Impact Assessments in the EU: room for improvement? - European Union Committee Contents


Influence on Commission thinking

62.  The justification for the production of impact assessments by the service which is also developing the associated proposal is that the two processes can feed into each other and policy development can take account of findings from the IA at an early stage. The Government raised a concern about the production of IAs by consultants, rather than in-house. They argued that it "brings risks, as it separates the development of an impact assessment from the development of the associated policy proposal. The Government believes that more should be done to ensure that the two processes (IA and policy development) are brought closer together" (p 44).

63.  The Impact Assessment Guidelines suggest that policy leads "should use [their] IA actively when presenting the merits of the proposal during the legislative process."[22] The Commission described how "[a]s Commission services become increasingly aware of the necessity to start the impact assessment work early, the influence of the impact assessment process on the content and quality of the proposals will increase accordingly" (p 25). Professor Radaelli argued that IAs are produced early enough to "stimulate learning among the different Directorates General" (p 2).

64.  As to the type of learning and influence stimulated by an IA, Professor Radaelli described "a kind of a trade-off between impact assessment as the document that contains only evidence and objectivity and impact assessment as support and justification for what an organisation wants to do" (Q 2). Malcolm Harbour MEP argued that "many impact assessments seem to be moulded to fit the Commission thinking rather than the other way around" (p 59).

65.  As early drafts of legislative proposals and impact assessments are not published, it is difficult to determine exactly how a proposal has been changed in the light of the IA. However, some of our witnesses suggested that the IA rarely leads to a proposal's being dropped. Open Europe said that they "have identified only three cases where an EIA [European Impact Assessment] has actually led to a proposal being aborted" (p 64). Professor Radaelli suggested that often "the 'do nothing' option and the 'non regulatory' option are sandwiched and superficially appraised only to show the superiority of the chosen regulatory option" (Q 6). The Government suggested that "there is little evidence that inclusion of the do-nothing option leads to its selection as the preferred policy option" (p 44). However, the Commission pointed to a proposal regarding witness protection, which had been included in the 2007 CLWP, in which the IA concluded action was not advisable, the findings of the IA being subsequently published as a Commission Working Document[23] in November 2007 (p 26).

66.  Analysis of the effect of IAs on Commission thinking would be difficult given that early drafts of IAs and legislative proposals are unpublished. The "do nothing" option will inevitably lead to a proposal's being dropped only rarely, due to the parallel production of the impact assessment and draft legislation. A comparison of Roadmaps and adopted draft legislation might shed some light on the issue, as would a comprehensive assessment of the weight given to the "do nothing" option in Commission IAs.

Influence on UK Government thinking and the production of UK IAs

67.  In addition to the EU impact assessment, the Government also produce their own, targeted at the impacts at UK level. These are a vital part of the process of parliamentary scrutiny. While we were conducting this inquiry we also scrutinised a proposal for a Regulation on Biocides, for which the UK impact assessment seemed to have been produced in an unusual way. That IA calculated the costs and benefits to the UK by using the figures provided by the Commission divided by six (because the UK has a sixth of the EU biocides market). We invited officials from the Health and Safety Executive to give evidence to this inquiry to explain why they had produced the IA in such a way, and, more generally, how they are normally produced, and subsequently used, at UK level.

68.  Steve Coldrick explained that the UK assessment was produced using those figures because the proposal "came forward a lot quicker than we actually anticipated, in the context of the intelligence we had had at that time" and that the IA was intended to inform a UK-level consultation exercise aimed at producing a more developed IA (Q 78). The consultation invited comments on those extrapolated figures and Mr Coldrick explained "the use we have made of the impact assessment firstly is in fact to help us get out as comprehensive a consultative document as possible, within the short time that we had. Secondly, to gauge the validity of the Commission's position and its assumptions". Within the HSE, such direct extrapolation from Commission figures had not been used for around 10 years.

69.  The production of the final UK IA on the Biocides Regulation is in hand, and we look forward to seeing it. However, the consultation process has not been straightforward, as Lord McKenzie of Luton, Parliamentary Under-Secretary of State for Work and Pensions, explained in his letter to us of 13 November[24]. We asked Robin Foster of the Health and Safety Executive why he thought the consultation had proved difficult. He said:

    "We elicited views from our stakeholders in two separate ways. We had a consultation exercise, and we had an open meeting of stakeholders, which was actually very positive and very useful … When it came to the section on impact assessment, I was there really trying to encourage people, 'Tell us what you think, give us information that we can use to improve it'. Although they had been really helpful throughout the rest of the meeting, in contributing to issues about the biocides directive, we just could not enthuse them, they had very little to say" (Q 83).

70.  We recognise that the Biocides Regulation was something of an unusual case and welcome the fact that the Health and Safety Executive are currently gathering their own evidence in preparation of the UK impact assessment. Clearly, the EU assessment should inform the Government's own impact assessment; but it is crucial that the Government assess the quality of data in the IA and gather their own data where appropriate, rather than simply taking the Commission's data and shrinking them to fit.

Impact Assessment of amendments to proposals

71.  The 2005 Inter-Institutional Common Approach to Impact Assessments, which is currently under review, states that the European Parliament and the Council will take into account the Commission impact assessment when examining proposals and that they will "carry out impact assessments, when they consider this to be appropriate … prior to the adoption of any substantive amendment".[25]

72.  Almost all our witnesses suggested that the European Parliament and the Council are not taking impact assessment as seriously as they should. The Government said "[b]oth Institutions have committed to assessing the impact of their substantial amendments, however this is not happening consistently in the Parliament, and not at all in the Council" (p 43). Professor Radaelli said that "[a]t the moment the common approach to IA of the three main EU Institutions is not working well. It is there in terms of desire but implementation is random, I would say" (Q 5).

73.  The Minister told us that, while the Commission had developed expertise in producing impact assessments, "I do not think that capacity is particularly there within the European Parliament as yet" and, in regard to the Council, "I am not sure whether the individual Member States would be seen as sufficiently impartial to be able to produce their own impact assessments, although I think that is better than having no impact assessment, quite frankly. So I think the source of the impact assessment is perhaps less important than the fact of the impact assessment" (Q 117).

European Parliament

74.  There are two aspects to the European Parliament's use of impact assessments. The first is the attention they give to the Commission assessment of each proposal. The second is the production of their own assessments of amendments they might wish to propose.

75.  Sub-Committee A of this Select Committee took evidence on the Alternative Investment Fund Managers proposal from Sharon Bowles MEP, Chair of the European Parliament Economic and Monetary Affairs Committee. That committee had examined the Commission impact assessment and found it wanting. They therefore commissioned their own assessments of the original, unamended, proposal. The first IA dealt with marketing and distribution issues, while the second was a cost-benefit analysis of the changes proposed by the draft Directive. Ms Bowles explained that her committee were not satisfied with the Commission IA as it had clearly been prepared in haste: "Several people mentioned this when we debated it in committee and because there was enough money in the budget[26] we decided that we would have an impact assessment of our own"[27]. Ms Bowles explained that the assessments did not take into account any amendments by the committee (which, at that point, had not been proposed) but suggested that "nevertheless it may well point to the way forward in terms of what those changes should be"[28].

76.  Malcolm Harbour, Chair of the Internal Market Committee (IMCO), reported similar activity:

    "IMCO has undertaken Impact Assessments, where the Commission have failed to produce one. For example, on the Consumer Credit Directive, the committee commissioned a study on the 'broad economic impact' of the proposal[29], because the Commission had refused to provide a new impact assessment for their modified proposal" (p 60).

77.  However, Mr Harbour argued that European Parliament committees are not consistently taking account of Commission impact assessments. He told us that "Parliament committees are still not taking their commitment to better regulation seriously" (p 60) and that "there is still some reluctance by all Committee Chairs to examine an impact assessment" (p 61). Although we have seen evidence that Mr Harbour's and Ms Bowles's committees are examining Commission impact assessments, it is not clear to us how often this happens across the Parliament as a whole.

78.  Mr Harbour suggested that the Parliament should not even consider proposals for which a comprehensive cost-benefit analysis had not been carried out by the Commission (p 61). The Federation of Small Businesses made a similar suggestion that any document which had not received an adequate impact assessment "should be sent back to the previous stage" and not considered by the Parliament (p 13).

79.  The Government argued that "the Council and the European Parliament should play a greater role in holding the Commission to account when it fails to comply with its guidelines and produce an impact assessment of sufficient quality" (p 43).

80.  In the Inter-Institutional Common Approach, the Parliament committed to carrying out assessments of its own substantive amendments, although it was left to the Parliament to determine, at a political level, which amendments it regarded as "substantive".

81.  The Parliament may commission an impact assessment at any stage of its consideration of a proposal. At first reading, this could be done before amendments have been adopted in a committee, between committee adoption and adoption by the plenary, or after adoption by the plenary. Assessments of amendments could also take place at second reading.

82.  Malcolm Harbour reported that IMCO had carried out such IAs, giving as an example the Nominal Quantities for Pre-Packed Goods proposal (p 60)[30]. However, it seems that this is a rarity. By December 2008, it seems that only seven such assessments had been carried out[31]. Professor Radaelli pointed out that the European Parliament had previously not had the capacity to carry out assessments, although "some capacity has been built … over the last two or three years or so" (Q 5). Currently, it seems that EP impact assessments are almost exclusively commissioned from external consultants, as in the in the examples cited by Malcolm Harbour and Sharon Bowles.

83.  Dr Italianer explained the Commission's position: "we think that when there are important amendments … it would also be of use for members of the Parliament and for Council ministers, for that matter, to use the same techniques to analyse what the impact of the various amendments would be" (Q 63). The Minister agreed that it is "hugely important" that IAs are performed on significant amendments because "the assessments have to be made on the proposal that is actually going to be implemented" (Q 116).

84.  Professor Radaelli spoke of the scale of the task assigned to the Parliament, saying it "is supposed to perform a task that is much more ambitious than the task asked of other parliaments of the EU, like the national parliaments. Note that we are asking the EP to perform an impact assessment of its own substantive amendments. That is a tall order" (Q 7). He suggested that the need was not for the EP to replicate the IAs of the Commission, "recalculating costs and benefits when the EP introduces substantive changes: I would say the EP would do this IA only to its own advantage", but that assessments should be performed on "whether the overall logic of intervention has been modified, whether the regulatory logic behind the option has been altered by the amendments" (Q 8).

85.  There was some suggestion that the Commission could make use of its capacity for producing IAs to assess certain of the Parliament's amendments. Dr Italianer thought this might be feasible, particularly where amendments touch on technical matters. He said that "[t]he Commission has already offered to look at such requests … Where for instance the changes in the Parliamentary process are changes in certain parameters that can be easily simulated with economic models, then that could certainly be done" (Q 64). He suggested that the other Institutions were not inclined to take advantage of this offer, because "they might perhaps have had the idea that the analysis might not be completely unbiased" (Q 64).

86.  The Minister made the same point, noting that "the suggestion may be made—certainly not by me—that [the Commission] might try to defend their earlier position rather than give informed advice on an amendment" (Q 116). However, the Government would welcome it if the Commission could play such a role (Q 116). Professor Radaelli warned about a "confusion of roles" which might be created if the Commission started undertaking such assessments "especially for amendments that go against the initial proposal of the European Commission; it would be either legal contortionism, a kind of torture, or a fiction for the policy officers of the Commission" (Q 8).

87.  We welcome the actions of certain European Parliament committees in commissioning impact assessments where they feel the Commission assessment is inadequate. This would constitute a "political" holding to account, in addition to the role of the Impact Assessment Board in assessing the adherence of IAs to the Guidelines. The relationship between the two may warrant further study.

88.  We are, however, concerned to hear that the Parliament may not be consistently taking the Commission IA into account when considering proposed legislation.

89.  The European Parliament does not seem to be adhering as fully as possible to its commitment in the Inter-Institutional Common Approach to perform impact assessments of substantive amendments. We recognise that this is an ambitious task and that the Parliament may not have the resources to do this consistently. However, we would encourage the Parliament to perform assessments in a proportionate manner where appropriate.

Council of Ministers

90.  As with the Parliament, the Council is also supposed to produce assessments of its proposed amendments. The Government told us that this is not happening at all (p 43). Given that the Council does not currently produce impact assessments of its amendments, we wondered what use Member States made in Council of the Commission impact assessment and of their own assessments.

91.  The Commission's written evidence referred to "the recent initiative by the Czech Presidency to promote the use of IA in Council Working Group discussions, following a similar initiative by the Austrian Presidency in 2006. The current Swedish Presidency has also indicated that it will make greater use of Commission impact assessments" (p 25). Other witnesses suggested that there is little discussion of IAs in the Council and working groups. The Government said that "[i]n Council working groups and committees of the European Parliament, debates regarding impact assessments remain the exception rather than the rule" (p 43), while Professor Radaelli said "working parties are too focused on negotiation and bargaining to switch to the evidence-based logic of impact assessment and take it seriously" (p 3).

92.  Robin Foster of the Health and Safety Executive confirmed this opinion. In regard to the negotiations on the Biocides Regulation he reported that:

    "[t]he Swedish Presidency invited comments [on the Commission IA], the UK submitted a six-page document, broadly critical of the impact assessment, but acknowledging all the work that had been done. Denmark submitted a one-page document broadly supportive of what the UK said … and that was it. When it came to discussing the impact assessment in the Council Working Group, this is the Environment Council, there was very limited discussion" (Q 78).

93.  As some Member States produce their own national impact assessments of proposals, we were interested to know whether these could, in some sense, be regarded as de facto assessments of certain amendments the Member States wished to propose. Dr Italianer said that to the extent it serves as a "pars pro toto[32] … it could give a good indication of an amendment and so in that sense I think it could help the decision-making in the Council" (Q 65). He warned, though, that "there are some policy proposals where there is a distribution issue between Member States". Citing a proposal to reduce CO2 across the EU he continued:

    "there was a distributive issue involved there: who is going to bear the cost; who has to achieve which CO2 reductions? In that case it is a little bit of a zero sum game. If one Member State has to reduce more than another Member State … in that case an individual impact assessment would perhaps be of less use because it would be to the detriment of another Member State" (Q 66).

94.  In reality, it seems that neither Commission nor Member State impact assessments are much discussed in Council. Robin Foster told us that "[w]e will certainly use [the UK IA] to formulate our amendments" but "we will be rather cautious about deploying cost-benefit assessment in the negotiating meetings themselves, and the reason for that is experience, which says that there are many Member States who will visibly recoil if the UK advances pure cost-benefit arguments in the meetings" (Q 81).

95.  When we asked the Minister for Better Regulation if impact assessments should be discussed in working groups and the Council he said, "I would be concerned if they did not" (Q 120). The Government subsequently provided us with two recent examples of the use of IAs in working groups: the Implementing Measures for External Power Supplies and for Simple Set Top Boxes (p 53). The Minister did, however, acknowledge that it is not always easy to discuss a proposal in such terms, stating that "I think that that creates some disagreements sometimes with other European countries because they do not think that that numerical cost-benefit analysis is as valuable as we think it is" (Q 99).

96.  The Minister told us that the Government had been trying to promote greater use of impact assessments in the Council and working groups and that he had been in contact with his German counterpart with regard to working more closely on Better Regulation. He also told us that the Government had met with European Parliament committee chairs (QQ 102 & 109).

97.  We conclude that the Council should produce impact assessments, but understand the political hurdles involved. With regard to Member State impact assessments, with due caution about their possible applicability Europe-wide, we would be interested to see how they might be used to assess amendments proposed by Member States in Council.

98.  We are concerned about the apparent lack of discussion of impact assessments in working groups and the Council. We urge the Government to promote their use further.

Ex-post evaluation

99.  The impact assessment process takes place at the beginning of policy formulation, but better regulation should also take into account the evaluation of regulation once it has been implemented. The Council of Ministers acknowledged this in May 2009, when it called for the Commission to provide for ex-post evaluation and "undertake comparison of intended and actual effects of approved EU legislation"[33].

100.  The FSB argued that "[p]ost implementation reviews of EU legislation should therefore happen in every Member State" (p 14) while Malcolm Harbour stated that "[l]egislation should be viewed by the European Commission as a circular process, in which ex-post audits should be an indispensable requirement" and that "the Parliament's committees should make better use of their resources, by providing some of the qualitative and quantitative evidence needed to review the impact of legislation" (p 60).

101.  The Minister told us that ex-post evaluation is not "happening as much as it needs to" but also suggested that "[i]t is something that we in the UK Parliament do not do very well, and that is assess the legislation that we have already passed[34] and really ensure that the effect that it was intended to have it has had and, if it has not, why not. I think our culture is such that we do not really think as politicians in those terms" (Q 122).

102.  Dr Italianer told us that the Commission are "strongly encouraging our departments to engage in ex-post evaluation because whenever they are reviewing legislation for which they are responsible, the natural starting point should be ex-post evaluation before they go into a new impact assessment" and that "I would expect that several years from now many of [the impact assessments] would be the subject of an ex-post evaluation when it comes to the policy review process" (Q 74). He also spoke of completing the "policy circle" in order to "get into place this chain of ex-ante and ex-post evaluations" (Q 76).

103.  We welcome the Commission's commitment to carrying out ex-post evaluations as they are crucial to improving legislation. However, it was not clear how often the Commission was carrying them out. We recommend that further work be done to determine their current use and how they are integrated into the legislative cycle. We urge the Government to promote the further use of ex-post evaluation and encourage the Commission to disseminate their evaluations more widely.

22   Op. Cit. p 11 Back

23   COM (07) 693 Back

24   EU Sub-Committee B, Correspondence with Ministers: Back

25   Inter-Institutional Common Approach to Impact Assessment (2005), paras 13-14: Back

26   In 2006 a specific budget line was created, initially allocating €500,000 to European Parliament committees to fund impact assessment-related activities. Since 2008, this has been subsumed into an "expertise budget", totalling €7,500,000 that year.  Back

27   Op. Cit. Q 375 Back

28   Op. Cit. Q 380 Back

29   Study requested by the European Parliament's committee on Internal Market and Consumer Protection, Broad Economic Analysis of the Impact of the Proposed Directive on Consumer Credit, April 2007:  Back

30   Impact assessment was requested by the European Parliament's Committee on Internal Market and Consumer Protection, Impact Assessment: Parliament's Amendments to a Commission Proposal on Nominal Quantities for Pre-packed Products, November 2: This was carried out before amendments were adopted by IMCO. Back

31   These were, at first reading, Priority Substances in Water Directive, Working Time Directive, Interoperability of the Community Railway System II Directive, Air Quality Directive, Nominal Quantities for Pre-Packed Products Directive; and at second reading, Waste Framework Directive and Batteries and Accumulators Directive. As reported in Impact Assessment: The European Parliament's Experience: A Stock-Taking Report of the Common Approach to Impact Assessment produced by the European Parliament Conference of Committee Chairs in December 2008. Back

32   Part for the whole. Back

33   Council conclusions on Better Regulation: 2945th Competitiveness Council Meeting, 28 May 2009: Back

34   There are, however, proposals for better post-legislative scrutiny in the UK. See, for instance, Post-legislative Scrutiny-the Government's Approach, Cm 7320 2008: Back

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2010