[First published as a post on 12 May 2016]
The EEA-EFTA nations have to adopt new single market EU acquis without having a vote on its enactment in either the EU Council or the European Parliament. Clearly, this is a major drawback to the EEA-EFTA (or ‘Norway’) Brexit option. But equally clearly, just how big a drawback this is depends greatly on how much of the total EU acquis we would have to adopt, or perhaps more to the point, on how much of the acquis we would have to adopt compared to how much we currently adopt, given our various opt-outs.
Some of the estimates that have been given are:
6.5%, from David Oddson, Icelandic Minister for Foreign Affairs, 2005. 1
9%, from Nei til EU, the Norwegian anti-EU campaigning organisation. See their articles here and here.
21% from Richard North here.
28% from the 2012 Norwegian government report ‘Outside and Inside’, at page 795 of the Norwegian edition.
‘Approximately three-quarters’, from the same Norwegian report, at page 6 of the official English version of chapter 1, here and here.
80%, from Halldór Ásgrimsson, Icelandic Minister for Foreign Affairs, 2003. 2
There are three main reasons for the divergence in figures:
The 9%, 21% and the 28% are all estimates of the proportion of legislative Acts adopted by EFTA-EEA states. The three-quarters derives from a count of the number of legislative sectors, not of Acts.
Richard North counts all legislative Acts, whereas the 28% from ‘Outside and Inside’ is a proportion of Directives and Regulations only.
Since old acts are sometimes replaced by new ones, the count may be either of the total number of acts adopted over a period of time (the gross number), or of the total number of acts currently in force (the net number). Both the 21% and the 28% estimates are of the proportion of Acts currently in force.
So far as I understand it, all these estimates are of the proportion of EU law that has actually been adopted. However, the EEA Agreement, notably under Article 1(2).f concerning cooperation in fields outside the four freedoms, allows for legislation to be adopted voluntarily. What we are most interested in is the proportion of legislation that we would be forced to adopt. The type of legislation that has been adopted on a voluntary basis has generally been in the form of Decisions, at least prior to 2010. 3 Excluding these could reduce Richard North’s 21% figure to some degree, but would not affect the 28% figure, for which Decisions were not counted, and which is based on 2008 figures.
The 28% estimate makes use of a 2009 study by Yves Bertoncini, entitled ‘What is the impact of EU interventions at the national level?’. Bertoncini provides the following table, (which is to be found also, in a Norwegian version, at page 794 of ‘Outside and Inside’):
Bertoncini acts table
It may be seen immediately that Decisions constitute the largest category of Acts, so that their inclusion or exclusion is a matter of some importance. According to the European Commission, a Decision ‘only deals with a particular issue and specifically mentioned persons or organisations.’ They are not generally applicable to EU states. It would prove nothing to show that an EEA-EFTA state like Norway only adopts a small proportion of EU Decisions, since the same is likely true of EU states too. So these are better excluded, as they are in Bertoncini’s total for ‘Internal Normative Acts’ of 9685 in the above table.
Bertoncini, rightly in my opinion, also excluded ‘Other Acts’ (including Opinions, Recommendations, Common Positions, et al), since these are not binding.
Finally, he excluded International Agreements. This was reasonable for his study, which concerned ‘the impact of derived law on internal law’. It is not obvious to me that they should be excluded for our purposes, which concern the extent to which the EEA-EFTA states have to adopt EU law. Sejersted (the Chairman of the EEA Review Committee which produced the ‘Outside and Inside’ report) also excluded them (page 795), but I am not sure that he was right to do so. After all, it is precisely their greater freedom to make international agreements in their own right, that is one of the features of the position of EEA-EFTA countries that is attractive to those who want to see national sovereignty restored to the UK. And the counterpart to this is the freedom not to be party to agreements made on our behalf by the European Union. So for me, these agreements should probably be included in a measure of the extent to which the EEA-EFTA nations are subject to EU law.
Sejersted took Bertoncini’s figures for EU Directives and Regulations as of 1 July 2008, and also obtained figures from the EFTA Secretariat for legislative acts adopted under the EEA Agreement and in force as of 31 December 2010. He points out that since there is a time delay of approximately 8-12 months between the adoption of acts by the EU and in the EEA, the time discrepancy is not as serious as it otherwise would be. He then presents the following table for Directives (Direktiver) and Regulations (Forordninger) in the EU and in the EEA (EØS):
The table shows the EEA-EFTA states as having adopted about 70% of EU Directives and 17.5% of EU Regulations, or 28% of the total number of Directives and Regulations combined. This therefore is the source of the fourth of the estimates listed above.
If International Agreements were to be included then the total number of EU Acts to be counted would then be 9685 + 3959 = 13644. I do not think that the EEA-EFTA states can be party to any of the acts listed as EU International Agreements. Then the number of EEA Acts as a proportion of EU Acts would be 2718 ÷ 13644 ≈ 20%, an appreciably lower figure.
Quantitative measures, based on raw numbers of legislative Acts, are of extremely limited value in showing the extent to which EEA-EFTA nations are obliged to adopt EU law. Quite simply, one Act is not of equal weight to another, and indeed a single piece of major legislation may be of greater political, social or economic importance than a great number of minor regulations. Some indication of the nature and scale of the problem is provided by the following table of Bertonscini’s (pp 17-18), which gives the number of EU Directives and Regulations in each sector, as of 1 July 2008. Acts are assigned to each sector which they impact upon, with the result that there is a degree of double counting, with the result that the numbers add up to 11,427, whereas the total number of Acts is 9,685 as above. The percentages given are with respect to the inflated totals, and so may be under-estimates in sectors where there is little double counting.
Acts by sector 1
Acts by sector 2
Immediately striking is the fact that 4,821 Directives and Regulations in force, 42.3% of the (inflated) total, were in the Agriculture sector, while another 703, or 6.2% of the total were in Fisheries, so that almost half (5524) were in Agriculture and Fisheries combined. In contrast, there were only 55 Directives and Regulations in force in the Energy sector. Is the impact of EU legislation in Agriculture and Fisheries anything like one hundred times greater than in Energy, or is Sejersted correct when he says (p. 798, first column, line 2) that there is no ‘correlation’ (sammenheng) between how important a sector is in the EU and the number of Acts?
Personally, I am inclined to think that the degree of correlation is very low. It seems to me that a single clause of one Article of one Act can have more impact on the national life than a thousand minor regulations. To take one example, Article 6 of the Working Time Directive (2003/88/EC), which reads in part:
Member States shall take the measures necessary to ensure that …
(b) the average working time for each seven-day period, including overtime, does not exceed 48 hours.
has had a substantive effect on the lives of millions of working people in the EU. In contrast, a whole Act may be devoted, for example, to the addition of another single biochemical substance to the list of those which are authorised. An arbitrary instance would be Commission Regulation (EC) No 1750/2006 of 27 November 2006 ‘concerning the authorisation of selenomethionine as a feed additive’.
Also, there are many Commission Regulations which are in effect subsidiary to Council Directives. An arbitrary example, taken from the same sub-sector of Plant Health, would be Commission Regulation (EC) No 645/2000 of 28 March 2000
setting out detailed implementing rules necessary for the proper functioning of certain provisions of Article 7 of Council Directive 86/362/EEC and of Article 4 of Council Directive 90/642/EEC concerning the arrangements for monitoring the maximum levels of pesticide residues in and on cereals and products of plant origin, including fruit and vegetables, respectively
This is not of course to belittle such legislation, or to denigrate the labours of those who formulate it, but I do think that a considerable number of Acts of these types could be weighed in the scales of national impact against a single major Directive or Regulation.
In similar vein, Eiríkur Bergmann of Bifröst University in Iceland, points out in his 2011 Report ‘Iceland and the EEA, 1994-2011’ (pp 12-13) that:
the variation of the scope covered by individual legal acts is huge
and with regard to the contrasting estimates of 6.5% and 80% made by Icelandic Foreign Ministers, argues that:
This statistical exercise performed by the two foreign ministers is meaningless to measure how involved Iceland already is in the European integration process.
The Iceland Progress Report
Iceland’s membership application to the EU provided Sejersted’s EEA Committee with an alternative way of measuring the level of Norway’s integration with the EU. The Accession Negotiations were divided into 35 Chapters, corresponding in large degree to the sectors made use of above, or of sub-sectors thereof. For each Chapter, an assessment was made of Iceland’s readiness for EU membership in that area of national life, and of the degree of alignment of her law with the EU acquis. The results were published in the Iceland 2010 Progress Report, and subsequently in revised form in the Iceland 2012 Progress Report.
For many Chapters, each Report draws attention to the role of the EEA Agreement in aligning Iceland with the acquis. Section 4.1 on ‘Chapter 1: Free movement of goods’ begins (2012 Report, p. 16):
Iceland has already reached a high level of alignment and, with a few exceptions, applies the acquis in this field due to its EEA membership.
Scetion 4.3 on ‘Chapter 3: Right of establishment and freedom to provide services’ begins:
Iceland has already reached a high level of alignment and applies a substantial part of the acquis in this field due to its EEA membership.
and specifies where there is less than complete adoption of EU legislation.
Section 4.12 on ‘Chapter 12: Food safety, veterinary and phytosanitary policy’ begins (2012 Report, p. 23):
Iceland partially applies the acquis on food safety, veterinary and phytosanitary policy due to its EEA membership.
and goes on to detail areas in which there was divergence in legislation or policy.
Section 4.16 on ‘Chapter 16: Taxation’ begins:
Iceland partly applies the acquis in the area of taxation.
Although there is partial alignment with the acquis, it is not stated that this is due to EEA membership.
Section 4.11 on ‘Chapter 11: Agriculture and rural development’ begins:
Iceland’s agricultural policy is not in line with the acquis.
In line with these five examples, I have divided the 33 Chapters into 5 categories, and tabulated them. In the first three, Iceland’s EEA membership has caused her to be aligned with the EU acquis, to a high degree, substantially, or partially. In the fourth category there is a greater or lesser degree of alignment, but this is not said to be due to EEA membership. (For Chapter 24, ‘Justice, freedom and security’, the alignment is said to be due to Schengen membership.) In the last category, which contains Agriculture and Fisheries there is said to be no or little alignment.
Bergmann provides a similar table in a 2012 article in the EFTA Bulletin: 4
Bergmann Chapter table
He claims that the table represents the degree to which Iceland has adopted EU law ‘through the EEA and the Schengen Agreements’. It is identical, apart from appearance, to Table 1 in his above-mentioned paper, and there he states that it is based upon the 2010 Iceland Progress Report. A few comments may be in order.
He does not have a category for partial participation. In particular, he places ‘Chapter 12: Food safety, veterinary and phytosanitary policy’, for which, according to the 2010 Report (p. 24) ‘Iceland is partially in line with the acquis’, in his ‘Mostly Participates’ category, which seems to me to be over-stating the degree of participation somewhat.
Since his middle category is ‘Mostly Participates’, one naturally assumes that ‘Participates’ is more or less full participation. But for Chapters 3, 4, 6, 7, 9 and 10, which are placed in the ‘Participates’ category, the 2010 Report says only that Iceland applies a ‘substantial part of the acquis’.
It is to be noted that Bergmann includes Iceland’s adoption of EU law through the Schengen agreement as well as through the EEA. He is therefore justified in placing ‘Chapter 24: Justice, Freedom and Security’ in his ‘Mostly Participates’ category.
With regard to ‘Chapter 23: Judiciary and fundamental rights’ the 2010 Report does not state that Iceland’s acknowledged high standards have any connection with either the EEA or Schengen. It seems to me therefore that for this Chapter, Iceland does not adopt ‘EU law through the EEA and Schengen Agreements’, and that it should therefore be placed in his ‘Does not participate’ category.
He is justified in placing ‘Chapter 20: Enterprise and industrial policy’ in his ‘Mostly Participates’ category, since the 2010 Report states that Iceland’s adoption of a substantial part of the acquis is due to is EEA membership. The 2012 Report spoke of ‘a high level of alignment’ but did not make any reference to the EEA, and for this reason I did not feel able to include this Chapter in any of my three EEA categories. I do note that Enterprise Policy is listed by EFTA as one of the Horizontal and Flanking Policy areas for which there is a ‘legal basis for the incorporation into the [EEA] Agreement of EU legislation’, but on the other hand, EFTA does not make reference to the adoption of EU legislation for this policy area into the EEA Agreement Annexes, in the way that it does for the Environment or Consumer Protection for example. I am therefore inclined to think that it may better left out of the EEA categories.
Bergmann summarises his analysis with the statement that:
Iceland already participates either fully or mostly in 23 out of the 33 chapters covering EU legislation, the acquis communautaire
noting that ‘chapers 34 and 35 do not cover any specific policy areas’. 5
An almost identical table appears again in Sejersted’s ‘Outside and Inside’ at page 799:
The Chapters are placed in exactly the same columns as they are in Bergmann’s table. The only substantive difference that I can see is that the ‘Delvis’ of the middle category translates as ‘partially’ rather than ‘mostly’.
Sejersted follows Bergmann in excluding Chapter 34 and 35, but goes further by leaving out Chapters 32 and 33 also, on ‘Financial control’ and ‘Financial and budgetary provisions’, as being of relatively minor importance. Accordingly, at page 803 of ‘Outside and Inside’, he concludes that Norway is fully or mostly aligned with the EU in 23 out of 31 areas (rather than Bergmann’s 23 out of 33), and that this gives the proportion 3/4.
23 ÷ 31 ≈ 74%, so it seems that Sejersted is giving the same value to the 11 Chapters in his middle column as he does to the 12 Chapters in the left-hand column. In addition, as I explained above, many of the Chapters place in the left-hand column seem to belong in the middle column. So I think that there is a degree of exaggeration in the result. It is important to note also that he is claiming this as a measure of the actual extent of Norway’s integration with the EU, and not of the extent to which Norway is obliged to adopt EU law through the EEA specifically.
I am inclined to think that the variation in weight and significance between one legislative Act and another is so great that measures based on the number of Acts adopted are of very limited usefulness. I prefer the sectoral approach and think that the Iceland Progress Reports provide a valuable assessment of the extent to which EEA nations have become aligned with EU law through their participation in the EEA.
Clearly, the same type of objection, that one sector is not of equal weight or significance to the national life to another, can be levelled against a quantitative measure based on a count of the number of sectors in the different categories. Perhaps the best approach is simply to read through the table and put one’s own weight on each Chapter, bearing in mind whether there is said to be full, substantial or partial alignment with the acquis. Here is my table again, which I believe keeps more closely to the actual wording of the Iceland Progress Reports:
For myself, I would say that under the EEA, we would have to adopt around half of EU law. Given our opt-outs from the euro, Schengen, and the area of Freedom, Security and Justice, I would say that we currently are required to adopt around, say, 80% of EU law. So, if we were to leave the EU, join EFTA, and remain in, or re-join the EEA, I would say that we would have to continue to adopt well over half, possibly getting closer to two-thirds, of what we currently adopt.
- ‘Outside and Inside’, Chapter 13, English translation (Official Norwegian Reports) p. 15 ↩
- ‘Outside and Inside’, Chapter 13, English translation (Official Norwegian Reports) p. 15 ↩
- Email from EFTA Secretariat Brussels, 3 May 2016. ↩
- Eiríkur Bergmann, ‘Iceland’s Position in the EEA’ (EFTA Bulletin, September 2012) p. 14. Link. ↩
- Bergmann, ‘Iceland’s position’, EFTA Bulletin 2 – 2002, p. 13 ↩