment by Revenue of the OMSP for used vehicles as follows:
The open market selling price will be calculated by reducing the open market selling price of the new vehicle in accordance with a depreciation table and incorporating further reductions for the condition and mileage. There are other factors that will be taken into consideration.
28. They observed in the course of their report:
The system will not always calculate the exact market value for each and every individual vehicle, but does go a long way to presenting an acceptable valuation on normal rule of thumb and methods of depreciating such vehicles Properly applied it would seem that the depreciation table would be a fair and reasonable method for the calculation of acceptable depreciation figures for calculating the new registration tax.
29. They also noted:
In very few instances are the proposed changes higher than that recommended in the Car Salesmans Guide which is a recognised parameter for vehicles within the motor market.
30. They concluded:
Finally, whereas we have made a few comments, we are of the opinion that in general the proposals put forward are sound, comparatively easy to operate and provide a fair basis for individuals importing second-hand cars and also taking into consideration the limitation factors that will be in force to ensure that the Government achieve the correct tax yield.
31. Edge Anderson were not told which cars were going to be on which schedule and were only given five schedules. The number of schedules proposed was increased to fourteen at the end of 1992 and was further increased in later years. That increase was said by Revenue to arise from further analysis of the values of used vehicles undertaken after the drafting of the August 1992 framework report.
32. In December 1992 Revenue produced a staff guide addressing the valuation system for new and used vehicles for the purposes of VRT. This detailed document broadly reflected the report prepared by Mr. Butler. The staff guide operated on the basis that the depreciation tables were central to the process, directing that the methodology would involve the reduction of the OMSP of a specific new vehicle in accordance with the relevant age depreciation table and incorporating further adjustments for condition and mileage where appropriate. The procedure for the valuation of extras was also explained, the value of these being added to the calculated OMSP of a similar new vehicle before the depreciation table was applied. The document provided a detailed calculation sequence for determining the OMSP for VRT purposes, directing the relevant officers to the order in which the relevant variables were to be inputted into the system in order to generate the appropriate value. The staff guide also explained the process of appeal. By this stage an additional nine depreciation tables had been added, these being built into the computer system at the end of November.
33. The depreciation tables were included in the staff guide. It explained that each and every vehicle would be categorised in accordance with one of those tables. It explained how they had been devised, as follows :
Through a process of trial and elimination fourteen depreciation tables were found to be adequate in describing yearly price variations for all vehicles available in the State. The technique and procedure outlined below are capable of expansion and enhancement to cater for additional models which will become available in the future. A comprehensive examination of a wide range of vehicle models listed in the Car Salesmans Guide was undertaken with the purpose of establishing whether or not regular patterns and trends existed for price variations from one year to the next. The Guide prices in respect of year one and information on as many subsequent years as were available were examined and the percentage depreciation as against the current recommended retail selling price (RRSP) were calculated. Where possible a cross check was made with advertised (newspapers) selling prices to ensure that there were no major disparities in the Guide prices.
34. The last event in the development of the methodology prior to the introduction of VRT was effected after publication of the staff guide and involved the adjustment upwards of a factor of 1.06 for all depreciation points across the depreciation tables. This followed a decision that, in relation to new cars, the OMSP should be fixed at 94% of the price declared by the distributors. This was done because the declared price did not reflect actual market prices - which were usually approximately 5% to 10% less than the advertised recommended retail selling price. The main distributors therefore advised Revenue that they would declare OMSPs at 94% of current prices. That figure of 94%, it was confirmed in evidence, was not uniform, rather it represented an average figure which, it was believed, was a reasonable reflection of the market.
35. In respect of used cars there was insufficient time to re-programme the depreciation tables before the system became live, and the 1.06% was accordingly applied to the tables by increasing the depreciation tables by 1.06 for each point on the scale of the 14 tables. The tables provided, it should be observed, a single first depreciation stage of three to twelve months. There was a monthly adjustment range running from + 5% in the case of VRT charged in January, to -5% in the case of VRT charged in December.
36. One feature of the system thus designed by Mr. Butler merits emphasis. The effect of that system was to take account of all the variables leading to the price of a used car except the price actually paid for that specific vehicle. Mr. Butler explained this in his evidence in terms that the only factors relevant to the price that were taken out were the negotiation skills of the buyer and negotiation skills of the seller and, possibly, the pricing policy of the particular dealer (Day 22 p. 76 Q.234). He said that this was done in order to ensure consistency in the valuation method (id.).
The introduction of VRT
37. On 1 January 1993 the VRT computer system went live. It had two functions - the calculation of VRT due on a vehicle being registered in Ireland for the first time and the registration of the vehicle and assignment to it of a registration number. The tax (which at that point was charged at a rate of 25.75% for cars under 2012 ccs, and 29.25% for cars over that threshold) was originally levied on the relevant percentage of the chargeable value of the vehicle or a prescribed minimum amount, whichever was the greater. Since July 2008, both the percentage and minimum amounts are determined by reference to the level of CO2 emissions at the time of manufacture.
38. The evidence tendered by Revenues witnesses as to fact suggested that the system introduced by Revenue in January 1993 for the calculation of VRT on used imported vehicles did not change in its essential structure over the following two decades - although as I explain later an important development occurred in November 1999 when a facility to create and apply hybrid depreciation tables was introduced.
39. The system so formulated and subsequently operated by Revenue corresponded generally with that proposed in Mr. Butlers report and developed and explained in the staff guide. It began with the depreciation tables. Revenues evidence was that these were developed by it to reflect identified patterns in the depreciation of used vehicles. The tables were, it said, validated against the used vehicle trade in the State. Following a number of iterations, the relevant officials determined that the tables accurately reflected market conditions at the time and became part of the process of valuation. Revenue witnesses said that the schedules were monitored and refined so as to ensure that they retained currency with the used car market. Mr. Butler described the system as a living organism that continues to reflect open market selling prices by means of a process of review.
40. An example was recited in a note prepared by Mr. Coyle of Revenue and included by the parties in the papers provided to this Court. Mr. Coyle did not himself give evidence at the trial (although as late as Day 18 he was identified as Revenues lead witness as to fact). The note was read into evidence by Revenues expert witness Dr. Bacon and formed the basis - at least in part - of his understanding of how the methodology operated. The example in the note was of Vehicle X presented for registration. The research into the vehicle shows that a 2 year old version depreciates on average by 68%, a 3 year old version by 62% and a 4 year old version by 53%. The table most closely matching that pattern of depreciation is Table 6, with rates of 69%, 61% and 52% for 2, 3 and 4 year old models respectively. Therefore, that table is allocated to vehicle X. If a 6 year old example of vehicle X is imported the VRT will be calculated based on 37% of the OMSP because according to Table 6 a six year old vehicle has depreciated to that level. In that way, Revenue systems could calculate the VRT due on all subsequent vehicles of that particular model or variant irrespective of the age of the model presented. It could do this as the valuation officer had already established the depreciation characteristics for that particular model. Mr. Coyles note, it should be observed, described the depreciation tables as part of the valuation process.
41. In March 1993, Revenue issued a Trader Guide to Vehicle Registration Tax. This made no reference to the depreciation schedules and contained significantly less information as to the process of valuation than was disclosed internally in the Staff Guide issued four months earlier. Para. 5.3 provided the only explanation of how the valuation of imported used cars was approached. It said :
42. While UCII objects in these proceedings to the use of a notional price calculated in this way as the basis for the OMSP, it was Revenues evidence that it offered to consider using the actual price paid by the eventual purchaser of a used imported vehicle as the basis for calculating the OMSP for this purpose. Thus, on 22 September 1993, a meeting occurred between Revenue and representatives of the trade, including both Mr. ODowling and Mr. Riordan. The meeting arose in a context where Revenue officials had visited a number of traders and, upon examination of their books and records formed the view that the system operated by Revenue was generating values that were under the actual open market selling price of Japanese imports. The purpose of the meeting was to explain Revenues view that Japanese used cars had been undervalued in the Revenue valuation system and to communicate its intention to increase the OMSPs by moving those vehicles from their existing depreciation tables to tables that would generate higher and, it was believed, more accurate OMSPs. Data was presented to the trade representatives outlining a series of under valuations by a range of imported Japanese vehicles. According to Mr. Butler, the values presented to the trade representatives were not disputed.
43. It was at that meeting that Mr. ODowling contended he first learnt of the depreciation scales which, he said was described as a sliding depreciation scale as in the old MVED system (although in fact correspondence produced to the High Court showed he had been advised by Mr. OHerlihy of the scales a year earlier). He gave evidence that he was told at that meeting that Revenue intended to further increase the VRT on used imports by up to 25%.
44. It was the evidence of Dr. Les Lennox, the head of Revenues excise administration branch, that at that meeting Revenue said that it could tax on invoice sales but that this prompted a negative reaction from the trade representatives. This was accepted by Mr. ODowling, who said that this was because to approach the matter in that way would involve taxing their added value in the form of dealer margins, pre-delivery servicing, and tyres. Revenue said that it indicated that some of this could be addressed through the preparation work being done after registration thereby excluding it from the VRT with, effectively, the production by the dealers of two invoices.
45. Dr. Lennox and Mr. Butler both said in evidence that the reason the traders did not want to use the invoice price as the basis for VRT was that this would require that Revenue be granted access to their accounts and workings so as to ensure that the invoice price was in fact the correct price. Revenue said it was made clear at this meeting that if it was going to levy VRT on invoices it would have to exercise very stringent control over the invoicing operations of used car dealers. That, it would appear, was not welcomed by those attending the meeting: according to Dr. Lennox the suggestion met with a very strong negative response. The implementation of the proposed changes was deferred following this meeting at the request of the trade representatives pending their meeting with the Department of Finance.
46. One feature of the system should, at this point, be re-iterated, and two others noted. The records incorporated into the valuation system included a statistical code, to which I have made reference above. The statistical code uniquely identified a particular make, model, and variant of a vehicle with an individual OMSP and depreciation group determined by a Revenue valuation official for that vehicle, the latter two of which were (Revenue said) frequently reviewed and thus liable to change. Traders were expected to identify the relevant statistical code when applying to Revenue which could then input that code to its computer system, thereby causing the selection of the appropriate depreciation schedule to reflect (Revenue said) the Car Sales Guide prices and other prices suggested by the trade information consulted by its officials.
47. Once a statistical code was generated, it remained permanently on the VRT computer system as the key identifying field for a particular vehicle. According to Mr. Butler, Revenue did not keep the statistical codes secret. His evidence was that it was always the policy and practice of Revenue to provide these to the public and the trade subject to relevant details of the vehicles being provided. It was Mr. ODowlings evidence that those codes were furnished to him in April 1993, although Mr. Riordan complained in evidence that he did not receive the manual explaining how these were to be applied until July 1995. Either way it is clear that the statistical code did not itself identify the OMSP of a vehicle : its primary utility was in obtaining estimates, and it was required for the purposes of registering the vehicle.
48. As I have explained, the system for valuation involved two different processes. One required the VRO to input the statistical code for the vehicle in question into Revenues system to produce the appropriate valuation or, where there was no statistical code in existence for a particular model, for CVO to generate one and allocate it - based on appropriate research - to a particular pre-existing schedule. Another involved inspection of the vehicle at the time of registration. As outlined by Revenues witnesses this inspection was directed to five features of the vehicle - that the particulars declared were consistent with those of the vehicle presented for registration, that the vehicle had been correctly classified for VRT purposes and the correct statistical code allocated, that chargeable extras had been identified and included in the OMSP, that the declared odometer reading and the corresponding depreciation allowance for excessive mileage were correct, and that the condition of the vehicle (good, fair or poor) was properly assigned and the condition depreciation allowance was correct.
49. Revenue also instituted from the commencement of the tax a system for the provision to any member of the public furnishing the relevant details of a used car of estimates of VRT payable on that vehicle. These estimates were not binding on Revenue, were valid for a two-week period following their issue, and were qualified by reference to the details provided by the requester and to the prospect of inspection of the vehicle. It would also seem that while appeals were not entertained against these estimates, in practice Revenue officials were (at least on occasion) prepared to entertain and act upon representations made in respect of estimates they issued. Revenues evidence was that in 1995, 75,000 estimates were supplied by it. Mr. ODowling agreed in the course of his evidence that in the period from 1993 to 1995 he regularly sought estimates in respect of vehicles he had already purchased in Japan.
The 1994 correspondence
50. On the 27 January 1994 the rate of VRT dropped by 2.55% from 25.75% to 23.2% for cars under 2012 ccs, and from 31.7% to 29.25% for cars over that threshold. However, and at the same time Revenue adjusted upwards the values of used cars imported directly from Japan on its system thereby increasing the VRT payable on registration of those vehicles. This was in accordance with the notice given at the meeting the previous September and (the evidence was) followed from information gathered by revenue officers in the course of inspection of the records of car sales outlets.
51. On 18 January 1994, the CVO wrote to UCII advising that a review of the OMSP prices of used cars imported directly from Japan had been undertaken, and that:
In view of the results of this examination the values of such cars have been adjusted on our computer system.
The new values and consequently new VRT duties will come into effect on 7th February 1994.
52. At the same time, a number of changes were made to the valuation system with a view to enhancing its precision. The fourteen depreciation tables were (effective from March 29 1994) replaced by twenty-two tables. This was done to cater for more accurate valuations at the high and low ends of the market. The existing tables were amended inter alia by providing for two depreciation stages within the first year of three to six months and six to twelve months. A new table was introduced to allow for the setting up of statistical codes for unusual cars, which might not depreciate at all or might do so idiosyncratically. For these there was to be no scale as such, the idea being that the vehicles would be valued manually. The monthly adjustment range was changed to + 3% in the case of VRT charged in January, to -3% in the case of VRT charged in December. The 1.06% increase included in December 1992 was simultaneously removed from the system and the assumed standard yearly average mileages were revised to 16,000 miles for diesel and 11,000 miles for other vehicles, the mileage allowance was increased to £60 per whole 1,000 miles in excess of standard, up to the lesser of £1,000 or 10% of the OMSP depreciated for age, with no mileage allowance being allowed for vehicles aged ten years or more.
53. Revenues letter of 18 January provoked correspondence from UCIIs solicitors objecting to the proposal which, the former noted, their client had been advised would result in across the board increases of between 5% and 30% over existing values. Their first letter (of 2 February 1994) not having been responded to, on 2 March 1994 UCIIs solicitors wrote :
I am further instructed by my client to demand from you full details how the system by which the Open Market Selling Price of imported vehicles is arrived at. Further, that the Revenue Commissioners publish and supply to my client or to this office full details of the values determined by you for all types of imported vehicles since the introduction of the VRT system, and as changes are introduced from time to time that full details actual values - would be supplied to my clients before the increase takes effect.
This information is absolutely necessary for my clients to plan and operate their business. It is impossible for them to operate under a tax system which is, in effect, secrete [sic.]
54. This was responded to on 21 March, Revenue saying:
In the operation of the system, each imported second-hand vehicle is examined by a Revenue official and the value is determined by reference to an established norm of values of similar vehicles sold in the State whether imported second hand or not.
The system of valuation has been in operation from 1 January, 1993. No arbitrary block adjustments are being made. The system by its very nature is subject to continuous adjustment based on the prices achieved for all used vehicles sold in the State. Any adjustment being made will be part of that continuing process and will, as indicated, apply to all similar vehicles equally irrespective of origin.
55. This explanation, it will be noted, made no reference to the use of depreciation schedules let alone to how those schedules were used to arrive at an OMSP.
56. UCIIs solicitors responded to this on 25 March. They asked Revenue to explain what the established norm of similar vehicles sold in the State was. They asked how that established norm of values was arrived at or determined by Revenue and what the relationship was between new car prices and that established norm of values. They enquired as to what form the established norm took, how often it was changed, whether it was calculated on a national or regional basis and asked that the list of the established norm of values would be furnished. They asked for confirmation that as changes in the VRT system were introduced that full details would be supplied to UCII, and they asked for a worked example of the method of calculation of VRT setting out in detail how it was determined.
57. Neither that letter, nor a subsequent reminder of 11 April was ever substantively responded to by Revenue.
58. I have already noted that s.138 FA 92 provided for an appeal against a valuation for the purposes of VRT. That facility was availed of at least twice by UCII in the course of 1993. On 19 July 1994, UCIIs solicitors responded to a letter from Revenue advising that an appeal lodged in respect of two such vehicles had been allowed. The former advised Revenue that UCII did not accept the decision (in their favour) and querying how very significant disparities between the initial valuation and that allowed on appeal arose. The letter interrogated the research that had led to that conclusion submitting that the value ought to have been ascertained by reference to the real market price established by a genuine commercial transaction conducted at arms length. The cheque Revenue had sent when advising of the successful appeal was returned.
59. On 31 August 1994 UCIIs solicitors sought again to pursue the queries they had raised in their letter of 25 March, repeating their objection to the secret system of valuation of vehicles and advising of communications with the European Commission DG 21 in relation to the matter.
60. On 24 October, Revenue responded in respect of the vehicle that had been the subject of the successful appeal earlier that year, noting that it had encountered difficulty in valuing the car as it was rarely traded and that there was a divergence of opinion amongst experts consulted. It advised that it proposed to use the transaction price as the basis for the valuation as it appeared bona fide. The letter recorded Revenues opinion that the method used by it to determine OMSP was correct and in accordance with law.
III THE PROCEEDINGS AND THE EVOLUTION OF THE SYSTEM FOR DETERMINING THE OMSP
The proceedings
61. When the proceedings were instituted in March 1995 UCII had been unsuccessfully pressing Revenue to provide basic information in respect of the functioning of the VRT system for twelve months. In May of that year UCII brought an application for a mandatory interlocutory injunction requiring the provision of that information. That was heard by Costello P. on 17 July 1995 and was refused, seemingly on the basis that the information sought could be obtained on discovery in the action.
62. The Statement of Claim, delivered shortly after the refusal of the injunction application, presented the following claims:
(i) That the manner in which the VRT system was operated was secretive, arbitrary and lacking in transparency and legal certainty;
(ii) That this uncertainty placed a severe burden on the plaintiffs ability to manage and plan its business;
(iii) That the values used by Revenue to calculate VRT were well in excess of actual prices capable of being obtained by the plaintiff resulting in VRT being a much higher proportion of the actual sale price than the then current rate for the tax;
(iv) That the values of imported used vehicles imposed by Revenue had been subject to arbitrary increases;
(v) That the system of valuation discriminated in favour of the domestic car trade and against the trade in imported used cars;
(vi) In particular, that it resulted in the charging of VAT on imported cars in excess of the residual VRT incorporated in the value of the domestic used vehicles of the same characteristics.
63. Following from those claims, the plaintiffs essential case was pleaded as follows:
VRT and the manner in which the tax is operated and administered by [Revenue] is unlawful and contrary to the provisions of the Finance Act 1992, Section 133 as amended by the Finance (No.2) 1992 [sic.], Section 9. Alternatively, the said VRT and the manner in which same is operated and administered by [Revenue] is contrary to the provisions of European Community law and in particular Articles 3(1), 3(c), 3(g), 7(a), 9, 12, 30 and 95 of the Treaty of Rome and the Sixth VAT directive 388/77 and to the principles of legitimate expectation, proportionality, legality, equality and legal certainty.
64. The Act and manner in which it is operated were further said to be unconstitutional in that it fails to adequately vindicate the property rights of the Plaintiffs, infringes their right to equality and is an impermissible delegation of a law-making power.
65. A mandatory injunction seeking publication of the full range of new and used motor vehicles and to publish revised values from time to time as they occur was sought, together with declaratory relief reflecting the contentions of law as pleaded. The Statement of Claim concluded seeking:
A Mandatory Injunction directing the return of all excess Vehicle Registration Tax paid by the Plaintiffs and if necessary, an Order for the taking of an account and enquiry.
Damages for conversion, detinue, negligence, wrongful interference with constitutional rights and rights in European Law.
66. The essential pleading of UCIIs case neither changed nor was significantly elaborated upon thereafter. Four sets of particulars (RTP) were delivered (20 November 1995, 14 November 1997, 3 June 2011 and 21 July 2011). The following features of the claim as explained in those replies are relevant here:
(i) The claim was based on the failure to take the invoice price (or actual sale price) as the chargeable value for VRT purposes;
(ii) The dealers margin referred to in the Statement of Claim was defined as the dealers net profit and added value to the difference between the purchase price and the sale price including that profit;
(iii) The residual tax referred to in the Statement of Claim was explained as involving all of the component elements, including tax, of the residual price being deemed to abate proportionately so that they bear the same proportion in the depreciated value as they did in the original price. Since VRT is deemed to abate proportionately as a car depreciates, UCII pleaded, the residual VRT in a domestic car is calculated by reducing the amount of VRT actually paid by a percentage representing the proportion by which the car has depreciated.
(iv) UCII objected to the establishment of a system of valuation and taxation based on the retail price, meaning that the amount of tax to be levied is included in the final price rather than calculated on the price less tax which leads to uncertainty.
(v) In the RTP of 3 June 2011, the losses of UCII sought to be claimed in the case were presented in the sum of 131M.
67. The defences delivered on behalf of the various defendants were, in large part, traverses. They pleaded in some detail that the method used by Revenue to determine the OMSP did not involve individual motor vehicles being valued by a computer-based system as had been alleged and denied that there existed tables of value of second hand vehicles. They denied an obligation to publish values for second hand vehicles. While neither defence pleaded the appeal process as a barrier per se to the proceedings, they did identify the facility for appeal - together with that for estimates - as a feature of the administration of VRT and relied upon it in connection with the contention that it was neither secretive nor arbitrary. The defence further asserted that a number of factors had tended to deprive UCII of advantages it previously enjoyed, including the claim that the single depreciation table previously used for the purposes of calculating MVED resulted in values for the purposes of that duty that were lower than the market value.
68. Although the defences were delivered (and, in the absence of any Reply, the pleadings closed) in September 1995 it took a further seventeen years for the case to come to trial. This has the remarkable consequence that the vast bulk of the time period to which UCIIs complaints relate post-dates the closing of the pleadings in the action. While the plaintiffs were inclined to attribute this to difficulties in obtaining proper discovery, it is not obvious that this explains the time taken in bringing this action to hearing. When Costello P. refused the application for an interlocutory injunction on 17 July 1995, he fixed a time frame for the delivery of pleadings and directed the making of general discovery by each party within six weeks of the delivery of a defence. While the defendants do not appear to have complied with this (their discovery was eventually made in 1997) a further application for discovery of specific documentation was then determined by Geoghegan J in June 1998. There, he ordered the defendants to make discovery of the Edge Anderson report file as well as a representative sample of declarations of new car values received from main distributors/importers. Six years passed before the Master of the High Court (in December 2004) made a further order for discovery by the defendants of inter alia a working copy of the desktop version of a valuation programme used by Revenue to calculate the VRT. Then, in March 2006 Laffoy J. made an order following an application by UCII which (it claimed) was necessitated by the failure of the defendants to discover the VRT valuation rules or the depreciation tables and their association with each vehicle valued. That order required the production of the VRT calculation rules and data relevant to the valuation process. This led to an inspection in Rosslare in June 2007 the events at which proved the source of some acrimony at the trial, principally because of the refusal of the Revenue representatives present at it to respond to requests for information as to how the OMSP and depreciation data was determined and maintained.
69. During that period the level of inactivity was such that three notices of intention to proceed were filed - in October 1999, February 2004 and October 2010. Even if one accepts at face value UCIIs claim that although promising in May 1997 to provide it with a record dump of the database of statistical codes and their associated OMSPs and depreciation table categories, Revenue persisted up to and after the inspection of the VRT computer valuation system in 2007 in refusing to provide this information, it is impossible to understand why UCII did not resort to the ample mechanisms provided by the law for the production pre-trial of information required to present a claim. It is unsurprising in these circumstances that Murphy J. in his judgement described the progression of these as somewhat dilatory (at p. 167). His comments echoed those of Laffoy J. who, in 2006, commented that the enthusiasm for a speedy resolution of the matter suggested by the early stages of the proceedings had dissipated.
70. Notice of trial having been served on 9 May 2011, particulars of loss quantified at 131m. were furnished on 3 June 2011. This led to the further round of particulars and replies to which I have earlier referred. At around this time an application was brought before the High Court for directions, eventually resulting in the trial date being fixed. That application was brought by Revenue.
71. One other feature of the events while the proceedings were pending featured prominently in the cross-examination of UCIIs witnesses, and was viewed by the trial Judge as important. In September 1995 - shortly after the application for an interlocutory injunction was refused - Mr. ODowling ceased to be a director and shareholder in UCII. The parting between himself and Mr. Riordan was described by both in evidence as amicable and was attributed by Mr. ODowling to the pressure their business was put under by VRT. In July 1995, and shortly before his resignation Mr. ODowling established a new business, Sports Car Centre Ltd. That company was involved in the business of importing Japanese sports cars and four-wheel drive into Ireland, exporting some of them to the United Kingdom. As of the date of the trial, it was trading profitably.
72. Sports Car Centre Ltd. instituted its own proceedings in March 2000 also challenging the legality of the methodology applied to determine the VRT. As of the trial before Murphy J. in March 2012, those proceedings had not progressed since the service of a then outstanding notice for particulars by Revenue in October 2000. However, Mr. ODowling was the driving force behind the within proceedings (he was described at one point by Mr. Riordan in evidence as the person who it was agreed at the time of the parting as going to handle the case (Day 11 p. 74 Q285)). Mr. ODowling said in evidence (Day 9 p. 153) that he personally funded the litigation from the sale of a property in Cork. It was Mr. Riordans evidence that Sports Car Centre Ltd. funded at least some of the costs incurred in bringing the case to trial via a loan to UCII (Day 11 p. 122 Qs. 489 to 500). In the course of his evidence Mr. ODowling stated that he had a verbal agreement with Mr. Riordan which he framed as follows I would come back in as a shareholder if the UCII case was successful (Day 10 p. 33 Q.105).
73. At the same time as Mr. ODowling established Sports Car Centre Ltd, Mr. Riordan established Munster Car Imports Limited. From the year 2001 it imported the vehicles and provided them to UCII, which in turn sold them on. This arrangement was attributed by Mr. Riordan to the difficulties then facing UCII in obtaining credit.
The controversy around publication of the valuation database
74. In the middle of 1997, Revenue gave consideration to the publication of the valuation database it used for the calculation of VRT. There were three reasons for this. One arose from these proceedings, to which such publication was seen as a possible response. A second was the coming into effect of the Freedom of Information Act of that year. Thirdly, this was suggested as advancing the general policy of the Revenue which favoured transparency in tax collection. In that context criticisms that the system then in operation was secretive and arbitrary had been noted, as was the volume of successful appeals. A Working Group was established to address these issues.
75. UCII placed considerable reliance in both the High Court and in this Court on the content of correspondence generated within Revenue at the time of these discussions. On 4 June 1997, Mr. John OLeary, Assistant Principal, referred to the reservations of the CVO as regards publication of information used to value used vehicles. Noting the limitations of the Car Sales Guide (and in particular the fact that it did not contain models that were more than eight years old), Mr. OLeary explained how publication of the data relied upon by Revenue in determining the OMSP of imported used cars would likely result in the demise of the Car Sales Guide ([w]hy buy the guide when the Revenue data is more comprehensive and free?). This was not a wholly altruistic concern:
The problem for the CVO is its total dependence on the Guide as a source of information. This publication is the key reference work within the CVO and forms the basis for most of our decisions in the valuation area. In the likely event of its demise, it would prove difficult and costly in terms of resources to produce replacement.
The existence of the Guide allows us to adopt a minimalist approach to research and analysis of the 2nd hand market. Anything else would not be possible with current CVO resources. Obviously this situation could not continue if the Guide ceases to exist.
76. Noting the inevitability that the data base will contain errors and the impact on the appeals process of publication of the information in question, he observed:
Heretofore, formal challenges were directed through our appeals system and required a party to have been personally wronged. This will no longer be acceptable when our data is made more generally available.
77. Meeting notes from the working group disclose similar concerns, CVO describing its reliance on the car sales guide to some extent as regards checking the validity of the data base noting that it also formed the basis for the establishment of the existing depreciation tables. When Mr. OLeary made his submissions to the working group, he said this:
it is vital that our new OMSPs be suppressed if the work is to be kept within manageable proportions. Many of these are questionable and would not survive a public challenge. There are numerous cases where two wrongs make a right. Non-disclosure removes the possibility of such challenges. The only test which can then be applied is to determine whether our outcome is accurate.
(Emphasis added.)
78. A note from the CVO generated at this time expressed the view that the system then used for the calculation of mileage did not reflect the marketplace. It recorded that the use of the monthly adjustment of plus and minus 3% was not the correct basis and that it should be plus or minus 5%. More fundamentally, it expressed the view that the application to a vehicle - particularly high value vehicles - of a single depreciation scale, was resulting in incorrect valuations. It said:
However, there are instances where it is not possible to fully reconcile a vehicle to an appropriate table. The difficulty will often be confined to one particular year. The current system cannot deal with this situation and the vehicle will be incorrectly valued in respect of that year. This is an inevitable consequence of the current computer dependence on single tables but puts the Commissioner in the position of knowingly failing to correctly value the vehicle.
(Emphasis added.)
79. The CVO submission to the Working Group recorded that there were 3,900 statistical codes which had not been reviewed for at least twelve months. There were a further 2,000 codes which had not been reviewed for six months. It said:
Experience within the CVO has shown that any code will become inaccurate if not reviewed at least twice within a year. The impact of such inaccuracy is obviously relative to the value of the vehicle and may be such as to fall within an acceptable margin of error. However, this determination could only be made on the basis of a review.
Prudence would suggest that the entire data base should be reviewed twice in any calendar year.
80. At the time of the trial Mr. OLeary was still working for Revenue and based in Nenagh. He was not called by Revenue to give evidence at the hearing.
The Report of the Working Group
81. The Working Group delivered its report in September 1997. The report provides a snapshot of the valuation process and of Revenues practices in terms of its disclosure at that point in time. As to the latter, it explained as follows:
Since its introduction policy in relation to the valuation system has been not to reveal how it works to the motor trade or general public, to issue estimates of the tax payable on specific used cars on request and to otherwise state when the issue is raised that used cars are valued for tax purposes on the basis of their market value in the State. As a result, members of the public often resort to the appeals procedures as a means of ensuring that the amount of tax charged is correct.
(Emphasis added.)
82. While the Working Group did recommend the grant of public access to some information, it concluded that the new OMSP (as either declared by distributors or determined by Revenue) should not be displayed even though this is used to arrive at the appropriate used chargeable value, and that the depreciation percentage will not be divulged.
83. Its conclusions in respect of weaknesses within the system itself, which had then been operating for a period of four years, are notable. The difficulties identified by the Working Group fell into three categories. First, having described the depreciation tables as being drawn up on the basis of the Car Sales Guide it said that the schedules have their limitations. Included in these limitations were the following:
The Car Sales Guide is confined to those models of cars sold from new in the State : used Japanese cars and certain models marketed only in other countries are excluded from the guide.
The depreciation of each model is confined to a particular table i.e where circumstances require, it is not possible to customise a depreciation table to suit that model; it is only possible to move a model from one complete table to another of the 23 tables available
84. Second, the Group further noted a number of flaws in the valuation programme as drawn to its attention by the CVO. These prompted some suggested modifications to the programme. The Working Group recorded these proposed modifications in an Appendix to its report. They had not been implemented as of the date of the Report. They included changes to the mileage allowance and monthly depreciation and the introduction of a facility to use a customised depreciation schedule where this was appropriate. It seems reasonable to assume that these proposed modifications were necessary because without them, the system was not always providing accurate valuations.
85. Third, it is clear that the Group concluded based, presumably, upon the submissions of the CVO - that many of the codes had not been reviewed and updated. It said:
The group agree that regular validation of system valuations by reference to the used car market in the State is critical if we are to stand over the material released to the trade and inspire public confidence in the system of assessment of the tax.
86. The report referred to 10,427 statistical codes for used cars, of which 3,915 had not been reviewed in the previous twelve months. Over 2,000 of the codes had not been revised in the previous six months. It recommended that the 3,915 codes should be reviewed as a matter of urgency and that the codes should be routinely validated by reference to either The Car Sales Guide or enquiries with the motor trade as appropriate twice per year.
87. The Working Group concluded that the existing valuation inquiry system was neither suitable for providing public access to the VRT valuation database nor amenable to the modifications necessary to render it so. It recommended the development of a new programme that would be suitable for release to the motor trade and for an interactive website on the internet. It felt that the most appropriate option was to make this available via a kiosk incorporating a touch sensitive screen. Mention was also made in the course of its report of the possible production and circulation on a regular basis, of a CD ROM.
Improvements to Revenues system
88. I have referred earlier to the development of the VRT system up to February 1994. Thereafter, further changes occurred. Many of these implemented the recommendations of the Working Group. In 1998, public touchscreen kiosks were introduced at three locations. These enabled the calculation of VRT due on specific vehicles by access to a snapshot of the dynamic database and facilitated the user in calculating the VRT payable on a particular vehicle by selecting the make, model, version and particulars of the vehicle. It was updated every five days. Revenue staff had available to them a desktop equivalent of the kiosk.
89. At approximately the same time Revenue introduced a CD VRT estimation service. This was available until 2001. The CD thus provided (and regularly updated) contained a version of the software similar to that of the public touchscreen kiosks but suitable for use as a standalone PC desktop. It enabled the calculation of the VRT thereby effectively providing those in the trade with an enquiry system similar to that used by Revenue officials. It thus facilitated access to the same information as could be obtained through the touchscreen kiosks. UCII subscribed to that service and received fifteen iterations of the CD as they were issued between May 1999 and July 2001.
90. In November 1999, the monthly adjustment range was restored from + and - 3%, to + and - 5%. The reason for this change was explained in the Working Group Report, as follows:
The problem here is that the existing tables depreciate a vehicle by reference to 30 June in any given year. The monthly adjustment was introduced to compute the proper value for any other month. Research carried out at the introduction of VRT indicated that plus or minus 5% was the correct variation in the value of a motor vehicle within any given year. When the programmes went live it was discovered that a fault in the mechanism of calculating mileage allowances was creating unacceptable anomalies in vehicle valuations. The impact of the fault could be minimised if the monthly depreciation rates were adjusted to plus or minus 3%. This was an emergency situation which was introduced on a temporary basis. There has never been any indication that the revised rates were correct. The rate should be returned to their original levels when the mileage issue has been resolved.
(Emphasis added.)
91. The assumed standard yearly average mileages were replaced with assumed monthly mileage allowances of 1,300 miles (diesel) and 900 miles (other). This also mirrored the suggestions made by the CVO to the Working Group. Provision was made, again in line with the recommendations of the Working Group, for the assignment of customised depreciation tables hybrids of the then existing 22 tables - to allow for more accurate reflection of apparent actual depreciation. The hybrids, as explained in the evidence of Mr. Campion of Revenue, allowed the valuation officer to assign a combination of tables. The example given was that the valuation officer might find that a table would show the depreciation accurately for years one to three, but that the table in year four might show a divergence from the market data. Hybridisation allowed him to select a table where the data for year four accurately reflected the market data and he could choose to use that table for years four onwards. In that way he would have a very large number of tables at his disposal to assign. Where this was done, the combination of tables applied to a particular class of vehicle was not published, but instead was built in to the statistical code for that vehicle, which would record (for example) that for years one to three depreciation Table A1 might apply, and for years four to seven Table B2 might apply.
92. I should digress here and emphasise that the evidence of Mr. Campion discloses that the facility for hybridisation (introduced with effect from November 15 1999) was viewed by Revenue as a critical feature of the valuation process. It meant that the system of valuation employed by it was not at all as it seemed. The depreciation tables which UCII believed to be both fundamentally flawed and to represent a central and rigid feature of Revenues methodology were, for most of the time covered by this action, not actually applied in the manner the information given to UCII suggested. It is not obvious to me that UCII can be faulted for operating under a significant misconception in this respect. It was Mr. Campions evidence that while Revenue would know from its computer record of a statistical code that it was a hybrid, a member of the public would not. If there was any document or other mechanism which advised taxpayers that Revenue operated this system of hybrid tables which I should stress was described by Mr. Campion as a very important, a very important development - it was not identified in the course of the thirty three days this case was at hearing. As of the date of the trial there were 1,043 customised codes of which 219 related to Japanese cars.
93. As of November 1999 the value of extras was depreciated by the system separately before being added to the OMSP; extras had been depreciated manually over a period of four years with effect from 10 May 1996. From this point, extras were depreciated over a seven year period (this being changed to five years in 2009). Then, with effect from 5 July 2002, the number of depreciation schedules was increased to twenty-four. By the time the case was heard, there were twenty-five schedules.
94. In 2004, an online VRT calculator was made available by Revenue. That facility, a new version of which was released in 2010, allowed the calculation of VRT on any vehicle that existed on the Revenue database. This was refreshed on a nightly basis. At the same time a system was introduced to enable the uploading by distributors of statistical codes to Revenue via its on-line system. At around this time - in September 2004 UCIIs solicitors were furnished with thirteen worked examples of the method used to establish the OMSP of identified vehicles including manual spreadsheets used in allocating the code.
95. In June 2010, the depreciation schedules were first published following a request made pursuant to the Freedom of Information Act 1997, as amended. In the meantime, UCII said, it did not know what the depreciation schedules provided and it did not know how one landed on one depreciation schedule rather than another. By the time of the trial it appears that the online calculator enabled access to the depreciation schedules and in the course of his evidence Mr. Kenny, a tax expert called by UCII, explained how he could access these via an icon on the Revenue Online System. Mr. Kennys evidence was that once that was made available you do get a very good sense of the methodology adopted by the Revenue for determining the VRT value or taxable value of your vehicle today. (Day 21 p. 20 Q. 35). However, his evidence in respect of the information available prior to this was highly critical. He said that there was no tax of which he was aware where the rules relating to the tax in question - including the methodologies for determining taxable value - were not published in any statutory form or supported by Revenue notices, guides or explanatory booklet. The consequence was that the taxpayer could not compute the tax due on any particular transaction. That evidence was never controverted. Indeed, while UCIIs expert witnesses each complained of the opacity of the system operated by Revenue (see for example the evidence of Mr. Yarrow Day 17 p.13 Q.47) it is of note that Revenue fielded no independent expert witness who was prepared to testify the contrary.
96. By 2010, according to Mr. ODowlings evidence, the system operated by Revenue enabled users to ascertain the OMSP of the vehicle as new, the depreciation code, the amount of depreciation, the monthly adjustment and the VRT figure. Although Mr. ODowling appears to have obtained a copy of a depreciation scale from an unidentified Revenue official in 1994, none of that information was officially available before then (Day 5 p.134 Q.575 et seq.).
IV REVENUES METHODOLOGY FOR THE DETERMINATION OF THE OMSP OF USED VEHICLES
UCIIs claim and Revenues methodology
97. The methodology adopted by Revenue for the determination of the OMSP was central to UCIIs claim, and a complete understanding of that methodology is critical to its resolution. The claims insofar based on rationality, lack of certainty, and the alleged breaches of EU law - which were the parts of the case most heavily dependent on the evidence - revolved around it. The evidence in relation to that methodology is also important to the complaint advanced by UCII that Revenues own explanation of its system as tendered to the Court was internally inconsistent.
98. That explanation falls to be reviewed bearing in mind two important aspects of these proceedings. First, it is a striking and significant feature of UCIIs case that it did not challenge the legality of any particular valuation applied by Revenue to any specific vehicle. There was thus no identified determination the legality of which was impugned. Instead, the challenge was pitched at a level of generality and abstraction, directed as it was to the manner in which the tax [was] operated and administered. It was a public law claim based not on any decision, but instead directed to a general, ongoing and (as the proceedings became more protracted) expanding state of affairs. It followed that the analysis the Court was required to undertake of that methodology was not targeted to any particular application of it (although examples of its operation were given in the course of the evidence).
99. The analysis of that methodology (and, perhaps more importantly, its effect) brings into focus a second important feature of the claim. The complaints advanced in the action extended over an astonishing period of time. Yet across the point from which VRT was introduced in 1993, to the date of the trial in 2012, Revenues system was far from static. The schedules prepared by Revenue incorporated an increasing number of statistical codes, and therefore of vehicle types. The number of depreciation schedules used increased, an exponential expansion of that resource occurring with the introduction of the facility for hybrid schedules. The range of information to which Revenue could turn in seeking to fix the values of the vehicles also expanded with the resources afforded by the internet. Review by Revenue of the applicable schedules based upon market changes and the results of appeals in individual cases meant that vehicles were, at different points, moved as between schedules. Changes were made to the mileage and monthly adjustments. The evidence of Revenue officials suggests that the frequency with which the information in the schedules was reviewed increased. The gradual making available to the public of more information about the system of valuation combined with the eventual ability of technology to deliver that information instantly, enhanced the overall transparency of the system. In respect of some of the complaints made by UCII, a situation in which aspects of those complaints have greater force at some times than at others renders adjudication on the validity of parts of its claim most challenging. In circumstances where UCII bore the burden of proof, these difficulties were capable of being surmounted to the extent that Revenues obligation was to meet the case presented against it, and no more, and the function of the Court was to adjudicate only on the claims so presented in the light of that response. Insofar as the onus was, for some parts of the case, on Revenue, however, the burden arising from the breadth of the claim was substantial to the point of being potentially unmanageable.
The documents
100. The Butler Report, the Edge Anderson Report, the staff guide, the Traders Guide, the report of the Working Group and the document prepared by Mr. Coyle all operate on the basis that the depreciation schedules are the key reference point in the valuation process. They suggest that trade information - that is the data in the Car Sales Guide and that available from other market sources - was relevant at three points. The first was when the tables themselves were initially devised in 1992, these being based on Revenues own analysis of the various depreciation trends in the market. The second was when a statistical code was first produced for a model and make of a vehicle of a particular vintage. In determining which depreciation schedule was to be applied to that particular vehicle, these sources were consulted. The third was when the codes were reviewed, when the same information (together with the product of appeals) was taken into account.
101. However, when a vehicle to which a statistical code had been allocated was valued on the ground, the automated process yielded the value for the vehicle, and that automated process functioned on the basis of the depreciation tables. There is no reference in any of these documents to the value produced by that automated process being subsequently cross referenced to any trade information nor was it suggested by any witness of fact that the OMSP generated upon input of the statistical code was reviewed by reference to the prices contained in the Car Sales Guide.
102. All of this is reflected in a document produced by Revenue shortly before the trial in December 2011. Reading this in its own terms, once the appropriate depreciation schedule had been designated for a particular vehicle, that formed the basis for subsequent valuations:
The OMSP and depreciation table relating to this vehicle will then be added to the Revenue database of used vehicles, so that the VRT charge for all future vehicles of this particular make, model, version and variant can be calculated at registration.
Mr. Butlers evidence
103. The position as recorded in these documents was mirrored in the oral evidence. Mr. Butlers evidence was that prior to the introduction of VRT, the staff at the CVO consulted the Car Sales Guide, newspapers, other magazines in the car trade and contacted individual dealers and thereby gathered together information as to the price obtained by various makes and models of used cars. From this information, spreadsheets were created recording the prices obtained for those vehicles across a number of vintages. This enabled them to use the spreadsheets (none of which by the time of trial were available to the Court) to build the depreciation tables by establishing patterns of depreciation of models available on the Irish market which they then sought to replicate on a computer system. He explained this as follows (Day 22 p. 84 Q. 254):
we translated the actual values into depreciation tables. And we analysed those depreciation tables to see whether we could build in a logarithm into a computer system so that when a particular vehicle was assigned a particular depreciation table that that depreciation table reflected the values of that particular model right along through its different ages . that was factor one in determining the open market selling price of a value
The other variables are dealt with subsequently in my report being mileage, condition, extras.
104. He explained the methodology leading to the development of the tables, as follows (Day 23 p. 113 Q. 346):
What we did was develop a series of depreciation tables which reflected the actual market values of models at the various stages of their age. And in applying a table what we did was we established which table in its entirety reflected the market values of that particular model at the various stages of its depreciation and then applied the table to that. It wasnt that we took a table and applied it to a new value in order to establish the market, it is actually that we looked at the value of a particular vehicle in its entirety right down through the different ages of the vehicle and matched that with the table that actually reflected those values.
105. Thus, when a vehicle was presented which had a statistical code, that was what the officers used to value it (see Day 23 p. 8 Q. 9; Day 24 p. 46 Q.114). When a vehicle of that make, model and age had not been previously valued and thus had no statistical code, the CVO would (Day 23 p. 7 Q.6):
research that car on the market immediately, identify which depreciation table matched the actual depreciation pattern of the actual market values and we assigned that depreciation table and we assigned it a statistical code in accordance with the structure we had in place.
106. In this way, as he put it, the chosen depreciation schedule would reflect the market depreciation pattern of that particular car and the values of that particular model in the market (Day 22 p. 51 Q. 130 : Day 23 p. 65 Q. 182). However, while the market information contained in the Car Sales Guide and other trade sources was stated to be critical to the process of identifying the depreciation schedule applicable to a vehicle and to the review of that designation when made, at no point in his evidence did Mr. Butler ever suggest that where a statistical code did exist for a particular make and model of a specific make that those inputting the statistical codes (the vehicle registration officers) checked the valuation produced by Revenues computer system against those sources.
107. The actual role of the trade information and its relationship with the depreciation schedules was described neatly by Mr. Butler at one point in the course of his cross examination, as follows (Day 23 p. 88 Q. 261) :
You see, the tendency is to think that we assigned a table to produce the value. It is actually the value that produced the table in the first instance. That is what the CVO did. And then the system reapplied the table to reproduce the value.
108. That begged the question why the depreciation schedules were required at all. If, as Revenue suggested, the application of the schedules did no more than mirror the information gathered from an analysis of the price of various vintages of makes and models, it should have been possible to simply look to the source information contained on the schedules or spreadsheets compiled as the information relating to those vehicles was collected. That question was put by Mr. Sreenan SC to Mr. Butler (Day 24 p. 93 Q. 294), and responded to as follows (Day 24 p. 99 Q. 307:
Because this system is much more efficient. Within the boundaries of the depreciation factors reflects that full extent of the Irish market. It allows 32 Vehicle Registration Offices to act in unison in relation to the same car. It is efficient, it is fast and, by the way, it allows central control and monitoring over the cars and the tables that they are assigned.
It wouldnt be effective or efficient to devolve the actual depreciation tables to individual VROs considering the way that the market changes all the time. It needed to be controlled centrally in order to be efficient, accurate and effective. So to devolve it out to where all these vehicles are registered would not be the most efficient or effective way of dealing with that.
109. Mr. Butler was pressed on this, it being suggested that his answer did not explain why the depreciation tables were needed at all, since any of the officers could, via the mainframe, simply get the actual value of any specific make, model and vintage of vehicle without going through the process of determining starting OMSPs and making per centage reductions to them (Day 24 p. 100 Q. 310). Mr. Butlers response re-iterated that the system as operated involving the depreciation tables was more efficient: the assignment of a depreciation schedule enabled the information gathered on first presentation of a particular vehicle to be preserved and used in all subsequent valuations of that vehicle until such time as the review team decided that the values had changed (Day 24 p. 100 Q. 310). It was put to him that the reason the actual hard data collected in the spreadsheets was ignored in favour of imputed values and depreciation scales was that it was just too much trouble to keep on collecting the data and was much easier just to have a few arbitrary scales that vehicles were arbitrarily assigned to (Day 24 p. 110 Q. 336). That was rejected by the witness: the tables did reflect the depreciation in the market and to actually have a separate depreciation for every single model on that market was completely unnecessary (id.).
The evidence of Mr. Campion and Mr. Fogarty
110. The explanation of the valuation methodology advanced by Mr. Butler was updated and clarified by the evidence of Mr. Campion. He was the Assistant Principal in the Central Vehicle Office, having been appointed to that position in 1995. Noting that he was absent for some of the time between then and the hearing of the case, he had experience of the functioning and development of the system over most, but not all, of the period to which the action related. For the purposes of addressing the legal issues presented by the case, his explanation of the system was broken down into a number of different phases.
111. The key starting point, he also explained, is the allocation of a statistical code to a make, model and variant of a motor vehicle. To do this an official (the valuation officer) makes two decisions. First, he determines the undepreciated OMSP of the vehicle. This is its price as new of the date of the exercise. Second, he then identifies in respect of that vehicle a depreciation table. Originally, this would have involved picking whichever one of the five, fourteen, twenty two or twenty four existing schedules (as the case may be) best reflected the rate of depreciation of that vehicle as demonstrated by the prices paid for different vintages and variants of that vehicle. Those prices were determined by reference to the prices paid for that vehicle or its nearest equivalent in the Car Sales Guide, newspaper advertisements, information made available by the trade, or the experience of appeals. Latterly, the valuation officer would have available to him the additional facility of choosing a hybrid depreciation table, and would have the further resource of the internet to guide him as to the appropriate price.
112. The next important feature of the process was the review of the depreciation tables. This was also done by the valuation officer Mr. Campion gave evidence of the number of statistical codes reviewed in the period between 2003 and 2010, these varying from a high of 27,876 in 2004, to a low of 14,045 in 2007, the number of statistical codes being amended consequent upon those reviews being presented at its highest in 2009 (20,312) and its lowest in 2007 (6464). To put these numbers in context, as of the trial (the evidence was) there were in the region of 40,000 statistical codes, of which between 28,000 and 29,000 were live or regularly used. Those amendments, Mr. Campion said in evidence, would typically have related to the depreciation table, with the table assigned to a statistical code being changed if the market data indicates a requirement to do so. Figures for reviews prior to 2003, it is to be noted, were not available.
113. In understanding - and resolving - some of the confusion that surrounded the role of the statistical code, the depreciation tables and the core source material used by Revenue to generate and validate these, it is thus important to distinguish on the one hand between the process of inputting the new OMSP and of deciding which depreciation tables or, latterly, combination of those tables should be applied to produce the value of a particular used vehicle, and on the other the process of determining the OMSP of a particular vehicle when presented for registration. The former was undertaken by the valuation officer in Rosslare and involved matching the particular vehicle to the price disclosed by the Car Sales Guide and other sources consulted. The latter was undertaken by the registration officer at one of the 32 registration centres located around the country which, since 2010, have been operated by the National Car Testing agency. That exercise did not involve consulting the Car Sales Guide, but merely inputting the statistical code. Thus, two features of the process of valuation predominate : the intention was that the prices would reflect those disclosed by the market, but that objective was not achieved by affording primacy to those sources at the point of valuation but instead by applying the depreciation tables - unless the make and model of vehicle was being presented for registration for the first time, in which case the valuation officer would use those sources to identify the appropriate schedule. Therefore, it was both true to say that valuation was based on the market prices and that the depreciation tables were used to identify that valuation.
114. This was made particularly clear during the cross examination of Mr. Fogarty of Revenue. He explained that all the registration officer could do is input the statistical code, condition, mileage and extras associated with a vehicle and apply the value produced in consequence. That officer had no capacity to change the valuation thus determined. If the owner of a vehicle were to arrive armed with a copy of the Car Sales Guide and to identify a value for the vehicle that differed from the value thus produced, his options - and his only options - were to pay the tax as thus calculated and appeal or to ask Rosslare to review the valuation in advance of registration (Day 28 p.136 to 137).
115. In the course of his evidence Mr. Campion (Day 25 p. 123 Q. 420) described the depreciation tables as simply a way that the system generates or regurgitates the valuation officers research. He outlined their role as follows (Day 25 p. 120 Q. 410):
The depreciation table and the undepreciated OMSP, they are tools in an algorithm on a computer system to deliver the result which the valuation officer has already determined by reference to the Car Sales Guide and other sources of information. This is why I said this morning that we begin with the answer. They are simply instruments in a computer programme. The actual OMSP of a vehicle in good condition with normal mileage has been determined by the valuation officer and he uses the depreciate [sic] table and the undepreciated OMSP to cause the system to spit out that price at the Vehicle Registration Office or on the VRT Calculator.
116. The end point of this evidence was crisply explained by Ms. OBrien SC in the course of her submissions to this Court. The depreciation schedules, she said:
were not used to determine the value of the vehicle, they were used to determine what the value of an equivalent or the same vehicle was as prescribed by the Car Sales Guide
V THE CASE
117. At its most general, UCIIs case as presented in the High Court reduced itself to four propositions. The first was that the system used by Revenue for determining the chargeable value of used vehicles had no statutory basis. That claim was based on the contention that the legislation required Revenue to take account of the price actually obtained for a vehicle and, if it did not, that the methodology used by Revenue to determine the OMSP was not bona fide, factually sustainable or reasonable because (in particular) it did not result in a value reflecting the price that the vehicle was actually likely to fetch. Reliance was also placed in this context on principles of legitimate expectation, legal certainty and proportionality.
118. Second, it was said that the legislation was unconstitutional, partly because the system operated by Revenue represented an unconstitutional delegation of law-making power and partly because the manner in which it operated resulted in an unjust and disproportionate interference with UCIIs property rights and/or right to equality. If that system was enabled by FA92, it was said, then it followed that that provision itself was contrary to the Constitution.
119. Third, it was contended that the application of Revenues methodology resulted in unlawful discrimination against UCII and in favour of those who sourced used vehicles in the State. That claim depended on the proposition that importers faced the imposition of VRT on their margin, while domestic dealers were placed at a competitive advantage by not having to pay the tax on theirs.
120. Fourth, it was said that that system was in breach of EU law, partly because of the discrimination to which I have just referred, but also because (it was said) the manner in which VRT was determined for imported used cars failed to ensure an equivalence with the value of an equivalent domestic vehicle and resulted in values that were higher than the prices yielded by the market. It was also claimed in this connection that the fact that VRT was calculated on a VAT inclusive OMSP rendered it contrary to the provisions of the Sixth VAT Directive.
121. These claims, while distinct, obviously presented overlapping themes. They were supported by expert evidence, and indeed an analysis of the defendants evidence which proposed that Revenues witnesses had presented inconsistent and/or incoherent accounts of how the OMSP of a used vehicle was actually determined, and in particular of the role of the depreciation scales in that process. UCIIs expert evidence from four economists, two IT experts, an accountant, a tax advisor and a valuation expert identified a series of flaws in the methodology as they understood it, evidenced in turn by the depreciation tables.
122. The principal contentions advanced by UCII via that expert evidence were helpfully summarised in a short note furnished by Mr. Dowling of DKM Consultants at the conclusion of his evidence. Insofar as relevant to this appeal, they were as follows:
(i) depreciation schedules of the type used by Revenue are useful only as forecasts of expected depreciation. If, it was said, the valuer has data on the market prices and those curves have been constructed by market prices, then he should use that data as the basis for valuing used cars;
(ii) no information had been provided to UCII to show how the depreciation schedules were constructed, and no information had been provided as to why the schedules became 14 in 1992 or 23 a year later or why they continued to be used 18 or 19 years after their introduction;
(iii) the data base used by Revenue consisted entirely of statistical codes for individual makes and models of cars and the depreciation schedule yet no information had been provided as to why a particular model was allocated to a particular depreciation schedule or as to the process by which a model was moved to a different depreciation schedule;
(iv) for most of the period from 1993 to the trial, it was impossible for an importer to know what depreciation schedule his car had been allocated to and accordingly what the valuation basis for VRT was;
(v) the depreciation schedules used by Revenue were in the main unrealistic and outside the international norm. They showed a far higher rate of depreciation especially in the nought to five year range than was plausible leading to a significant over-valuation of used cars. They were mainly linear when it was universally accepted that used cars depreciate in non-linear fashion with substantial depreciation taking place in the early years;
(vi) by using retail prices achieved by main car dealers as market prices, Revenue overvalued used car prices;
(vii) it was not consistent for Revenue to base VAT on the retail price actually achieved for a car (as it did) and at the same time to claim that it is reasonable to hold that the car would retail at a higher price;
(viii) the scheme discriminated against used car importers by effectively applying VRT on the margin of used car dealers. It was Mr. Dowlings contention that the only price that could guarantee that the embedded VRT in a domestic used car is not less than the embedded VRT in an imported used car is to base VRT on the landed price. VRT on a price that is higher than that discriminates against dealers in used car imports compared to dealers in domestically sourced used cars;
(ix) it would have been possible for Revenue to construct its own, real, data base to inform its opinion with respect to the price of used car transactions;
123. Underlying all of these complaints was the claim that the system was arbitrary and uncertain. Mr. McDowell explained this as follows (Day 17 p. 104 Q.394).
It appears to rely on an extraordinary number of depreciation schedules the purpose of which escapes me, given that where there is any doubt under appeal, appeal is made to the Car Sales Guide which seems to me as being sort of the long stock, that the depreciation schedule works as long as nobody upsets it by waving a Car Sales Guide under their noses saying its not getting that price and then they change it again and they reallocate cars between schedules in a way which I dont understand.
124. UCII contended that as a result of these various features of Revenues system, it had suffered loss (including consequential losses) of 131M. The calculation of that sum (which occupied a significant part of the evidence) was based exclusively on its Japanese imports. While both Mr. Riordan and Mr. ODowling gave evidence that it was intended that UCII would import cars from other EU states, and while evidence was given of imports of forty-four vehicles from Northern Ireland, all of the focus in the calculation of loss was upon the tax paid on imports from outside the EU.
125. The case was described by counsel in the course of oral submissions in this Court as daunting. This may have been an understatement. It spanned a time period of seventeen years and was directed at a system which was processing in the region of 40,000 vehicles per annum. The trial Judge had to negotiate a significant volume of documentary and oral evidence tendered over the course of the thirty-three days the matter was at hearing before him. Underlying this was a wide spectrum of issues of domestic and European law, a volume of questions of fact spanning that period of almost two decades, and complex and at important points conflicting expert evidence as to applicable principles of economics, econometrics and forensic accounting. These he sought to resolve in a lengthy and detailed judgment.
126. Not all of the arguments advanced before the High Court were pursued on appeal. I will address the issues that were addressed in UCIIs oral and (insofar as not abandoned at the hearing) written submissions in this Court in the order in which they seem to me to most logically present themselves. I will outline the relevant findings of the trial Judge as they thus arise.
VI THE STATUTORY APPEAL
The Appeal and its operation
127. As originally enacted, s.138(1) FA92 provided for an appeal in the following terms:
Any person who has paid or is liable to pay vehicle registration tax may appeal to the Commissioners against the amount of the tax charged
128. The section required that the appeal be made in writing (ss.(2)), be lodged within 21 days from the date on which the tax became due (ss.(3)) and be decided within 21 days of its being lodged (ss.(4)). The Commissioners were required to give their decision in writing (ss.(5)). The provision assumed that the tax had to be paid even if an appeal had been brought, with the section specifying that where successful the excess would be refunded, and where an appeal resulted in the VRT being increased the taxpayer would be required to pay the excess (ss.(6) and (7)).
129. These provisions were replaced in the Finance Act 1995 (which repealed s.138 FA92) with an appeal procedure which applied generally to excise duties. Section 104 of that Act provided for a right of appeal framed in similar terms to that provided for in s.138 FA92, extending the periods for appeal and for determination to 30 days. Section 105 introduced an onward appeal from a decision of the Commissioners under s.104 to the Appeal Commissioners, with the provisions of the Income Tax Act 1967 governing appeals in respect of income tax applied, with some modification, to the appeal under s.105. These, in turn, provided for both a further appeal to the Circuit Court and an appeal to the High Court on a point of law. Section 106 expressly mandated payment of the duty notwithstanding the bringing of appeal under either ss.104 or 105. The Finance Act 2001 broadly reflected the same structure, providing for appeals within two months of payment (s.145(5)(a)), and applying similar provisions of the Taxes Consolidation Act 1997 to the proceedings before the Appeal Commissioners to those stipulated in the 1995 Act (s.146(3)).
130. Mr. Butler gave evidence as to the operation of the appeal process in the time he was at the CVO. He explained that the taxpayer was required to produce independent evidence in relation to the value of the car. This could comprise, he said, anything that was reasonable and was not obviously faked including engineers reports, reports from an independent dealer or data from a publication that established the value. It would appear that account was also taken of the invoice price of the vehicle.
131. Then, the car would be inspected by an officer at one of the Vehicle Registration Offices, who would forward the appeal with the supporting documents to the CVO. From there, contact was made with three independent dealers on the basis of whose valuation the CVO would adjudicate on the appeal. Essentially, as it was explained by Mr. Butler, if the resulting value was within 5% of the value on which the VRT was determined the appeal would be disallowed. If more than that it would be allowed, and the VRT refunded. The valuation of the vehicle in question would then be reviewed across all its years and adjusted if considered necessary. There was no charge for the appeals. According to Mr. Butlers evidence, approximately 1% of valuations were appealed, of which approximately 80% of the appeals were successful.
132. The evidence of Mr. Campion explained how the appeal process operated in practice during the time he was at the CVO. Thus, he said that a review officer operated from the CVO office in Rosslare. Once an appeal was lodged, that review officer would note the Car Sales Guide, check sources such as carzone.ie to find advertisements for the make and model in question, and go to the market and obtain five valuations from dealers. He would compare those five valuations and if they were consistent, he would take the median valuation of the five. If any of them were inconsistent, he would ignore them. He then made a report to an appeals officer appointed by Revenue under the Act who determined the matter.
The High Court judgment
133. Throughout his judgment, Murphy J. was critical of the failure of UCII to appeal OMSPs which it claimed in the action were too high. The appeal process is referenced by him in a number of different ways and contexts. On many occasions, the Court recorded that it was taking the appeal process into account, was having regard to it (see pp.84, 210) or made it clear that the Court viewed the facility for an appeal as generally relevant to the proceedings (at p.182). It is clear, in particular, that Murphy J. viewed the appeal as providing the mechanism for dealing with anomalies which the evidence had (he also found) identified in the system (at p.214), and that he believed the availability of an appeal was in part a reason the tax did not amount to an unjust attack on UCIIs property rights (at p.218).
134. At one point, Murphy J. related the fact of the appeal to the entitlement to seek relief from the Court at all (at p.196):
There was little evidence on the plaintiffs part regarding appeals. The plaintiffs [sic] submitted that the successful appeals demonstrated that the OMSPs were unfair. The initial evidence was that the plaintiff had not been successful and that the plaintiffs [sic.] did not pursue many appeals as it would have to pay the VRT as a condition of appeal. Given the volume of imports and the resolution of appeals within a reasonable period, the court finds that the plaintiff did not avail of its right to challenge the opinion of the Commissioners. They did not seek a remedy by way of appeal at the time. They did not exhaust that remedy.
135. In a related vein and noting the comments of Keane CJ in Glencar Explorations Ltd. v. Mayo County Council [2002] 1 IR 384 that the remedy of a person affected by the commission of an ultra vires act was an order of certiorari or equivalent relief, Murphy J said (at p.204):
In the present case the remedy available to the plaintiff was to appeal the determination.
136. Similarly, but perhaps distinctly, while noting that the court had to be satisfied that the decision reached was factually sustainable, the Court observed [e]very decision can be appealed where there is error and can be judicially reviewed if the correct procedures are not followed (at p. 201). Elsewhere, he indicated that the availability of the appeal in providing a channel for challenge to Revenues opinion, was an answer to the claim of irrationality or ultra vires.
137. At other points, Murphy J. made clear his view that the availability of an appeal obviated errors in the process and that it was an inherent part of the valuation system. He said, for example (at p.85):
The court is not satisfied that the Irish system did systematically overvalue the OMSP of second hand vehicles. Whether, on occasions, it was not very close to their actual value is a matter which could have been attested [sic.] by way of appeal which, in turn, would have formed part of the basis of the valuation system.
138. Thus, he stressed, the requirement for consistency in and a basis for a valuation could be tested on appeal (at p.195). These were, the Court held, free, expeditious and the majority of them were successful (id.). Indeed, while saying that the Court had no doubt but that there were teething problems with the underlying system Murphy J. again underscored the availability of both estimates and an appeal (at p. 179). He referred to the availability of an appeal and facility for estimates in the context of the claim by Mr. ODowling that he could not calculate VRT on five year old models that this was somewhat of a generalisation, given that he had the facility for getting an estimate and, of course, for appealing (at p.177). He also expressed the view that the absence or paucity of appeals was an evidential defect insofar as complaints based upon a lack of clarity of the depreciation schedules or an argument as to legitimate expectation were concerned (at p.210).
139. The existence of an appeal facility and failure of UCII to exhaust it was also relevant at an evidential level: this, the Court held, deprived UCII of what the High Court believed to be necessary evidence of price deviation (at p.219). The failure of UCII to appeal was further viewed by Murphy J. as an answer to the claims for damages for conversion, detinue and negligence (at p.238). The legal submissions advanced by counsel for Revenue in his closing argument in the High Court, it should be observed, was to the effect that the failure to appeal was a failure to mitigate loss.
140. Murphy J. specifically noted this in relation to forty-four vehicles which UCII said it had imported from Northern Ireland in the early years of VRT. In that connection he rejected the contention that the appeal process took too long noting that the evidence was that in relation to an appeal lodged in February 1993 this was dealt with in favour of the plaintiff within a month (at p.79). Throughout his judgment, Murphy J. referred to the absence of evidence that appeals were unworkable or of the time an appeal took, or of the number of appeals or success rate of such appeals. He also noted the failure of UCII to give any satisfactory evidence as to why the practice of lodging appeals by UCII was abandoned by it (at pp., 168, 175, 235).
141. UCII contends that the trial Judge placed inordinate significance on the fact of an appeal. While both Mr. ODowling and Mr. Riordan were cross-examined as why they did not avail of the facility for appeals (their essential response - that they were time consuming and inconvenient - was not accepted by the trial Judge) Revenue, as I have noted, did not contend that the existence of an appeal precluded UCII from advancing any part of its case. Insofar as it invoked the fact of the appeal in the course of its submissions, it did so somewhat diffidently. In its High Court written submissions, Revenue noted that it was open to UCII to appeal the valuations if only to mitigate the size of its alleged losses. In his closing oral submissions at the conclusion of the evidence before the High Court, counsel for Revenue but briefly referenced the availability of an appeal, observing that if UCII had a problem with the way Revenue formed their opinion, they could appeal the decision in the normal way, and that that appeal was to the benefit of the taxpayer. In countering the claim that the method adopted by Revenue was factually unsustainable, counsel referred to an appeal suggesting (although not actually contending) that the failure to exhaust that remedy affected the entitlement of UCII to agitate at least part of its complaint, elsewhere describing the appeal as an alternative remedy. At another point it was said that the failure to appeal was relevant to UCIIs obligation to mitigate its loss and thus fatal to any claim for either recovery of the tax, or for damages.
142. In its written submissions in this Court, Revenue similarly observed that if there were individual valuations that were found to be excessive, there was such a right of appeal which the High Court had found involved neither excessive cost nor unreasonable delay, and it invoked the appeal in observing that valuation is an inexact science, and that the evidence disclosed anomalies in the system operated by Revenue. At no point before this Court or the High Court was any authority cited as to the relevance of an appeal to the entitlement of UCII to maintain this claim.
The relevance of the appeal
143. While the trial Judge thus made frequent reference to the availability of an appeal, his judgment suggests that he felt it relevant in four different ways. First, in precluding UCII from seeking declaratory relief in respect of (at least) some parts of its case. Second, in adjudicating upon the reasonableness of the mechanisms put in place by Revenue to form its opinion as to OMSP. Third, in insulating the system and provisions in question from challenge on the ground that they were anomalous and fourth in affecting the entitlement of UCII to claim damages on the basis that by failing to appeal it did not mitigate its loss.
144. For reasons to which I will later return, it is not necessary here to address the fourth of these propositions. Insofar as the trial Judge accepted the first, in my view he erred. Insofar as his judgment might be interpreted as accepting that the fact of an appeal conducted according to one method in some sense rectified the illegality of an original decision reached by reference to another, he also erred. However, the reason this is so is important is that it emphasises, and in part explains, inherent limitations in aspects of UCIIs case while at the same time confirming the relevance of the third aspect of the trial Judges consideration of the appeal, as I have thus categorised it.
145. The availability of an appeal against an administrative decision will frequently operate to preclude a party from seeking judicial review of that determination. The fact of the appeal engages the courts discretion, and that discretion is often properly exercised against granting relief in the teeth of an alternative statutory remedy for the matters complained of (a) because to do so may undermine the applicable statutory regime by channelling a decision which the Oireachtas has determined should be reviewed administratively, into the court process and (b) because enabling such review will sometimes involve the court in adjudicating prematurely on a matter that, if resolved one way by appeal, might never require judicial determination. Thus, review may be precluded not merely of a decision that may be appealed, but the court may also decline to entertain challenges to the validity of underlying legislation if an appeal could resolve the issue animating the proceedings so as to render such a challenge unnecessary. These principles have a particular resonance in the specific context of attempts to pre-empt through court application the exercise by Revenue of its powers of tax collection and assessment. These are conferred within a statutory context intended to provide through the process of assessment and appeal an exclusive machinery for the ascertainment of a taxpayers liability, Criminal Assets Bureau v. Hunt [2003] 2 IR 168, at p. 185. While UCIIs action did not comprise an action by way of judicial review pursuant to Order 84 of the Rules of the Superior Courts, they sought in part to declare unlawful the exercise by Revenue of its statutory discretionary powers, and I can see no basis on which the same principles would not apply (see by way of analogy ODonnell v. Dun Laoghaire Corporation (No. 1) [1991] ILRM 301).
146. Both of the elements to which I have just referred - an alternative statutory remedy and a challenge to the validity of underlying statutory provisions - were present in Habte v. Minister for Justice [2020] IECA 22. There, this Court refused in its discretion to entertain certain challenges to the decision of the respondent to embark upon the process enabled by s.19 of the Irish Citizenship and Nationality Act 1956, as amended, for the revocation of the applicants certificate of naturalisation. The challenge was brought before the procedure of adjudication enabled by the section had concluded. Following the decision of the Supreme Court in EMI Records (Ireland) Ltd. v. Data Protection Commissioner [2013] IESC 34, [2013] 2 IR 669, the Court explained (at para. 122) that the onus is on the party seeking relief by way of Judicial Review to establish either that the alternative remedy is not adequate, or that there is a particular exigency which renders it unjust that it should have to be exhausted. It outlined four circumstances in which the authorities suggested that judicial review might be available in this circumstance, none of which were present in that case. The circumstances thus identified were where the ground of challenge includes a case based on fair procedures, where the issue is jurisdictional, where the case presents a net issue of law so that judicial review is particularly appropriate, or where there is a legitimate basis for perceiving an unfairness in forcing the applicant to proceed within the statutory scheme. These, it should be said, were not presented as being either exhaustive or universally applicable. They are reflected in the case law considering the relationship between relief by way of judicial review and statutory tax appeals as demonstrated by Stanley v. Revenue Commissioners [2017] IECA 279, [2018] 1 ILRM 397 (judicial review available where assessment alleged to be ultra vires and dependant on issue of statutory interpretation) and Menolly Homes v. Appeal Commissioners [2010] IEHC 49 (judicial review available where it is claimed tax inspector never had reason to believe tax due).
147. It follows that where a party seeks to challenge in declaratory proceedings the reasonableness of an administrative measure in a particular case or cases, it will in many circumstances face an insuperable obstacle in obtaining that relief where there exists a statutory appeal that could produce an outcome favourable to the complainant. However, this was not the relief claimed in this action. As I have already noted UCII did not challenge the legality of individual decisions of Revenue to value specific vehicles in a particular way. Had it done so, the challenge would have been clear and focussed on the methodology applied to and outcome of, a particular case. It would have been no answer to such a claim (were it permitted notwithstanding an appeal) that the outcome was anomalous or that the system operated by Revenue inevitably presented anomalies.
148. Instead, UCII framed its challenge at a level of abstraction, claiming that what was unlawful was the manner in which and extent to which Revenue assessed VRT and the manner in which the tax is operated and administered. An appeal is not necessarily an alternative remedy for such a systemic challenge because the appeal will not address the operation of the system as a whole and will merely focus on its application in an individual case. It may well be arguable that such a systemic challenge should not be permitted at all where the claimants real complaint arises from the application of that system to the valuation of individual vehicles and he has available to him an appeal which could obviate the need for proceedings, and it is at least arguable that an abstract challenge to a legislative or administrative scheme ought not to be permitted where divorced from a specific application of that scheme (see, by analogy, the comments of Charlton J. in Sweeney v. Ireland [2019] IESC 39 at para. 5 and following). Indeed in Brennan v. Attorney General [1983] ILRM 449 both the High Court (at p.486) and the Supreme Court ([1984] ILRM 355 at pp. 359 to 360) emphasised in determining the plaintiffs constitutional challenge in that case to the provisions of the Valuation Acts 1852 to 1864 that they had no mechanism other than the constitutional challenge to agitate their complaints.
149. In the course of her oral submissions to this Court Ms. OBrien SC suggested that it was not appropriate for the Court to enter into an inquiry in relation to what she phrased as the bona fides or the factual sustainability of a system as operated as opposed to an instance of the engagement of that system. The point was made faintly, was not referenced to authority and was not - beyond a general reference to the separation of powers - developed further. That was understandable in a context where no specific objection to the form of the case was articulated before the High Court Judge or in his judgment beyond very general statements that the Court was not conducting an inquiry. In this case there having been no legal objection raised to such a broadly framed - and at points abstracted challenge, the fact of an appeal in specific cases did not preclude the High Court from considering it.
150. As I explain later, a plaintiff who frames its claim in this way might thus avoid the consequence of a statutory appeal (at least where the respondents do not object to such a claim in the first place). However it does so at a cost, as it limits the case it can make in important respects. Revenue is entitled to react to the claim that generally its system is unlawful by replying that generally it is not. The very abstraction that renders an appeal an unviable alternative remedy thus limits the ambit of the challenge. The issue in this case was not whether the system was such that there may have been cases in which it produced anomalous or inconsistent results. It was whether the fundamental components of the overall methodology were lawful by reference to an identified standard. The level of scrutiny a court brings to bear on an abstracted question of this kind is inevitably blunted by the generality of the question it must answer. This becomes important when the challenge on the basis of irrationality is addressed. Further, to the extent that UCII complains of the potential for anomalies within the system of valuation, the availability of an appeal becomes relevant in confirming that the system as a whole provides a mechanism for dealing with those anomalies. For it to succeed in its challenge to the rationality of the system UCII must do more than show it has the potential to produce anomalous results in particular cases. I return to this later.
151. It is also important that this relief was combined with other claims that were not suitable for determination within the appeal system. In particular, a significant feature of UCIIs case depended on the claim that the inclusion of the dealers margin in the price used to calculate VRT was contrary to EU law (which would mean the provisions of s.133 itself were either invalid or required to be disapplied), and it was part of its case that the charging of VRT on a figure that included VAT was contrary to EU law. The decision in Petecel v. Minister for Social Protection [2020] IESC 25 suggests that in those circumstances, the court should not (at least in some cases) exercise its discretion against entertaining the claim. There, the appellant (a Romanian national resident in Romania) sought relief by way of Judicial Review of a decision of the respondent to refuse his application for disability allowance, a decision which was amenable to an internal statutory appeal process from which there was on onward appeal to the High Court on a point of law. The claim presented two relevant grounds. First, the appellant contended that the disability allowance question had been incorrectly classified in the relevant European Union Instrument intended to co-ordinate the social security systems of Member States (Regulation 883/2004 of the European Parliament and of the Council of 29 April 2004 on the coordination of social security systems). He claimed that in consequence the Regulation was invalid. That was a determination that could be reached only by the CJEU and could be brought before it only on foot of a Reference from the High Court. Although those administering the appeal process might have had the power to disapply the measure if they determined that it was contrary to EU law, this was not a determination of invalidity and - in any event - the Court believed it unlikely that those hearing any appeal would have felt other than bound to apply the legislation as it was (at para. 109). Thus, OMalley J. (with whose judgment the other members of the Court agreed) determined that the case came within the exception operative where for jurisdictional reasons the statutory appeal process cannot provide the remedy sought (at para. 110).
152. It is notable that the second issue in that case - the question of whether the appellant was (as required by the relevant legislative scheme) habitually resident in Ireland - could have been determined on appeal and had it been resolved in favour of the appellant would have obviated the need to consider the first ground. OMalley J., noting this, said (at para. 111):
the judgments in Koczan and EMI expressly refer to the possibility that a statutory appeal process might not, for jurisdictional reasons, be able to deal with all of the issues in a given case. It would, I think, be wrong (and possibly would breach principles of equivalence and effectiveness) if a claimant were to be required in all such cases to split up the issues and pursue different forms of litigation, with the increased time and costs that would necessarily be involved. It should be borne in mind that court procedural rules are intended to assist in the administration of justice, and not to put litigants through an extended series of obstacles.
153. In this case, UCII sought to agitate issues of EU law that - at least at the time the proceedings were instituted - could properly have been seen to be outside the scope of the jurisdiction of those determining any appeal. This may have been changed consequent upon the decision of the CJEU in Minister for Justice and Equality v. Workplace Relations Commission Case C-378/17 EU:C:2018:979). UCII also sought damages. If UCII could not be precluded from challenging the provisions governing VRT on EU grounds, Petecel would suggest that it should be allowed to maintain its claim on the other domestic grounds in the same action. For much the same reason, it might be said that because a substantial part of UCIIs claim was directed to a claim for damages, the imposition of a requirement to appeal before proceeding with such a claim might be unjustified.
154. I believe that this also disposes any suggestion that in some sense the existence of an appeal precluded the High Court from determining whether the method adopted by Revenue for determining the OMSP was reasonable. UCIIs claim was that that method was unlawful because it did not take into account a consideration mandated by statute, and because it was neither bona fide, factually sustainable or reasonable. Those claims were either well placed or they were not. The fact that there was an appeal in an individual case from a decision made according to that methodology, which appeal was determined according to an entirely different methodology, does not in and of itself affect the entitlement to assert and have determined the illegality (if such it be) of the initial valuation process.
VII THE OBLIGATION TO TAKE ACCOUNT OF THE ACTUAL SELLING PRICE
The argument
155. It was undisputed that Revenue did not, in determining at first instance the chargeable value of a used vehicle, make any reference to the price obtained for the actual vehicle in question - although regard appears to have been had to this in the appeal process. By determining the OMSP without taking account of that price, UCII contended, Revenue had failed to give effect to the legislation. Section 133, it said, required that VRT be calculated by reference to the value of the vehicle not by reference to the value of particular makes and models of vehicle generally. Ms. Barrington SC in her clear and helpfully structured submissions argued that although the section did not so provide where there was an invoice in respect of the sale of a vehicle prior to registration, this was the obvious proof of the envisaged Open Market Selling Price. Where there was not such an invoice, she accepted, there must be another mechanism. The Act did not, thus, only require that the price be related to that on an invoice. It did, however, mandate that Revenue have regard to that price where it could be proven.
156. In a related vein, UCII pointed to other provisions in the tax code that required Revenue to form an opinion as to the market value of an asset and to the fact that when operating these provisions Revenue interpreted its mandate as being to ascertain the actual market value of the asset in question. Uniquely in relation to VRT, it was said, Revenue had interpreted the text as providing a statutory mandate to establish a centralised system of imputed administrative values through the use of a series of schedules on depreciation and kilometre adjustments.
The High Court judgment
157. In rejecting the argument that Revenue was required to effect its valuation by reference to the price actually obtained for a specific vehicle, Murphy J. attached some importance to the fact that in respect of new vehicles, the statutory valuation did not correspond with the price that would have been obtained for the vehicle. The High Court noted in particular that the OMSP thus defined related to the manufacturers delivered price and could well be higher than the market price [h]owever artificial this may be as compared to actual prices negotiated by a buyer it is the statutory basis for depreciation (at p.194). Later in his judgment, he referenced this argument again, and specifically the complaint of UCII that Revenue did not take account of the actual market selling price of the vehicle and did not make any effort to value the vehicle under assessment but rather charged the tax on a notional value for similar vehicles (at p. 204). Thereafter, the Court emphasised that the expression in the opinion of the Commissioners as it appeared in the governing provision was a fundamental element in determining the open market selling price (at p. 205). The interposition of that discretion, in the Courts view, was not consistent with an obligation to fix the notional price envisaged by the legislation by reference to the actual consideration received following a specific sale. Murphy J. put it as follows (at p.196):
The statutory provision regarding the determination of OMSPs is unambiguous. It is based on the opinion of the Revenue Commissioners on the basis of the declared price. The evidence does not establish an arbitrary or secret system. While there is a provision which allows for regulations, there was no suggestion as to what regulations were necessary.
The legal test
158. The argument requires close attention to the terms of s.133(3)(c). At the relevant times it provided three separate definitions of OMSP:
(a) in the case of a new vehicle referred to in subsection (2), the price as determined by that subsection,
(b) in the case of any other new vehicle, the price, inclusive of all taxes and duties, which, in the opinion of the Commissioners, would be determined under subsection (2) in relation to that vehicle if it were on sale in the State following supply by a manufacturer or sole wholesale distributor in the State,
(c) in the case of a vehicle other than a new vehicle, the price, inclusive of all taxes and duties, which, in the opinion of the Commissioners, the vehicle might reasonably be expected to fetch on a first arms length sale thereof in the State by retail and, in arriving at such price
(i) there shall be included in the price, having regard to the model and specification of the vehicle concerned, the value of any enhancements or accessories which at the time of registration are not fitted or attached to the vehicle or sold therewith but which would normally be expected to be fitted or attached thereto or sold therewith unless it is shown to the satisfaction of the Commissioners that, at that time, such enhancements or accessories have not been removed from the vehicle or not sold therewith for the purposes of reducing its open market selling price, and
(ii) the value of those enhancements or accessories which would not be taken into account in determining the open market selling price of the vehicle under the provisions of subsection (2) if the vehicle were a new vehicle to which that subsection applied shall be excluded from the price
159. Subsection (2) as referred to here, provides as follows :
(a) For a new vehicle on sale in the State which is supplied by a manufacturer or sole wholesale distributor, such manufacturer or distributor shall declare to the Commissioners in the prescribed manner the price, inclusive of all taxes and duties, which, in his opinion, a vehicle of that model and specification including any enhancements or accessories fitted or attached thereto or supplied therewith by such manufacturer or distributor, might reasonably be expected to fetch on a first arms length sale thereof in the open market in the State by retail.
(b) A price standing declared for the time being to the Commissioners in accordance with this subsection in relation a new vehicle shall be deemed to be the open market selling price of each new vehicle of that model and specification.
(c) Notwithstanding the provisions of paragraph (b), where a price is declared for a vehicle in accordance with this subsection which, in the opinion of the Commissioners, is higher or lower than the open market selling price at which a vehicle of a similar type and character is being offered for sale in the State at the time of the declaration, the open market selling price may be determined by the Commissioners for the purposes of this section.
(d) Where a manufacturer or sole wholesale distributor fails to make a declaration under paragraph (a) or to make it in the prescribed manner, the open market selling price of the vehicle concerned may be determined by the Commissioners for the purposes of this section.
160. In analysing this aspect of UCIIs case in the light of the section it is important to bear in mind the distinction between matters that are so central to the exercise of a discretionary power that they must be taken into account in its exercise and matters which are not irrelevant and therefore may be taken into account. While there is no doubt but that Revenue could legitimately take into consideration the price obtained on the sale of a vehicle in deciding the OMSP (and indeed seemingly did so when determining appeals), that is not sufficient for UCIIs case. Unless another ground of irrationality can be identified it is only where matters in the former category are not taken into consideration that a decision is vitiated (R(NAHS) v. Department of Health [2005] EWCA Civ 154 at para. 63).
161. The decision of the Divisional Court in Dellway Investments Ltd. v. National Asset Management Agency [2010] IEHC 364, [2011] 4 IR 1 at paras. 69 to 71 identifies the principles by reference to which a court should determine if a matter must be taken into account in the exercise of a statutory discretionary power. Thus, as that decision makes clear, the Court is concerned to ascertain what the Act in terms requires to be taken into account or what may be said to be required by implication by virtue of the subject matter, scope and purpose of the legislation concerned. Here s.133(3) does not expressly require Revenue to take into account the price actually fetched for the vehicle. The issue reduces itself to whether such an obligation should be implied having regard to the subject matter, scope and purpose of the provision.
162. In approaching that question, the statute falls to be construed in accordance with well established principle. The Court is concerned to ascertain the intention of the legislature having regard to the language used in the Act but bearing in mind the overall purpose and context of the statute. While UCII placed some reliance for these purposes on the fact that the Court was concerned with the construction of a taxing statute, I do not see that this affects the essential inquiry in this case. In construing such an Act in Dunnes Stores v. Revenue Commissioners [2019] IESC 50 (at para. 63) McKechnie J. observed context is critical: both immediate and proximate, certainly within the Act as a whole, but in some circumstances perhaps even further than that. More recently, in confirming the relevance of legislative purpose to the construction of a revenue statute, ODonnell J. in delivering the judgment of the Court in Bookfinders Ltd v. Revenue Commissioners [2020] IESC 60 has explained (para. 52) :
It is not, and never has been, correct to approach a statute as if the words were written on glass, without any context or background, and on the basis that, if on a superficial reading more than one meaning could be wrenched from those words, it must be determined to be ambiguous, and the more beneficial interpretation afforded to the taxpayer, however unlikely and implausible.
163. The rule of strict construction with which Revenue legislation is identified (and assuming that the principle applies to provisions dealing with the mechanics of calculation of liability as opposed to those imposing the charge in the first place), operates (according to Bookfinders), as follows (para. 52) :
If, after the application of the general principles of statutory interpretation it is not possible to say clearly that the Act applies to a particular situation, and if a narrower interpretation is possible, then effect must be given to that interpretation,
Application
164. In resolving this question, the most obvious and important feature of the definition of OMSP relevant to used cars (that in s.133(3)(c)) is that it does not relate the chargeable value in any way to the price actually obtained for the vehicle. Instead, the focus within the Act is not on what the vehicle has fetched on a particular sale but upon what it might reasonably be expected to achieve in a first, arms-length retail transaction in the State. The legislation thus envisages an exercise in prediction that is divorced from any particular transaction and is instead confined only by the requirement of reasonable expectation.
165. That focal point is mediated through the opinion of the Commissioners. The OMSP is not the price the vehicle has fetched or the price that it might be reasonably expected to fetch, but that which the Commissioners believe it might reasonably be expected to fetch. In this regard the Commissioners must, clearly, act rationally in the formation of that opinion. However, the anchoring of the chargeable value in their opinion and the delimitation of that opinion by reference to their assessment of what price might be reasonably expected removes the inquiry envisaged by the Act from that into the price obtained, and firmly locates it within Revenues forward-looking judgment. The legislation was well described by Ms. OBrien SC in her submissions for Revenue, as envisaging a degree of estimation, an objective not immediately consistent with a limitation by reference to the price actually obtained.
166. Moreover, that exercise in prediction is clearly directed by an over-riding preference for consistency. As Revenue contended, the section envisages a uniform method of valuation (as suggested by the reference to the State) rather than one dictated to by the particular features of an individual sales transaction between particular parties at a specific point in time or place. UCIIs argument if well placed would thus fundamentally alter the nature of the envisaged valuation process. It would mean that the rate of VRT chargeable on a vehicle would be mandatorily dependent on when, where and by whom a vehicle was sold rather than by Revenues judgment as to what it should obtain on the market. That the exercise in valuation envisaged by the Act was to occur in a context in which there was never going to be any sales value available for the many imports comprising second hand vehicles purchased abroad for private use, rendered the need for such consistency all the more pressing.
167. None of this is changed by the feature of the section around which UCII structured its argument - the reference in the provision to the vehicle. While it is certainly the case that the Act is concerned with the price a particular vehicle would be reasonably expected to fetch, that does not mean that Revenue are constrained to what that vehicle has actually fetched. What it means is that Revenues assessment of the price the vehicle might reasonably be expected to fetch must take account of the particular features of the vehicle itself - hence the requirement to make the adjustments stipulated in s.133(c)(i). Revenue complies with this obligation by undertaking an inspection of the vehicle and having regard to the condition of the vehicle and its specific features. However, the use of the definite article in the provision does not mean that Revenue, in forming an opinion as to the price the vehicle might reasonably be expected to fetch is required to take account of the price that vehicle has obtained or is precluded from approaching that exercise by establishing a scale of values. UCIIs argument, wrongly, elides these two separate questions the price the vehicle might reasonably be expected to fetch, and the price the vehicle has actually obtained. The clear obligation to determine the first, does not necessarily entail an obligation to take account of the second.
168. Viewed in that light, the context and purpose of FA92 are revealing. MVED was charged on both new and used vehicles by reference not to the price obtained for a vehicle, but to a notional price, framed in the case of new vehicles by the RRP and in the case of used vehicles by that price as reduced by reference to an identified scale. Nothing in the 1992 Acts suggests that in this respect the legislature was adopting a different approach. On the contrary, none of the other OMSPs as defined in FA92 are based on a price actually fetched. The OMSP of new vehicles is determined by the declaration of the opinion of the manufacturer or sole wholesale distributor of the price it believes that vehicle might reasonably be expected to fetch. Where the Commissioners form the opinion that that price is higher or lower than the price at which the vehicle is offered for sale the Commissioners may - but are not required to - themselves determine the OMSP.
169. Furthermore, when one applies these factors to the contention advanced by UCII - that Revenue must take account of the price obtained - a series of imponderables follow. For a start, if required to take account of the price obtained, then in reality it follows that Revenue must, absent good reason for not so doing, use that price as to OMSP. There is, in reality, no other way that that can be meaningfully taken account of. Of course, good reason for not using that price might include concern that this was not the actual price paid, or a change in the market between payment of the price and registration, or (perhaps) a belief that one or other party had driven a better bargain than the other should have accepted. But the legislation - and in particular the exercise in estimation which properly construed it envisaged - gives no sense that it was intended that Revenue be mandated to put in place an inquiry structure of this kind. Having regard to the sheer scale of the task envisaged by the Act, it would be surprising if an obligation to take account of price and thus an implied requirement to establish and operate a framework that would enable an individualised assessment of the transactions underlying every tendered price, was envisaged by the Oireachtas.
170. Thus, it seems to me that everything - the language of the section, its focus on the opinion of Revenue and the benchmark of reasonable expectation, the legal and factual context and consequence of the obligation suggested had it been imposed - points to the Oireachtas envisaging exactly what the Revenue has done. It did not mandate the determination of OMSP by reference to the price actually paid for a vehicle and, accordingly, necessarily sanctioned the establishment by Revenue of general criteria by reference to which the price that might reasonably be expected for a used vehicle in the State could be ascertained. Those criteria could include the price actually obtained on a sale (it will be recalled Revenue offered to approach the matter in this way, but the trades response - for whatever reason - was not enthusiastic), but it did not have to do so.
VIII THE TEST FOR DETERMINING THE LEGALITY OF REVENUES METHODOLOGY
171. That begs the question as to what test should be applied to assess the legality of that methodology. Both parties proceeded in the High Court, and in this Court, on the basis that the applicable test for determining whether Revenue had properly exercised its discretionary power to determine the OMSP of vehicles was defined by the decision of the Supreme Court in State (Lynch) v. Cooney [1982] IR 337. There, the Court was faced with a combined challenge to the constitutional validity of section 31(1) of the Broadcasting Act 1960, as amended, and to the validity of an order made by the Minister for Posts and Telegraphs under that section directing RTE to refrain from broadcasting any matter made by, on behalf of or inviting support for the Sinn Fein organisation. The section stated that such an order could be made where the Minister is of the opinion that the broadcasting of a particular matter would inter alia tend to undermine the authority of the State.
172. OHanlon J. struck down s.31 because, he concluded, the power given to the Minister on the face of the section was absolute, and thus it was not open to the Court to examine whether there were any reasonable grounds for the formation of such an opinion. He said that he was not aware of any case where the use of the expression is of opinion had admitted of judicial review with regard to the reasonableness of the decision made (although he did agree that an opinion that was not bona fide formed could be so reviewed). Thus construed, the section enabled the Minister to form an opinion on wholly unreasonable grounds thereby prohibiting access to broadcasting time by persons but without there being any jurisdiction vested in the Courts to relieve against such an exercise of the power. For that reason, the High Court held, the provision was invalid.
173. The comparatively brief judgment of the Supreme Court addressing the constitutional question rejected the absolute interpretation of s.31 adopted by the High Court, and it accordingly upheld the section. OHiggins CJ shortly stated (at p.361):
The Court is satisfied that the sub-section does not exclude review by the Courts and that any opinion formed by the Minister thereunder must be one which is bona fide held and factually sustainable and not unreasonable.
174. In the course of her submissions, counsel for UCII suggested that this three-limbed formulation - bona fide held, factually sustainable and not unreasonable - presented the appropriate test for judicial review of an administrative power defined by the formation of an opinion and differed from the generally applicable test of unreasonableness reflected in OKeefe v. An Bord Pleanala [1993] 1 IR 39 as more recently developed in Meadows v. Minister for Justice, Equality and Law Reform [2010] IESC 3, [2010] 2 IR 701. It imposed, it was contended, a much more intrusive and rigorous review than had been envisaged in OKeefe, although it was suggested that since the decision in Meadows it was a question of nuance.
175. The difference asserted was itself two fold. First it was said that the test was more demanding in that it was not sufficient to show some basis for having formed the opinion, because what was termed the factually sustainable leg of The State (Lynch) v. Cooney appeared to impose a higher threshold than the test in OKeefe and required a degree of scrutiny of the factual basis for the opinion. Second it was said that the burden was on the respondent - although that was modified in the course of argument to suggest the decision merely meant that a justification of the opinion was required. The argument thus presented focussed on two features of the decision - the fact that it was concerned with the review of an opinion and the consideration that under the relevant legislation that opinion involved the consideration of certain factual matters.
176. I do not believe the suggestion that State (Lynch) v. Cooney establishes a particular standard of review operative wherever the power of an administrative authority is defined by the formation of an opinion as a precondition to administrative action is well placed. In fact, the most significant feature of the decision was the conclusion that a statutory power framed by reference to an opinion was no different in terms of its amenability to review than a power framed by reference to whether a Minister was satisfied of a particular matter. It would be remarkable if the Court had actually - for no immediately obvious reason - concluded that such a power was more prone to review than the exercise of other differently phrased statutory discretions. When the Court referred to the requirement that the Ministers opinion be factually sustainable, it was doing no more than reflecting a requirement suggested by the common law authorities that where a statutory provision predicates the exercise of a discretionary power on the existence of certain facts, the Court in resolving a challenge to its exercise in a given case must inquire into whether those facts exist. Indeed, it may be that this aspect of the Courts formulation reflected the comments of Lord Wilberforce in Secretary of State for Education and Science v. Tameside Metropolitan Borough Council [1977] AC 1014 at p. 1047 and cited by OHanlon J. in the High Court:
If a judgment requires, before it can be made, the existence of some facts, then, although the evaluation of those facts is for the Secretary of State alone, the court must inquire whether those facts exist and have been taken into account, whether the judgment has not been made upon a proper self-direction as to those facts, whether the judgment has not been made upon other facts which ought not to have been taken into account. If these requirements are not met, then the exercise of judgment, however bona fide it may be, becomes capable of challenge
177. This describes the precise issue that presented itself in State (Lynch) v. Cooney: the Ministers power to issue a prohibition order was dependant on the broadcast inter alia undermining the authority of the State, and it was to that factual issue that his opinion had to be directed. In considering the legality of his decision, the Court was required to consider whether that factual proposition was sustainable, albeit in a context where the evaluation of the facts was a matter for the Minister. It follows that while the former may at least in some cases (specifically those involving the exercise of the power to impose a very significant constraint on the exercise of constitutionally protected rights) fall to be determined according to an objective test the latter is, obviously, subjective (see the judgment of Henchy J. at p. 380 to 381 and the discussion of this aspect of the case in Hogan and Morgan Administrative Law, 5th Ed. 2019 para. 17-11). Thus, in his judgment on the non-constitutional issues, OHiggins CJ (with whom Griffin and Hederman JJ. agreed) described the nature of the review, as follows (at p. 365):
the Courts only concern is to enquire whether, in using his statutory powers, the Minister acted reasonably in accordance with the factual situation as he saw it.
178. The other judges phrased their analysis in similar terms. Walsh J. (with whom Hederman J. also agreed) expressed the view that the evidence was sufficient to enable a Minister reasonably to form the opinion that any broadcast made on behalf of that party .. could itself amount to an advancement of a cause and of methods which are inimical to the legitimate authority of the State (at p.372). Henchy J. (whose judgment presents the major statement of the applicable principle, Hogan and Morgan para. 17-11) suggested an objective test for the determination of whether a factually based statutory pre-condition had been established in cases where a decision making power which affected the exercise of personal rights was made conditional on a prescribed opinion, but proceeded from there to explain the limits of a statutory power in entirely conventional terms (at p. 380 to 381):
This means, amongst other things, not only that the power must be exercised in good faith but that the opinion or other subjective conclusion set as a pre-condition for the valid exercise of the power must be reached by a route that does not make the exercise unlawful - such as by mis-interpreting the law, or by misapplying it through taking into consideration irrelevant matters of fact or through ignoring relevant matters.
179. It follows that in determining how this test is applied to the exercise of any particular statutory power, it is necessary to begin with an understanding of the power itself. As I have noted, in State (Lynch) v. Cooney, the Ministers opinion under the statute in question was directed to whether a broadcast would have a particular effect, and in that case this depended on his appraisal of certain readily identifiable facts, effectively the nature and objects of the organisation whose broadcast it was proposed to prohibit. Here, the object is quite different. Revenue must form an opinion as to what a vehicle might be reasonably expected to fetch on a sale. Any opinion of value will depend on facts, usually the value afforded to other similar assets. However, the element of judgment in the determination of value is wide, and the compass of fact to be considered less determinate, less circumscribed and (unlike the situation in State (Lynch) v. Cooney) is not specifically directed as a jurisdictional pre-requisite to the taking of a particular step. The valuer must exercise judgment in determining the assets with which the item to be valued should be compared, in deciding what sources of information it will consult in identifying those comparators, in adjusting the value to reflect differences between the assets, in taking account of changes in the market and - in this case - in addressing depreciation, condition of the vehicle, kilometrage and the value of extras. There is no fact that must exist before Revenue may exercise this function, although there are many facts to which it may have regard in discharging it.
180. In deciding whether the conclusion Revenue has reached having exercised that judgment is reasonable as that term is used in that case and (as arises here) whether the methodology it has adopted can be so categorised, the Court must, of course take account of the factual basis on which Revenue has fixed on that methodology and must be satisfied that that basis was itself reasonable. However, in this case I do not think it either accurate or helpful to address this in terms of whether that basis is factually sustainable. That description may be apt to the process of interrogating a decision of the kind addressed in State (Lynch) v. Cooney, which, properly interpreted, may in certain circumstances require an objective test to the ascertainment of facts that comprise factual preconditions to the exercise of a statutory power. Here, the issue is one of approach and analysis rather than of fact per se, and if this decision is to be used as the benchmark for review (and it was the only authority referred to by either party in their written submissions to this Court or the High Court in that connection) it seems more apposite to look at the test actually applied to the administrative law challenge in that case whether the decision maker acted reasonably in accordance with the factual situation as he saw it, whether the evidence was sufficient to enable a Minister reasonably to form the opinion and whether the decision maker misapplied his discretion through taking into consideration irrelevant matters of fact or through ignoring relevant matters.
181. The decisions in State (Keegan) v. Stardust Compensation Tribunal [1986] IR 642 and OKeefe v. An Bord Pleanala as developed in Meadows v. Minister for Justice, Equality and Law Reform (both of which were briefly referred to in oral argument) continue to define the general approach to the determination of reasonableness as it is used in these passages. Given that neither party made any detailed submission to this Court as to the precise relationship between these decisions, or the exact parameters of the power of review after them, I will limit myself to this, I think, uncontroversial summary. The burden of proof is on a party seeking to review a decision on the grounds of unreasonableness. Nothing in State (Lynch) v. Cooney purports to articulate a different burden of proof, and nothing in Keegan, OKeefe nor in Meadows suggests the Court in imposing that burden was acting inconsistently with State (Lynch) v. Cooney. The inquiry as to whether that burden has been discharged is defined by what the Chief Justice in OKeefe described as the limited scope for judicial intervention on the basis of irrationality (at p.71), and the anchoring of that power of intervention by reference to fundamental variance with plain reason and common sense (at p.70). In OKeefe the Court also said that in determining whether the decision is unreasonable as thus understood, the Court was concerned inter alia with whether the decision-making body had before it any relevant material that would support its decision, although it is notable that that part of the Chief Justices judgment was not referred to in Meadows. It is not necessary in this case to apply that aspect of the OKeefe test - which may well have been intended to operate only in the context of the type of quasi-adjudicative power with which the Court was concerned in that case. Insofar as Meadows affected this, its impact is widely viewed as directed to the introduction - at least in certain circumstances - of a requirement to consider proportionality in the context of decisions affecting some constitutional or ECHR rights (see, Efe v. Minister for Justice, Equality and Law Review [2011] IEHC 214, [2011] 2 IR 798 and most recently, Friends of the Irish Environment v. Government of Ireland [2020] IESC 49 at 5.54). I will return to proportionality insofar as relevant to this case, further below.
182. In this case that test fell to be applied at a systemic level. In other words, the question was whether the method developed by Revenue to determine OMSP was reasonable as understood in the authorities. It was not whether a particular decision to value a specific vehicle met the test, nor whether the method was incapable of generating anomalous results. In essence it was whether, in principle, the use of a system designed to deliver the value of a used vehicle as it appeared in the Car Sales Guide and other sources identified by Revenues witnesses afforded, having regard to how it actually operated according to the evidence, a reasonable basis as thus understood on which Revenue could form its opinion as to the OMSP of that car in the manner envisaged by FA92.
IX ALLEGED INCONSISTENCIES IN REVENUES EXPLANATION OF ITS SYSTEM
The argument
183. All of this presents the issue of what Revenues methodology actually was. That question occupied a significant aspect of UCIIs submissions in this Court in respect not merely of the reasonableness of the valuation process in domestic law, its sustainability having regard to the requirements of EU, and the principles of legal certainty applied in both legal regimes, but also its contention that the trial Judge had not, as he was required to do, properly engaged with the evidence. Central to all of these complaints was the proposition that Revenues case had been internally contradictory as to the role of depreciation tables in the methodology and that the trial Judge had never resolved that contradiction.
184. The contention that Revenue had presented inconsistent accounts of how it conducted the process of valuation was, as presented in UCIIs submissions to this Court, based upon a contrast between two aspects of Revenues evidence. The first was the evidence of Mr. Butler. Citing the trial Judges summary of the evidence, UCII presents Mr. Butler as having said that the OMSP was calculated by reducing the open market selling price of the new vehicle in accordance with a depreciation table and incorporating further reductions for the condition and mileage. Then, UCII says, on Day 27 Dr. Bacon testified that the description provided by Mr. Butler was not in fact how Revenue valued cars for VRT purposes at all, that his evidence was that the cars were valued against prevailing used car prices as published in the Car Sales Guide, and that he had in fact said (Day 27 p.51) the Revenue dont use depreciation schedules at all. The trial Judge, UCII complains, never addressed the implications of the fact that the defendants own witnesses provided two differing descriptions of the process for determining the chargeable value of the used vehicles. It is suggested that this was particularly important given UCIIs case that if in fact the depreciation tables were relied upon to determine value, these failed to comply with European law and were out of sync with international trends thereby vitiating that process. Furthermore, it is said, given that Revenues own witnesses could not agree as to how the chargeable basis of the tax was calculated the trial Judge could not have concluded that the system was transparent or complied with principles of legal certainty.
185. I think it fair to describe this contention as a centrepiece of UCIIs appeal. It is the first point made in its written submissions, presented with the exclamation that it would be difficult to overstate the significance of this divergence in the Defendants evidence, and it was the subject of some considerable focus at the hearing of the appeal.
The evidence
186. So stated, this argument suffers from three infirmities. To understand these, it is necessary to revisit what the evidence disclosed Revenues methodology to be, and the role the depreciation schedules played in that methodology. In short, and as I have explained in more detail earlier, the depreciation schedules presented a set of scales by reference to which Revenue determined how the OMSP of a car when new was reduced to reflect the reduction in the value of that car over time. For some of the time covered by this action the fourteen, twenty-two or twenty four schedules (as the case may be) provided for fixed reductions to that value for each year. The role of the Revenue official in Rosslare in appropriating a statistical code to a make or model was to identify which of these schedules reflected the reduction in value as evidenced by market experience, that market experience being ascertained from the Car Sales Guide and other sources. The evidence was that this information was updated from time to time, ensuring that the schedules as applied to a particular make and model of vehicle continued to produce the market price. The input of the statistical code enabled a valuation to be identified for the make model and year, to which reductions were then applied following inspection by the official. As I have also emphasised in my earlier summary, at no point was it suggested in Revenues evidence that the process of valuation of an individual vehicle involved cross-checking the valuation generated through the input of the statistical code with trade information.
187. With the introduction of the facility for hybrid schedules, that process became more nuanced, as it was possible to move in and out of different schedules to provide a more accurate reflection of the data collected from the market. However - and it seems to me that this is the key point - it was Revenues case that the schedules were chosen to reflect the information from the market. Thus, the depreciation schedules were chosen so as to ensure that the input of the statistical code produced the same valuation as would have been obtained had the official available to him or her a compendium of the market prices for all models and makes of vehicle. While that was the object of the methodology, the question of whether it succeeded was another matter, dependent upon the depreciation schedule chosen correctly aligning with the market price as disclosed by these sources, and it being reviewed with sufficient frequency to ensure that it continued to reflect those market prices.
Analysis
188. This leads to the first difficulty with this aspect of the case urged by UCII. Its argument depends on the issue being viewed in binary terms, with Mr. Butler recording the centrality of the depreciation schedules, and Dr. Bacon viewing the market information from the Car Sales Guide as the only relevant data source. Once the evidence given by Revenue is considered in its totality, however, it becomes clear that the two cannot be separated in this way. Indeed, the evidence of Mr. Butler did record the relationship between the depreciation schedules and the information derived from the Car Sales Guide and other sources. I have quoted that evidence earlier but see in particular my reference above to his evidence at Day 22 p.113 Q.346. It was most pithily reflected in his statement that while the tendency was to believe that Revenue assigned a table to produce a value it was in fact the value that produced the table in the first instance (Day 23 p. 88 Q.26). This was echoed in the evidence of Mr. Campion. It will be recalled from my earlier summary that he explained that the depreciation schedules were tools in an algorithm on a computer system to deliver the result which the valuation officer has already determined by reference to the Car Sales Guide and other sources of information.
189. The evidence of both witnesses corresponds with Dr. Bacons explanation of the methodology : rather than starting with the schedules the actual process starts with used car prices as published in the Car Sales Guide (Dr. Bacons Report at para. 3.17). Indeed it was presumably for this reason that Mr. Sreenan SC specifically asked Mr. Butler in cross-examination why, if his account of the function of the depreciation tables was correct, it was necessary to have the depreciation tables at all.
190. The response - that this was more efficient, allowed the 32 offices to work in unison and enabled central control and monitoring over the cars and tables as they were assigned - was not inconsistent with what Dr. Bacon had said in his evidence. Dr. Bacon said that the cars were valued on the basis of the Car Sales Guide and that depreciation tables were used as an administrative tool to ensure consistency and efficiency. That was what Mr. Butler had said - the depreciation tables were created to generate the Car Sales Guide prices and were used for that purpose (rather than simply maintaining a data base of price for each model, make and vintage of vehicle) in order to obtain such consistency. In fact, it will be recalled from my outline of the evidence on this issue that counsel cross-examined Mr. Butler on the basis that the depreciation tables were used because it was just too much trouble to keep on collecting the data and it was much easier just to have a few arbitrary scales. That was rejected by the witness.
191. While UCII places some emphasis on its claim that Dr. Bacon testified that Revenue did not use depreciation tables at all, this is not merely inconsistent with what he actually said in his report (which was that they used them for administrative purposes and with a view to consistency) but is not a complete account of his actual evidence. The context of the statement made by him under cross-examination makes it clear that his evidence was that Revenue did not use the depreciation tables to arrive at a value. He introduced the issue (Day 27 p.7 Q.8) by recording his finding that Revenue do not use such schedules for the purpose of calculating VRT liability. That framed the observations made later upon which UCII focussed. The full quotation (Day 27 p. 51 Q.71) was as follows:
It seems to me that what is said in Mr. Dowlings report is that the Revenue Commissioners use depreciation schedules to arrive at a value but if you use a different depreciation schedule you would get a lower value of the vehicle than you would do if you use the depreciation schedules which the Revenue use. But the Revenue dont use depreciation schedules at all to arrive is my understanding.
(Emphasis added.)
192. While not using the tables to arrive at a value, Dr. Bacon was clear that Revenue did use the tables. He said (Day 27 p.26 Q. 32) :
they dont use the depreciation schedules from the point of view, in order to establish the base. They use the retail price. And whatever the retail price is, it seems to me they go back through the depreciation schedules and say there is a - for the administrative purposes for which they use the depreciation schedules - they look at their schedules and say well there is the depreciation that corresponds with that.
193. In a similar vein - and as it happens reflecting closely what Mr. Butler had said in his evidence - Dr. Bacon observed (Day 27 p.48 Q. 69) :
Depreciation schedules, to the extent that they were used, were basically used in a backwards way, that they took the retail price and there is the original price then the depreciation was that [indicating] not that depreciation was the basis for, or any schedules of depreciation was the basis on which the VRT was established.
194. Thus, when the explanation tendered by Mr. Campion as to how the tables were used by Revenue was put to Dr. Bacon, his response was clear (Day 27 p.74 Q. 170) :
That does not contradict my understanding of what I was told in the sense that we take the Car Sales Guide price and the new price and in effect take a depreciation table that links the two.
195. The depreciation table, he explained, is a schedule of depreciation which takes it to the Car Sales Guide price (Day 27 p.81 Q.198).
196. In fairness to UCII, it must be said that at points Dr. Bacon - while stating that the depreciation tables were used by Revenue - did come across as being unclear to the exact use to which they were applied. On occasion he referred to a methodology by which Revenue found the retail price in the Car Sales Guide and from there (for no immediately apparent reason) looked at the depreciation schedules to see well which one gives that (Day 27 p.56 Q. 95). This suggests some confusion between the proposition that the schedules are assigned by reference to, and updated to take account of, the Car Sales Guide price (which was what Revenues evidence posited) and a process where the Car Sales Guide was looked at every time a car was valued with the depreciation schedules being consulted thereafter (which, emphatically, did not occur). Both meant that Revenue was using the Car Sales Guide prices, but in different ways - albeit to achieve (at least in theory) the same result. The confusion is, perhaps, confirmed by the fact that Dr. Bacon confessed in cross-examination that he did not actually know the administrative purposes for which the schedules were used (Day 27 p.64 q.128), and that he did not have the faintest idea why Revenue retrofit a depreciation table to the Car Sales Guide price (Day 27 p.76 Q. 180). Furthermore, and similarly, Dr. Bacon did suggest in the course of his report that a depreciation schedule could not be used if it produced a wrong value (at para. 3.18). He put it as follows:
if a schedule is wrong as is implied by the submissions then it simply cannot be used - it cannot produce a value for a particular car that is not the same as in the CSG. By implication, a schedule that is used for any particular car must be correct.
197. If this statement was intended to suggest that the Car Sales Guide could override the schedules, that was wrong. Mr. Fogartys evidence was clear that a vehicle owner could not at the point of registration dispute the OMSP produced by inputting the statistical code by reference to a different price in the Car Sales Guide. He could, however, seek to have the OMSP changed before registration by making representations to CVO in Rosslare, or he could appeal (Day 28 p.136 Q.564). As it happens Dr. Bacon was not cross examined on this statement (and it did not form part of the case on appeal as advanced in oral and written legal submissions). In any event, I have been at pains to note in the course of my consideration of Revenues evidence, that it was most emphatically not the evidence of Revenues witnesses of fact that in an individual case the application at first instance of the depreciation schedules would be arrested by production of a different price from the Car Sales Guide, and it is not a version supported by any documentation. In that regard I think it important that on the second day of Dr. Bacons evidence he produced and read into the record a note prepared by Mr. Coyle explaining Revenues methodology. I have summarised the contents of that note earlier. There could have been no doubt from that document how Revenue attributed and used the depreciation tables.
198. However, even if Dr. Bacons evidence is interpreted as demonstrating a belief that whenever a vehicle was registered its value was determined by reference to the Car Sales Guide with the depreciation tables operating to provide some form of undefined administrative back up to that exercise (and I do not believe that properly construed it can be so interpreted) it was not correct to present this as a conflict that had to be resolved in the manner repeatedly claimed by UCII. An expert witness cannot give evidence of fact, irrespective of where he says he received his information from. The Court can only act on the basis of the factual evidence of witnesses of fact, and the opinion evidence of experts. Insofar as the latter is dependent on the experts recitation of fact, if those facts are not admitted and not proved as a matter of fact, the opinion cannot be acted on. Where the facts recited by the expert conflict with the facts as proven, the facts as proven prevail. All of this reinforces the concerns I later express in respect of the failure of Revenue to make their methodology generally known, but it does not affect the trial Judges essential conclusion on methodology, which was more than established by the evidence, with all of the evidence of fact being entirely consistent as to the method used by Revenue. What is important here, however, is that the High Court never concluded that the Car Sales Guide over-rode the depreciation tables (if this was what Dr. Bacon was implying), and Dr. Bacons opinion was based on the alignment between the Car Sales Guide prices and the output of the depreciation tables, not on the former in some sense cancelling out the latter. That alignment represented the focal point of Revenues evidence, and it was accepted by the trial Judge as a matter of fact. The conflict might have been relevant to the Courts assessment of Dr. Bacons evidence (which is quite different from an internal conflict between Revenues own evidence of fact), but the Court did not have to do this if it was not germane to its conclusions. Given that no matter how one interpreted Dr. Bacons evidence it was based on the proposition that the valuation methodology (whatever it was) was intended to yield the price recorded in the Car Sales Guide, it is not obvious that it was necessary to address the issue.
199. That leads to the second point of importance. In its written submissions, having presented the issue as one of Dr. Bacons evidence as against Mr. Butlers evidence UCII suggests that it was only on Day 27 of the trial that the position recorded in Dr. Bacons evidence was expressed by Revenue. The implication of a dramatic last-minute shift in Revenues position was repeated in the oral submissions to this Court. Dr. Bacons evidence involved (it was said) a surprising twist. In fact, UCII knew in advance of Day 27 of the the trial what Revenues evidence in this regard as to be. Putting to one side that UCII had Dr. Bacons report in advance of his actually giving evidence (it is dated December 2011), Revenues High Court written submissions (also delivered in December 2011) made it absolutely clear from the outset what Revenues case (not Dr. Bacons evidence) would be. These stated as follows (at paras. 2.10 and 2.11):
The Plaintiff in its submissions asserts that the OMSP is derived from the depreciation tables. This assertion is entirely incorrect. Rather, the Commissioners commence with the second hand value of a particular model and choose a particular depreciation table to fit the depreciation trends of that model having regard to its second hand value at a point in time. This enables the Commissioners to provide OMSPs readily and consistently.
The Plaintiffs submissions also claim that the valuation system is entirely computer based. This assumption is also groundless. To value models of vehicles, the Commissioners form an opinion of the valuation of the particular model based on market material and research. The valuations that they reach on this basis are consistently applied by means of depreciation codes. Consequently the Commissioners opinion on OMSP is expressed through the operation of the system.
(Emphasis added.)
200. Revenue made its position in relation to the depreciation schedules clear in the course of the cross examination of UCIIs witnesses. It was put by counsel for Revenue to Mr. Yarrow, one of UCIIs expert witnesses that depreciation tables arent the central data source in the valuation process in this country (Day 17 p. 58 Q.276). And more importantly, as I have said, other witnesses of fact confirmed that this was how the system operated. UCII knew from the start that this was the position adopted by Revenue, and indeed Mr. Dowling of DKM (UCIIs first witness) alluded in the course of his evidence to the suggestion of a contradiction in Revenues explanation of its methodology. It was open to UCII to cross examine Revenues witnesses around this suggested contradiction and indeed - at least indirectly - UCII did so. The response of the Revenue witnesses was clear and robust. They attested to the relationship between the values of vehicles in the market and the depreciation schedules as I have outlined it, and clearly explained why those schedules were used to that end. Dr. Bacons report was never put to Revenues witnesses of fact, and it was never suggested to them that their evidence was contradicted by the written statement of evidence suggested by Revenues expert evidence.
The High Court Judgment
201. Once this is understood, the error in the assumption underlying UCIIs case that the trial Judge failed to resolve a conflict within Revenues own evidence is in my view clear. Aside from it not being evident to me what that conflict actually was, what is important - and this is the third point of significance - is that the trial Judge made unequivocal findings as to what methodology was actually adopted. He found that the depreciation tables were based on the Car Sales Guide, small motor ads, communications with dealers and the results of appeals. He said that the evidence did not disclose a significant deviation from the values contained in the Car Sales Guide. In the conclusion to his judgment he repeated this : the evidence was that reliance on sources including the Car Sales Guide provided the OMSP. That corresponded with the explanation provided by Revenue in the course of its evidence.
202. Thus, Murphy J. accepted the fact as contended for by Revenue that the depreciation tables were based upon the Car Sales Guide, small motor ads and, indeed, the result of appeals (at p. 200), to which he later added communications with dealers (at p. 201). He recorded the following finding at the conclusion of the judgment (at p. 239):
The evidence was that reliance on sources, including the Car Sales Guide, provided the open market selling price.
203. That is why - as I explain in the next section - he directed his entire analysis of the rationality of Revenues method not to the effect of the depreciation tables in any particular case, but to (a) the correlation of Revenue OMSPs with the values in the Car Sales Guide and (b) the question of whether the Car Sales Guide itself afforded a reasonable point of reference for this purpose.
X REASONABLENESS
The High Court judgment
204. The court held that the legal parameter within which this opinion of Revenue had to be formed was framed by the requirement that [t]here must be consistency in, and a basis for, the valuation (at p. 195). Referring to the decision in State (Lynch) v. Cooney , Murphy J. said that the test to be applied by the Court was limited in reviewing the opinion of the Commissioners, that it was bona fide held, factually sustainable and not unreasonable (at p. 202). The latter required a basis of support or a justification or the exercise of rational powers [and] consistency (at p. 223).
205. The Court expressed the view that there was no evidence that the opinion formed by Revenue was not bona fide held (at p. 200). It also concluded that it was factually sustainable (invoking, again, the facility for appeal):
The Court is satisfied, from the evidence, that the determination of the open market selling price was based on a system that was factually sustainable even if there were anomalies. There was always the right of appeal in relation to those anomalous valuations. There was no cost involved and no unreasonable delay. Moreover there was a high success rate for the relatively small proportions of valuations appealed. The revised OMSPs then became precedent.
206. At the same time the trial judge acknowledged grounds for criticism in relation to inconsistencies of valuation and in the preparation and production of depreciation tables (at p. 200). However, Murphy J. was clearly conscious of the administrative difficulties in producing a database that ensured consistency for all importers in every Vehicle Registration Office (at p. 200). In that regard, he observed in the course of his judgment that the main criteria for the price paid for a car was that of supply and demand. He said [t]he system obviously cannot cater for this, but it can establish a reasonable OMSP for a vehicle using criteria that are acceptable, consistent and fair (at p. 167).
207. Further, Murphy J. concluded, given the evidence before him, that the determination of the valuation was not unreasonable in the sense applicable to judicial review cases. He followed that finding by saying this:
Data relating to age, mileage and condition was inputted by Revenue officials and, while the intention of the system was to harmonise valuations for like cars, it did have some differences that were not entirely explained to the satisfaction of the court. The court accepts the evidence of the plaintiffs experts that the depreciation scales did not reflect the market. However the court acknowledges that the plaintiff itself, given the volume of imports of three to five year old Japanese cars, had the ability to and did, on occasions, challenge by way of appeal such anomalies and was in a better position to do so using its own database than an individual or a one-off importer.
208. Thus, and referring to the decision in Brennan v. The Attorney General [1984] ILRM 355, Murphy J. said that the system operated by Revenue could not be compared to the eccentric and ludicrous system of land valuation condemned in that case.
209. Noting at para. 33 of the judgment in Commission v. Greece C-74/2006 [2006] ECR I-06585, Murphy J. said that the basis of calculation of the taxable value by reference to a guide indicating the average prices of second hand vehicles in the national market or to a list of average current prices used as reference in the sector, was allowed. He recorded that evidence had been given by Revenue of use not merely of the Car Sales Guide, but also of reliance on newspaper advertisements and liaison with dealers in the trade (at p. 84).
210. Having decided that the depreciation tables were based upon the Car Sales Guide, small motor ads and, indeed, the result of appeals (at p. 200), to which he later added communications with dealers (at p. 201), the Car Sales Guide, the Court said, was a reasonable reference that was used in a widespread way in the industry and corresponded to the declared prices of new cars discounted in respect of used cars (at p. 204) and was a matter to which Revenue was entitled to make reference (at pp. 205, 219). He said that the Court (at p. 205):
Was not satisfied that the evidence considered by the court showed a significant deviation from the values contained in the Car Sales Guide.
211. Later in his judgment Murphy J. stressed that evidence of broad correlation with the Car Sales Guide and with OMSP as defined in the Act was not controverted by any statistical analysis (at p. 210) noting as significant Mr. Dowlings evidence that OMSPs were close to the Car Sales Guide prices which were not market prices (at p. 219). He observed the plusses and negatives of the use of this Guide in the formulation by Revenue of the OMSP (at p. 224):
The Car Sales Guide was partly relied on. It had the advantage of being available to traders and the disadvantage of reflecting values that the main dealers attributed to used cars. There was no evidence given to the Court as to the research underlying the data.
212. He noted as follows of Mr. Dowlings expert evidence (at p. 224):
DKM tested prices achieved at car auctions (closer to wholesale rather than retail prices), which DKM said were less that [sic] 60% of Revenue OMSP. Prices available from dealers and individual sellers were 86% lower. Prices in the Car Sales Guide were about 2% below the valuers generated by the Revenue model.
However, wholesales [sic.] prices are not retail arms length prices which, with dealers margin, would be higher. Significantly prices in the Car Sales Guide were close to OMSP.
213. In his conclusions, he said (at p. 234):
The reliance on the annual Car Sales Guide is an appropriate reference for the assessment of current used car sale prices, even if it reflects main dealers prices. It is comprehensive and used widely as an indication of used car prices over a wide range of models and ages of cars. Prices in the guide were 2% below the values generated by the Revenue Model.
214. While Murphy J. was of the view that the depreciation schedules were the cause of some confusion, equally clearly he did not believe that that confusion affected the rationality of the method used by Revenue to determine the OMSP. This was because he accepted that in fact the baseline for the determination of the OMSP used by Revenue was the Car Sales Guide combined with the other sources identified by the Court. This is evident from the above comments, but most clearly perhaps the following finding at the conclusion of the judgment (at p. 239):
The court notes that these plenary proceedings are not proceedings by way of judicial review which, in any event, would require the decision to be, inter alia, irrational or ultra vires. The court is not satisfied that the opinion was irrational or ultra vires. The evidence was that reliance on sources, including the Car Sales Guide, provided the open market selling price.
UCIIs case
215. Here, it is necessary to disentangle some separate albeit closely related complaints presented across both the domestic law challenge, and the claim insofar as based upon European law. UCII begins in its written submissions by pointing to the comments made in Revenues internal documentation to the effect that the Commissioners were knowingly failing to correctly value the vehicles and that some aspects of the system were questionable and would not survive a public challenge. It says that in the light of these comments the trial Judges finding that the opinion of Revenue was bona fide held is unsustainable. The Commissioners, UCII says, were aware of the problems inherent in the system and frustrated those seeking access to the detail of how the system operated in order to ward off any challenge.
216. Separately, UCII placed in connection with this part of its case, very considerable reliance on the evidence of Mr. Yarrow. That evidence, UCII says, was to the effect that the depreciation tables had the effect of overestimating the value of a used vehicle by as much as 40%, and typically 30%. Thus, UCII says, the conclusion of the trial Judge that the defendants system did not systematically overvalue the OMSP of second-hand vehicles is inherently contradictory, unsatisfactory and unsustainable.
217. These contentions were elaborated upon and supplemented by Ms. Barrington SC in the course of her oral submission. Emphasising the evidence of Mr. Yarrow (which, it was stressed, was not contested by Revenue) Ms. Barrington explained that the reason for the gap alleged by Mr. Yarrow to exist between the value of a used vehicle and the value determined by Revenue based on its tables, was that the Car Sales Guide was based on prices obtained by main dealers only. Those dealers cost more to deal with, because they provided warranties or something else of value in return for the price paid. She noted evidence to the effect that prices obtained at auction were less than 60% of the Revenue estimate of value. Prices available for dealers and individual sellers on websites were approximately 86% of the Revenue value. Thus, it was contended, if Revenue was not required to take account of the price actually fetched for a vehicle and used instead an administrative scheme to determine value, that had to take them to an end point that reflected the market.
218. Ms. Barrington also contended that the application of the depreciation schedules operated arbitrarily. She relied in this regard on the evidence of Mr. Dowling of DKM. He said that if, as Revenue claimed, the price is determined by a combination of the original OMSP, and the allocated depreciation schedules are continuously compared with actual market prices, there would be no need for the model. Continuously updated price data would form the basis for valuing the used car retail price. It was, Mr. Dowling said, precisely because continuously updated data was not available that Revenue relied on a model to generate notional retail value. The arbitrary allocation of a particular model to one of 25 depreciation schedules combined with the arbitrary nature of the schedule made the determination of used car values for VRT itself unreliable. Reflecting the evidence of Mr. Yarrow, Mr. Dowling said that the depreciation schedules used in Ireland had no international equivalent, the international experience showing much higher depreciation values in earlier years.
219. UCII from there emphasised the following statement by the trial Judge:
Data relating to age, mileage and condition was inputted by Revenue Officials and, while the intention of the system was to harmonise valuations for like cars, it did have some differences that were not entirely explained to the satisfaction of the court. The court accepts the evidence of the plaintiffs experts that the depreciation scales did not reflect the market.
(Emphasis added.)
220. This, counsel contended, could only be a reference to Mr. Yarrows evidence. From there, the conclusion of the trial Judge that the Car Sales Guide was a reasonable reference and that the evidence did not show a significant deviation from the Car Sales Guide was noted (and in respect of the latter, accepted). However, it was contended, there was a fundamental disconnect in the reasoning of the trial Judge: on the one hand he accepted that Revenue was entitled to rely upon the Car Sales Guide, while at the same time finding that the end point did not result in market prices.
221. In the course of the hearing further complaints were made, directed more to the granularity of the process. It was said to be wrong that Revenue used the present day price of the car as new as the starting point for its exercise in depreciation. It was noted that it followed from the report of the Working Group that schedules had not been updated as of the date of its report and that the monthly adjustments and adjustments for mileage did not reflect Revenues own view of how these matters should feature in the determination of the OMSP.
The correct legal analysis
222. Had this case been concerned with a challenge to the application of a particular methodology to the case of a specific vehicle or vehicles and had such a challenge been permitted notwithstanding the availability of an appeal, the steps in the legal analysis of the claim would have been straightforward. The valuation method applied to the particular vehicle would have been identified and the outcome of that valuation would have been known. Had the application of the depreciation schedules resulted in a value that differed from the value produced by the Car Sales Guide and/or other trade sources, it might have been said that the method did not produce its intended result and was to that extent deficient. If the application of the method produced a result that was aligned with the value produced by the Car Sales Guide, UCII might have said that this was ultra vires because that value could not reasonably have been believed to be the price the vehicle might reasonably have been expected to fetch on the open market having regard to the fact that it reflected only the prices expected by main dealers. UCII could have called evidence of what the vehicle in question and other similar vehicles had actually achieved on the market at the relevant time, in order to substantiate its arguments. If the price as determined by Revenue was significantly higher than that revealed by any identified source relied upon by UCII, UCII might have said that the decision was disproportionate.
223. Necessarily, all of these issues would have arisen against the background of an actual valuation fixed by Revenue and a proposed valuation contended by the taxpayer to be the correct one. They would have also allowed an examination of Revenues methodology at a clear and fixed point in time. To mount his claim, the taxpayer would have to identify the features of Revenues valuation at that time which rendered it unreasonable, and the reason why the price he suggested was the proper one. In these various situations, the Court would been answering a specific question - was the particular decision as to the OMSP unreasonable having regard to these various considerations?
224. This was not the question presented by the relief sought by UCII in this case. Instead, the question was whether (to quote again the Statement of Claim) the manner in which the tax is operated and administered was unreasonable. That was, as I have explained earlier, an abstract question, and it was a question that could only be answered on an abstract level. The question before the Court was not whether the methodology was such that in some individual case an unreasonable decision could in theory arise from its application. That could only be resolved in such an individual case. Of course, it could be that the evidence would point to a system which could lead to an accumulation of individual instances in which unreasonable decisions would unavoidably be reached. That could operate so as to condemn the system as a whole. Whether that was so would depend on how fundamental the alleged errors in the design of the system were, whether they were errors that infected the valuation of all vehicles, or only some, and if only some, what proportion of valuations would be so affected. While prepared to accept for the purposes of argument that in theory UCII might succeed in its claim if it were in a position to identify a basis on which a substantial proportion of Revenues valuations would be unreasonable, the most usual circumstance in which a claim pitched at this level could result in the grant of any relief is where it is proven that there is an illegality which is so fundamental that it strikes to root of all valuations conducted by Revenue. Clearly, success in such a claim would depend on the identification by UCII (which it must be re-stated bore the burden of proof on the domestic law challenge) of specific grounds of alleged irrationality in the operative methodology at specific points in time. The ultimate fate of contentions of this kind was thus dependant on the evidence adduced by UCII and, in relation to some of its evidence-based contentions, the reliability and extent of the sampling exercises that underpinned them.
225. Thus, in summary, adding the foregoing to the examination of the applicable rationality principle I have outlined earlier, the question before the Court in this case was not whether the system that was put in place by Revenue presented anomalies, nor whether it might happen that in individual cases its application could result in an OMSP that bore no rational relationship to the price a vehicle might be expected to fetch on the market in an arms length sale in the State. It was whether no reasonable authority taking all relevant considerations into account could have reasonably concluded that the methodology would generally produce a reasonable estimate of what individual vehicles might be expected to fetch in a sale by retail. That inquiry was properly directed to the fundamental features of the methodology affecting all, or possibly a very substantial proportion of, the vehicles to which that system was applied.
Refining the grounds of challenge and the depreciation tables
226. I do not believe the contention that this Court should interfere with the trial Judges conclusion that Revenues methodology was operated by it bona fide to be well placed. While the Court did not hear evidence from the author of the documents in which concerns were raised about aspects of the methodology and disclosure of the information used to operate it (a point to which I return later) Murphy J. did hear evidence from those directly involved in the implementation of the VRT system from its inception until the date of trial. Those witnesses had the documents prepared within Revenue in 1997 put to them, and they disagreed with the proposition that they disclosed flaws in the system (Mr. Campion described what was proposed as enhancements). The trial Judge was fully entitled having heard from those actually involved in the valuation process to find on the basis of that evidence that Revenue acted bona fide in the devising and operation of the scheme.
227. Similarly, I am of the view that the suggestion that Revenues method was unlawful because for a period it applied incorrect margins to allowances for mileage or to allow for monthly variations, or because for a time it failed to update the information contained in those schedules, or because of the manner in which or time over which it depreciated extras, or the manner in which the criteria relating to condition (which did not appear to affect UCII at all, given that the vehicles imported by it were in good condition) were applied, fall to be viewed not as a systemic illegality but as factors which could, in some specific cases, have produced an overvaluation. It follows from my earlier observations that this is not in itself sufficient to generally render the system devised by Revenue irrational. As I have explained above, before this could arise, UCII would have to establish that the errors were so fundamental in nature and significant in effect that they applied either to all valuations, or to such a substantial number of valuations, as to render the methodology systemically flawed and unlawful. These complaints do not fall into that category.
228. Once these elements of this part of UCIIs claim are stripped from the case, the essential claim advanced was that the manner in which the tax was operated and administered was irrational as it yielded market valuations that deviated from the actual prices achieved over the market as a whole. In resolving that issue, the finding of the trial judge that the methodology was intended to and did produce an OMSP for each vehicle valued that corresponded with the value for a car of that make, model and vintage as established by the Car Sales Guide and other market sources, is key. I do not see any basis on which either conclusion can be challenged. In particular, UCIIs own evidence was that the use of the online calculator produced by Revenue resulted in valuations that differed by 2% - within the margin of error - from those provided for in the Car Sales Guide. This was the evidence of Mr. Dowling, and it reflected the evidence of Mr. Kenneally, an economist also called by UCII (the OMSPs track Car Sales Guide prices fairly closely ) (Day 20 p.65 Q.185). A great deal follows from that.
229. For a start, it meant that complaints that the depreciation tables failed to reflect international norms or that they failed to reflect depreciation trends, or indeed that a correct starting point was not used in respect of the OMSP of the used vehicle when new, necessarily fall away. What was critical was the product of the system, and the evidence adduced by UCII itself established that the end result of the methodology employed by Revenue was the production of the valuations they were intended to produce, and that those valuations were derived from an identifiable and objectively verifiable source.
230. In that regard I think it important that UCIIs expert evidence in relation to the depreciation tables did not take account of hybridisation : this was presented in Revenues evidence as a very important development, the consequence of which was to enable the tables to be applied to fit the data, and to allow (as Mr. Campion explained in the course of his evidence) the valuation officer to start with the answer (Day 25 p.7-8 Q. 13). Viewed thus, UCIIs evidence as to the asserted deviation from international standards reflected in the depreciation tables (while certainly relevant pre-hybridisation) was not germane to a system in which the tables did not have to be and (the evidence was) were not rigidly applied, being instead mixed and matched. Instead of working to fourteen or twenty five tables, Revenue was operating what was described in evidence as an infinite number of tables. Not only did UCII not address the consequence of this in its expert evidence, but Mr. Yarrow - the principal witness called by UCII to address this question - accepted in cross examination that he was not even aware of the fact that as a result of appeals or as a result of observing prices in the market the depreciation table applicable to a given make, model or year of car were changed by Revenue (Day 17 p.60 Q. 287). He does not appear to have addressed the use of hybrid tables at all. I hasten to add that Mr. Yarrow is not necessarily to be criticised for not being aware of these matters. I return to the opacity of Revenues methodology later.
Reliance upon the Car Sales Guide
231. That presented the issue of whether recourse to the Car Sales Guide was itself unreasonable. UCIIs case was that the Car Sales Guide reflected transactions on the part of main dealers, and they in turn were said to comprise only 20% of the market. Other components of the market - sales by auction or direct sales - were not taken into account. When they were taken into account, UCIIs case was that the OMSP determined by Revenue was significantly higher than that produced by the market as a whole. It relied upon evidence from Mr. Dowling of DKM who, based on market tests conducted by his firm, found that prices achieved at auction were less than 60% of Revenue estimates of value, while those available from dealers and individual sellers on websites were at 86% of Revenues OMSPs.
232. The evidence - and common sense - suggested that motor vehicles could be sold in one of a number of different ways and that the price obtained on such methods would vary. Once Revenue determined to select an identified type of retail sale to ground its market opinion of the price a vehicle was likely to fetch in accordance with the statutory estimate it was required to make, inevitably it could be said that the source was unreliable because it reached a conclusion that differed from the other possible methods by which such a sale could be effected.
233. However, the reasonableness of a particular source thus chosen could not be simply a product of the number of vehicles sold according to a particular method in any specific year. It depended at least to some extent on the ability to obtain reliable information about prices in the market, and on the degree to which that information was itself comprehensive. It also depended on whether the particular source was viewed by the market itself as a reliable reference point for the determination of likely price. All of this has to be appraised in the light of the reality - confirmed by the report of Mr. Yarrow, an expert on vehicle pricing called by UCII - that [p]ublished information on used car values in Ireland is limited.
234. The findings made by Murphy J. in respect of this aspect of UCIIs claim were clear. He found that the Car Sales Guide was a reasonable reference and was used in a widespread way in the industry and corresponded to the declared prices of new cars discounted in respect of used cars (at p. 204). It has not been suggested that in and of itself this conclusion was wrong, or against the weight of the evidence. It was supported by the oral testimony, with Dr. Bacon describing it as as representative as you can get (Day 27 p. 92 Q. 248). In fact UCIIs own evidence did not controvert this: Mr. Yarrow accepted that the report provided a meaningful benchmark for the market (Day 17 p. 54 Q. 263). He described it as the most commonly used guide to prices in Ireland (id. p. 57 Q.274). The evidence was that UCII itself relied upon the Car Sales Guide in its dealings with Revenue as a reference where it sought to prevail on Revenue to alter OMSPs it had determined or estimated and, indeed, in the course of Mr. ODowlings cross-examination at least one instance was identified in which the price obtained by UCII for a vehicle coincided with that in the Car Sales Guide. I cannot see how the use of a reference point of that kind could in itself be said to be unreasonable at the systemic level with which this challenge was concerned. That is all the more so given, as was accepted by counsel for UCII in the course of the trial, there was no statistical evidence as to the proportion of cars bought through main dealers and those bought through other media (Day 27 p.91 Q. 248).
235. Moreover, the evidence of Revenue and findings of the Court were clear in this respect also - the Car Sales Guide was certainly the primary source used, but it was not the only source to which recourse was had. Taking this as being the case (and the finding to that effect is nowhere challenged) the fact that that the end product of Revenues research was, more or less, the prices recorded in the Car Sales Guide is consistent with the conclusion that those other sources themselves point to the Guide as a reliable benchmark of the market.
236. It follows that for the purposes of the rationality analysis with which the Court is concerned as a matter of domestic law, UCII has not established that the use of the Car Sales Guide and other material to which Revenue had regard in formulating its valuation of used cars in itself renders the manner in which the tax is operated and administered unlawful. It is hard to see how the use of a Guide consistently described by the experts for both parties as a standard market reference point and used throughout the market as the Car Sales Guide was, was in itself unreasonable as a matter of law. In order to succeed in its claim that it was, UCII would have had to establish that recourse to the Guide and other information consulted by Revenue had resulted in a systematic and significant overvaluation of motor vehicles.
237. Yet the trial Judge also rejected precisely that claim. The terms in which he expressed that conclusion are important and bear repetition. They were made in the specific context of his discussion of the decision of the CJEU in Commission v. Greece (to which I return in detail further below) in which the CJEU held that for the purposes of determining whether there had been compliance with Article 110 TFEU it was necessary for the Member State to ensure that the values used for the purposes of determining motor registration tax were very close to their actual value. The Court decided in that case that for that purpose the Greek authorities were entitled to use the wholesale price as of the date a vehicle was put into circulation as their starting point, and in which it also said that Member States could refer to a guide indicating the average prices of second-hand vehicles in the national market or to a list of average current prices used as a reference in the sector. Having recorded this, Murphy J. said (at p. 94 to 95) :
The defendant relied on the Car Sales Guide. The evidence given also related to reliance on newspaper advertisements and conferring with dealers in the trade.
The court is satisfied that the Car Sales Guide is the reference guide used in the sector for average current prices of new vehicles.
It would appear from Mr. Yarrows report that in the Hellenic Republic, as in Ireland, the prices of new vehicle are determined by the manufacturers and distributors. That was also acknowledged in the Commission v. Greece judgment which did not find that such determination discriminated against used car dealers.
The Court also takes into account that an appeal procedure was available and, indeed, was used by the plaintiff to its advantage at times.
The court is not satisfied that the Irish system did systematically overvalue the OMSP of second hand vehicles. Whether, on occasions, it was not very close to their actual value is a matter which could have been attested [sic.] by way of appeal which, in turn, would have formed part of the basis of the valuation system.
(Emphasis added.)
Systematic overvaluation
238. If this finding is to stand, UCII cannot succeed in its claim that by reason of the values attributed by Revenue to used vehicles it acted unreasonably in the manner in which it operated the system of determining OMSP. Thus, UCII contends that this aspect of the decision is inherently contradictory, unsatisfactory and unsustainable.
239. The claim that the finding is contradictory was grounded in UCIIs claim that the Judge accepted evidence led on its behalf to the effect that the depreciation scales did not accurately reflect the market as they overvalued second hand vehicles for VRT purposes. In oral argument it was contended by UCII that this was a reference to the evidence of Mr. Yarrow, and that the Court was therefore accepting that evidence. Given that he was accepting that evidence (the argument went) he could not have rationally also found that Revenues methodology did not systematically overvalue used cars.
240. It is in my view clear that throughout that part of the judgment in which this statement of the trial Judge appeared, he was drawing a clear and firm distinction between systematic overvaluation, and anomalies. Having regard to the nature of UCIIs case, this was an entirely proper distinction to draw. There was no doubt but that he was finding that there was no systematic overvaluation and it is equally obvious that he was deciding that there were anomalies. This is evident not merely from the passage I have quoted earlier from page 95, but from the paragraphs immediately preceding that section of the judgment on which UCII heavily relies in this regard. The first paragraph of the relevant section begins with this statement (at p. 200):
The court is satisfied, from the evidence, that the determination of the open market selling price was based on a system that was factually sustainable even if there were anomalies
(Emphasis added.)
241. The next following paragraph (at p. 200) is that relied upon by UCII :
Data relating to age, mileage and condition was inputted by Revenue officials and, while the intention of the system was to harmonise valuations for like cars, it did have some differences that were not entirely explained to the satisfaction of the court. The court accepts the evidence of the plaintiffs experts that the depreciation scales did not reflect the market. However the court acknowledges that the plaintiff itself, given the volume of imports of three to five year old Japanese cars, had the ability to and did, on occasions, challenge, by way of appeal such anomalies and was in a better position to do so using its own database than an individual or a one-off importer.
(Emphasis added.)
242. So the judgment, when the relevant passages are read together and in context, is in my view quite clear. The depreciation scales did not reflect the market, because there were anomalies. Those anomalies did not amount to a systematic overvaluation. At no point did the court say or remotely suggest that it was accepting Mr. Yarrows evidence that there had typically been a 30% deviation between Revenues values and those of the market.
243. That presents the question of whether, having regard to the evidence the trial Judge was entitled to reach this conclusion. Here UCII in its submissions in this Court focussed on the evidence of two witnesses - Mr. Yarrow and Mr. Dowling, also referring in passing to that of Mr. Kenneally. This evidence must be viewed in the light of what it had to establish were UCII to succeed in this aspect of its case. To repeat what I have said earlier, it was not sufficient for them to prove that for some vehicles the application of Revenues methodology was liable to result in an overvaluation. They had to prove that the methodology was so flawed and unreasonable that its effect was systemic - affecting either all vehicles valued by Revenue or (perhaps) such a significant proportion of the cars to which it was applied as to vitiate the entire methodology (as opposed to its application in individual cases).
244. Mr. Yarrow was an expert in the analysis of European vehicle pricing. In his evidence, he expressed the view that typically the value of a vehicle drops by 30% of its original list price within the first twelve months of its life (emphasis added). Comparing the depreciation scales used at a number of different points (1984, 1993, 2003 and 2006) and noting the differences in the scales in question, he concluded that there was a significant gap between the Car Sales Guide prices and the prices yielded by the tables. That gap, he said, was most significant for vehicles up to 12 months old. He explained how based on a review of 13 worked examples of valuations supplied for specific vehicles, for 12 of these the valuations provided were above the range of likely Car Sales Guide values. In the course of his report, he said that it is widely accepted in Ireland that Car Sales Guide values are up to 20% higher than reality. His conclusion, based upon his assessment of the tables and the examples he had, was that Revenue over-estimated the value of a used vehicle by as much as 40% but more typically 30%.
245. A number of problems arise from UCIIs reliance on this report in this appeal. Mr. Yarrows opinion as to this alleged deviation (which it must be noted was not constant from year to year) had two components. The first was the view that there was a variation between the Car Sales Guide and the product of the depreciation schedules. In the example given by him in evidence - a three year old car - this accounted for an overvalue of 23%. However, his opinion in that regard departed from the position adopted by other witnesses and accepted by the trial Judge that as a matter of fact Revenues valuations were proximate to those in the Guide. One such witness was Mr. Dowling of DKM whose report, as I have previously noted, concluded that the Car Sales Guide and Revenue OMSPs closely aligned. When asked in cross examination if he had any comment on that, Mr. Yarrows response was none whatsoever (Day 17 p. 54 Q. 267). Mr. Yarrows conclusion in this regard, it must be observed, was based on sampling of far fewer cars (forty) than the other experts called by UCII. He looked at one edition of the Car Sales Guide and one depreciation table out of twenty five. He was also unaware that Revenue used hybrid tables.
246. The second assumption was that the Car Sales Guide did not itself reflect market values. For the purposes of his example, he assumed that the Car Sales Guide values were up to 20% higher than the market. This took the alleged difference in the case of a three year old vehicle to 33% (representing the basis for the conclusion that in Revenues valuation was 30% out). However, he never gave, and presumably was not in a position to give, evidence as to the relationship between the prices in the Car Sales Guide and the market generally in Ireland. In his report, what he said was that such a deviation could present itself depending on the seller and the condition of the vehicle and that it was widely accepted that that gap existed (Mr. Yarrows report paras. 23 and 37). In evidence (Day 17 p. 23 Q. 110) he said that this was his view although he expressed no specific basis for that.
247. Bearing in mind that the evidence was that the tables were adjusted from time to time, that for most of the period with which the case was concerned Revenue could operate hybrid tables, and that Revenues depreciation tables did not take account of mileage or condition (these being addressed separately) for Mr. Yarrows evidence to prove anything he had to establish that the end product - the prices in the Car Sales Guide - were themselves overvalued not in individual cases but to such an extent as to properly ground UCIIs systemic challenge. While Mr. Yarrows evidence of an overvaluation was certainly based on the assumption that the Guide overstated values, he provided no evidence of this.
248. Mr. Dowlings report was heavily relied upon in oral submissions to this Court as demonstrating that the Car Sales Guide prices were themselves not reflective of the entire market. He said in his report that probably less than 20% of used car transactions in Ireland were conducted through main dealers. His report also stated in its summary that prices achieved at auction were less than 60% of the Revenue estimate of the OMSP (although those prices excluded auctioneers commission which could be as much as 15% of the price) and prices from dealers around 86%. This was the source of Mr. Yarrows evidence on the point.
249. But these percentages are clearly variable and were the product of, as Mr. Dowling himself characterised it, a small study (Day 5 p. 16 Q. 12). I do not see how auction sales prices afforded reliable evidence of the price obtainable on the retail market. It was his evidence that auction prices were closer to wholesale prices and, indeed, he accepted in cross examination that auctions are not necessarily retail transactions at all wholesalers also purchase at auction (Day 16 p. 45 Q. 120). In point of fact as his report makes clear - Mr. Dowlings invocation of auction prices was clearly heavily influenced by his belief that the price should not be fixed by reference to the retail market in the first place (which is what the legislation requires) but instead by the price paid by the dealer.
250. The block graph contained in DKMs report (at p. 68) records the figures for internet sales (referred to in evidence as Car Buyers Guide prices) and auction sales at slightly above 90% and 55% respectively of the OMSP as determined by Revenue. The latter do not compare at all with UCIIs own recorded sales experience, which posited a ratio of its sale prices and Revenues OMSPs of close to or above 90% for its sales of two, three, four and five year old cars (88%, 91%, 95% and 98 % respectively) and higher sale prices than the Revenue OMSP for years six to ten. The figure, it should be said, for one year old cars was lower at 79%. The reliance by UCII on internet sales, if anything makes the point about choosing a comparator source: UCII itself was promoting its cars and aiming them at the top of the market (Day 16 p. 49 Q.140).
251. In summary, having regard to the small studies informing this evidence and the period of time they purported to address, remembering that Revenues own OMSPs were 2% above that in the Car Sales Guide, and bearing in mind that even if averaging was the appropriate way of accommodating different prices charged by different sources in the market, the average of the Car Sales Guide price and those obtained from other parts of the retail market (excluding auctions) would not have necessarily revealed a very substantial and systemically relevant difference from the prices ultimately fixed upon by Revenue. They certainly would not have resulted overall in a very large deviation from the prices obtained by UCII. That being so it is hard to my mind to see how the deviations in price from, in particular, information from dealers and internet sales could on any version be said to ground the claim that the trial Judge erred in finding that UCII had not met the test of irrationality in respect of a review of the kind in issue in this case in establishing (as the trial Judge put it) systemic over valuation rather than anomaly in price.
Were the Courts conclusions as to the reasonableness of Revenues methodology sufficiently reasoned?
252. The conclusion of the trial Judge that UCII had not established that Revenues system was irrational, albeit subject to anomaly, was thus based on three fixed points - that vehicles were valued by reference to the information in the Car Sales Guide and other trade information, that the Car Sales Guide was a reasonable reference point for this purpose, and that UCII had not established that its application systematically overvalued the vehicles. True it is that Murphy J. did not explain in the course of his evidence why he accepted neither the evidence of Mr. Yarrow that in thus valuing the vehicles Revenue did effect an overvaluation, nor that of Mr. Dowling that the Car Sales Guide did not reflect the market. However, it is clear that he did not accept either of these conclusions.
253. The test to be applied by this Court in considering whether the trial Judge sufficiently reasoned his conclusions is clear and is succinctly summarised by MacMenamin J. in Leopardstown Club Ltd. v. Templeville Developments Ltd. [2017] IESC 50, [2017] 3 IR 707 at para. 109 to 110. Save where there is clear non-engagement with essential parts of the evidence an appeal court may not reverse the decision of a trial judge by adverting to other evidence capable of being portrayed as inconsistent with the trial judges primary findings of fact. Non-engagement with the evidence must mean that there was something truly glaring which the trial judge simply did not deal with or advert to and where what was omitted went to the core or essential validity of his findings.
254. The application of these general principles in any given case requires the striking of a balance between two competing considerations. On the one hand the trial Judge must give reasons for his decision, and those reasons must be such as to allow the parties - and an appellate Court - to understand the basis on which the trial Judge reached his or her decision (O Driscoll v Hurley [2015] IECA 158 at para. 19). At the same time, it is in the interests of neither the parties nor the resources of the Court that the judge get lost in the undergrowth of the evidence or the arguments made (McDonald v. Conroy [2020] IECA 239 at para. 34).
255. Reasons, of course, can be endlessly interrogated and a plausible argument constructed that each component part of any train of logic itself requires to be further explained. This is particularly so in a case such as the present which sprawled across manifold issues of fact and law, documentary and witness evidence - and time. Key to the striking of the balance between these considerations is the same essential determinant that applies to reasons challenges in the case of administrative agencies or quasi-judicial bodies: do the reasons allow the parties to understand why they won or lost on a specific issue and in the case of the losing party to judge whether they have a basis for challenging that decision?
256. The single judgment of the Court delivered by Clarke J. in Donegal Investments Group plc v. Danbywiske [2017] IESC 14, [2019] 1 IR 150 specifically addressed itself to the application of these principles and factors to expert evidence. The case was different from the instant proceedings in two respects - the trial Judge was presented with competing evidence from experts on an issue and proceeded to adopt a conclusion which did not correspond precisely with the conclusions of either. However, noting that in some cases involving expert evidence the parties and the appeal Court will have the benefit of a trial record, transcripts of evidence, expert reports and legal submissions, Clarke J. observed (at para. 8.8):
Against that backdrop it will often be possible readily to infer why a particular finding was made even if there is no express statement in the judgment. The parties will know how the case ran. An appellate court can read the record of the case. The judgment needs to be read in the light of the case as made and defended before the trial judge.
257. The key question was helpfully framed by Clarke J. in the next following paragraph of his judgment - whether it is possible to ascertain with reasonable confidence the reasons why the Judge decided the issue as he did. There is no doubt why Murphy J. decided that the methodology adopted by Revenue was reasonable - it was based upon a source that was widely and properly used and that source did not overvalue the vehicles. It follows that he rejected the evidence adduced by UCII insofar as it posited otherwise. He was entitled to reject that evidence, because (as I have explained above) it simply did not establish a systemic overvaluation. Mr. Yarrow could not give evidence of a deviation between the Car Sales Guide prices and the market. Mr. Dowlings evidence showed that it might be possible to obtain lower prices on other platforms in the retail market. Interrogation of those sources viewed in the light of the prices UCIIs customers were prepared to pay (insofar as they were in fact representative of the retail market) did not show that there was for all or most classes of vehicles an enormous disparity in price. Each of these points was made by Revenue in cross examination, and each is a matter of analysis not of appraisal of fact or credibility. At the end of the process the question the trial Judge had to answer was not complex: had UCII established that the methodology as a whole was one that could not reasonably be believed to deliver a reasonable prediction of the price a particular vehicle would obtain on an arms-length sale in the State. The answer to that question was clearly expressed by the trial Judge, and the three building blocks that led to it plainly identified. I can see no basis on which this Court can conclude that the trial Judge erred, based upon his appraisal of the evidence, in so concluding.
XI PROPORTIONALITY
258. UCIIs written submissions propose that, in accordance with the decisions in Iarnroid Eireann v. Ireland [1996] 3 IR 321 and Daly v. Revenue Commissioners [1995] 3 IR 1, a proportionality test should be applied to assess the legitimacy of the system of assessing VRT as administered by the Revenue Commissioners. It complains that while the Judge stated in his decision that he was not satisfied that the principle of proportionality had been infringed, he never engaged with UCIIs claim. Before the High Court, the complaint based on a breach of the principle of proportionality was referenced to the claim that VRT was driven by a desire to protect the Irish motor trade from competition, and the assertion that dealers in imported used cars were treated differently from dealers in used cars sourced in the State. The former seems to have been envisaged as a challenge to the validity of the legislation. The latter was not, as I explain later, made out.
259. The contention in UCIIs written submissions in this Court that a proportionality test should have been applied to the system of assessing VRT lacked focus or definition, and the argument received little attention in the course of oral submissions. Although now usually referenced to the decision of Costello J. in Heaney and McGuinness v. Ireland [1994] 3 IR 593, the requirement that a Court in adjudicating on the constitutional validity of legislation alleged to infringe constitutional entitlements assess the proportionality between the extent of legislative incursions on constitutional rights and the objective sought to be achieved by those restrictions has featured in legal analysis from the early stages of the development of constitutional law in this jurisdiction (see Ryan v. Attorney General [1965] IR 294 at 313 per Kenny J.). In Meadows (to which the trial Judge was not referred) the application of the principle in at least some cases to the assessment of the validity of administrative action which affected constitutional rights was enabled through its absorption within the rubric of the common law test for unreasonableness. However, if proportionality is to be applied as a coherent legal principle by reference to which the legality of administrative action is to be tested it is necessary for a claimant to identify and substantiate (a) a constitutionally or ECHR protected right (or perhaps other legal entitlement) that has been impacted by the administrative action complained of, (b) a basis on which it is said that that impact on that right or entitlement arising from the action either advances no legitimate interest or goes further than necessary so to do and (c) a coherent theory by reference to which that lack of proportion between impact and objective renders the administrative action outside the scope of the statutory power pursuant to which the administrative action is taken. In each case, this must be presented on a basis that takes account both of the nature of the constitutional right or other entitlement alleged to be impaired and the over-riding mandate that it is the function of the Court to determine the legality of the impugned measure, not to assess its merits.
260. Whatever the role of such a proportionality test in respect of administrative action affecting fundamental rights such as that of the person or family rights, it is not at all clear to me what role it can meaningfully play in cases involving alleged illegality in the design of administrative schemes for the levying and collection of tax in accordance with primary legislation. This is supported by a consideration of the case law addressing the constitutional validity of legislation on the basis of property right infringement, to which I turn in the next section. It is also illustrated by the questions that would arise were such a claim to be formulated. What, exactly, makes the application of a fiscal measure disproportionate ? Is it that too much tax is required on foot of it? Too much by reference to what metric? If that metric is the price that the vehicle might reasonably be expected to obtain on the market, by reference to what criteria does the Court determine whether this is too much? If the claim is based on differential treatment when the position of the claimant is compared with that of other taxpayers, what is the applicable standard by reference to which such differences fall to be assessed? What is the difference between such an argument and a case based on ordinary common law unreasonableness?
261. It is difficult to see how these questions could be answered in the context of, or a case of this kind could ever be presented in respect of, a system of valuation, as opposed to a specific decision in a particular case. In any event no developed case along these lines having been presented to the High Court, I do not believe that the High Court can be criticised for not addressing it in any more detail that it did. The case based on proportionality as recorded in the written legal submissions was a repetition of the claim based on unreasonableness which, for the reasons to which I have referred, was ill founded.
XII THE CLAIM BASED ON BREACH OF CONSTITUTIONAL RIGHTS
Findings of the trial Judge
262. Murphy J. referred to a series of cases dealing with the constitutional validity of taxation and social welfare statutes, observing that given the complex task of administering the provisions of the Finance Acts, latitude must be allowed to Revenue in administering the legislation governing and directing the raising of taxation (at p. 218). He concluded that VRT did not differentiate unfairly against the owners of vehicles, did not constitute an unjust attack on property rights and said:
The court is of the view that the Revenue Commissioners valuation has been determined in a reasonable manner which is subject to challenge by way of an appeal procedure by which that valuation can be tested.
263. In the concluding part of his judgment, Murphy J. said that a corporation such as the plaintiff is not entitled to property rights (at p. 236). In that latter respect, clearly, the trial Judge erred (Iarnroid Eireann v. Ireland).
The claim
264. The invocation by UCII of its constitutional right to property was related to three separate propositions. The first - the issue I have just addressed - is the issue of proportionality and the suggestion that its rights had been disproportionately impacted by the scheme used by Revenue. The second was that the operation of the scheme represented an unjust attack on UCIIs property rights, for which it was entitled to claim damages. I will address this in the context of the claim for damages.
265. The third was to the effect that if the scheme operated by Revenue was in fact lawful and intra vires, the legislation represented an unconstitutional violation of UCIIs property rights. In UCIIs written submissions, the claim that there was a disproportionate or discriminatory restriction on property rights was presented as follows:
(a) In circumstances, it was said, where the High Court accepted that the system of valuation suffered from anomalies and that there were defects in the system - including that the depreciation scales understated depreciation so as to overvalue the vehicles for VRT purposes, it was inconsistent to conclude that there was no unfair discrimination against UCII.
(b) Further, the trial Judge also erred, it was said, in failing to adjudicate on UCIIs claim that the restriction on property rights as importers of second-hand vehicles was unjust in the light of the lesser restriction imposed on dealers of second hand vehicles who sourced those vehicles in the State.
(c) Finally, it was contended that the respondents acted in breach of UCIIs constitutional rights in enabling a system which introduced a centralised model of assessment which does not record differences in the value of individual vehicles presented for valuation for VRT purposes.
266. While it may be understandable, if only as a counsel of prudence, that UCII included in its administrative law challenge a claim that the underlying statutory provision was contrary to the Constitution, in the context in which it was presented here, such a claim was always bound to be superfluous. The Act conferred on Revenue a broadly worded discretion in connection with its power to determine the OMSP. If any one of the three propositions advanced by UCII in support of its claim that the provision was unconstitutional were correct it could not have been reasonable for Revenue to exercise that discretionary power as it did. It can never be reasonable to achieve an outcome through the exercise of a statutory discretion which, were that outcome mandated by the legislation conferring that discretion, would have rendered the legislation invalid. The legislation would have to be read so as to exclude that outcome from the scope of the permissible exercise of the discretion.
267. Each of these contentions was thus properly viewed not as a basis for challenging the validity of the legislation, but of contending that the exercise of the power to fix the OMSP was ultra vires. The first and third, in fact, have been considered by me above in that context. A system which presents anomalies is not for that reason alone unreasonable (although in a particular case an applicant who is for whatever reason excused from the obligation to exhaust the facility for an appeal of that decision may be in a position to establish that that decision was ultra vires). For exactly the same reason the fact that the system may fail to reflect differences in the value of individual vehicles is not unreasonable - subject to the same proviso. Had a discrimination between the position of importers of used vehicles and dealers who acquire the same vehicle domestically been established (and as I explain later is has not been) this would not have been an unreasonable differentiation. Importers and those trading only in domestic vehicles are clearly in different positions, the former necessarily trading subject to the entitlement of the State to tax imports and to do so differentially. Outside the ambit of European law (to which I will return) there is no obligation to treat imports and domestic products as equivalent.
Relevant principles of constitutional law
268. None of this is changed by presenting the same argument as one of constitutional violation. If a party chooses to express their claim in the language of constitutional law, they must accept the restraints and limitations imposed by the attendant jurisprudence. Those principles are clear and firmly established. As OHanlon J. put it in Madigan v. Attorney General [1986] ILRM 136, at p. 151 a party challenging the validity of a taxation statute faces a very uphill battle. As he also explained, tax laws are in a category of their own. Very considerable latitude must be allowed to the State in the enormously complex task of organising and directing the financial affairs of the State (id.). Where fiscal legislation has been condemned as invalid having regard to the property rights protected by the Constitution, the Courts have so held not because there is tax imposed in a particular context, or because the tax is too high, but because the tax itself has operated in a manner that is arbitrary or irrational (see, in particular Blake and ors.v. Attorney General [1982] IR 117) or has through manifest unfairness produced a hardship which was unnecessary having regard to the objective of the legislation (see Daly v. Revenue Commissioners [1995] 3 IR 1). Even then, the bar is set high. It is not insignificant that the various challenges to the validity of fiscal legislation as contravening property rights that have succeeded have involved the very particular position of legislation that was outdated and in which over time anomaly had become the rule rather than the exception (Brennan and Blake being cases in point) or in which a scheme originally conceived as transitional and to deal with a particular exigency had become permanent (as was the case with the collection mechanisms for withholding taxes in issue in Daly).
269. In particular, the fact that A believes himself to be in a similar to position to B and is yet taxed differently from B will not in itself justify the conclusion that the imposition of that liability unconstitutionally interferes with property interests. Taxes are imposed on categories of persons and the very necessity to establish categories means that there will be persons who can plausibly claim that in relation to their particular positions their situation is closer to those of a person taxed at a lower rate. Anomalies are an inherent aspect of any process of taxation. Proving that a system of taxation operates in some circumstances anomalously has never in itself successfully grounded a claim of constitutional invalidity. To sustain a claim that a scheme of taxation put in place by a statute of this kind is invalid, it is usually necessary to establish that it is unreasonable, arbitrary, or constitutes hostile discrimination against particular persons or classes (Madigan at p. 154). The arguments advanced by UCII do not meet this threshold.
XIII TRANSPARENCY, LEGAL CERTAINTY, STATUTORY DISCRETIONS AND FIXED POLICIES
The Argument
270. In their written submissions, UCII contend that the trial Judge erred in failing to consider whether the system of administration of VRT complied with requirements of legal certainty in domestic law. The essential complaint suggested there, and re-iterated in the oral submissions to this Court, was that Revenue was determining the value of vehicles by means of a secret system, and that it deliberately refused to disclose the methodology by which that value was fixed because it knew that to do so would expose its flaws. All of this was directed to the claim that the failure to make the methodology publicly known was unlawful, the predicate of which contention was that Revenue was under a legal obligation to make its methodology known, at least to those who asked for it.
The High Court judgment
271. Murphy J. rejected the contention advanced by UCII that the system for the administration of VRT was undertaken in a manner that was secret and was for that reason unlawful: the administration of VRT was reasonably transparent (at p. 214). Throughout his judgment, various considerations relevant to that conclusion are identified. Thus, in rejecting the claim that UCII did not understand the system by which VRT was calculated by Revenue, Murphy J. noted the involvement of Mr. Herlihy and the fact that Mr. ODowling had said that he was conversant with the system (at p. 174). He stressed that there was no evidence that when Mr. OHerlihy was acting for UCII any complaint was made of uncertainty or of an inability to ascertain the VRT due (at p. 182). He said that the Court was not satisfied that the plaintiff was purchasing blind or that the estimates could not be relied upon (at p. 175), or that goalposts were moving all the time (at p. 176). He repeatedly stressed his view that Revenue staff had been both open and flexible in relation to the determination of open market value by way of the depreciation tables (at p. 200).
272. Murphy J. quoted with apparent approval the evidence of Dr. Bannister (at p. 214):
that the Revenue did what it reasonably could to make the information available to the public but it was not under an obligation to do so any more than an inspector of taxes is to explain publicly how the valuation of property or land is assessed.
273. At the same time, the Court accepted that there were difficulties with the use of the schedules, concluding that these were remedied by the facility for appeal (at p. 210):
The court accepts that the Revenues use of depreciation schedules was unclear even when they were made available to the plaintiff. Cars which moved from one scheme to another without any published or disclosed notice cannot, however, amount to a representation or breach of legitimate expectation. The absence or paucity of appeals to test the basis and consistency of OMSP for imported cars is an evidential defect.
274. At one point, Murphy J. recorded confusion as to the origin of the depreciation tables - [t]he court was not clear on what basis these scales were introduced (at p. 224). He returned to this at the conclusion of his judgment, stating that [t]he depreciation of the price declared by way of an increasing number of schedules has caused confusion and misunderstanding (at p. 234). He said (at p. 238):
The court acknowledges that the issue of depreciation tables, which appears to be a hangover from the MVED system was a source of some confusion, even from the point of view of the Revenue officials themselves.
275. However, the court said that while there may have been problems that did not of itself render unlawful the administration of VRT. Insofar as UCII had relied upon Mr. OLearys internal memorandum of 14 July 1997, Murphy J. recorded the evidence of Mr. Campion to the effect that it was Mr. OLearys habit to persistently press his superiors for additional resources, and proceeded to state that the Court had no doubt but that more resources within the Vehicle Registration Office would have assisted Revenue in establishing a comprehensive data base of relevant motor vehicles (at p. 206). The Court also underlined the developments in information technology over the twenty-year period between the introduction of VRT and the hearing of the case.
276. Murphy J. observed the absence of any complaint as to the transparency of the system in the complaints made to the European Commission in 1993, the Court concluding from this and from Mr. ODowlings comment that it was after the increases in 1994 that the real system kicked in, that it was the increases in the rate that was of concern (at p. 179). Murphy J. noted that as Revenue developed the database more information was provided to the public by way of estimates, calculators and CD ROMS.
Relevant legal principles
277. UCII referred to the decision of Costello J. in McCann v. Minister for Education [1997] 2 ILRM 1 in support of its claim insofar as based on legal certainty. There, the Court was concerned with a challenge to the validity of two decisions of the respondent Minister refusing applications by the applicant for incremental credits for her previous service as a secondary school teacher and refusing to recognise for that purpose service by her as a teacher in the United Kingdom, the grant of which applications would have increased the salary payable to her. The issue arose because the applicant had been teaching while holding a certificate of education from a teacher training college abroad, and in a context where the scheme operated by the Minister for the award of those credits depended on teacher having a recognised qualification. The qualification held by the applicant was not so recognised.
278. The two impugned decisions were made within a framework of non-statutory schemes and circulars which had been amended from time to time. While the Court refused to grant the applicant the relief she claimed, finding that the administrative schemes and decisions made by the Minister were neither irrational nor arbitrary, and that they were not in breach of EU law, Costello P. proceeded to make some observations in respect of administrative schemes the operation of which affected the rights of those to whom they were addressed. His comments bear repetition in full (at p. 15):
The law should be certain and it should be readily accessible. The same applies to non-statutory administrative measures. In the case of primary and secondary education hundreds of millions of pounds are administered annually by means of a large number of administrative measures whose existence is known only to a handful of officials and specialists, which are not readily available to the public and whose effect is uncertain and often ambiguous. If administrative ministerial rules and regulations were dated, if they were identified by reference to the sub-head in the book of estimates to which they relate, if amendments before the same reference and were dated by reference to the minister order which made them, if a register was kept of the original measure and amendments to it, if the original measure and amendments were regularly consolidated and meanwhile made available in loose leaf form to members of the public, those would be one way of obviating the danger of injustice which is inherent in the present highly informal procedures.
279. While these observations did not in terms posit a legal obligation to put in place the measures suggested, it is clear from the opening sentence of the passage I have quoted that some form of access to non-statutory administrative measures of the kind in issue was an entitlement of those whose rights were affected by them. Indeed, the observations were preceded by the suggestion that they were made by way of suggestion to the legislature and executive to avoid injustice. That being so, it is notable (and perhaps explicable by reference to the intervention of Freedom of Information legislation) that it appears to have been some twenty years before that issue next raised its head in the case law as it did in Luximon v. Minister for Justice, Equality and Law Reform [2015] IEHC 227, [2016] IECA 382, PO v. Minister for Justice, Equality and Law Reform [2015] IESC 64 and DE (An infant) v. Minister for Justice, Equality and Law Reform [2018] IESC 16, [2018] 3 IR 326, each of which presented the question of whether, and if so when, executive agencies were obliged to publish the criteria by reference to which statutory discretions were exercised by them. In Luximon and DE, the arguments advanced by the applicants to the effect that the respondents were under, and failed to comply with, such an obligation were based on the decision of the United Kingdom Supreme Court in R (Lumba) v. Secretary of State for the Home Department [2011] UKSC 11, [2012] 1 AC 1.
280. Lumba arose from an extraordinary practice that had developed within the respondent Home Office. Essentially, a scheme was published by the executive identifying the basis on which its statutory discretion to release convicted foreign national prisoners whose sentences had been served, from detention pending their deportation. That scheme presumed that these prisoners would be released unless their continued detention could be justified. However, in fact, the Home Office operated a secret and unpublished policy that they would be detained unless their release could be justified. The applicants sought, and obtained, declaratory relief that their detention pursuant to that policy was unlawful because it was on foot of a blanket policy the details of which had not been published, and in respect of which, accordingly, the prisoners had been deprived of the right to make effective representations. They failed in their action for damages for false imprisonment, because the evidence was that had the discretion been correctly exercised, each would in any event have been detained.
281. It is the basis for the declaratory relief that is relevant here. Lord Dyson explained the position as follows (at para. 34 to35):
The rule of law calls for a transparent statement by the executive of the circumstances in which the broad statutory criteria will be exercised. Just as arrest and surveillance powers need to be transparently identified through codes of practice and immigration powers need to be transparently identified through the immigration rules, so too the immigration detention powers need to be transparently identified through formulated policy statements.
The individual has a basic public law right to have his or her case considered under whatever policy the executive sees fit to adopt provided that the adopted policy is a lawful exercise of the discretion conferred by the statute There is a correlative right to know what that currently existing policy is
282. Putting to one side (but in no sense to minimise) Lord Dysons general appeal to the rule of law, two specific and related considerations drove the conclusion that the applicants had the right to know what the policy was - so that they could make relevant representations in relation to it (at para. 35) and so that they could challenge an adverse decision when made (at para. 36). While these features presented themselves in a particularly significant context for the applicants - the proper application of the policy could mark the difference between their detention and release - and while the overall background was strikingly irregular (they had been misled as to what the true policy was) the reasoning was in no sense limited to these circumstances. He suggested that the principle applied in respect of any scheme which imposes penalties or other detriments or which confers benefits (at para. 36), endorsing the broad principle articulated by Stanley Burton J. in R (Salih) v. Secretary of State for the Home Department [2003] EWHC 2273 at para. 52:
it is in general inconsistent with the constitutional imperative that statute law be made known for the government to withhold information about its policy relating to the exercise of a power conferred by statute
283. It is hard to see that the theory underlying these principles is other than applicable in its entirety within our constitutional framework. It was well expressed in, if not pre-empted by, the remarks of Costello P. in McCann to which I have earlier referred and, to take one of many examples, is demonstrated by the comments of Laffoy J. giving the judgment of the Court in Cussens v. Brosnan [2015] IESC 48. There the appellants sought to amend their notice of appeal in a case concerned with the operation of the European law doctrine of abuse of rights in respect of a scheme intended to minimise their VAT liabilities on the disposal of certain properties. The amendment was intended to agitate the effect of derogations from the Sixth VAT Directive sought by and granted to the State. They explained their failure to raise earlier the issue they sought to include in their appeal by reference to the refusal of the respondents to provide the derogations when requested in the course of the proceedings. Laffoy J. said this (at para. 35):
it is difficult to understand how a State body could consider that it had the entitlement to withhold the derogation, which, in reality, is a legal instrument which regulates the charging of VAT and can be regarded as part of the law of the land, from a taxpayers advisor It should have been furnished when it was sought in the context of this appeal.
284. Of course, a policy governing the exercise of a statutory discretion for the time being and a legal instrument are not the same thing. For a start, the executive is not universally required to have a policy governing the implementation of every statutory discretion vested in it, and indeed in certain circumstances it will act unlawfully by tying the exercise of such a discretion to a pre-determined policy. That was the issue in Luximon, in PO and in DE. In Luximon, Barr J. decided, based on Lumba, that the Minister acted unlawfully in failing to publish the policy applied by him in deciding whether to exercise his statutory discretion to allow certain timed out foreign national students permission to reside in the State. He said (at para. 199) that the Minister failed to comply with the principles of natural and constitutional justice, and basic fairness of procedures, in failing to publish the criteria that would be taken into account by the Minister, or an immigration officer acting on her behalf, when determining an application from a timed out non-EEA student for a change of immigration permission to a stamp 4 permission pursuant to s. 4(7) of the Immigration Act 2004. The appeal against this part of Barr J.s judgment was allowed in this Court, Finlay Geoghegan J. deciding (at para. 62) that the Ministers obligation was to determine each application on its own facts and merits having regard to general guidelines which had in fact been made available.
285. While the issue was not pursued in the appeal against the decision of this Court in Luximon, the application and limits - of the principle articulated by Lord Dyson was referenced by Clarke CJ in DE. There, the Court was concerned with an appeal against the refusal of the High Court to grant leave to seek judicial review of a decision of the respondent refusing to revoke a deportation order having regard inter alia to a serious medical condition from which the applicant suffered. One of the grounds urged by the applicant was the claim that the respondent operated a de facto policy to grant residence in the State to persons subject to deportation orders who had been in the State for a minimum of five years. He claimed that he was unable to make submissions in respect of that policy, as he was unaware of its operation.
286. Clarke CJ held that on the facts of the case before him, he was not satisfied that any practice which existed was not sufficiently known to the applicants advisors to allow them to make representations. He furthermore held that there were not substantial grounds for the contention that the Minister was obliged to adopt a policy or practice in relation to the exercise of the very broad residual discretion which the Minister had to allow persons remain in the State for humanitarian reasons. However, while at the same time noting the extreme circumstances in Lumba, while stating that he was prepared to assume that the principle applied in that case was recognised in the law in this jurisdiction and while stressing that decision maker exercising a statutory power could not fetter his discretion, he said the following (at para. 31):
On the other hand, there can be little doubt that, where a policy or criteria have been developed for the exercise of a particular power, transparency would require that persons who may be engaged by the power concerned should be able to familiarise themselves with the relevant guidance or criteria so as to be able to make a case for the exercise of the power concerned in a manner favourable to their interests. In circumstances where there actually is guidance or criteria and where the law would require that a person who might be affected by the exercise of the power concerned would have a right to make representations, it would greatly devalue the benefit of that entitlement if the person concerned had no means of finding out the guidance or criteria by reference to which the decision affecting their interests might be made.
287. Later in his judgment Clarke CJ repeated that while he did not think it arguable that the Minister in that case in exercising a broad residual discretion was obliged to narrow down the field by defining the criteria by reference to which that discretion had to be exercised, what was arguable was that he must make details available of any criteria or guidance which are actually adopted (para. 36).
288. In this case, the issue which presents itself in this regard is in one sense narrow. The applicant requested Revenue to provide it with details of the basis on which the OMSP of imported used cars was determined. Revenue refused to do so. It adopted this position in a context in which it had developed a sophisticated system for gathering information as to the value of such vehicles, reflected that system in depreciation schedules developed for the purpose of enabling a value to be ascribed to vehicles of a specified make, model and age, operated that system as a fixed methodology for the determination of the OMSP and yet refused to provide that information on request, while also operating a defined framework for determining depreciation for extras, taking account of the condition of vehicles, and allowing for the kilometrage of cars it was valuing.
289. Unlike DE, this was not a case in which the applicant was aware of how that system was formulated or applied to any specific vehicle, and in particular it did not know which schedule was applicable to which vehicle in which circumstances. Nor did the taxpayer know the manner in which the price was adjusted to take account of the condition of the vehicle (evidence of Mr. Butler Day 23 p. 101 Q. 299). In the unlikely event that Revenue had not developed any system for the valuation of such vehicles and had simply left it to the discretion of individual officers to value a vehicle on the ground then the comments of Clarke CJ in DE might well suggest that it had no obligation to advise persons required to pay VRT how the OMSP had been determined. That, however, is not what happened. Revenue did have a policy, that policy came in the form of its valuation methodology as explained, in the first instance in the staff guide. The only conclusion that could have been rationally drawn from the evidence before the Court was that from 1993 until 2010 Revenue deliberately decided - for whatever reason - that it would not share with those who had to pay VRT the basis on which it proceeded to determine the OMSP of their vehicles (and therefore the amount of VRT they were required to pay).
290. It refused to make this information available, and it refused to do so in a context where it was the undisputed evidence of Mr. Kenny that unlike all other taxes levied in the State, the rules relating to VRT - including the methodologies for the determination of taxable value and the resulting calculation of tax due - were not published in statutory form and/or supported by Revenue Notices and/or guides or explanatory booklets explaining in detail how the tax operates. Indeed, when counsel for Revenue was asked in the course of the hearing in this Court whether an importer of used cars would have any way of knowing what methodology Revenue had applied to calculate the VRT of a vehicle, her response was that she did not believe that they would. The evidence before the High Court left no doubt but that this was the only response that she could have tendered to that question.
Revenues arguments
291. In the course of its written and oral submissions, Revenue advanced four arguments as to why it was not legally required to make the information sought by UCII available to it. None of these, I should say, were related to any legal authorities, none involved an explanation of why the comments of Costello P. in McCann were either wrong, or inapplicable, and none engaged at all with the evidence of Mr. Kenny.
292. First, it contends that UCIIs case was that it should know what the VRT was fifty days before it became due and that the risk that VRT might increase after an estimate was no different from the risk of currency fluctuations. However, this confuses an entitlement to know the exact tax payable on a transaction before it is entered into, with the right of a taxpayer to understand the basis on which that tax has been calculated. Revenue is correct in submitting that there is no right to the first of these, but its contention in respect of the second goes no further than the observation that valuation is in exact and imperfect. This is true. However, it misses the point. Revenue had determined to discharge that function of reaching the inexact and imperfect by reference to a specific and highly developed methodology. The inexactitude of the objective does not detract from the entitlement of a person affected by its implementation to know how it is sought to be attained.
293. Second, Revenue says that there is and was no legislative obligation on it to explain every step it takes in forming its reasonable opinion regarding the OMSP for any vehicle and that there was no legal basis for such a proposition. It underlines the provision by it of estimates, notes the trial Judges reference to the evidence of Mr. ODowling that he had a reasonable idea of the VRT liability on the vehicles he imported and restates that if UCII was unhappy with any aspect of the valuation, it had the right to appeal.
294. However, it is not suggested that a person affected by the exercise of a statutory discretion must be told of every step in the formation of an opinion. The obligation suggested in the case law is operative in respect of a policy where that policy has been formulated and is relied upon by those charged with the operational exercise of that discretion : what must be published is that which a person who is affected by the operation of the policy needs to know in order to make informed and meaningful representations to the decision maker before a decision is made (Lumba at para. 38). The existence of an appeal, it might be added, far from representing a basis on which Revenue is released from the obligation to make its policy known attests to the necessity for disclosure in the first place. A party who does not know the basis on which a decision is made could, depending on the facts of a particular transaction, be constrained when contending that it was wrongly reached.
295. Third, Revenue says that the methodology used by it was evident from clause 5.3 of the Traders Guide. I have quoted this earlier. This made no reference to the depreciation tables, did not advise of the sources of information used to construct them, did not explain the operation of hybrid tables (which of course came later), did not elucidate how mileage and condition were factored into the process, or of how the new OMSPs from which the valuation process commenced, were identified. Yet Revenue had a clear policy detailing how all of these considerations were factored in to the valuation process. UCII did not know what that policy was, and indeed did not find out the critical features of Revenues methodology until the delivery of Revenues evidence in this case - sixteen years after it had been initiated. Even then there was evident (and, I believe, entirely understandable) confusion on the part of its witnesses, including its expert witnesses, throughout the trial as to how the valuation method actually operated.
296. The fourth contention is a variant of the first. Taxpayers knew, it was said, that Revenue used the values in the Car Sales Guide and other sources to reach a value for each vehicle to be taxed. The depreciation schedules, it was said, did not function to depreciate the car, they operated instead to identify what the retail value of the vehicle was as prescribed by the Car Sales Guide. This was not, accordingly, actually a case in which Revenue operated an administrative scheme. Instead, it did what paragraph 5.3 of the Traders Guide said it would do.
297. I cannot accept this argument. For a start, it was repeatedly said by Revenues witnesses that they did not only use the Car Sales Guide to value vehicles being taxed. Again and again in the evidence Revenues witnesses were at pains to stress that other sources were consulted. This only made sense if the information in the Car Sales Guide was being supplemented. More fundamentally, UCII was told that Revenue made some use of depreciation schedules and indeed it was central to its case and evidence that this is precisely what it did. While it may have been true to say that the depreciation schedules were not used to value the cars, it was to the depreciation schedules that Revenue looked to identify the values, even if in theory these were intended to do no more than replicate the information that would have been derived from consulting the Car Sales Guide and other sources. And finally, and I think most importantly, it necessarily followed from the fact that it was to the depreciation tables that Revenue had recourse in identifying the market value (as opposed to a database of what those values actually were) that those tables might not in every case have accurately tracked the Car Sales Guide price, or might not at a point in time have been updated appropriately or in a sufficiently timely way, that those paying tax under this regime had a right to know what the methodology employed actually was. This, I should observe, is aside from the fact that this argument fails to answer in any way the right of taxpayers to know how extras were depreciated, how condition was accounted for, how kilometrage was brought into the valuation, how the starting point for the OMSP was identified, and how issues such as monthly variances were at identified points in time taken into account.
Conclusion
298. It is inherent in the rule of law in general, and the mandate of legal certainty in particular, that where a public body is conferred with a statutory discretion or power and formulates detailed internal criteria on the basis of which that discretion or power will be exercised, those affected by the exercise of the discretion are entitled to be informed of those criteria. As explained by Green J. in R(Justice for Health Ltd.) v. Secretary of State for Health [2016] EWHC 2338, at para. 141, that obligation serves a number of functions. A law or policy should be sufficiently clear to enable those affected by it to regulate their conduct. It should also be sufficiently clear so as to obviate the risk that a public authority can act in an arbitrary way. Publication thus both enables the affected person to identify when the policy is being properly applied to them and incentivises (a) those who formulate it to ensure from the outset that it is lawful and (b) those who operate it to comply with its terms in a consistent way. Clear notice of a policy or decision is required so that the individual knows the criteria that are being applied and is able to both make meaningful representations to the decision maker before the decision is taken and subsequently to challenge an adverse decision. Publication also allows persons affected by the exercise of those discretions and powers to challenge the legality of the policy itself.
299. The importance of an obligation of this kind is well demonstrated by the facts of and evidence in this case. The evidence left no doubt but that at the time of the report of the Working Group in 1997 Revenue had a policy of not making the most basic of information available to taxpayers regarding the operation of the VRT system. Absent evidence from Mr. Lynch, there were strong grounds for concluding that this was motivated at least in part by a concern that perceived frailties in the system would be exposed through disclosure. Evidence from other persons that Mr. Lynch might have been simply exaggerating matters to obtain funding are no more than their opinion as to his motivation - admissible from witnesses of fact to prove nothing. While matters undoubtedly improved following the report of the Working Group and coming into effect of the Freedom of Information Act 1997, the entirely understandable confusion on the part of UCII and its expert witnesses at trial as to the role of the depreciation schedules in the entire process was most unfortunate and in no sense was attributable to any default on their part. A review of the documentation obtained by them on discovery would inevitably have led them to believe the system operated as they explained it in their reports. The statement by the trial Judge that the depreciation of the price declared by way of an increasing number of schedules has caused confusion and misunderstanding was if anything an understatement. As I have previously noted, it does not appear that the facility for hybridisation (operative since November 1999) was made known to the public at all - even though Revenue itself viewed this facility as very very important.
300. Neither UCII, its experts or its lawyers should have found themselves in a position where the first time they discovered how the system of determining VRT actually operated was in witness statements delivered by Revenue at the start of a trial. Taxpayers should not have to litigate to understand how their liabilities to Revenue are determined in circumstances where Revenue itself operates firm but undisclosed criteria by reference to which it arrives at that calculation.
301. It is trite to say that the nature and extent of the obligation to disclose a fixed and detailed policy of the kind in issue in this case will vary according to each statutory power and the specific criterion by reference to which the relevant authority has determined to exercise it. As the judgment in DE makes clear, the obligation does not arise in respect of broad, residual discretions. This case was not concerned with such a discretion, or at least not with a discretion that was treated by Revenue as such. In this case, it is unnecessary to go further than to say that UCII should have been informed of the following upon request:
(i) The sources from which Revenue determined the OMSP of a vehicle as new;
(ii) The source of the information used to devise the depreciation schedules;
(iii) The depreciation scale (or combination of such scales) applied to each model, make and age of vehicle legitimately requested by UCII;
(iv) The methodology applied to the valuation of extras on a motor vehicle;
(v) The criterion used by Revenue to reflect condition and kilometrage of a vehicle in that process of valuation.
302. I have deliberately omitted from this the actual OMSP of the vehicle as new. It was suggested in the course of evidence that this was commercially confidential information, and if that was so there would be a rational basis for excluding it from the information to which the public was entitled. There does not appear to have been any concrete evidence of this, and there was no decision by the Court on the issue. The evidence indicated, indeed, that it was possible on the basis of the information presently available to the public to reverse engineer the OMSP of the vehicle as new (Day 29 p.11 Q. 23), suggesting that it is not in fact commercially sensitive. It is unnecessary to decide the question here.
303. While developments in technology are such that (iii) are now readily available, I have seen nothing to indicate that from the introduction of VRT until the use of such technology that these scales could not have been provided to those who asked for them. The evidence of Dr. Murphy tendered on behalf of UCII that information as to how OMSP and depreciation data was determined was easily available and could have been supplied by the Revenue does not appear to have been seriously disputed (see the evidence of Mr. Fogarty Day 29 p.33-34 Q.s 115 and 116).
304. While it was submitted by Revenue that UCII had the information it required to understand how VRT was determined - and heavy reliance was placed by counsel on para. 5.3 of the Traders Guide - as Ms. Goodman SC for Revenue fairly and properly acknowledged in the course of her well reasoned submissions, what UCII did not know was how Revenue would arrive at its opinion. It knew that depreciation schedules were used, and it knew the relevant statistical codes. It did not - until shortly before the trial - have the depreciation schedules nor did it know which schedules were applied to which statistical codes.
305. By failing to provide UCII with the information I have identified when requested in this case, Revenue acted unlawfully. UCII is entitled to declaratory relief to that effect. Given that there is no specific challenge to the valuation of any particular vehicle in this case, the issue of any further relief consequent upon such a declaration does not arise. The fact that Revenue failed to make the information sought by UCII available to it does not in and of itself invalidate the levying of the tax. In the absence of a particular case in which it is established that UCIIs ability to make representations or to appeal because of the absence of such information was actually impaired in such a way as to vitiate the tax demanded by Revenue it cannot be said that there is any causal connection between the failure of Revenue to comply with this obligation and the tax actually imposed on a particular vehicle. It would not be possible to reach such a decision in the abstract and without reference to a specific vehicle and specific valuation, and particular evidence as to how the ability to obtain a correct valuation was impaired by the failure to provide the relevant information to UCII. Finally, this aspect of my finding is limited to the time period up to 2010. From then on, Mr. Kennys evidence makes clear, UCII had sufficient information to understand the methodology by which the tax was determined.
XIV WAS UCII ENTITLED TO MAINTAIN A CLAIM UNDER EU LAW ?
The High Court judgment
306. In the course of his oral submissions in the High Court, counsel for Revenue submitted that the fact that UCIIs business was limited to the importation of cars from outside the EU, precluded it from presenting a case based upon breaches of EU law. He submitted that if the plaintiffs business was importing cars from Japan it could not make an argument as if it was importing cars from Germany. In addressing that contention, Murphy J. stressed that in general the registration tax fell outside the scope of EU law, and that those authorities which held that there is a breach of Article 110 TFEU where the amount of tax levied on an imported second-hand vehicle exceeds the residual tax incorporated in the value of a similar second-hand vehicle already registered on national territory, were concerned with tax insofar as it burdened products originating from other member states (at pp. 80, 186). He re-iterated that for trade with non-member countries the Treaty did not include any rule similar to that laid down in Article 110 (Simba SpA v. Ministerio Delle Finanze Cases C-228/90, C-234/90 and C-353/90 [1992] ECR I-03713; OTO v. Ministeria Delle Finanze C-130/92 [1994] ECR I-03281 ). He re-emphasised that domestic taxes are not regulated by the Communities (at p.183).
307. He made it clear in that context (referring to the decisions in Madigan v. Attorney General, Norris v. Attorney General [1984] 1 IR 36, and A. v. Governor of Arbour Hill Prison [2006] 4 IR 88) that UCII was confined to arguments based upon circumstances that affected it directly, that arguments as to jus tertii could not be made and that the plaintiffs claim was based on importing cars from Japan (at p. 181). The judgment of the High Court appears to operate on the basis that while UCII did import forty-four vehicles from the UK, that because the Court had been given no satisfactory analysis as to the overvaluation of OMSP in relation to those imports, UCII was precluded from asserting a breach of EU law at all (at p. 192). At the same time, it appears to have been the view of the Court that because the particulars of loss given by UCII related solely to imports from Japan, the case fell to be determined solely by reference to Irish domestic law. It was, he said, an important evidential requirement that damages be proven to have arisen under domestic law (at p. 208).
308. Murphy J. expressed some concern with regard to the extent of importation of used cars from EU member states, noting differing accounts in the evidence as to the exact number of vehicles imported from the United Kingdom (at p. 177). At one point in his judgment (at p. 190), he suggested that evidence from Mr. ODowling to the effect that the cost of Japanese three to five-year old cars was cheaper and accordingly profitable, indicated that the plaintiffs decision was to import Japanese used cars. Furthermore, he said that the evidence in relation to the intra community imports of 44 cars imported from Northern Ireland did not establish that the values as a general rule were not very close to their actual value (at p.190). He quoted Mr. Riordans evidence on the subject (at p. 190):
While Mr. ODowling did not give specific details, Mr. Fintan Riordan said that between the 23rd September 1993 and the 7th February 1996, the plaintiff imported 44 cars from Northern Ireland. He said that valuations assigned to cars coming from Northern Ireland were high, but not as high as those placed on Japanese cars. He said that if they had not been as high the plaintiff would have increased the numbers of imports from the UK. No satisfactory comparison of OMSPs was given.
309. Murphy J. also, on a number of occasions, noted that the Commission had not instituted proceedings against Ireland based on complaints made by UCII, but that it had done so against other member states (at pp. 88, 172, 188 to189). He also attached significance to his conclusion that many aspects of what was reported to the Commission in fact concerned the business of Sports Car Centre Ltd.
Analysis
310. There was and could be no dispute between the parties that insofar as imports from outside the EU are concerned, principles of EU law are not applicable. However, in my view UCII is correct when it says that the trial Judge erred when he decided that this meant that it was precluded by jus tertii from asserting that the system operated by Revenue was in breach of EU law. As Mr. Sreenan SC submitted in the course of the hearing, UCII was not seeking to invoke the rights of any third party. It was relying upon its own rights - as a business established for the purposes of importing vehicles into the State from the EU, and as an entity which actually did so.
311. The trial Judges conclusion in this regard was affected by a conflation of two distinct issues - whether UCII had established losses which followed from a breach of EU law, and whether it had an entitlement to assert a breach of EU law in the first place. Undoubtedly UCII could not maintain a claim in damages consequent upon a breach of EU law in respect of losses sustained by an overcharging of VRT on Japanese imports, and this was properly accepted by Mr. Sreenan SC in the course of his oral submissions.
312. However that does not mean that because it could not establish financial loss it was precluded from seeking declaratory relief to the effect that the system that was operated by Revenue was in breach of EU law. These two propositions are, conceptually, entirely distinct. In concluding that UCII could not assert EU law rights because it had not established loss as a consequence of the operation of the relevant provisions on EU exports, the trial Judge overlooked that difference. The question of whether UCII could assert the EU law rights in question was a matter to be determined solely by reference to whether it had standing so to do.
313. I do not believe it can be seriously contended that UCII lacked that standing. The starting point in the analysis of that issue is the fact that while it is, in principle, a matter for national law to determine an individuals standing and legal interest in bringing proceedings, EU law requires that national rules do not undermine the right to effective judicial protection and that the application of national rules cannot render virtually impossible the exercise of the rights conferred by EU law (Vervolen v. SVA Cases C-87/90, C-88/90 and C-89/90 [1991] ECR I-3757 at para. 24). In Irish law the litmus test for standing is whether the person who asserts it has inter alia a bona fide concern or interest in the provisions sought to be impugned (Digital Rights Ireland Ltd. v. Minister for Communications [2010] IEHC 221, [2010] 3 IR 251 at para. 48). It has most recently been expressed simply in terms of whether the impugned measure affects the plaintiff as a matter of fact (Mohan v. Ireland [2019] IESC 18, [2019] 2 ILRM 1 at para. 16.)
314. The undisputed evidence was that UCII was incorporated with a view to importing cars from the EU. The fact that it engaged in that business was not disputed either. Indeed, in oral submissions Ms. Goodman SC properly accepted that UCII had locus standii to raise an issue in relation to the 44 cars under European law. This meant, and only could mean, that UCII was entitled to challenge the methodology by which and legislation pursuant to which VRT was calculated in respect of those vehicles and accordingly to challenge by reference to EU law the legality of that system and legislation generally. Therefore, the fact that it could not establish pecuniary loss as a consequence of the operation of those rules was neither here nor there. Its business in general, and its importation of vehicles from Northern Ireland in particular, was subject to a regime of taxation which it contended was contrary to EU law. I can see no basis on which it could be said it was precluded by an absence of proven financial loss as a result of that alleged breach, from advancing that case.
XV THE CLAIM IN EU LAW
The principles
315. There is no rule of EU law which prevents Member States from introducing a general system of internal taxes charged in accordance with objective criteria on a particular category of goods. However, Article 110 TFEU conditions how this can be done in one important respect. It provides as follows:
No Member State shall impose, directly or indirectly, on the products of other Member States any internal taxation of any kind in excess of that imposed directly or indirectly on similar domestic products.
Furthermore no Member State shall impose on the products of other Member States any internal taxation of such a nature as to afford indirect protection to other products.
316. The objective underlying this provision is to preclude internal taxation of such a nature as to afford indirect protection to domestic products (Bergandi v. Directeur general des impots Case 252/85 [1988] ECR -01343 at para. 23). Operating as it does in a context in which fiscal measures alleged to constitute barriers to imports must be assessed only by reference to Article 110, it supplements the provisions on the abolition of customs duties and charges having equivalent effect with a view to ensuring free movement of goods between the Member States in normal conditions of competition. It achieves this objective by the elimination of all forms of protection which may result from the application of internal taxation that discriminates against products from other Member States (Commission v. France Case 168/78 [1980] ECR 347; Commission v. Denmark Case 171/78 [1980] ECR 447. Thus, Article 110 is infringed where the taxation on the imported product and that on the similar domestic product are calculated in a different manner on the basis of different criteria which lead, if only in certain cases, to higher taxation being imposed on the imported product (Commission v. Luxembourg Case C-152/89 [1991] ECR I-03141 at para. 20).
317. It is therefore unsurprising that the provision is applied stringently with a view to securing that objective. Article 110 must guarantee the complete neutrality of internal taxation as regards competition between domestic products and imported products. It must be interpreted widely. It covers all taxation procedures which, directly or indirectly, conflict with the principle of equality of treatment of domestic products and imported products. The prohibition is operative whenever a fiscal levy is likely to discourage imports of goods originating in other Member States to the benefit of domestic production (Bergandi at para. 24). At least in circumstances where the taxation system at issue lacks transparency, the burden in establishing that there is no possibility of such discrimination between domestic and imported products is imposed on the Member State: the system of taxation at issue can be considered to be compatible with the first paragraph of Article 95 only if it is proved to be so arranged as to exclude any possibility of the imported product being taxed more heavily (Commission v. Luxembourg at paras. 21 and 25).
318. A substantial body of case law has developed in which the CJEU has grappled with the application of these general principles to the specific problems presented by the process of valuation of used imported cars for the purposes of domestic vehicle registration (and other) taxes. The objective is clear and follows from the foregoing: registration tax should not burden products originating from other Member States more heavily than similar national products (Nadasdi and Nemeth (Case C-290/05 and C-333/05), [2006] ECR I-10115).
319. These decisions have sought to reconcile the entitlement of the Member States to impose such tariffs with the requirement that where applied they do not create for the domestic used vehicle a competitive advantage vis a vis its imported equivalent through the imposition of higher taxation on the import. To do this, it has been necessary to develop a theory by reference to which such vehicles can be valued, and to then ensure that the Member States operate a valuation methodology which reliably implements this. At the same time, the valuation of any asset represents an exercise in prediction that can rarely produce more than an estimate. In the context of used motor vehicles, the wide range of variables that can impact on value renders both exercises difficult. The end point of the case law in seeking to reconcile these realities insofar as relevant to these proceedings - can, I believe, be summarised as follows.
320. First, the overriding principle is that the tax regime applicable to motor vehicle registration must not discriminate against imported vehicles by favouring directly or indirectly vehicles from that Member State. To thus ensure the neutrality of the tax, the value of the imported second-hand vehicle taken as the basis of taxation must reflect the value of a similar vehicle already registered on national territory (Weigel v. Finanzlandesdirektion fur Vorarlberg Case C-387/01 [2004] ECR I-04981 at para. 71). As it was put by Advocate General Sharpston in her opinion in Tatu v. Statul roman prin Ministerul Finantelor si Economiei Case C-402/09 [2011] ECR I-02711 (at para. 37), this is the overriding principle, from which all other considerations follow.
321. Second, the very nature of a first registration tax is that it is not imposed on the domestic vehicle at the same point in time as it is being imposed on the import. The tax will usually have been imposed on the domestic vehicle contemporaneous with the first sale. To ensure where establishing equivalence between the value of the domestic product and the import that like is being compared with like for the purposes of the assessment of value, the CJEU has applied a method approved in Staatssecretaris van Financien v. Schul Case 47/84 [1985] ECR 149 in the context of VAT on imports. This is based upon the construct that at any point in time the value of an asset on which VAT has been paid incorporates an element reflecting the amount of that tax. This is the residual tax. In determining the residual amount of value added tax incorporated in the value of goods for the purposes of calculating what, if anything, the State into which second hand goods are imported should take into account in determining VAT payable on those goods, the Court in Schul required that regard be had to the market value of the asset on the basis that the amount of residual tax declines in proportion to the value of the goods. As applied to motor vehicle registration duties, the residual tax was explained by Advocate General Mischo in Commission v. Denmark Case C-47/88 [1990] ECR I-04509 as being the amount of the duty paid at the time of registration of the car when new, reduced by an amount equal to the actual depreciation in the value of the car (at para. 36).
322. Third, central to the rules developed by CJEU to ensure that there is no discrimination between domestic products and imports is the requirement that the amount of tax on imported second-hand vehicles does not exceed that of the residual tax incorporated in the value of similar second-hand vehicles already registered in the national territory (Ministerio Publico Antonio Gomes Valente v. Fazenda Publica Case C-393/98 [2001] ECR I-01327 at para. 41).
323. Fourth, in order to give effect to that mandate, the national authorities must operate a methodology which renders a true value of the import for the purposes of assessing the chargeable basis for the tax. This must have regard to every factor relevant to such value, and in particular must take account of the vehicles make, model, method of propulsion, age and kilometrage, its condition, and the fact that annual depreciation in the value of motor vehicles is generally more than 5% and is not linear - particularly in the first years (Commission v. Greece Case C-375/95 [1997] ECR I-5981 at para. 22, Gomes Valente at para. 24 and Tatu v. Statul roman prin Ministerul Finantelor si Economiei Case C-402/09 at paras. 45 and 47).
324. Fifth, it is obviously open to the national authorities to operate a straightforward system of inspection of vehicles and consequent process of individualised valuation for this purpose, provided of course that it operates within these constraints (Tatu at para. 46). However, it is not necessary to do this. The Court made it clear in Gomes Valente that Member States are in principle entitled to apply fixed scales provided they take account of the vehicles age, kilometrage, general condition, method of propulsion, and make or model and provided that the value thereby produced would as a general rule be very close to their actual value (at para. 24). The burden in this regard was on the state authorities (at para. 29) and this entails an obligation on those authorities to establish the basis on which the relevant scales had been devised (at para. 30). However - reflecting the analysis I have earlier suggested of the position in domestic law - the Court is concerned to ensure that the alignment with actual value be achieved as a general rule rather than in every conceivable case.
325. Sixth, Gomes Valente posited two important conditions relevant to this case on the use of methods of valuation of this kind. Where a scale reflected a general trend of depreciation but did so in an imprecise manner it could be rendered compatible with Article 95 if the owner of the vehicle had the opportunity to challenge the application of the scale before a court (at paras. 31 and 32). However, it was a precondition to this saver that the criteria or criterion on the basis of which the scale is calculated are brought to the attention of the public (at para. 34). As explained by the Court, it appears that the appeal thus mandated has to be one in which the taxpayer had the right to adduce evidence that the fixed scale was inadequate to determine the real value of the second-hand vehicle imported by him (at para. 35).
326. Seventh, while the availability of an opportunity to challenge the application of a scale is a precondition to compliance with Article 110, it is not the case that by enabling an appeal to be determined in accordance with completely different criteria an otherwise unlawful process of initial valuation is thereby rendered compliant (Commission v. Greece Case C-74/06 at para. 40).
327. Eighth, the CJEU acknowledges throughout the decided cases the obvious reality that no system of valuation of assets of this kind can be perfect. Necessarily, it can at best be based upon the reasonable approximations inherent in any system of that type (Gomes Valente at para. 26). Thus, in ascertaining value described by the Court in Weigel v. Finanzlandesdirektion fur Vorarlberg Case C-387/01 as the price the imported vehicle would achieve on the domestic market (at para. 79) - the CJEU has had no difficulty with the operation of systems which take account of both wholesale prices, and retail prices as demonstrated by trade magazines (Gomes Valente at para. 25, Weigel at para. 74 and Commission v. Greece Case C-74/06 at para. 33).
328. A further, more general principle is, as it happens, iterated by one of the motor registration tax cases. In Gomes Valente, it was argued that because the Commission had first initiated and thereafter discontinued enforcement proceedings against Portugal this impacted on the obligation of a court of last instance of a member state to refer to the Court of Justice a question of Community law in relation to the legislation the subject of that infringement action. A question was referred to the Court of Justice as to whether that contention was well placed. In answering that question in the negative, the Court observed (at para. 18):
the Commission is not empowered to determine conclusively, by opinions formulated pursuant to Article 169 of the EC Treaty or by other statements of its attitude under that procedure, the rights and duties of a Member State, or to give its guarantees concerning the compatibility of a given line of conduct with the Treaty, and that, according to Articles 169, 170 and 171 of the EC Treaty the rights and duties of Member States may be determined and their conduct appraised only by a judgment of the Court of Justice.
329. It follows that the view taken by the Commission as evidenced by its instituting only limited enforcement action against Ireland in respect of the VRT regime (that action having related to the non-application of depreciation to cars less than three months old or with less than 3000 kilometrage) is in no sense determinative of the issues before this Court. The position of the Commission was repeatedly noticed by Revenue in cross-examination of UCIIs witnesses and in submissions and was heavily relied upon by the trial Judge in the conclusions he reached. This did not take sufficient account of the fact that the obligation of the High Court was to assess the arguments advanced by UCII on their own merits, and not by reference to what the Commission said or did not say about the matter.
330. UCII sought to apply these various principles in this case in three different ways. First, it referred to and relied upon the decision of the CJEU in relation to the transparency of the methodology used by Revenue. Second, it raised an issue around the residual value referred to in the CJEU case law which, it said, demonstrated that the basis of valuation mandated by s.133 was contrary to EU law. Third, it said more generally that Revenue had failed to demonstrate that its method of valuation complied with the requirement underlying all of these decisions to deliver a value for the purposes of VRT which approximated to the actual value of taxed vehicles in the market.
Obligation of transparency in EU law
331. The conclusions I have reached above in relation to the failure of Revenue to comply with domestic law requirements to disclose information relating to the manner in which the OMSP of used cars was calculated points strongly to the conclusion that the cognate obligation imposed by European law has also been breached. A consideration of the terms in which the CJEU has expressed, and applied, the requirement of such notice, leaves no doubt in that regard.
332. The requirement was first referenced in Gomes Valente in the context of depreciation scales: the possibility of using those scales (the Court held) is in any event subject to the condition that the criterion or criteria on the basis of which the scale is calculated are brought to the attention of the public (at para. 34). The system operated in that case was condemned inter alia because the factors on the basis of which the percentages of reduction in the tax were determined are not identified in the law in question (at para. 36). This has been repeated consistently since (Tulliasiamies and Siilin Case C-101/00 at para. 87). As it was put in Commission v. Greece Case C - 74/06 at para. 46 the criteria on which the fixed method of calculating the depreciation of vehicles is based must be made known to the public.
333. I do not see that this requirement can be limited to an obligation to disclose the basis on which the scales themselves are calculated. The obligation referred to by the CJEU stems from the more fundamental consideration that in determining whether a tax imported second-hand motor cars is discriminatory, it is necessary to have regard to the basis of assessment and the detailed rules for levying the tax (Commission v. Denmark Case C-47/88 at para. 18). It follows that the person discharging the liability is entitled to know both. Thus, in Tatu the Advocate General framed the requirement in terms that the criteria on which the method of depreciation is based must be made known to the public (at para. 42).
334. This is evident, in particular, from the decision in Commission v. Greece Case C-74/06. This presented a challenge to a registration tax using a single criterion of depreciation based on vehicle age, being a rate of 7% for vehicles between 6 and 12 months old which had covered more than 6,000 kilometres, or 14% for cars more than one year old, thereafter increasing incrementally based on age up to 80% where the vehicle was between fifteen and sixteen years old. That depreciation was applied to the wholesale price of the vehicle when put into circulation on the international market, as provided by the relevant distributors. The Greek government declined to divulge any information as to the taxable values adopted by it, citing administrative and commercial confidentiality. Furthermore, it stressed that before customs clearance of an imported vehicle the owner could request that he be informed of all factors influencing the determination of the taxable value of that vehicle.
335. The system also provided for an appeal to a committee of experts who were charged with determining the taxable value in such a way that the registration tax payable was equal to the amount of the residual tax incorporated into the value of a vehicle same type. That committee of experts could examine the vehicle and take into account its condition and, the Greek government said, could consider retail prices published in magazines in the event that the owner claimed that there was a substantial difference between those prices and the value adopted on customs clearance. A fee of 300 was payable before the appeal could be invoked.
336. The complaint made by the Commission that this system of calculating the registration tax was based on a single criterion of depreciation - the age of the vehicle - was upheld. The essential reason for this was that because the legislation did not take mileage into account in the calculation of depreciation the value of the vehicle concerned would be set at an overly high level (see paras. 40 and 43). Three additional propositions relevant to this case emerge from the judgment.
337. First, while deciding that the appeal that was available met, notwithstanding the small fee payable, the obligation to have a mechanism to challenge the valuation (at para. 58), the Court held that the fact that there was available an appeal did not affect its conclusion insofar as the deficiency identified by it in the underlying methodology was concerned. It addressed this very briefly, stating (at para. 40):
Contrary to the Hellenic Republics assertion, the fact that, in such cases of intensive use of a vehicle, it is open to the owner to challenge the application of the fixed scale is not sufficient to prevent a system of taxation from contravening Article 90 EC.
338. This might be said to follow from the fact that the conditions articulated in the case and in particular in Gomez Valente are cumulative. It would make little sense to impose obligations that the system deliver a valuation that is close to the market value and that there be an appeal mechanism, if the latter obviated the necessity for the former.
339. Second, although the factors taken into account by the Greek authorities in determining value were not generally published, the criteria for calculating depreciation and the factors applied by the complaints commission were. More importantly, the importer could obtain on request the wholesale price by reference to which the vehicle was depreciated (at para. 48), and in this way the wholesale price must be regarded as being brought to the knowledge of the public (at para. 48). For that reason, it was held that the obligation to disclose the criteria on which the fixed method of calculating depreciation values was based, was fulfilled.
340. Third, while the Court held that the Greek authorities were entitled to use as their starting point the wholesale price as of the date a vehicle was put into circulation, it also said that Member States could refer to a guide indicating the average prices of second-hand vehicles in the national market or to a list of average current prices used as a reference in the sector (at para. 33). It would appear from the Courts summary of the submissions made by the Commission that this was a reference to lists of retail prices (at para. 12).
341. The case provides, both in its similarities and differences, a pathway to the conclusion that while the overall structure of the system operated by Revenue was capable of complying with the requirements of EU law - in particular in being based on an analysis of market price - it demonstrates, if only by way of contrast, the shortcomings of the approach adopted by Revenue to disclosing the basis on which valuations were obtained. There is no obvious reason why Revenue could not - during the period with which this action is concerned - have made available (at the very least on request) the source of the values it was attributing to a vehicle on its being taxed or details of the base price and depreciation schedule applied to obtain it. By not providing that information Revenue created the same impediment to the exercise of EU law rights, as it did to domestic entitlements. It potentially prevented the taxpayer from meaningfully questioning the basis for the valuation and could have erected a hurdle to its ascertaining whether that value did in fact equate to the value of a domestic used vehicle presenting the same essential features. While the question of whether such an impediment was in fact created so as to actually preclude the exercise of EU law rights could only be determined in an individual case, in creating the potential for this to occur Revenue breached EU law.
The issue around the residual value: the arguments of the parties and findings of the High Court
342. UCII, however, also maintained a more fundamental attack on Revenues approach and, more particularly, on the governing legislation. This was focussed on the fact that under FA92 the OMSP was fixed by reference to the retail price of the imported vehicle. Thus, UCII said, this meant that the dealers margin was included in the price (and thus the chargeable value of the car). However, when a car dealer sold a used vehicle which it had sourced domestically, no VRT was payable as the car was already registered. While VRT had been paid on first registration, that was fixed by reference to an OMSP which did not include the margin of the second vendor of the car. Thus, it was contended, the tax was discriminatory as between the same used car when imported from another Member State and its equivalent when sourced in Ireland and sold second hand.
343. Central to this argument was the residual VRT to which I have referred above. This concept (described throughout the trial and hearing of this appeal as the embedded VRT) was helpfully elaborated upon by Mr. Sreenan SC, with an example.
344. If, he explained, a motor vehicle is purchased new for 10,000.00 and VRT is 20%, there is embedded in the price of that vehicle VRT of 2000.00 with the remaining 8000.00 representing the price of the car without VRT. If the vehicle is sold to a dealer five years later for 5,000.00 the embedded VRT remains at 20% of the price, and thus represents a sum of 1,000.00. Therefore, if the dealer sells the car for 6,000.00, his margin is 1,000.00 but the embedded VRT is still 1,000.00. When he charges VAT, the VAT is on the margin. However, there is no VRT on the margin.
345. However, in the case of a used imported car, UCII says, the position is different and that difference presents an unlawful discrimination. So, if the same used vehicle as in Mr. Sreenans example is imported into the State and sold by UCII for 6,000.00 VRT is charged on that sum. His competitor - the dealer selling the same domestic used car - is not paying VRT on his margin. Thus - according to this argument - the imposition of VRT on anything other than the price at which the dealer purchased the vehicle was in breach of EU law. It would follow from this argument, if it were well placed, that the Irish legislation in providing otherwise was invalid having regard to EU law.
346. Revenues position - reflected in the expert evidence of Dr. Bacon - was that the VRT will always be a constant proportion of the value of the car as it depreciates. It was Dr. Bacons evidence that the value of the car for these purposes was the retail value of the car to the end customer. This was so because it was the end customer who pays the VRT. VRT, Revenue says, is never paid by an intermediary, and it is never paid at any stage during the course of any wholesale transaction in relation to the car. It is only, ever, paid once and exists only at the level of retail price. Where a trader buys it, Revenue contends, there is no retail price, but a trade price. Furthermore, Revenue emphasised, VRT is assessed on sale of a new car on inter alia the dealers margin. There was, Revenue argues, no evidence that there was discrimination as between the two.
347. UCII - in both the evidence of the four expert economists it tendered in evidence and in the submissions of its counsel - portrayed Dr. Bacons evidence as depending on the theory that, staying with Mr. Sreenans example, where the domestic vehicle was sold by the dealer for 6,000, the VRT was re-embedded in the vehicle, so that when the vehicle was sold for 6,000 20% of that entire sum would be treated as being VRT, even though the additional VRT (that is the difference between 20% of 5,000 and 20% of 6,000) was never actually paid to Revenue at all. UCIIs experts emphatically dismissed Dr. Bacons evidence based on that assumption. Mr. Sreenan SC pithily summarised their position in the terms that residual VRT depreciates, it does not ever increase, because there is a fixed amount that has been paid in the first instance.
348. At the root of the difference between the expert witnesses thus lay a dispute as to whether the embedded VRT is properly reflected in the prices in the Car Sales Guide (the main dealers prices), or whether (as UCII put it in its High Court closing submissions) the embedded VRT is reflected by the sale by the owner of the car to the trader who may be the main dealer or may sell on to a main dealer. The former was the position adopted by Dr. Bacon, while the latter was the analysis adopted by all the other economists who gave evidence - Mr. Dowling, Mr. McDowell, Mr. Massey and Mr. Kenneally.
349. One example from the evidence illuminates this distinction. If, it was put to Mr. Dowling, a car is sold for 20,000 with VRT of 25% (5000) and sold four years later for 10,0000, the embedded VRT is 2,500. Everyone agreed this much. If the vehicle is then sold by a trader for 12,000, UCII says that the embedded VRT remains 2,500, while Revenue says it is 3,000. Mr. Dowlings rationale was that the VRT could not increase because no additional tax was being paid. This is the case even though the sale by the trader shows (or at least strongly suggests) that the market value of the vehicle is 12,000. So, one way of looking at the question between the parties is to ask whether the embedded VRT is permissibly determined by reference to the lowest price ever paid for the vehicle (the terminus of UCIIs argument) or by the price a willing purchaser is prepared to pay for it on an arms length sale by retail at any given point in time (the end point of Revenues case). Importantly, the conclusion derived by UCII from its analysis of the embedded VRT is, as explained by Mr. Dowling in his evidence, that the only price at which it could be guaranteed that the embedded VRT in a domestic used car is not less than the embedded VRT in an imported used car is if the VRT is based on landed price. VRT based on a price higher than that, he said in evidence, discriminates against dealers in used car imports compared to dealers in domestically sourced used cars.
350. Given the High Courts conclusions in relation to the non-applicability of EU law to the case, it is perhaps understandable that there is little analysis in the judgment of Murphy J. of the underlying merits of this aspect of UCIIs case. He did, however, appear to accept Revenues contention in respect of the embedded VRT, saying at one point that that it was clear from the judgment of the Court in Tatu Case C-402/09 that the market value is a percentage of the original value determined by depreciation. It is not a fixed sum. (at p. 81). That, he said, was the approach adopted by Dr. Bacon. He said the residual value of the tax diminishes proportionally with the depreciation of the vehicle (at p. 81).
The case law
351. In oral argument before this Court, both parties argued this issue by reference to the expert evidence. In the High Court and in its written submissions to this Court, Revenue contended that the question was one not of expert evidence at all but of law. I agree with that latter position. While the expert evidence adduced by each party is helpful in understanding the operation of the underlying concepts, the net issue presented by the dispute is, in my view, properly determined as a matter of law, and the legal principles are those articulated in the CJEU case law. It is those legal principles, not the interpretation attached to them by the five expert economists who addressed the issue, that determine the question.
352. Although there had been earlier cases addressing the relationship between Article 95 and various motor taxes (see Commission v. Greece Case C-132/88), the application of these principles to the particular issues presented by taxes on domestic and imported used motor vehicles first fell for consideration by the Court of Justice in Commission v. Denmark Case C-47/88. There, the enforcement proceedings arose from the manner of calculation of motor registration duty which applied to new vehicles and to imported used cars. In the case of imported used vehicles the taxable value was 100% of the price of the vehicle as new where less than six months old, and 90% of that price where it was more than six months old. The sale of vehicles already registered in Denmark did not give rise to a duty. The conclusion of the Court that Denmark had thereby failed to fulfil its obligations under Article 95 of the EEC Treaty was shortly expressed, as follows (at para. 20) :
Accordingly, even if it appears that by reason of the very large amount of tax levied on new cars the portion of the duty still incorporated in the value of the vehicle is written off more slowly in Denmark than in other Member States which levy a lower duty, that does not prevent the levying of a registration duty for which the basis of assessment is at least 90% of the value of the car when new from constituting generally manifest over-taxation of the vehicles in comparison with the residual registration duty in the case of previously registered used cars bought on the Danish market, whatever their age or condition.
353. As I have already noted, it is clear from the opinion of Advocate General Mischo in the case that the reference here to residual registration duty is to the method approved by the Court in Staatssecretaris van Financien v. Schul (No.2) Case 47/84 [1985] ECR 1491 in the context of VAT on imported second hand goods. Schul arose from the import of a second hand boat from France into the Netherlands. The issue was whether the Dutch authorities could, consistent with Article 95, impose VAT on the importation. The vessel had been purchased by a private individual from a private individual. Had the sale occurred in the Netherlands, no VAT would have been payable on the transaction. In Schul (No.1) Case 15/81 [1982] ECR 1409 the CJEU decided that the imposition of VAT on the importation of goods did not breach Article 95 provided a similar domestic transaction was also subject to the same tax. Where, however, the domestic transaction was not subject to that tax and where the imported product remained affected by the VAT levied when new in its country of origin, it was necessary that the amount of VAT levied on import be reduced by the residual part of the VAT still incorporated in the value of the imported product.
354. The application of that principle proved problematic in the national Courts, and those difficulties resulted in Schul (No.2) where the CJEU addressed how the residual tax should be calculated. Schul contended that the correct approach was that the rate by which the original tax bill should be amortised would be reflected in the depreciation in the market value of the goods which had occurred since the VAT was levied. A 50% reduction in value would, on this basis, have resulted in the residual tax being 50% of the original price paid. This was the analysis ultimately favoured by the Court. On the facts, the value of the boat had increased, and therefore the residual value was the full original amount. I note in passing that the issue of how an increase in value from the point at which VRT was paid would be accommodated was inconclusively debated in the evidence in this case.
355. Thus, and based on the approach thus adopted by the Court in Schul, Advocate General Mischo explained in Commission v. Denmark that the residual tax was the amount of the duty paid at the time of registration of the car when new, reduced by an amount equal to the actual depreciation in the value of the car (at para. 36). However, he clearly envisaged some degree of flexibility in this regard, saying (at para. 37):
Different methods may be used to apply that principle. The estimated value of such vehicles could be progressively reduced, for example, or the value of the vehicle could be disregarded completely, and registration duty charged at a fixed rate, based on the residual amount of duty still deemed to be contained in the price of a car of the same type and age offered for sale on the Danish used-car market.
356. The critical role of the residual value in setting a basis for equivalence between the tax charged on registration of an imported used car, and the tax paid on first registration of a new car and later sold as used, was acknowledged again in Nunes Tadeu C-345/93 [1995] ECR I-479. There, the Court was concerned with a Portuguese Motor tax, applicable to imported new or second-hand vehicles, or vehicles assembled or manufactured in Portugal. It did not apply to second hand domestically manufactured or assembled vehicles, on the basis that the tax would already have been paid in respect of them. The tax varied according to cylinder capacity in accordance with a table annexed to the legal instrument imposing it, which provided that for imported cars first registered for than two years previously, the tax would be 10% less than the amount resulting from that table.
357. The essential reason for the Courts conclusion that the imposition of the tax on second hand vehicles breached Article 95 was stated in two paragraphs of its judgment (at paras. 13 to 14):
it is common ground that in the case of imported second hand vehicles, the tax charged under the decree-law may not be less than 90% of the tax charged on a new car regardless of their age or condition, whereas the residue of the tax incorporated in the value of a second-hand car purchased in the national territory may be less than that since the residual value of the tax diminishes proportionately with the vehicles depreciation.
As a result, a motor vehicle tax on imported second-hand vehicles which is not less than 90% of the tax charged on new cars is manifestly excessive for those vehicles compared with the residue of the tax in second-hand vehicles already registered and purchased on the Portuguese market.
(Emphasis added.)
358. Thus, because the tax restricted the reduction in the amount of the tax charged on new cars to 10% without having regard to the vehicles actual depreciation, it discriminated against imported second-hand cars (at para. 15). The residue of the tax incorporated in the value of the second-hand vehicle purchased on the Portuguese territory diminished automatically with the depreciation of the vehicle, whereas the tax charged on imported second hand cars was never less than 90% of the tax charged on the car as new (at para. 16).
359. Again, the opinion of the Advocate General (in this case, Mr. Jacobs) is valuable in explaining the concept and relating it back to the decision in Schul. Both Schul and Nunes Tadeu were, he explained at para. 13, concerned with determining the residual amount of tax incorporated in the value of second-hand goods and it was accordingly logical that the principles articulated in the former, should also apply to the latter. He continued (at para. 14):
Thus, if the value of a new car placed on the market in Portugal in 1987 was ESC 1,000,000 including car tax of ESC 200,000 and if in 1990 the value of the car was ESC 600,000, the residual tax incorporated in that value amounts to ESC 120,000. That is the maximum amount of tax that may be charged when a comparable car is imported into Portugal.
(Emphasis added.)
360. Advocate General Jacobs made one further observation in Nunes Tadeu of relevance to this case. He explained that the importing State is not required to base the tax on the price paid for the car by the importer or on its value in the exporting State; it is entitled to take into account the value in the importing State. That follows from the requirement that the tax charged in the importing State must not exceed the residual amount of tax incorporated in the value of a domestic used car of the same characteristics. If the importing State were required to base the tax on the lower value in the exporting State, he said, that would not merely preserve any competitive advantage arising from that lower value but would in addition give the importer a fiscal advantage which would be inconsistent with fiscal neutrality (at para. 18). While the Court in its judgment emphatically rejected an argument advanced by the Portuguese Government that the tax was lawful because imported cars were so much cheaper than domestic vehicles that even with the tax imposed they remained competitive, the contention that the member states are entitled to look to the price the vehicle will obtain in the domestic market and not the market of origin, remains correct and was confirmed subsequently by the Court in Weigel, to which I will return.
361. I stress this because this proposition undermines a feature of the evidence of UCIIs witnesses - in particular of Dr. Massey - who proceeded on the basis that the residual VRT was determined by reference to the intermediate price paid for the vehicle, irrespective of where it was purchased. He explained the incongruity that might arise where cars depreciated faster in another jurisdiction by stating that over time one would expect that the domestic and imported prices would tend to converge (Day 20 p.132 to 133 Q.347). In fact, the correct analysis based on the case law is to address what a willing purchaser would pay at any particular point in time for the vehicle in this jurisdiction.
362. In the course of his opinion in Gomes Valente Advocate General Fennelly adopted the analysis of residual value suggested by Advocate General Jacobs in Nunes Tadeu. The critical variable in the determination of whether there was an equivalence of treatment between imported used cars and the same vehicle when sold domestically, he explained, was value. Noting that there may in fact be no exact identifiable comparator, that, in consequence, the residual tax element in the value of a vehicle was infinitely variable, and there might be no reliable means of ascertaining its precise value he explained that that value was in practice equal to the amount the buyer is willing to pay and the vendor to accept as consideration (at para. 23).
363. The analysis adopted by the Court in Weigel v. Finanzlandesdirektion fur Vorarlberg Case C-387/01 reflects the definition of value for these purposes as articulated by the Advocate General in Gomes Valente. There, at issue was the effect of an Austrian measure which calculated a base tax payable on registration of a motor vehicle by reference to fair market value. That value was fixed by the relevant national law as the mean of the price paid by the dealer and the price at which it was sold (see para. 21 of the judgment). The taxpayers contended that because second hand car prices were lower in Germany (where they had purchased their vehicle) than in Austria, the tax in question eliminated the competitive advantage which those vehicles would hold on the Austrian market in its absence. In rejecting this claim the Court adopted the position that the tax did not have this effect as it fixed the fair market value at a level that corresponded with the price the imported vehicle would achieve on the domestic market. Therefore, that price tended to correspond with the residual tax incorporated in the value of used cars already registered in the State and did not accordingly prevent the trader from taking advantage of price differences between second-hand cars in the Member States. It explained the position as follows (at para. 79) :
the fair market value is a set value that corresponds, broadly speaking, to the price the imported vehicle would achieve on the domestic market. Consequently, the NoVA base tax charged on imported used cars tends to correspond to the residual NoVA incorporated in the value of used cars already registered in the State.
(Emphasis added)
364. A series of propositions relevant to the debate in this case emerge from these decisions. They are as follows:
(i) The residual tax is defined as the amount of duty paid at the time of registration of the car when new, reduced by an amount equal to the actual depreciation in the value of the car (Advocate General Mischo in Commission v. Denmark at para. 36).
(ii) Thus, the residual value of the tax diminishes proportionately with the vehicles depreciation (Nunes Tadeu at para. 13).
(iii) It is consistently described in the authorities as a function of the value of the vehicle at any point in time (Advocate General Jacobs in Nunes Tadeu (at para. 14); Advocate General Fennelly in Gomes Valente (at para. 23); Weigel (at para. 79).
(iv) That value is the amount the buyer is willing to pay and the vendor to accept as consideration (Advocate General Fennelly in Gomes Valente (at para. 23); Weigel (at para. 79).
(v) The question of what the value of the vehicle is for this purpose may be determined by reference to value in the importing State. The importing State is not required to base the tax on the price paid for the car by the importer or on its value in the exporting State (Advocate General Jacobs in Nunes Tadeu) so that it may fix the fair market value at a level that corresponds, broadly speaking, to the price the imported vehicle would at the relevant time achieve on the domestic market (Weigel para.79).
Analysis
365. I have already noted that the end point of UCIIs argument is that a registration tax levied by reference to price or value and imposed on anything other than the landed price of imported vehicles is contrary to EU law. Once the margin of the dealer is added to the taxable value, the import is suffering a charge that is not imposed on the domestically sourced vehicle (the argument runs) and it is this that creates the discrimination against imports and in consequence the illegality. Thus understood, it is notable not merely that CJEU has never so held, but that it has repeatedly sanctioned the determination of taxable value by reference to the prices of second-hand vehicles in the national market or lists of average current prices used as a reference in the sector (Gomes Valente at para. 25). In Weigel the basis for valuation was the mean of the dealers purchase price and selling price. While in Commission v. Greece the tax was based on the wholesale price of the corresponding vehicle current at the time the imported vehicle was put into circulation, the Court made it clear that this was not the only basis on which the tax could be imposed, confirming that regard could be had to average prices in the national market (at para. 33). These were all retail prices; yet it was UCIIs case that this was impermissible because the retail price in the domestic market necessarily involved some profit for the retailer and this profit was being taxed for the import, but not for the domestically sourced vehicle.
366. Moreover, any of the cases to have articulated an explanation of the embedded tax have done so by reference to value and have referenced that value to the price a willing purchaser will pay for the asset in the jurisdiction levying the tax, not the price an importer has paid in another jurisdiction. A consideration of the authorities and principles I have outlined in the previous section of this judgment demonstrates why this is so. Obviously, the residual value of a tax arises where that tax has actually been paid in respect of an asset. When that happens, the residual duty is at every point in time thereafter referable to two variables - the proportion that the tax as paid bears to the original value of the asset (which never changes), and the value of the vehicle in this jurisdiction (which may change). Thus, staying with Mr. Sreenans example, the residual value of the VRT in the vehicle to which he referred is always 20% of the value of that vehicle unless the value of the vehicle has increased from that at the point the tax is imposed, in which case Schul appears to suggest the residual value of the duty equals the full original amount.
367. The objections to the suggestion by Revenue that it was possible for the residual value of a tax to fluctuate as the price paid for the asset fluctuated failed to acknowledge that the residual tax is an artificial construct designed to enable a comparison to be made with the tax levied on the sale of what is deemed to be an equivalent asset, and that it is a product of the price a willing purchaser will pay for that asset. It is concerned with the value of the vehicle in this jurisdiction and it is concerned with that value only at the point at which a comparison needs to be made between the value of the domestic second-hand vehicle and the value of the imported used car. That point is the point at which registration tax on the import must be paid.
368. It is for that reason that the contention underpinning UCIIs expert evidence - that in identifying the embedded VRT in a vehicle that is acquired by a main dealer and subsequently sold by him the focus was properly on the first of these transactions (Mr. Dowling Day 4 p.87 Q. 238; Mr. McDowell Day 18 p.64 Q.70) - was mistaken. The level of the embedded VRT notionally paid by each subsequent purchaser matters not. What matters is the amount of VRT embedded at the point at which the imported vehicle is taxed. It is at that point that the embedded VRT in the domestic car must be no less than the VRT paid. It is simply not relevant at any other point and therefore other notional sales of the vehicle and the amount of embedded VRT paid by them (whether they increase or decrease) are neither here nor there (save and insofar as they might provide evidence of what a willing purchaser is prepared to pay for the vehicle). The fact that no further tax is paid to Revenue where one sale is at a higher price than another is similarly irrelevant : the entire construct is artificial. The tax, as I have said, is only ever paid once.
369. So, in summary, the residual tax was that residing in the domestic product, and that only became relevant when a comparison fell to be made with another imported and equivalent vehicle at the point of a taxable event. The vehicle in which the residual tax resided never underwent any other sale of relevance, because tax was only paid once. When that tax was paid, it was determined by reference to retail value. Throughout the life of the vehicle, the embedded VRT continued to bear the same proportion to the value on the basis of which it was calculated - the retail value. When the tax was imposed on the equivalent import, it was also imposed on the retail value. The object of Revenues valuation methodology was to align the valuation of the import with that of the domestically sourced equivalent product. Both values incorporated a dealer margin in the price on which the tax was paid. That was the essential point made by Dr. Bacon, and it is both logical and reflects the purpose of the construct as explained in the cases.
370. Of course, that whole process depends for its efficacy on what is meant by value and how that is determined. The focus of the inquiry is whether the imported and domestic products are differentially burdened (Nadasdi and Nemeth C-290/05 and C-333/05), not whether the intermediaries selling those products are put in precisely the same position. If VRT is charged on the domestic vehicle on first registration by reference to one definition of value and then charged on the used imported vehicle by reference to another, there is no equivalence. However, UCII never established that Revenue was wrong in asserting, and never refuted the evidence adduced by it establishing, that there was no such differential. When the tax is paid, it is determined on the basis of the OMSP of the vehicle as new. That price incorporates what it costs the purchaser to obtain the vehicle from a dealer. Dr. Bacon gave evidence that that price included the dealers margin. That was not contradicted - Mr. McDowell actually said that this was certainly possible. In point of fact, as Revenue contends in its written submissions in this Court, the dealers margin on the sale of the new vehicle will normally be greater than the margin on it when used and second hand. Either way, the same base level applies when the vehicle is sold second hand. UCII never explained in its case why that train of reasoning was wrong.
The compatibility of Revenues valuation methodology with EU law: the High Court judgment
371. I have explained earlier why I have concluded that UCII has not established that the trial Judge erred in his conclusion that the system operated by Revenue for the purposes of determining the OMSP of used imported cars was irrational. Bearing in mind the burden imposed on UCII to establish that asserted illegality, the trial Judge was entitled to conclude on the basis of the evidence before the Court that Revenue acted reasonably in establishing a methodology based on the values in the Car Sales Guide and other sources, and that it had not been established that the operation of that methodology produced OMSPs that were sufficiently divorced from the market value of those vehicles to meet the test of irrationality imposed by Irish law on the determination of whether the exercise of an administrative power was unlawful.
372. This, however, does not determine the issue of European law that arises from the operation of Article 110. The Court was presented with a claim advanced by a party enjoying standing to make it that the methodology operated by Revenue did not in fact result in valuation for imported cars that equated to the value of equivalent domestic models. It follows from the principles I have considered earlier that the onus thereupon fell on the respondents to establish that the system put in place by Revenue did in fact result in a valuation for used imported vehicles which would, as a general rule, be very close to their actual value.
373. Noting my earlier finding that Revenue breached EU obligations of transparency in connection with its operation of the discretion conferred upon it by FA92, other important features of the methodology operated by Revenue for the determination of the OMSP of imported used cars meet many of the criteria identified by CJEU as preconditions to compliance with Article 110. The evidence of Revenue as accepted by the trial Judge was that it accorded values to such vehicles based on their make, model and age. Those values were based on trade information of the kind to which I have referred above, in particular, the Car Sales Guide. The Judge concluded that the use of the depreciation schedules resulted in an outcome that, within a small margin of error, reflected the prices in the Car Sales Guide. The process of individual inspection enable account to be taken of condition, kilometrage and extras. There was an appeal process in the course of which the original valuation could be challenged.
374. Furthermore, when the actual manner in which the valuation process was conducted is properly understood, some of the objections raised by UCII to the granularity of the procedure fall away. Once the choice and application of the depreciation schedules was properly directed to producing as its end point the valuation for an imported vehicle of the make, model and age reflected in the Car Sales Guide and other trade information, it matters not whether the starting point for the application of those scales was the price of the vehicle as new when put into circulation, or the price of the vehicle as new at the time of the valuation.
375. However, the trial Judge did not rule on the particular inquiry mandated by European law as to whether the methodology used by Revenue resulted in a value for second hand vehicles which, as a general rule, would be very close to their actual value. Instead, for the purposes of the domestic challenge, he ruled that UCII had not established that there was a differential between the two such as to render Revenues methodology unreasonable. These are not the same.
376. Again, bearing in mind the conclusion of the trial Judge that UCII was not entitled to pray in aid EU law rights, it is unsurprising that the findings in respect of the alleged breaches of those principles, are sparse. The findings in his judgment that are relevant to that question, appear as follows:
(i) Citing Commission v. Greece C-74/2006 paras. 24 to 29, Murphy J. concluded a Member State can establish a fixed scale in relation to models and makes of cars and that there is no requirement for an assessment or expert examination for each car.
(ii) Insofar as UCII had relied upon the fact that VRT did not take into account subsequent changes in tax rates, Murphy J. said that there was no discrimination arising out of the fact that tax rates may change from year to year, referring to the decision in Nadasdi and Nemeth C-290/05 and C-333/05.
(iii) He observed later (at p. 172) that [c]ommunity law did not require Member States to apply the real value of the vehicle as the taxable amount for this type of tax is not harmonised at European level. The scale will always be somewhat arbitrary.
(iv) Murphy J. said that he was of the view that the straight line depreciation commented on by Mr. Yarrow did not accord with international market data which shows the high depreciation in earlier years of most models. However the statutory definition of new cars which extends to cars of up to three months sold and with mileage of 3,000 km is a plausible explanation of cars being held in bond and not taxed until they are older (at p. 166).
(v) At p.85 of his judgment Murphy J. said that the court was not satisfied that the Irish system did systematically overvalue the OMSP of second hand vehicles. He reached a similarly framed conclusion at p.210 :
the court is not satisfied in relation to the evidence given by the plaintiff that the chargeable value of a used vehicle bore no relation to the open market selling price. The court has already referred to the development of the Revenue database, the reference to the Car Sales Guide and the attempt to streamline through depreciation tables which were an effort by the Commissioners to have a uniform opinion in relation to the values.
(vi) Murphy J. also said of the 44 cars imported from Northern Ireland that the evidence does not establish that the values as a general rule were not very close to their actual value (at p. 83) The assertion of a 20% differential was not, he said, substantiated. This, the Court felt, was because UCIIs witnesses had simply been asked to apply the same figures as in the Japanese direct imports.
377. The last two findings are the closest the Court came to this issue. However, they did not resolve it. First, the Court was viewing the matter from the vantage of a burden placed on UCII, when the burden was in fact on Revenue. The trial Judge was correct to approach the matter in this way insofar as the domestic law challenge was concerned, and it is clear from his judgment as a whole that it was through the lens of such a claim that he was appraising the evidence adduced (noting in particular in regard to the burden of proof his insistence at p.219 of the judgment that this was borne by UCII at p. 219). However, this was not the burden imposed by EU law, and nor was the legal test applicable to the domestic law challenge (that of irrationality) the correct calibration of the scrutiny required under EU law. Second, the latter holding was addressed exclusively to the forty-four vehicles imported from Northern Ireland. That, with respect, missed the point. UCII brought a systemic challenge and which, by the time the case came for trial, was directed to Revenues methodology over a period of seventeen years. Revenue did not before the High Court object to UCIIs entitlement to mount such a case, making only very brief reference in oral argument in this Court to a general concern about the scope of the claim, and it did seek to either arrest or constrain the case by reference to any particular time period. Yet the Court neither determined the issue underpinning that challenge nor decided that it was too broadly framed to adjudicate upon.
378. If applied across the entire period covered by these proceedings, resolution of that issue would depend on Revenue establishing inter alia (a) that the system operated by it did in fact ensure that the values afforded to vehicles reflected those in the Car Sales Guide, (b) that the methods used to update that system ensured that this process was at any relevant point in time reasonably current, (c) that the use of the Car Sales Guide itself produced as a general rule values that were very close to their actual value bearing in mind that CJEU has strongly suggested that this requirement is fulfilled where the resources used to this end identify the average prices of second-hand vehicles in the national market or a list of average current prices used as a reference in the sector and noting the evidence that the Car Sales Guide reflected only 20% of relevant transactions (d) that the use of the standardised tables prior to the development of the facility for hybrid scales (and bearing in mind the number of such scales at various points in time) in fact produced outcomes that were reflective of those in the Car Sales Guide, (e) that these tests were met notwithstanding the use of plus and minus 3% rather than plus and minus 5% for the period during which this was the method applied to monthly reductions, and (f) that the reductions made in respect of kilometrage and condition operated to bring about that alignment of value between imported and domestic vehicles.
379. While this Court was in a position to adjudicate on the contention that, as a matter of law, the imposition of VRT on the dealers margins was a breach of European law, it is not immediately apparent to me that this Court can determine the remaining issues of European Law to which I have referred in the absence of relevant findings by the High Court. However, I am of the view that further submissions should be received from the parties as to how the matter should be addressed from here having regard, in particular, to the fact that the Court proposes to make declaratory orders that Revenue acted in breach of EU law in one specific respect, to the fact that UCII has no basis for claiming either damages or return of taxes paid under EU law, and to the broad, abstracted nature of UCIIs challenge and the absence of any proper temporal limitation to it. To be clear, and having regard to the length of time that has elapsed since the institution of these proceedings and the events to which they relate, this Court will entertain and consider any submissions the parties choose to make as to the breadth and nature of UCIIs claim, the utility and fairness of a remittal of this aspect of the action, whether it is in the interests of justice that the proceedings be so remitted (including any issues the parties choose to agitate as to the passage of time since the events in issue, any resulting prejudice and the reason for that passage of time) and in the event that they are to be so remitted whether (and if so how) the scope of UCIIs claim can or should be refined or defined.
The VAT Directive
380. Article 78 of the VAT Directive, Council Directive 2006/112/EC of 28 November 2006 on the common system of VAT provides as follows:
The taxable amount should include the following factors:
(a) Taxes, duties levies and charges excluding the VAT itself;
(b) Incidental expenses, such as commission, packaging, transport and insurance costs, charged by the supplier to the customer
381. Section 10(1) of the Value Added Tax Act 1972 provides:
The amount on which tax is chargeable by virtue of section 2(1)(a) shall, subject to this section, be the total consideration which the person delivering goods or rendering services becomes entitled to receive in respect of or in relation to such delivery of goods or rendering of services including all taxes, commissions, costs and charges whatsoever, but not including value added tax chargeable in respect of the transaction
382. The point made by UCII arising from the foregoing is simple. These provisions, it says, mandate that VAT is the last tax chargeable and must be the last element in the calculation of the total tax payable on a vehicle. Thus, it must be charged on the VRT. However, the system as operated in Ireland is such that the customer pays the dealer the retail price of the vehicles inclusive of all taxes, including VAT. VRT is then calculated on this amount. This, it says, is in direct contravention of both EU law and national law both of which mandate that VAT be the last calculation. The argument was not addressed at all by the learned High Court Judge.
383. I believe the argument to be misplaced. The purpose of Article 78 is to identify specific expenses that must be included in the taxable amount and to that end to ensure that costs paid by the purchaser as an inherent part of the sales transaction are incorporated into the price on which VAT is calculated. However, central to the provision is the taxable amount and there is nothing in Article 78 that suggests that it was intended to fundamentally affect the meaning of that term. It is defined in Article 73 of the Directive, as follows:
In respect of the supply of goods or services other than as referred to in Articles 74 to 77, the taxable amount shall include everything which constitutes consideration obtained or to be obtained by the supplier in return for the supply from the customer to a third party, including subsidies directly linked to the price of supply
384. Section 37(1) of the 1972 Act (which was the provision in force at the time relevant to these proceedings) transposes this provision into domestic law, stating:
The amount on which tax is chargeable by virtue of section 3(a) or (c) shall, subject to this Chapter, be the total consideration which the person supplying goods or services becomes entitled to receive in respect of or in relation to such supply of goods or services, including all taxes, commissions, costs and charges whatsoever, but not including value-added tax chargeable in respect of that supply.
385. It follows that consideration is central to the definition of the taxable amount for the purpose of Value Added Tax. VRT is not part of the consideration which the person supplying the goods or services becomes entitled to receive in respect of or in relation to such a supply. It is instead charged, levied and paid on the registration of the vehicle by the person who registers it. An individual who presents a vehicle for registration is thus liable to pay VRT not by way of consideration to which the supplier of the vehicle is entitled, but by the provisions of FA92. Although a tax or charge within the meaning of Article 78, it is not properly viewed as a component of the consideration for the sale. It arises from an entirely distinct obligation imposed on the owner of the vehicle at the point of registration. While s.136 FA92 (upon which UCII relied in this regard) effectively requires an authorised dealer in motor vehicles to pay VRT before delivering an unregistered vehicle to a third party, that does not convert VRT into a tax on the sale transaction. It is and remains a tax on registration alone.
XVI DAMAGES AND RECOVERY OF TAX
The causes of action for damages and the High Court judgment
386. In the course of her submissions, Ms. Barrington SC identified three causes action as relevant to UCIIs claim for damages - misfeasance in public office, a claim for damages for breach of constitutional rights and a claim for damages for breach of European law in accordance with the principles identified by the CJEU in Francovich. To these Mr. Sreenan SC added a claim for damages for conversion and detinue arising from the collection by revenue of tax which, it was said, should not have been levied.
387. Murphy J. observed that even if UCII had established that the actions of Revenue were ultra vires, the liability of the defendants for damages would not follow. He said there was no specific reference to damages resulting from the law being ultra vires and no calculation upon which damages could be assessed (at pp.196 to 197).
388. He proceeded to refer to the decision of the Supreme Court in Glencar Explorations plc v. Mayo County Council (No.2) , and to the formulation of those circumstances in which ultra vires action will ground a claim in damages as approved in Pine Valley Developments v. Minister for the Environment [1987] IR 23, concluding that none of those situations arose in this case. Revenue, he said, had statutory power, there was no evidence of malice, and the issue of whether a recognised tort had been committed does not arise (at p. 197).
389. As already noted, the High Court rejected the claims for conversion, detinue and negligence seemingly noting the failure of UCII to avail of a remedy by way of appeal. Insofar as repayment of VRT was sought by the plaintiff, this was also refused in part because the Court was not satisfied that all payments of the tax were made under protest, but also because the evidence did not establish whether it was the plaintiff or authorised dealers who paid the tax (at p. 239). The Court emphasised in the latter respect the absence of any evidence from dealers or documentary evidence of any such claims having been made (at p. 239).
390. Leaving to one side the difficulties the Court clearly saw in proof of the alleged losses, the critical reason Murphy J. held that UCIIs claim for damages under EU law failed, was that they were based upon a contention of discrimination in connection with Japanese imports which were, for the reasons I have summarised earlier, not in breach of European law (at p. 189).
391. Murphy J. also concluded that the quantification of the claim for damages is based on discrimination at EU law (at p. 189) and viewed this as material to his rejection of the claims insofar as they were based on Irish law. There was, he said, no quantification of damages that would arise as a result of headings of legitimate expectation, proportionality, legality, equality and legal certainty (at p. 208).
Assessment
392. Given that insofar as the claims in domestic law are concerned I have found that Revenue did not, save in one respect, act illegally, only one of these claims arises. The others can thus be dealt with briefly. The collection of a tax unlawfully (if this were established) does not give rise to any claim in detinue or conversion and no authority that remotely so suggests was opened to the Court. Where monies are paid on foot of a facially valid legislative provision or administrative decision, they are lawfully transferred and thus not amenable to claims based on either of these torts. If the relevant provision or decision is subsequently declared to be void ab initio so that there was, in actuality, no legal basis for demanding payment, the taxpayer is entitled to recovery of the monies, subject to any available defences such as laches. The appropriate remedy to secure that return is restitutionary. The claim for damages for misfeasance in public office is in large part negated on the evidence: the trial Judge made an express finding that there had been no malice.
393. The claim for damages for breach of constitutional rights does not arise, as I have found that there was, in fact, no such breach. Even were that not the case, the judgment of Baker J., delivering the decision of the Court, in M.C v. Clinical Director of the Central Mental Hospital [2020] IESC 28 shows that such a claim would not be cognisable. There, the Court addressed the applicants claim that she should be entitled to damages consequent upon the decision of the Court that the respondents had acted unlawfully in connection with the circumstances of her detention at the Central Mental Hospital. The illegality arose from the failure of the respondent clinical director to put in place arrangements necessary to give effect to a conditional order proposed by a review board that she be released from detention. Having decided that, on the facts, the respondents had acted bona fide and that there was thus neither a basis for a claim in breach of statutory duty nor for misfeasance in public office, the Court proceeded to determine the claim for damages for breach of constitutional rights.
394. Addressing the well-established principle that such a claim for damages will only arise where the claimant proves that the existing common law does not afford a constitutionally adequate - or effective - remedy (see, most recently, Blehein v. Minister for Health and Children [2018] IESC 40), Baker J. explained as follows (at para. 133):
What is effective in that sense is not tested by reference to whether a plaintiff can establish the case, but whether the elements of the tort or . Its parameters are present, and would establish a cause of action.
395. Thus described, the process involves fixing on a general categorisation of the claimants specific complaint and determining whether the common law of tort (or for that matter any applicable statutory remedy) operates to provide a cause of action for that category of complaint. Of course, key to this is the level of specificity with which the complaint is thus categorised. In M.C the Court looked at the essential grievance - that the relevant agencies had wrongly failed to undertake a process in accordance with their statutory obligations which, had it been undertaken, would have resulted in her release from detention. She then matched that category of complaint against the existing common law, finding that it was addressed through the tort of misfeasance in public office. The applicant could not bring herself within the constraints of that tort because she could not establish malice. However, that did not mean that the tort was basically ineffective, this being the trigger for intervention through constitutional damages (Hanrahan v. Merck Sharpe and Dhome (Ireland) Ltd. [1988] ILRM 629, at 636). Thus, Baker J. explained (at para. 133):
The substance of the present claim is for damages for breach of statutory duty and for misfeasance in public office, even if it is not so pleaded. The facts relied on by Ms C to ground her claim for damages fall within these recognised categories, even though she cannot on those facts establish an essential ingredient or element of the torts. It is only when, to use the language of McKechnie J., the suggested cause of action cannot attract an appropriate or effective remedy that a court would fashion a remedy or right. What is effective in that sense is not tested by reference to whether a plaintiff can establish the case, but whether the elements of the tort or what McKechnie J. calls its parameters, are present, and would establish a cause of action. When a remedy does exist under common law, under statute, or in equity, and no new alternative or exceptional remedy is required.
396. I labour this not merely because it is relevant in this case: for precisely the same reason as MC had no cause of action in constitutional tort for the failure of the respondents there to comply with their statutory obligation, the finding of the trial Judge that Revenue had acted bona fide means that UCII has no cause of action arising from the operation by Revenue of its statutory discretion in the determination of the OMSP. The conclusion is also important to the overall structure of public torts and their relationship with claims for constitutional damages. It follows from M.C that a claim for damage for maladministration falls only to be determined within the rubric of common law torts of breach of statutory duty, misfeasance in public office or (where relevant) negligence and that claims for damages for breach of constitutional rights have no role in that arena. This is clear from paras. 136 and 137 of the judgment:
The actions complained of against the Clinical Director fall, quite clearly in my view, within the recognised category of misfeasance in public office or breach of statutory duty: the Clinical Director failed to take steps required of him by the statutory regime under which he could be, and was, directed by the Review Board to assess and then make the necessary arrangements to support the compliance, supervision, and enforcement by Ms C of the altered conditions. In doing so the Clinical Director breached his statutory obligations, and what is in issue is whether that breach by him offers a sufficient legal basis on which a cause of action could be maintained by Ms C. The answer to this question must be yes, and therefore the action against the Clinical Director is to be assessed in the light of the constituent elements of those torts. One element cannot be established on the facts, that element might broadly be called mala fides, and therefore while a cause of action exists, Ms C cannot succeed.
The conclusion that follows from the case law I have analysed must be that Ms C is not entitled to frame her action as one for breach of constitutional rights as she has available to her an effective remedy at common law, albeit she was unable on the facts to establish either mala fides or knowledge by the Clinical Director that his actions were in breach of his statutory powers and obligations.
397. This leaves two discrete issues in terms of the claims for damages. The first arises from my finding that Revenue acted unlawfully in failing to provide UCII with the criterion by reference to which it determined the OMSP. The trial Judge did not make any finding that in that respect Revenue acted bona fide, and it would have been difficult to do so without a detailed interrogation of Mr. Lynchs correspondence. However having regard to the conclusions of the trial Judge in respect of loss (to which I return), I cannot see how a claim could ever be made out even if UCII had formulated and related losses attributable to such a claim (which it never did).
398. The second arises from the claim for Francovich damages. Insofar as such a claim might be predicated on the failure to comply with transparency obligations, it fails for the reason I have just identified. Insofar as such a claim might follow from any finding that Revenues methodology did not comply with the requirements of EU law, based on the findings of the High Court such a claim would be arid. A direct claim could not be based upon anything other a breach of UCIIs European law rights. Yet, there was no loss established on the evidence in respect of any such breach. The trial Judge castigated what he described as the lack of seriousness and precision in relation to a claim that might otherwise have formed the basis for discrimination within the Member States (p.79). The possibility of a claim for consequential loss on the theory that but for any incompatibility between EU law and the system operated by Revenue UCII would have embarked upon the business of importation from the EU (which on its face seems at best highly speculative) is negated by the trial Judges findings in respect of loss. Murphy J. reached firm findings rejecting the fundamental underpinnings of a claim for consequential loss, emphasising in particular that there was no evidence to ground an expectation that the plaintiff would ever have reached the capacity assumed by its claim and the findings of UCIIs own experts that market characteristics operated against the development of a major wholesale business in the distribution of UK sourced cars in the Irish market. As I explain in the final section no basis has been disclosed on which those findings can be interfered with by this Court.
The claim for recovery of VRT paid
399. It follows from the conclusions I have reached and explained above that there is no basis in law for the recovery by UCII of taxes paid consequent upon breaches of domestic law. Nor, insofar as such a claim was predicated upon the contention that s.133 was in some way in breach of EU law and thus invalid so that all taxes paid in reliance upon it are recoverable, can such a claim be sustained. The provision is not invalid.
400. Even if UCII were to have established that some aspect of the manner of calculation of VRT was contrary to EU law, this would not avail it in any claim for recovery of taxes paid by it. It can only recover taxes paid consequent upon a breach of EU law that arose from breach of its EU law rights. It had, as Murphy J. held, no right in EU law to have taxes on Japanese imports calculated in any particular way. While it did have such a right in relation to taxes paid in respect of the Northern Ireland imports, the onus is on UCII to establish that it overpaid tax on those imports. While UCII contended in its submissions to this Court that the trial Judge had accepted that it had been overcharged in the amount of £14,997 in respect of its Northern Ireland imports, I do not see such a determination in the judgment. At p. 79 Murphy J. said that he was not satisfied as to the quantification of this estimated overvaluation of Northern Ireland imports. No basis has been advanced on which it could be concluded that that decision was in error.
XVII LOSSES AND REMEDIES
The findings of the High Court as to loss
401. Murphy J. was clear and emphatic in his conclusion that UCII had failed to establish any loss consequent upon the matters of which it complained. The Court did not accept the underpinnings of the claim for consequential loss, noting that there was no evidence to ground an expectation that the plaintiff would ever reach the capacity assumed by this claim given that the number of cars imported by the plaintiff had decreased from a high of 850 in 1990 to 600 in 1993 and 1994 and further gradually declined from September 1995 when Mr. ODowling left the plaintiff. Murphy J. concluded that it was unclear what was the cause and effect of the declining proportion of the plaintiff [sic.] imports (at p. 170). He noted in this context the large volume of imports of used cars from the United Kingdom (35,000 in 1993), and the fact that this must have depressed prices of used cars significantly given evidence that a total of used vehicles was during the period 1996 to 1998 on average 46,000 (of which, on average, 17,000 were from Japan (at p.170)). Evidence given as to projections suggested by Mr. Murakami, a businessman involved in the shipping of used cars from Japan and called as a witness for UCII, was similarly rejected on the basis of the overall capacity of the Irish market (at pp.169 to 170). Murphy J. furthermore felt that there was no convincing evidence why the plaintiffs competitors, under the same VRT system, increased their imports during that period (at p. 166) and that the Court had difficulty in understanding the evidence of Mr ODowling that the plaintiffs performance had suffered because of the VRT system (at p. 178). He said that there appeared to be other variables that account for the difference of performance of the plaintiff compared to competitors. Indeed, the Court even expressed a difficulty in ascertaining the number of cars imported by UCII in the period from 1990 to 2000, as the figures given by Messrs. ODowling, Riordan and OBoyle did not correspond (at p. 226). Murphy J. said that there remained a lack of clarity on the substance of the claim in relation to the basic figures (at p. 227).
402. The Court expressed the view having regard to the foregoing considerations and in the light of the evidence it had heard that it was the departure of Mr. ODowling and the establishment by him of a business competing with that of UCII which, in combination with the decrease in the price of used cars due to competition, was decisive in the decline of UCII (at p. 171). Murphy J. stressed in this regard the fact that it was Mr. ODowling who had everything in place from finance, shipping, dealers and product, and the fact that Mr. Murakami did not know Mr. ODowling had ceased to be a director and shareholder and that he continued to ship vehicles to Sports Car Centre Ltd. Murphy J. said that he had difficulty in understanding why Mr. ODowling set up Sports Car Centre Ltd, which he said appeared to be in competition with the plaintiff and noted that the business was deprived of his experience and connections (at p. 229). He said of the business of Sports Car Centre Ltd. that this profitable venture could have been replicated within the plaintiff company to mitigate any losses that the plaintiff suffered (at p. 229).
403. The trial Judge was, in particular, critical of the manner in which the claim for losses in respect of the vehicles imported from Northern Ireland was agitated. That was presented on the basis of the same rate of overcharge as was alleged to have been imposed in respect of the Japanese vehicles. That, Murphy J. said, did not provide the Court with any reliable evidence of the actual overcharge: he said that this demonstrates the lack of seriousness and precision in relation to a claim that might otherwise have formed the basis for discrimination within the Member States (at p. 79). Throughout he emphasised that the evidence tendered by UCII (in the view of the Court) was based exclusively on tax discrimination under EU law (at p. 88). He determined that the evidence in relation to the forty-four cars from Northern Ireland did not establish that the values as a general rule were not very close to their actual value, was critical of the failure of Mr. ODowling or Mr. Riordan to give evidence of overvaluation of the OMSPs on cars imported by the plaintiff from the UK or, specifically, from Northern Ireland and he underlined the absence of any specific analysis of the resulting loss of opportunity (at p. 191). Because no analysis of the forty-four cars was given, Murphy J. said, the court did not have an evidential basis of determination of loss: the cars were (he said) an afterthought in the claim (at p. 233). DKM, he noted, had said in their report:
Imports from the UK might also have been expected to grow but market characteristics operated against the development of a major wholesale business in the distribution of UK sourced cars in the Irish Market.
404. Overall, throughout the judgment Murphy J. variously and consistently described UCIIs case as exaggerated (at p. 173), not substantiated (at pp. 173, 174), and that he was not satisfied that Mr. ODowling, the plaintiffs main witness had established a basis for the DKM analysis or Mr. OBoyles calculations of 2.12m excess VRT paid to Revenue. He was also critical of the expert evidence as to loss tendered by UCIIs expert, DKM, finding that the assumptions in its report on loss did not take into account the ability or resources of the plaintiff to grow as dramatically as suggested. No account was taken, Murphy J. said, of competition from new as well as from used car suppliers. The resulting change in market price for both was not considered (at p. 219). The assumption that UCII would maintain a market share of 20% was not supported (at p. 221), and the continued growth of the business of Sports Car Centre Ltd. deprived the plaintiff of some market share, turnover and profitability (at p. 221). Indeed, far from being stunted by VRT, Murphy J. concluded that the evidence disclosed a healthy market for Japanese cars during that period (at p. 231), a growth which was accompanied by a drop in UCIIs imports and consequent reduction in its market share (at p. 232). Murphy J. concluded that this may be explained by UCIIs inability to maintain market share while competition in the sector was obviously increasing and by the departure of Mr. Dowling (at p. 232). To attribute that departure, the Court found, to the burden of VRT makes less sense in the light of the plaintiffs declining market share (at p. 233).
405. It followed that the plaintiff could not have maintained a market share of at least 20% as the DKM report had assumed, and thus could not have sold an additional 93,633 cars from 1993 to 2009, as had been contended in the court below (at p. 232).
406. Further, the DKM analysis, did not take account of the legislative basis upon which Revenue had to assign OMSP, an apparent reference to the fact that its calculations were based on a price which did not take account of dealers margins (this being UCIIs contention as to how the exercise ought to have been conducted under EU law) (at p. 223). Having regard to all of these circumstances (and noting in particular the success of Sports Car Centre Ltd.) Murphy J. said that it was difficult to understand the plaintiffs claim that VRT was the cause of its losses.
407. No credible basis has been suggested on which this Court could interfere with any of these conclusions based, as they were, on the trial Judges appraisal of the oral evidence. It follows that irrespective of the view I have formed as regards the causes of action enjoyed by UCII, UCII had no basis on which it could recover damages from the respondents arising from the administration of VRT.
The claim for declaratory relief
408. Given that he determined that Revenue had not acted unlawfully, there is no detailed consideration in the judgment of Murphy J. as to whether the case presented was one in which, had such illegality been established, it would have been appropriate to grant declaratory relief. That said, on one view Murphy J. also may have concluded (as Revenue in its closing submissions to the High Court urged him to do) that no purpose would be served by making any other form of order, insofar as he referred to the decision in Blanchfield v. Hartnett [2002] 3 IR 207, in which the High Court and Supreme Court had declined to quash orders admittedly made without jurisdiction for this reason (at p. 198). Perhaps in a similar vein, he noted later in his judgment that the case comprised a private action being taken against the Revenue Commissioners and the State for damages. It is not a judicial review nor a public inquiry into vehicle registration tax (at p. 199).
409. Insofar as I have determined that Revenue acted unlawfully, both in accordance with the requirements of domestic and of European law, in failing to provide UCII with details of the basis on which the OMSP of used motor vehicles was determined, in my view it follows that declaratory relief should be granted to that effect. While the Courts discretion to refuse declaratory relief does encompass the power to refuse to make any order even where illegality had been determined where such relief would serve no useful purpose, the factors which might in some circumstances justify such forbearance are not applicable to the complaints advanced by UCII in relation to the failure of Revenue to make known the methodology applied by it to the determination of the OMSP.
410. In Blanchfield v. Harnett the applicant claimed, and respondents accepted, that orders made under the Bankers Books Evidence Act 1879 had issued without jurisdiction. By the time of the proceedings, the orders were spent and, insofar as the objective of the proceedings was to secure the exclusion of evidence obtained on foot of the orders from the trial of certain criminal proceedings against the applicant, that issue of admissibility was one that could be determined by the trial Judge. Fennelly J. delivering the judgement of the Supreme Court decided that the trial Judge had acted within his discretion in refusing any relief in those circumstances. However, that decision was informed by the general undesirability of the Superior Courts intruding via their supervisory jurisdiction into the process of a criminal trial. Thus, Fennelly J. relied upon and approved the remarks of Gannon J. in The State (Glover and Mulligan) v. McCarthy [1981] ILRM 47, at p. 51 (themselves cited in the context of the refusal of judicial review on discretionary grounds in Berkeley v. Edwards [1988] IR 217 and Byrne v. Grey [1988] IR 31) which directed the discretion to refuse relief by the objective of achieving a just solution of the matters at issue with minimal inconvenience consistent with regularity of judicial procedures. Fennelly J. may also have been concerned that the issue of an order of certiorari would in some sense affect how the Judge with seisin of the criminal proceedings would approach the issue: the illegality of such an order is not determinative of the issue of admissibility. Taking the issue out of its proper context may create a misleading impression as to its impact.
411. No such countervailing factor applies here insofar as the failure of Revenue to make known the features of its methodology to which I have referred. While it is undoubtedly the case that the declaratory relief sought in this case relates to events of some time ago, I do not see that in and of itself as affording a reason not to grant it. This is the case for a number of distinct reasons.
412. First, the passage of time presented no difficulty in addressing this issue, which was dependant solely upon the application of the law to uncontroversial facts. Although the period of time to which the complaint related was substantial, and although the extent of the information disclosed by Revenue changed over that period, it was not until 2010 that the depreciation tables were made available and until that was done Revenue was not in compliance with the legal obligation I have identified. There was no controversy around the facts relevant to this : the matter was exclusively one of law.
413. Second, the issue seems to me to be of general importance. There is a real point in making clear the obligation on Revenue to make a policy of the kind in issue in this case publicly known. In this regard I attach some significance to the failure of Revenue to respond to the queries raised in UCIIs correspondence of 1994 and the adherence by Revenue to an apparently deliberate policy of non-disclosure of the depreciation schedules which, for the reasons to which I have referred, was unlawful.
414. Third, in determining whether to grant or refuse declaratory relief (which is of course discretionary), the Court should properly have regard to whether the determination of the issue in question is liable to result in an unfairness to the defendant. In the case of a wide-ranging and ill defined claim requiring a defendant to justify its actions over a lengthy period of time, there are certainly circumstances in which this could be so. This aspect of UCIIs claim - that of an admitted failure to disclose certain information - did not present any such unfairness.
XVIII CONCLUSIONS AND ORDERS
415. My conclusions in respect of the issues I have identified and addressed in the course of this judgment and, in brief, the reasons for them can be summarised as follows:
1. If and insofar as the trial Judge placed reliance on the availability of a statutory appeal against individual decisions of Revenue determining the OMSP of imported used vehicles as a basis for precluding the declaratory relief claimed in this action, he erred. This was a systemic challenge to the manner in which Revenue operated VRT, which was combined with claims for damages, and declarations as to the validity of the governing statutory provisions having regard to the Constitution and European law. The appeal did not provide a remedy in respect of those complaints.
2. UCII is wrong in contending that Revenue is obliged to have regard to the price obtained on the sale of a vehicle in determining the OMSP of that vehicle. The language of s.133 of the Finance Act 1992, its focus on the opinion of Revenue and the benchmark of reasonable expectation, the legal and factual context in which the provision was intended to operate and the consequence of the obligation suggested had it been imposed points to the Oireachtas envisaging the general approach to the determination of the OMSP put in place by Revenue. It did not mandate the determination of OMSP by reference to the price actually paid for a vehicle and, accordingly, necessarily sanctioned the establishment by Revenue of general criteria by reference to which the price that might reasonably be expected for a used vehicle in the State could be ascertained. Those criteria could include the price actually obtained on a sale but did not have to do so.
3. UCII is also mistaken in its contention that there was a fundamental conflict in the evidence given by Dr. Bacon on the one hand and Revenues witnesses of fact on the other as to the role of the depreciation tables in Revenues methodology. The contemporaneous documentary evidence, and the testimony of Mr. Butler and Mr. Campion were absolutely clear that the depreciation tables were constructed from and reviewed by reference to the prices contained in the Car Sales Guide and other market sources. The depreciation table applied to any one vehicle was chosen so as to ensure that the input of the statistical code produced the same valuation as would have been obtained had the official available to him or her a compendium of the market prices disclosed by those sources for all models and makes of vehicle. That corresponded generally with the evidence of Dr. Bacon. Insofar as aspects of Dr. Bacons evidence suggested a lack of clarity on his part as to the exact purpose to which the depreciation tables were applied, given that Revenues witnesses of fact gave clear, consistent and unequivocal evidence of Revenues methodology, there was no conflict that required to be resolved. Dr. Bacon was an expert witness not a witness of fact.
4. Because this was an abstract and systemic challenge to the general operation by Revenue of VRT, the burden assumed by UCII in asserting that Revenues methodology was unreaonable was substantial. The question before the Court in this case was not whether the system that was put in place by Revenue presented anomalies, nor whether it might happen that in individual cases its application could result in an OMSP that bore no rational relationship to the price a vehicle might be expected to fetch on the market in an arms length sale in the State. It was whether no reasonable authority taking all relevant considerations into account could have believed that the method would generally produce a reasonable estimate of what vehicle might be expected to fetch in a sale by retail. The trial Judges conclusion that UCIIs challenge to the rationality of that methodology failed was based upon three fixed points (a) that vehicles were valued by reference to the information in the Car Sales Guide and other trade information, (b) that the Car Sales Guide was a reasonable reference point for this purpose, and (c) that UCII had failed to establish that its application systematically overvalued the vehicles. UCII has not established that the trial Judge erred in any of these conclusions. It follows that its irrationality challenge had to fail.
5. Insofar as UCII based part of its claim on breaches of the principle of proportionality or of breach of its constitutional rights, these did not add to the claim based on unreasonableness and properly failed for the same reason as did that contention.
6. It is inherent in the rule of law in general, and the mandate of legal certainty in particular, that where a public body is conferred with a statutory discretion or power and formulates fixed and detailed internal criteria on the basis of which that discretion or power will be exercised, those affected by the exercise of the discretion are entitled to be informed of those criteria when they request them. That obligation serves a number of functions. A law or policy should be sufficiently clear to enable those affected by it to regulate their conduct. It should also be sufficiently clear so as to obviate the risk that a public authority can act in an arbitrary way. Publication thus both enables the affected person to identify when the policy is being properly applied to them and incentivises those who operate it to comply with its terms in a consistent way, as it does those who formulate it to ensure its legality. Clear notice of a policy or decision is required so that the individual knows the criteria that are being applied and is able to both make meaningful representations to the decision maker before the decision is taken and subsequently to challenge an adverse decision. Publication also allows persons affected by the exercise of those discretions and powers to challenge the legality of the policy itself. Until depreciation tables were made available to UCII in 2010, Revenue failed to comply with that obligation in respect of the criteria it applied in determining the OMSP of used vehicles pursuant to s. 133 of the Finance Act 1992.
7. There was and could be no dispute between the parties that insofar as imports from outside the EU are concerned, principles of EU law are not applicable. However, UCII is correct when it says that the trial Judge erred when he decided that this meant that it was precluded by jus tertii from asserting that the system operated by Revenue was in breach of EU law. UCII was not seeking to invoke the rights of any third party. It was relying upon its own rights - as a business established for the purposes of importing vehicles into the State from the EU, and as an entity which actually did so. The trial Judges conclusion in this regard was affected by a conflation of two distinct issues - whether UCII had established losses which followed from a breach of EU law, and whether it had an entitlement to assert a breach of EU law in the first place. Undoubtedly UCII could not maintain a claim in damages consequent upon a breach of EU law in respect of losses sustained by an overcharging of VRT on Japanese imports. However it does not follow from the fact that it could not establish financial loss, it was precluded from seeking declaratory relief to the effect that the system that was operated by Revenue was in breach of EU law. These two propositions are, conceptually, entirely distinct. The learned trial Judge erred in eliding them.
8. EU law requires that where fixed criteria are used to determine the value of imported used vehicles those criteria must be brought to the attention of the public. That obligation stems from the fundamental consideration that in determining whether a tax on imported second-hand motor cars is discriminatory, it is necessary to have regard to the basis of assessment and the detailed rules for levying the tax. It follows that the person discharging the liability is entitled to know both. For the same reasons as Revenue breached the cognate obligation I have identified in domestic law, it breached the similar requirements imposed by EU law.
9. I have identified from the decisions of the CJEU and explained in the course of this judgment five principles of law that govern the application of residual VRT to the taxing of imported used vehicles. They are first that the residual tax is defined as the amount of duty paid at the time of registration of the car when new, reduced by an amount equal to the actual depreciation in the value of the car. Second, that the residual value of the tax diminishes proportionately with the vehicles depreciation. Third, that it is a function of the value of the vehicle at any point in time. Fourth, that that value is the amount the buyer is willing to pay and the vendor to accept as consideration. Fifth, that the question of what the value of the vehicle is for this purpose is determined by reference to value in the importing State. The importing State is not required to base the tax on the price paid for the car by the importer or on its value in the exporting State. Once these five principles are properly understood and applied, UCIIs claim that the State was required to tax imports on their landed price is shown to be misconceived. Its case was based on the mistaken belief that successive sales of an asset fixed a residual tax reflecting the lowest price paid for the asset, which residual tax could never thereafter increase. Subject to the rule expressed in the case law that the residual tax in an asset does not increase where the value of the asset increases above that at the time the tax is imposed, the residual tax is an artificial construct and is a variable function of value at one point in time - the point at which VRT is levied on an equivalent import. It follows that UCIIs assumption was incorrect. The question was not whether the residual tax in a vehicle increased or decreased, but whether the tax on the domestic vehicle and on the import were imposed by reference to the same base line. Revenues evidence established that they were.
10. While I have determined that UCII has failed in this appeal to establish that the conclusions of the trial Judge in respect of its reasonableness challenge were wrong, this does not automatically mean that UCII failed in its EU law challenge to that methodology. The principles of EU law that I consider in this judgement dictated that the onus in this case fell on the respondents to establish that the system put in place by Revenue did in fact result in a valuation for used imported vehicles which would, as a general rule, be very close to their actual value. That issue was not determined by the trial Judge, who decided only the distinct and different question of whether UCII had failed to discharge the burden imposed on it to establish that the methodology was irrational.
11. UCIIs claim insofar as based upon the VAT Directives fails. VRT is not part of the consideration which the person supplying the goods or services becomes entitled to receive in respect of or in relation to such a supply. It is instead charged, levied and paid on the registration of the vehicle by the person who registers it. An individual who presents a vehicle for registration is thus liable to pay VRT not by way of consideration to which the supplier of the vehicle is entitled, but by the provisions of FA92. Although a tax or charge within the meaning of Article 78, it is not properly viewed as a component of the consideration for the sale.
12. UCII has made out no claim in tort in domestic law. Nor does it enjoy any credible basis for a claim in Francovich damages. It cannot claim such damages consequent upon the import of Japanese vehicles. It was found by the trial Judge to have failed to specify how it suffered loss as a result of the VRT levied on the Northern Ireland cars and the claim that it would, but for the manner in which VRT was operated, have established a business importing cars from the EU was rejected on the facts.
13. Nor, for same reasons of proof, has it any basis for claiming recovery of taxes paid.
14. The findings of fact of the trial Judge in respect of loss cannot be disturbed in this appeal.
416. The Court will now receive submissions from the parties as to the orders that it should make consequent on these findings. It is my provisional view that the Court should make very specific declarations as to the unlawfulness of Revenues actions as a matter of domestic law and of EU law in failing to provide UCII with the information to which I have referred in this judgment. I have explained in the course of this judgment why I do not believe any other relief follows from this determination.
417. The Court will direct submissions from the parties as to what order it should make having regard to the fact that the trial Judge did not address the question of whether Revenue had established that the methodology used by it to determine the OMSP of imported used cars resulted generally in values that were close to their actual value. Those submissions should be directed to how the matter should be addressed from here having regard, in particular, to the fact that the Court proposes to make declaratory orders that Revenue acted in breach of EU law in one specific respect, to the fact that UCII has no basis for claiming either damages or return of taxes paid under EU law, and to the broad, abstracted nature of UCIIs challenge and the absence of any proper temporal limitation to it. Having regard to the length of time that has elapsed since the institution of these proceedings and the events to which they relate, this Court will entertain and consider any submissions the parties choose to make as to the breadth and nature of UCIIs claim, the utility and fairness of a remittal of the this aspect of the action, whether it is in the interests of justice that the proceedings be so remitted (including any issues the parties choose to agitate as to the passage of time since the events in issue, any resulting prejudice and the reason for that passage of time) and in the event that they are to be so remitted whether (and if so how) the scope of UCIIs claim can or should be refined or defined.
418. The respondents should deliver their written submissions on these issues within 21 days of the date of this judgment. UCII should respond to those submissions within 14 days. Neither partys submissions should exceed 5,000 words. Upon review of those submissions, the Court will determine whether a further hearing is required in this matter.
419. The question of costs cannot be determined without a final decision being reached as to whether, and if so on what terms, the issue as to the compatibility of Revenues methodology for the determination of the OMSP was compliant with the one aspect of EU law I have referred to above is to be either decided by this Court or remitted to the High Court.
420. Donnelly J. and Faherty J. are in agreement with this judgment and the orders I propose.
Result: Appeal Allowed in Part