First-tier Tribunal (Tax)

You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Amino Communications Ltd v Revenue & Customs [2015] UKFTT 0035 (TC) (26 January 2015)
URL: http://www.bailii.org/uk/cases/UKFTT/TC/2015/TC04248.html
Cite as: [2015] UKFTT 0035 (TC), [2015] UKFTT 35 (TC)

[New search] [Printable PDF version] [Help]

Amino Communications Ltd v Revenue & Customs [2015] UKFTT 0035 (TC) (26 January 2015)
CUSTOMS DUTY
Classification - nomenclature

[2015] UKFTT 0035 (TC)

[image removed]

TC04248

Appeal number: TC/2013/7027

CUSTOMS DUTY - whether IPTV set-top boxes are ‘video tuners’ when they are incapable of receiving radio-frequency signals but instead receive data via the Internet – relevance of EU’s Treaty obligations – appeal allowed

FIRST-TIER TRIBUNAL

TAX CHAMBER

	AMINO COMMUNICATIONS LTD	Appellant

	- and -

	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

TRIBUNAL:

JUDGE BARBARA MOSEDALE

Sitting in public on 10-11 December 2014 at Royal Courts of Justice, London

Ms V Sloane instructed by International Trade Solutions Ltd for the Appellant

Mr M Donmall, Counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents

DECISION

1. The appeals concerns the appellant company’s importation of internet protocol TV (“IPTV”) set-top boxes of various specifications into the UK from 1 November 2009 to 30 June 2011. On 13 September 2013 HMRC on review upheld its refusal to repay to the appellant a claim for £743,565 in customs duty which the appellant says it overpaid because it paid duty at 14% on the set top boxes in that period whereas it contends it should not have paid any duty at all. The appellant appealed that review decision.

The issue

2. The case turns on the proper classification of the set-top boxes (“STBs”). There was no dispute that the rate of duty is dictated by the EU’s Combined Nomenclature (“CN”). There is also no dispute between the parties that the set top boxes in question fell under the classification heading 8528 71 of the CN:

“reception apparatus for television, whether or not incorporating radio-broadcast receivers or sound or video recording or reproducing apparatus”

3. Within that heading was a sub-heading which again the parties were agreed that the products fell within:

“not designed to incorporate a video display or screen”

4. The dispute was on the third level, the sub-sub-heading. One sub-sub-heading was “video tuners” and the other was “other”. The “other” heading was CN code 852871 90 and was chargeable at 14% duty. The sub-sub-heading “video tuners” had itself 3 sub divisions. The parties were agreed that, if the set-top boxes fell within the heading “video tuners” then they would fall in the sub-sub-sub-heading 852871 13:

“Apparatus with a microprocessor-based device incorporating a modem for gaining access to the Internet, and having a function of interactive information exchange, capable of receiving television signals (“set-top boxes with communication function”)”

That heading was charged at 0% duty. The appellant had paid duty on the basis the goods were categorised under 852871 90 “other” at 14% and were not “video tuners”.

5. So the dispute between the parties can be summarised as whether the IPTV set top boxes imported by the appellant during the relevant period were ‘video tuners’ or not.

6. In brief, the appellant’s position is that while a ‘tuner’ could be given what it considered to be the word’s original and narrow meaning as equipment which received and isolated radio or TV data carried on a ‘frequency’ (or wavelength) of electromagnetic waves, the word ‘tuner’ also possessed a more modern, less technical, meaning of simply receiving TV/radio broadcasts.

7. HMRC did not agree that ‘tuner’ had such a meaning and considered that a tuner by definition had to be able to receive signals on a frequency/wavelength. Mr Donmall’s position was that receiving signals on a frequency was intrinsic to a tuner: this is what ‘tuning in’ meant.

The evidence

8. Mr D M Cormie, commercial manager at Amino, gave evidence in the form of an unchallenged witness statement. In this short statement he confirmed that the STBs at issue in the appeal had the functionality that the expert witness, Mr Jones, had been instructed that they had.

9. In brief, the functionality of the STBs in issue was as follows. The STBs would be imported by a broadcaster with an internet protocol television (“IPTV”) network. The broadcaster would supply the STBs to end users (its customers) who signed up (presumably for a fee) to receive its IPTV network.

10. The STBs would be connected by the end user, on the one hand, to his TV and, on the other, to the internet via an Ethernet cable. The STB would be supplied containing software that would enable the STB to connect to the network of the IPTV service provider to which it was dedicated, and once connected to that network, to download from a server on that network the software it needed to operate.

11. So its first operation when connected to the IPTV network would be to download its operational software from that IPTV network. Once operational, it would then download from the network up to date channel information. The end user would then be presented (on his TV screen) with a list of available channels on the particular network of the IPTV service provider who had supplied the STB. The user would select the channel he wanted to view. The STB would receive via the Internet the signals transmitted by the particular IPTV network, and then isolate from those signals the data for the particular channel selected by the end user. It would convert the data to video data, and then transmit the data to the TV for the viewer to watch.

12. Whenever the operational software was upgraded on the network, the STB would automatically upgrade itself. It would also update its information on available channels whenever connected to the IPTV service provider’s network.

Split between box and software

13. From Mr Cormie’s evidence it was clear that the STBs were imported without the software that they needed to operate as an STB. The reason for this was manufacturing efficiency: it reduced manufacturing time and also avoided the risk of supplying STBs with software which was out of date by the time it reached the end user, thus needing an immediate upgrade.

14. The appellant’s case, which was not disputed by HMRC, was that the fact the STBs at the time of import did not contain the software to allow them to function as STBs was irrelevant to the customs classification. The boxes were sold to a TV network at a combined price for the box and the operating software.

15. Applying the decision in the CJEU case of Medion AG C-208/06, the parties accepted, and I agree, that for the purpose of customs classification I must consider the STB as if it contained on importation the software with which it would be installed as soon as it was connected to the internet. I do not refer to this issue again.

The expert evidence

16. The expert evidence was given by Mr Gareth Jones. HMRC accept that he was a suitably qualified and independent expert so I do not set out his background in detail. Suffice it to say he holds degrees in electrical and electronic engineering and has worked on circuits and digital technologies since 1986. For at least the last ten years or so he has helped design and develop various kinds of set top boxes and video recorders as well as worked on a TV network’s broadcast platform including its IPTV platform and ‘video on demand’ platform.

17. Much of Mr Jones’ technical evidence was unchallenged but his opinion on the meaning of ‘video tuner’ in the CN was not accepted by HMRC. I agree with HMRC that Mr Jones’ opinion on this is not one which should influence the tribunal: while he could (and did) give evidence on what the term ‘video tuner’ was understood to mean in the industry, he could not tell the Tribunal what the term meant where it was used in the CN. He was cross examined at length on the meaning of ‘tuner’ in the industry and other related matters and I set out my findings following his evidence below.

Receivers of TV broadcasts

18. Originally a TV broadcaster (such as the BBC) would broadcast analogue data using radio frequencies, one frequency for each channel (or programme) broadcast. So far as I understand it, a ‘frequency’ refers to a particular wavelength of electromagnetic waves. Originally radios, and later TVs, were manually tuned by the user to the particular frequency for the channel they wished to hear/watch. The tuning was done manually using a dial controlling physical apparatus.

19. Shortly before the turn of the last century, technology had developed so that the ‘tuning’ function could be carried out by software running on a microprocessor: the software would automatically locate the various frequencies on which a network broadcast its various channels and store them in a memory to be accessed by the user whenever required. It was no longer necessary for the user to manually tune the TV.

20. Technology also moved on in that the data from a broadcaster could be transmitted digitally rather than in analogue form. At this point, it became possible for each frequency to carry the data for a number of different channels. So the job of the tuner become two-fold: firstly, it needed to find and isolate the various frequencies on which the network broadcast its digital data. Secondly, it then needed to filter the data to isolate the various channels on each frequency. This entire process would be carried out by software running on a processor; that processor would also run the software for the other operations such as displaying the selected channel to the user. No longer could a specific piece of apparatus be identified as carrying out exclusively the function of tuning. Nor was tuning solely a function of receiving and isolating a particular frequency.

21. The above description of technologies were all based on terrestrial or satellite broadcasts of the signal, whether analogue or digital. But another advance in technology was that it became possible to transmit a digital signal using a frequency down cables rather than through the air/atmosphere as with terrestrial or satellite signals. The cable receiver would have software which would enable it to ‘tune’ into the frequencies used by the particular network, and then to filter and isolate from each frequency the data on the various channels carried by that frequency, in order to display the chosen one to the user.

22. A further advance in technology was the advent of internet protocol TV (“IPTV”). Internet Protocol TV does not ‘tune’ to a frequency. The signal which IPTVs receive is in the form of ‘data packets’ transmitted over the Internet. All parties were agreed that IPTV STBs do not receive a signal on a frequency. But, as with the other products mentioned above, the IPTV STBs receive data which has been broadcast or, at least, multicast (the meaning of which is discussed below). The STBs then filter the data and isolate from it just the data relating to the TV channel selected by the user. Like the more technologically advanced products which do ‘tune’ to a frequency, this entire operation is done by software.

Modern meaning of tuner

23. The gist of Mr Jones’ evidence on the meaning of ‘video tuner’ was that it was not a phrase that would be used in the industry to refer to set-top boxes of any description, nor was it a phrase used to describe anything else.

24. So far as the word ‘video’ was concerned, I understood from him that it was really synonymous with ‘TV’ or ‘audio and sound’.

25. It was his evidence that the word ‘tuner’ would be used in the context of terrestrial analogue and digital and satellite broadcast products but only ‘loosely’ in the context of IPTV products because it was entirely a software function. But his understanding of the industry was that the meaning of ‘tuner’ was not now limited to isolating a frequency and that a ‘video tuner’ would be understood to be

….at a functional level … a means for selecting a specific video or TV channel form a set of available broadcast channels and the process of ‘tuning’ a TV involves selecting one of those TV channels for display….

26. He was referred to various definitions of various products on the internet. There is such a product actually called a ‘TV tuner card’ (or ‘plug in tuner card’) which was something added to a computer which isolated a TV signal carried on frequencies, and then isolated from the data stream on that frequency the required channel and transmitted the data to a PC which converted the data to a video output. TV cards differed from STBs because they left it to the computer to convert the isolated data to a video output: STBs in contrast would convert the data to video output and transmit the video output to the TV.

27. Mr Jones was also shown a description on the internet of a DTV Tuner (ie digital TV tuner) and he did not consider the description entirely accurate as it failed to distinguish between the two processes, firstly isolating a signal frequency, and secondly selecting a channel from all the data transmitted on that frequency. It was also his opinion that the article was wrong in implying that the ‘tuner’ would be an identifiable physical component of a TV tuner as for digital TV both those processes were probably done software run on hardware which carried out a number of other functions as well.

28. Mr Jones was also shown a Wikipedia definition of a ‘DTV receiver’ (a ‘digital TV receiver’). Again, Mr Jones did not agree that the description was entirely accurate, in particular, because it failed to make clear that for digital TV there are two processes involved in isolating a channel.

29. In conclusion, in so far as it was HMRC’s case that the definition of these other, related items indicated that Mr Jones’ view of the industry understanding of the meaning of ‘tuner’ was wrong, I do not agree. I accept Mr Jones’ criticisms which amount to saying that they were not reliable as the writers did not really understand how the equipment worked.

30. I find that ‘video tuner’ is not a term used in industry but I accept Mr Jones’ evidence that it would be understood as having the meaning he set out at §25 above. I accept this as Mr Jones was consistent in his evidence given under cross examination and what he said made sense. In a very brief summary, he said that originally TV receiving apparatus had to ‘tune’ to a frequency, with the original meaning of becoming receptive to a particular wavelength. But as technology developed, the term ‘tuner’ continued to be used for a process that was no longer simply ‘tuning’ to a frequency, but now included, once a frequency carried more data than just one channel, to isolating a single channel from that signal. This part of the process was no longer ‘tuning’ in the original meaning. It also makes sense that when the data was no longer broadcast on a frequency at all, but by another means, nevertheless the process of acquiring and isolating a single channel from that broadcast of a number of channels would still be referred to as ‘tuning’ albeit the process no longer had anything to do with frequencies. This is what Mr Jones said had happened to the meaning of ‘tuner’ and I accept that.

Broadcast/multicast/unicast

31. Mr Jones went on to draw a distinction between broadcast, multicast and unicast. He described broadcast as a signal made available to anyone to receive, if they have the necessary apparatus. Analogue and digital terrestrial TV signals were ‘broadcast’ as were satellite signals, through the air. Multicast was a signal made available to anyone who connected to the appropriate network, whether cable or internet, but could not be received by anyone else. Unicast was a signal which was sent to a single user at that user’s request.

32. Mr Jones saw broadcast and multicast as very similar as both involved the distribution of data by one person to many recipients. There was a distinction in that broadcasts were theoretically available to all as the process involved was sending a signal over ‘airwaves’, although in fact the signals could only be utilised those with equipment which could receive and decode the signal. Multicasts were only available to all those connected to broadcaster’s network. In other words, broadcasting over a cable network was properly ‘multicasting’ (although never described as such) as only equipment attached to the cable network could receive the signal. The expression ‘multicasting’ was in practice only used for IP TV. IP TV broadcasters had a signal which could only be accessed by those authorised to join its network via the Internet.

33. But so far as the broadcaster or multicaster was concerned, it broadcast (or ‘multicast’) the signals once. A user could pick up the signal at any time but would see what all other users would see. A broadcast or multicast intrinsically had no ‘on demand’ facility. If a film was broadcast (or multicast), the broadcaster only transmitted it once, and the user had to watch it from the advertised start time if he wanted to see it from the start.

34. Unicast, on the other hand, was ‘on demand’ by its very nature. It was one to one. While the transmitter might offer a selection of viewings to the user, only the one selected by the user would be transmitted. And it would be transmitted when selected so the user could select the start time. Unicast did not offer the transmitter/broadcaster efficiencies and economies of scale, involving as it did individual transmissions to individual users. A device which enabled a user to watch unicast did not, and did not need to have, an ability to select a frequency or to select and isolate a particular channel. The selection was made by the user, transmitted to the provider, and only the selected data packet would then be transmitted to the user’s computer for display.

35. Mr Jones accepted that broadcasts and multicasts were not identical; one distinct difference was that the multicaster would broadcast into sections (‘segments’) of a network, and if, in one section, none of the users wanted a particular channel, that channel would not be broadcast into that segment. But as soon as one user attempted to select that channel it would be broadcast (multicast) to all users in that segment and available for viewing. Broadcasters did not have the same ability. Nevertheless, it was his view, as I understand it, and which I accept that ‘broadcasting’ and ‘multicasting’ were very similar and I find the terms might be used synonymously in some situations so that, for instance, an IPTV network provider might be referred to colloquially as broadcasting TV whereas technically they were multicasting TV.

Dictionary and other definitions

36. Both parties referred me to dictionary definitions of ‘tuner’ and ‘to tune. HMRC relied on the Oxford English Dictionary and in particular the definition of ‘to tune’. The original meaning seems to be adjusting musical instruments for pitch (which I understand to be about selecting a wavelength). When radio was invented, it gained a new meaning which was to ‘make sensitive’ a radio and then a TV to a chosen signal frequency or wavelength.

37. The appellant does not dispute this. Ms Sloane’s point was that the most recent source used by the OED for these definitions was 1975. While the appellant accepts that tuning a radio or TV to a particular wavelength is still a valid meaning of a ‘tuner’, the word ‘tuner’ has, on its case, gained at least one more meaning since 1975.

38. The appellant referred me to online dictionaries for what it considered to be more modern meanings of ‘tuner’. Ms Sloane pointed out that the WTO panel (referred to below at §82 below) also used a Your Dictionary definition for modem. The definitions I was referred to included as follows:

Cambridge Dictionaries Online

‘the part of a radio or television that allows you to choose the broadcasting station you want to listen to or watch’

Macmillan Dictionary

‘the part of a radio or television that receives broadcast signals’ –

Computer Desktop Encyclopedia

Of TV tuner “the electronic circuit in an analog or digital TV that latches onto to TV channel and filters out the signals” but of

‘tuner’ “an electronic circuit that locks on to a selected carrier frequency (station, channel)…the tuner filters out all but the desired frequency and passes it on to other stages in the device to extract the audio, video or data. See TV tuner.”

Wiktionary

TV tuner “1. a hardware component that allows a computer to receive television signals”

Tuner: “the component of an audio system that receives radio broadcasts”

American Heritage Dictionary

“a device for tuning, especially an electronic circuit or device used to select signals at a specific frequency for amplification and conversion to video, sound, or both.”

Websters New World College Dictionary

“the part of a radio or TV receiver that detects signals”

39. Some of these definitions refer to isolating a frequency; others only require the apparatus to isolate a signal, in most cases a broadcast signal. Overall they support the appellant’s contention that, while isolating a frequency is a possible definition of tuner, in recent years, in common useage, it has not been the only valid definition of tuner, and a device can be a tuner as long as it receives a broadcast of a number of TV signals from which it isolates one channel, even if the broadcast is not transmitted on a frequency.

The law

GIR 1

40. The CN contains general rules of interpretation of itself. These are referred to as GIRs. Both parties were agreed that only the first GIR was relevant and this provided:

“The titles of sections, chapters and sub-chapters are provided for ease of reference only; for legal purposes, classification shall be determined according to the terms of the headings and any relative section or chapter notes….”

The relevant headings (see §4) were ‘video tuners’ or ‘others’ and, as I have said, the parties were agreed that the question for me was whether the products in issue were video tuners.

41. So the question is what the CN meant by the term ‘video tuners’. I was asked to consider various matters:

· Intended use

· Other language versions of the CN

· Explanatory notes to the HS and CN

· EU Treaty obligations

Case law on interpretation – intended use

42. The appellant referred me to the case of Neckerman C-395/93 [1982] ECR 2493. In that case, as with this case, the question was the meaning of an undefined term used in the CN. In that case, it was the meaning of ‘pyjamas’. In this case it is ‘video tuner’. The CJEU said at [7]:

“In the absence of such a definition, the objective characteristic of pyjamas, which is capable of distinguishing it from other ensembles, can be sought only the use for which pyjamas are intended, that is to say to be worn in bed as nightwear.

….

…It suffices if that is the main use for which it is intended”

43. However, I am unable to derive much assistance from this interpretative principle in this case. In that case, the parties agreed on what the intended use ought to be but disagreed over whether that use had to the main use or the exclusive use for which the item was intended. Here, the parties do not agree on the objective characteristics of a video tuner. There is no ‘main’ or ‘exclusive’ use disagreement. If, in order to be a video tuner the apparatus must tune to a frequency, then the IPTV STBs in issue did not do that at all and could not then be video tuners.

44. So this Tribunal needs to determine what are the objective characteristics of a video tuner. While the CN carries no definition, all language versions of the CN are equally valid so I was referred to other language versions.

Other language versions

45. In passing I note that, while in the UK ‘video’ often has or had colloquially the connotation of a machine which recorded programmes off television onto tape (a video recorder or VCR), the word ‘video’ actually refers to the image, or image and sound, output of a TV, and that is clearly the meaning it has in CN and elsewhere in this case.

46. Some other language versions of the CN, such as the German, actually used the English expression ‘video tuners’ and therefore nothing can be gained by considering them. The appellant relied on the French and Spanish versions of the CN, respectively, as follows:

“recepteur de signaux videophoniques (tuners)”

“receptors de senales de imagen y sonido (sintonizadores)”

47. The appellant translated these as, and HMRC did not challenge the accuracy of the translation:

Receivers of video signals (tuners);

Receivers of image and sound signals (tuners)

48. The appellant accepted that in the Spanish version the word in brackets “sintonizadores” translated as ‘tuners’ and again HMRC did not challenge that. Therefore, both language versions could be loosely translated as:

“receivers of TV signals (tuners)”

49. If the word ‘tuners’ in brackets describes rather than limits the definition of ‘receivers of TV signals’ then the French and Spanish versions would cover not just set-top boxes but IPTV streaming devices too (ie devices which receive unicasts). In other words, as most computers connected to the internet are capable of receiving TV signals transmitted over the internet, they could, therefore, say HMRC, on this understanding of the French and Spanish versions of the CN, be described as ‘receivers of TV signals’ and therefore as ‘video tuners’. Neither party suggested that this was a meaning intended by the CN.

50. On the other hand, if the word ‘tuners’ in brackets limited the definition of “receiving TV signals” then I am back to the question faced by the English language version of the CN, and that is what is meant by ‘tuners’?

51. Therefore, I am unable to derive assistance from considering the foreign language versions of the CN, and must look elsewhere for assistance with the meaning of ‘tuner’ in the CN. I was next referred to the explanatory notes of the CN (“CNEN”) and the Harmonised System (“HSEN”)

Explanatory notes to the HS

52. The rate of duty is dictated by the CN. Nevertheless the CN is the EU’s implementation of its international obligations to use the Harmonised System (“HS”). The HS is more general than the CN and does not contain the level of detail in the CN. Nevertheless, notes to the HS (“HSEN”) are accepted by both parties as relevant to interpretation of the CN.

53. HMRC point out that the HSEN 85.21, which contains a definition of recording equipment, describes a ‘tuner’ as:

Something “which enables selection of the desired signal (or channel) from the frequency bank of signals transmitted by the television transmitting station.”

And then at 85.28 (reception apparatus for television) the HSEN provides:

“(1) receivers of television broadcasts (terrestrial, cable or satellite) which do not include a display device…These apparatus receive signals and convert them into a signal suitable for display. They may also incorporate a modem for connection to the Internet.

These receivers are intended to be used with video recording or reproducing apparatus, monitors, projectors or televisions. However, devices which simply isolate high-frequency television signals (sometimes called video tuners) are to be classified as parts….”

54. From this I find that the HSEN does not therefore provide clear guidance on the meaning of ‘tuner’. The definition at 85.21 (first paragraph) refers to frequencies but the definition at 85.28 does not. While the 85.28 definition does refer to high-frequency signals, that is only in reference to an exception to the definition for a distinct part, begging the question whether a complete apparatus which simply receives a broadcast and isolates a particular signal is a tuner irrespective of whether or not it contains such a part. And as Mr Jones’ evidence, which I accept, was that even STBs which receive a frequency often do not now have distinct parts which isolate the frequency (but do all their operation by software), that only makes it less likely the HSEN intended only to define as ‘tuners’ apparatus which contained a distinct ‘tuner’ part.

55. So I derive little assistance from the HSEN; it is not dealing directly with the point in issue, and is somewhat ambiguous, and not binding. Nevertheless, if only weakly, the HSEN supports, the appellant’s case that ‘receiving TV broadcast signals’ is a possible (but not the only) definition of a ‘tuner’.

Explanatory notes to CN

56. Turning to the explanatory notes to the CN, I was shown that the CNEN were amended in 2008 to include a definition of the sub-sub-sub-heading 852871 13 (which was under ‘video tuners’ – see §4). The sub-sub-sub heading was:

57. The definition of this sub-sub-hearing given by the Commission was as follows:

“This subheading covers apparatus without a screen, so-called ‘set top boxes with communication function’, consisting of the following main components:

- a microprocessor

- a video tuner

The presence of an RF connector is an indicator that a video tuner may be present.

- a modem

Modems modulate and demodulate outgoing as well as incoming data signals…..An indication of the presence of such a modem is an RJ 11 connector.

Devices performing a similar function to that of a modem which do not modulate and demodulate signals are not considered to be modems. Examples of such apparatus are ISDN-, WLAN- or Ethernet devices….

The modem must be built into the set-top box. Set-top boxes which do not have a built-in modem but use an external modem are excluded from this subheading (eg a set consisting of a set-top box and an external modem).

The Transmission Control Protocol/Internet Protocol (TCP/IP) must be present as firmware in the set-top box.

Set-top boxes of this subheading must enable the user of the apparatus to access the Internet. The apparatus must also be able to run Internet applications in an ‘interactive information exchange’ mode such as an email client or a messaging application using UDP or TCP/IP sockets.

Set-top boxes which incorporate a device performing a recording or reproducing function (for example, a hard disk or DVD drive) are excluded from this subheading (subheading 8521 90 00).”

58. HMRC’s case on the CNEN is that to be within this category of the CN, the set-top box must include a video tuner, and by that word, it is clear the CN mean something which can tune to a frequency as (says HMRC) that is the only possible meaning of ‘tuner’ in this context and in any event the CNEN itself refers to an “RF connector” and that is a reference to a radio frequency connector.

59. The appellant’s case is that neither the CNEN nor HSEN are legally binding and should be disregarded to the extent that they restrict the scope of a heading within the CN. For this proposition the appellant relies on the cases of Develop Dr Eisbein C-35/93 [1994] ECR I-3989 and Sony Computer Entertainment Europe Ltd T-243/01 [2003] ECR II-4189:

“[21] The Court has stated on many occasions that the [CNEN] constitute an important means of ensuring the uniform application of the Common Customs Tariff by the customs authorities of the Member States and as such may be considered a valid aid to the interpretation of the tariff. However, those notes do not have legally binding force so that, where appropriate, it is necessary to consider whether their content is in accordance with the actual provisions of the Common Customs Tariff and whether they alter the meaning of such provisions.” Develop Dr Eisbein. The same was said at [116] of Sony Computer Entertainment Europe Ltd.

60. The CNEN are not helpful in resolving this appeal. If the CNEN are accepted as offering reliable guidance, then, as I think the appellant accepts, the STBs at issue in this appeal are not ‘video tuners’. This is because the CNEN indicate that a video tuner requires a radio frequency connector, thus indicating that a video tuner is something that receives frequencies. However, that begs the question of whether the CNEN does offer reliable guidance in this context. The CJEU rejects reliance on the CNEN where the notes alter the meaning of the CN. So that brings me full circle back to the question of what the CN meant by ‘video tuners’. If the CN has the restricted meaning put by HMRC then the CNEN is reliable, confirming what would be plain from the CN, that the appellant’s STBs in issue are not video tuners; on the other hand, if the CN has the meaning put by the appellant, then the CNEN are not reliable, as they give a more restricted meaning than the CN, and they can be rejected.

61. So the question remains, what are ‘video tuners’ in the context of the CN? And for the above reason, the CNEN offers no direct help.

62. However, some general points can be made. In BSkyB and Digitalnet the CJEU did reject certain elements of the same CNEN. Nevertheless, no one suggested that because parts of the particular explanatory note in question have been found to be unreliable that that necessarily meant other parts of the same part of the CNEN carried less weight than otherwise.

63. In BSkyB C-288/09 [2011] ECR I-02851 the CJEU concluded that the restriction within the CNEN (see §57 above – last section) which excluded STBs with recording/reproducing capabilities from the duty free category contradicted the CN itself and therefore the CNEN had to be disregarded: [82].

64. Digitalnet C-320/11 (2012) concerned STBs which had been denied duty free treatment on the grounds they were said not to contain modems. They had software which ran the TCP/IP protocol and the software did not modulate and demodulate signals.

65. The CN did not define ‘modem’ but did describe its function as ‘for gaining access to the Internet, and having a function of interactive information exchange’ (§4 above). The CJEU ruled that function was ‘central’ to the definition ([42]) because of this description and because the rules of interpretation applicable to the CN are that ‘the intended use of a product may constitute an objective criterion for classification if it is inherent to the product, and that inherent character must be capable of being assessed on the basis of the product’s objective characteristics and properties.’ [43]

66. The CNEN, however, (see §57 above – third indented paragraph) gave a technical definition to ‘modem’ requiring it to modulate and demodulate signals, thus excluding other devices which also allowed TV signals to be received over the internet but by a different means of internet connection. So the CJEU ruled that the CNEN was wrong:

“[45]…by excluding from the concept of a ‘modem’ devices which fulfil similar functions to a modem because of technical considerations, whereas only the objective of the capacity for gaining access to the internet is relevant for the purpose of classification, the [relevant CNEN] have restricted the meaning of that term. Those notes therefore contradict the CN on that point and must be disregarded.”

67. In other words, the CJEU found that the CN used the term ‘modem’ on the basis of functionality and not technical mode of operation:

“…a ‘modem for gaining access to the internet’ is a device which is capable of accessing the internet and of ensuring interactivity or an exchange of information in both directions. It is solely the capacity to gain access to the internet, and not the technique used to achieve this, that is relevant for the purposes of classification.”

68. The appellant’s case is that Digitalnet offers guidance on how the CJEU would interpret ‘video tuner’. The word ‘modem’, which was used in the CN and CNEN and was the word at issue in that case, is a relatively new word formed by a contraction of ‘modulate-demodulate’. The term was (at least originally) used to refer to apparatus which modulated and demodulated analogue to digital signals and back. Doing so enabled a computer to connect to the Internet. But, as the CJEU found, advances in technology meant that different methods of digital communication evolved, some of which did not involve digital to analogue conversion. Nevertheless, apparatus which permitted a computer to connect to the Internet without modulation/demodulation might still be referred to as a modem. Therefore, the CJEU ruled as set out in the above paragraph. The definition of ‘modem’ in the CN was not limited to apparatus which modulated and demodulated.

69. The appellant drew a parallel with video tuners. Originally, as the name implies, a tuner had to tune to a frequency. Now, says the appellant, the term ‘tuner’ is properly used to describe receiving broadcast TV signals even if that signal is not conveyed on a frequency. The meaning of the word, says the appellant, has moved beyond its original etymology, just like the word ‘modem’.

70. Moreover, says the appellant, it is clear that the CJEU is concerned with the function achieved and not the technicalities of how it is achieved. A modem was something which allowed the computer to access the internet; a video tuner by analogy is something which allows a channel from a TV broadcast signal to be isolated. It does not matter whether the ‘tuner’ tunes to a frequency to achieve that function any more than it mattered whether the modem actually modulated and demodulated signals.

71. HMRC does not agree the comparison is apt. They point out that the CN gave ‘modem’ a functional definition (a ‘modem for gaining access to the internet’) whereas no definition, functional or otherwise, is given for ‘video tuner’. I agree that the CN does not contain a functional definition for ‘video tuners’. There are just two headings, ‘video tuners’ and ‘others’. And while ‘video tuners’ has sub-headings, one of those sub-headings itself is ‘other’ so these sub-headings can’t be used to explain what was meant by ‘video tuner’.

72. So while the CJEU has ruled ‘the intended use of a product may constitute an objective criterion for classification if it is inherent to the product…’ [43] of Digitalnet, the problem for this Tribunal is what the intended use of a video tuner actually is. Is its intended use to receive and isolate TV signals transmitted on a frequency or is its intended use to receive and isolate broadcast TV signals? Digitalnet does not provide guidance on how that question is to be answered because the CN gives no guidance on what the function of a video tuner is intended to be.

73. In conclusion, the CNEN and the cases on other aspects of the same CNEN are of no help in resolving the question before this Tribunal.

Must there be an identifiable ‘video tuner’ part?

74. The appellant’s position is that the CN merely requires the product to be a ‘video tuner’. It does not require the product to contain an identifiable part that is dedicated to the operation of a video tuner. In other words, the appellant’s case is that the STBs in question are not prevented from being video tuners simply because they comprise a processor which runs software, without any identifiable element that is dedicated to receiving and isolating the video signals.

75. The CNEN, on the other hand, appears to indicate that a video tuner will be an identifiable part of the apparatus as it says (see §57 above) a video tuner is something ‘consisting of the following main components’ including a ‘video tuner’.

76. The expert evidence, which I accept, was that many STBs which receive TV signals on a frequency are technologically much more advanced than original radios and televisions and carry out the operation of receiving those signals and isolating the desired channel using software running on hardware. That hardware (a microprocessor) runs all the software used by the apparatus (see §20 above). In other words, even STBs which HMRC appear to accept are within the definition of ‘video tuners’ are likely not to contain an identifiable element dedicated to receiving frequency signals.

77. Mr Donmall’s position on this seemed to be that the CNEN was right and that to the extent it did require an identifiable part operating as a video tuner this was satisfied by the combination of hardware (not dedicated) and dedicated software. However, even if HMRC were right on this, it would not exclude the appellant’s STBs from the definition as they too had a combination of hardware (not dedicated) and software. So the fundamental question remained whether or not the term ‘video tuner’ by definition excluded STBs which did not receive frequency signals.

78. So far as the CNEN was concerned, I consider in so far as the CNEN required a ‘video tuner’ to have an identifiable component that was a video tuner then it did restrict the CN, which only required the apparatus as a whole to be a ‘video tuner’. The better interpretation of the CNEN is that it did not require such an identifiable, dedicated component.

EU’s international obligations

79. I was also referred to an international Treaty and the CJEU’s case law on how that should affect interpretation of the CN.

80. By Council Decision of 24 March 1997 (no 97/359/EC) the European Union approved the Agreement on Trade in Information Technology Products. This agreement was an agreement between the signatories, including the EU, that all customs duties would be eliminated by 2000 on a great many products, including (on page L 155/14 of the Official Journal):

“Set top boxes which have a communication function: a microprocessor-based device incorporating a modem for gaining access to the Internet, and having a function of interactive information exchange.”

81. This definition of STBs entitled to the zero rate does not require the STBs to be or incorporate a ‘video tuner’. HMRC do not want me to actually decide whether, if Amino’s products were excluded from the 0% duty rate in the period 2009-2011 the EU was in breach of its international obligations; they say I should simply precede on the assumption that a failure of the products at issue in this appeal to benefit from the 0% duty rate would be a breach of the EU’s international obligations. Certainly it was the appellant’s case that a decision against them in this appeal would mean the EU was in breach of its international obligations and HMRC did not suggest otherwise. I therefore accept that the STBs at issue in this appeal fell within the above definition in the Agreement on Trade in Information Technology Products and that therefore if I find that the appellant’s STBs at issue in this appeal are not ‘video tuners’ then my conclusion is only consistent with the EU being in breach of its international obligation to remove duty on these products as from 2000.

82. WTO panel report: I was referred extensively to the very lengthy report of a panel of the WTO following a complaint to the WTO about the CN and CNEN on STBs and other items. So far as STBs were concerned, the complaint was that the CNEN excluded certain STBs from the duty free category by excluding STBs with either or both:

· modems which did not modulate/demodulate signals (eg Ethernet modems); and/or

· recording or reproducing capability.

These two issues arose, of course, in the BskyB and Digitalnet cases.

83. There was no complaint to the WTO about the use of the word ‘video tuners’ in the CN nor the definition given to it in the CNEN. As with the Digitalnet case, the complaint (so far as STBs with modems were concerned) was that the the CNEN (see §57 above) defined ‘modem’ as something which modulated/demodulated signals and expressly excluded from the duty free category ISDN-, WLAN- or Ethernet devices on the basis that they therefore were not modems.

84. The WTO panel considered the various dictionary definitions of modem (§7.877-7.880). They noted that the original meaning of modem, still reflected in some dictionaries such as the Shorter Oxford, was that it was a device to convert analogue signals to digital (and vice versa) over a telephone line. However, as I have already said, technology advanced and different methods of digital communication evolved, some of which did not involve digital to analogue conversion. Their conclusion was:

“Given the foregoing, the plain meaning of the term ‘modem’ can include devices other than those that convert a digital signal to analogue for the purposes of information transfer over a telephone line.” (§7.880)

85. They went on to say that the trade agreement was concerned with functionality:

“In our view, the phrase ‘for gaining access to the Internet, and having a function of interactive information exchange’ informs the nature of the ‘communication function’ referred to in the language of the concession preceding the colon”

86. So a modem was something with:

“a communication function enabled through the ability to access the Internet in an interactive manner”

87. In other words, they considered in this context a modem was something which enabled a connection to the Internet through which information could be sent and received, irrespective of whether any modulation or demodulation of analogue signals was involved. They reached in substance the same decision on this as the CJEU in the case of Digitalnet.

88. What is the significance of this report to this case? The report led to a change in the CN. This is Regulation 620/2011 which had effect from 1 July 2011. In brief, the effect of the new regulation was to expressly state that the ‘video tuners’ category included certain devices with recording/reproducing functionality; and to include a duty free category in the ‘other’, non-video tuner category. This has meant that the products at issue in this appeal have been free of duty since 1 July 2011, and this explains why the claim is not for any period after 30 June 2011 (§1).

89. The Regulation was expressly stated not to be retrospective in effect nor to provide interpretative guidance on a retroactive basis. It is therefore irrelevant to the issue in this appeal which concerns the interpretation of the earlier CN.

90. So far as the panel report itself is concerned, it is of no authority in this Tribunal either, neither under UK law nor under EU law. The latter is clearly stated, for example, in [83] of British Sky Broadcasting Group plc C-288/09:

“…the provisions of a treaty such as the ITA are not such as to create rights upon which individuals may rely directly before the courts under European Union law”

91. The appellant relies on it by analogy and in particular considers that ‘video tuner’ like ‘modem’ should be given a functional meaning and that its modern usage of the word should be taken into account.

92. HMRC’s position, as I understand it, is that the WTO panel report led to a change in both the CN and CNEN thus indicating by analogy (as the point on ‘video tuners’ was not considered) that the earlier version of the CN did not allocate the products at issue in this appeal to the duty free category.

93. I consider, however, the report, although I was referred to it extensively, does not really have any relevance to this issue. The panel did not consider the meaning of ‘video tuner’ although inevitably it was a phrase it referred to: this is not surprising as the phrase appears in the CN but not in the Treaty. In so far as it was relied on as an analogy, for the reason given at §§71-72 there is no analogy: in the Treaty (as in the CN) ‘modem’ is given a functional definition. The term ‘video tuner’, which only appears in the CN, is given no definition at all, begging the question of what its function is meant to be. And while the appellant’s products are now accepted to be duty free, as the ‘other’ category is now duty free, that does not answer the question of whether they are video tuners in any event.

94. Finally, the CJEU’s decision in Digitalnet and the WTO panel report do demonstrate that a word can change its meaning over time and in particular that a word whose etymology refers to a particular technical process (such as modulation and demodulation in the case of ‘modem’) can gain a wider meaning no longer tied to its original technical function. Nevertheless, that gives no support by analogy or otherwise to the appellant’s case as the word in question was ‘modem’ and not ‘tuner’. Whether the same process of technological advances cutting a word adrift from its original meaning has happened with ‘tuner’ as with ‘modem’ is something that the appellant must prove has happened. It cannot do that by relying on the CJEU decisions and WTO panel opinions on ‘modem’.

95. Effect of breach of international agreement: What is relevant to the case is the Treaty. As I have said, I proceed on the basis that, if the CN excluded the STB boxes at issue in this appeal from the duty free category up to 1 July 2011, that was a breach of the EU’s international agreement. What is the effect of that?

96. The appellant relies on the case of EC Commission v Germany (re International Dairy Agreement) C-61/94. The facts of the case are irrelevant in this context. The ruling of the CJEU was:

“…the primacy of international agreements concluded by the Community over provisions of secondary Community legislation means that such provisions must, so far as possible, be interpreted in a manner that is consistent with those agreements.”

This ruling has been repeated in other cases, such as Monsanto Technology [2010] C-428/08 and British Sky Broadcasting (above) at [83].

97. The appellant’s case on this simply is that there is more than one possible definition of video tuner, and in order not to breach international treaty obligations the wider, more modern definition should be adopted for the CN. HMRC’s case is that there is simply no such wider definition and that by any definition a video tuner must receive and isolate frequency signals.

98. From Mr Jones’ evidence at §23, it is clear that ‘video tuner’ is not a term used in the industry. It is clear that the CN uses ‘video’ to refer to TV. Mr Jones’ evidence at §25 was that ‘tuner’ would in the industry now be understood to be ‘a means for selecting a specific video or TV channel form a set of available broadcast channels and the process of ‘tuning’ a TV involves selecting one of those TV channels for display’ and in particular would not require the available broadcast channels to be on a frequency, and, his evidence implies, the channels might be multicast rather than broadcast on airwaves.

99. The dictionary definitions referred to §§38-39above support the appellant’s case that it is not a misuse of the English language to refer to an apparatus which selects from a signal containing a number of broadcast/multicast channels, a channel for viewing, even if that apparatus does not receive the signal containing the channels over a frequency. While limiting ‘tuner’ to something which receives the signals over a frequency is still a possible use of the word, I find relying on the expert evidence of industry practice and dictionary definitions reflecting normal usage, that a less restricted meaning is now, and has been for some time, also a proper use of the term ‘tuner’.

Appellant’s definition too broad?

100. HMRC’s position was that ‘tuner’ could not have a new meaning which included IPTV STBs because that would bring into the ‘video tuner’ heading most computers, as most computers were capable of receiving a TV signal via the internet.

101. The appellant’s response was that a ‘video tuner’ must be understood as something which receives TV broadcasts/multicasts and it would not include IPTV streamers which simply receive a data packet of a requested TV channel which has been ‘unicast’. This is, says the appellant, a quite distinct functionality. To be a video tuner the apparatus must have capacity to receive a broadcast or multicast signal, whether or not sent on a frequency, and then isolate from that signal including many TV channels the one particular channel sought by the user and transmit only the video output for that channel to the TV which then displays it to the user. Ordinary computers do not have the software or the licences which enables them to do this.

102. HMRC’s response to this is that they consider IPTV STBs and internet streaming to be a distinction without a real difference. From the user’s perspective, Mr Donmall said, the experience is virtually identical. In both cases the user selects a channel and is able to watch it.

103. I do not agree that the evidence shows that the experience is identical for the user. On the contrary it is clear that §33) using an IPTV STB to select a channel inevitably limits the user (subject to recording ability) to watching the programmes when actually broadcast/multicast: using a computer to watch programmes (whether available free from, say, BBC’s iPlayer or from a pay-to-view operator) permits ‘on demand’ viewing albeit that a user might choose to watch the programme live (eg if a sports match)(§34). From the point of view of the operator, the evidence is that the position is different: there are economies in multicasting that do not exist with unicasting (see §§32-33).

Conclusion

104. ‘Video tuner’ should be defined by its function but there is no help in the CN as to what that function is. While it is possible that the function of a tuner is to select a frequency, it is, and has been at least since the advent of IPTV, a valid use of English language to refer to a tuner as acquiring broadcast TV signals from any source, whether or not transmitted on a frequency. As the EU’s international obligations required the EU to give a duty free rate to IPTV STBs as from 2000, and that only this wider definition of ‘video tuner’ would under the CN in force 2009-2011 put IPTV STBs within a duty free category, then the CN must be interpreted on the basis of that wider definition.

105. It seems to me that that definition must be adopted even if (as HMRC claims) it would lead to more things gaining the duty free rate than IPTV STBs. However, I do not agree that that is the outcome of using the wider definition which requires the apparatus to receive a broadcast (or multicast) TV signal from which it isolates a single channel. I do not think such a definition would cover IP streaming and unicasts.

106. Adopting this wider definition of ‘video tuner’ for the CN means that the appeal is allowed.

107. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.