House of Lords - European Union - Minutes of Evidence

Select Committee on European Union Minutes of Evidence

Memorandum by Open Europe

1. What should be the Objectives of the EU's Structural Funds? How can the Funds become more effective in supporting public policies in Member States and regions? What mechanisms of delivery could make the policy more performance-based and more user-friendly?

1.1 For the Structural and Cohesion Funds (SCF) to become more effective in supporting public policy, at least five major reforms need to take place:

(i) Better targeting of funds both amongst and within regions to focus effort on the worst off areas.

(ii) Lower deadweight costs and less bureaucracy both for public and private actors.

(iii) Better project selection to focus on genuinely growth and employment enhancing programmes.

(iv) Better coordination with, and less constraints on, national policies.

(v) Stronger systems to reduce fraud and mismanagement.

1.2 We argue that these flaws predominantly stem from the very involvement of the EU in regional policy in the first place, and will be very hard to fix through micro-reforms. Rather, we suggest the complete repatriation of regional policy to the UK. We also suggest that all Member States should have the option to opt out of the SCF.

1.3 First, the poor targeting (both in terms of projects and areas) has its roots in the EU's budget negotiations themselves. Historically and today still, the evidence suggests that the SCF are treated more as "political sweetener" than a development tool. The influential Sapir report for the Commission rightly argued that,

"National political constraints mean that each government worries more about being able to flag a negotiation success (ie obtaining a significant share of EU money to be spent in its own territory) than about being sure that money is spent on worthwhile projects, let alone those fostering convergence in the EU as a whole".

1.4 Likewise, as often pointed out in academic literature, the classic EU horse-trading often makes the SCF function as a "side-payment", as with the so-called "Integrated Mediterranean Programme" from which Greece was promised some €2 billion, in return for its vote on the accession of Portugal and Spain.[2] The Cohesion Funds was likewise set up to "compensate" for the forthcoming monetary union. Hence, the SCF are not a needs-driven policy, but more based on contingencies in negotiations. This, in turn, paves way for arbitrary allocation processes. Should control of regional policy be brought back to the Member States which so choose, at the very least, the bargain-driven element of the SCF would be become less significant.

1.5 The poor targeting is also tied to EU-specific rules on the rate of spending. The European Court of Auditors (ECA), amongst others, has pointed to problems with the Commission's N+2 rule. The rule means that the commitment by the member states to spend the allocated funds must lead to payments within two years of being entered into the budget or the money will be cancelled. The objective behind the rule is to prevent unused funds from storing up and ensuring that the spending is well planned. However, the ECA argues, it provides a tremendous incentive for the member states to spend where the money can be quickly absorbed, rather than where it can lead to long-term and sustainable gains.

1.6 An ECA report concluded that for the 2000-06 financial period, "budgetary allocations were determined less by a well established development strategy and the effectiveness of the Structural Funds than by maximising likely take-up of funding".[3] It found several instances where money had been reshuffled from projects that were deemed effective and deserving, to projects that simply could absorb the funds.

1.7 A better solution—which would make the delivery of the SCF substantially more result-based—would be to leave governments free to choose the investment project to be financed by EU transfers, but oblige them to declare beforehand the expected results of the project. Disbursement of the funds could then be made in lump sums, and would depend upon reaching these results—rather than ability to spend. This is essentially the solution the Sapir-report and others have suggested. However, from such a system there is a relatively small step to bringing back regional policy to the national level altogether.

1.8 Moreover, if the SCF truly are to add value to domestic policy, the heavy bureaucracy of the SCF must be downsized and the deadweight loss drastically reduced. The cumbersome bureaucracy involved in the SCF is universally acknowledged.[4]

1.9 On the EU level, substantial resources are spent on administration and overlapping committees. For example, the Committee of the Regions (CoR), costing some €140 million a year; some €70 million is spent under the label "Working in Europe: Social dialogue and mobility" and another €120 million on "Employment, social solidarity and gender equality". There is also the Assembly of the European Regions. Also the EU's Economic and Social Committee, costing some €120 million a year, has a specific unit for the SCF. It is far from clear what all these committees and groups actually achieve. All in addition to the admin costs involved in DG Regio.[5]

1.10 On the national level, there are no less than seven different Whitehall departments involved in the SCF—the DWP, BERR, Treasury, DEFRA, CLG, DFT and DCMS. For the DWP, the yearly administrative cost involved in the ESF is reported at £10 million.[6] For the other departments, such figures have not been possible to obtain but could well be of the same magnitude.

1.11 On top of this the regional tier adds substantial costs. An evaluation of the Peace 2 programme in Northern Ireland concluded that out of the £641.2 million that was allocated for the programme, £57.1 million—or 11%—was absorbed by administration.[7] A report from the Scottish Parliament put the annual figure for administering the SCF in Scotland at £30.9 million, equivalent to 10% of the funds allocated.[8]

1.12 The English GOs and RDAs (and at present regional assemblies) cost £366 million to administer. It has not been possible to distinguish between the administrative costs created by the SCF and other spending. Nor would it be possible to get rid of the regional tier while the EU's current spending framework stays in place. Regardless, it is clear that also at the regional level, the bureaucratic cost is high.

Region
Admin cost,
RDA
Overseas
spending, RDA Admin cost,
Government Office
for Region Grants,
Regional
Assembly

Total

North East 22,607 836 13,055 2,073 38,571

North West 38,144 577 17,400 3,045 59,166

Yorks & Humber 19,290 606 13,286 2,339 35,521

West Midlands 20,000 1,056 14,972 2,518 38,546

East Midlands 16,000 375 11,496 2,507 30,377

East 10,900 180 12,672 2,469 26,221

South West 19,290 672 14,766 2,470 37,198

South East 21,222 443 14,095 3,771 39,531

London 29,500 17,881 65(admin) 112,381

Total 166,063 4,745 129,624 21,436[9]
366,868

Sources: GLA, Greater London Authority Consolidated Budget 2006-07, February 2006, p 9; Hansard, 19 July 2007, Column 494W; Hansard, 22 May 2006, Column 1460W; Hansard 10 May 2006, Column 376WA onwards; Hansard, 11 January 2007, Column 738WA.

1.13 The combined SCF-related administrative costs of the EU, national and regional policies suggests that even while spending the same amount of money on regional/regeneration policies as at present, it would be possible to save hundreds of millions in administrative costs by sweeping away the costly bureaucracy necessitated by the EU's regional policies.

1.14 Beyond this, cutting out the EU-level would presumably lead to significant savings also for recipients, given that the EU's bureaucracy and regulations, almost universally, are perceived as more cumbersome than its national counterparts.[10]

1.15 Again, as widely acknowledged, the EU's regulations are today overdone and too inflexible to accommodate the specific needs of 27 countries and hundreds of regions—at different stages of their development. Something that is made worse by the lack of "opt-outs" for smaller-sized projects, or projects with special needs. The requirement to keep records for 12 years, regardless of project size, is a case in point. As has been noted elsewhere, it seems odd that a small café offering intermediate job opportunities for the mentally ill should be under the same accounting rules as a multi-million pound infrastructure project.[11]

1.16 The obvious risk is that the cost of both the extra bureaucracy and regulations are passed down to the stakeholders at the bottom of the chain, hitting smaller players that have a much harder time to absorb the costs, ie small or struggling firms and community or voluntary organisations. There are examples of small projects that drown in this paper work, such as a training organisation funded by the ESF with only 12 trainers that had to employ five support staff in order to fulfill all the administrative requirements of the SCF.[12] This runs completely contrary to the supposed aim of the SCF, in as far as the policy is meant to encourage local initiatives and subsidiarity.

1.17 The entire structure of administrative control and audit seems to be both disproportionate to the small amount of funds given and out of touch with economic reality. As a DTI/ODPM report pointed out, there is little evidence that the rigorous regulations have actually led to better and more efficient spending.[13] It would of course make sense to create a system of opt-outs for smaller actors, and regulate with more attention paid to specific, local needs—such as including social housing among the eligible funding activities.

1.18 However, notwithstanding some flexibility for the UK in 2007-2013 in terms of reporting requirements, British proposals to simplify and streamline the regulations seem often to conflict with the Commission's ideas and rules. For example, a type of fast-track procedure for projects of less than £50,000—which would make life a lot easier for smaller players—did not comply with the Commission's rules on open and competitive tendering. Similarly, the UK attempted to drop the requirement to keep records for 12 years, but was found to be in breach of EU-regulations.[14] Gordon Brown has acknowledged that, "There are many things that we want to do to encourage local skills and research and development, and local businesses, but we're not able to do because of the existing rules".[15]

1.19 All of this seems to support the conclusion that the best way for the SCF to support public policy in the UK is through the complete repatriation of the funds.

1.20 In addition, corruption and mismanagement remain at large, undermining sound public spending and policy. The ECA found that at least 4 billion euros of the money that the EU Commission handed out in 2006 "should not have been". Of the projects the ECA audited, only 31% "were found to be free from error". The ECA warned that there was a "high risk" that the project costs were "overstated" and that there were large numbers of claims for "ineligible expenditure." The report stated that there was generally "a lack of evidence to support the calculation of overheads or the staff costs involved".[16]

1.21 But the ECA usually only monitor some 90 projects, and the Commission only audits 5% of the projects given money. Its figures could well be underestimates. For example, Italy's tax and fraud investigator, Guardia di Finanza, noted in its latest annual report that €433 million worth of EU money was subject to outright fraud in Italy alone in 2006.[17]

1.22 As the ECA also has pointed out[18], the SCF are particularly prone to errors, due to the set-up. There are several reasons for this:

(i) The large number of programmes and projects which are implemented over several years.

(ii) The large number of eligibility conditions which are hard to follow and sometimes open to divergent interpretations.

(iii) The variety of entities and actors which for different reasons intervene in the management process.

(iv) The large number of diverging countries and regions subject to the same centralised regulations.

1.23 These are all issues relating to the scale and complexity of the SCF. By bringing back regional policy to the national level, the number of instances where fraud and mismanagement could take place would radically decrease. And the spending would be easier to audit.

2. Do Structural Funds meet the principle of subsidiarity? Could the same cohesion objectives be met through repatriation of these funds?

2.1 The complex regulatory culture of the SCF can undermine both localised, small-scale projects (see above), as well as the ability of regional and local authorities to channel money to where the needs are and where the funds can do most good.

2.2 A well-known example is that the SCF cannot be used for schooling or social housing projects—paradoxically as this was initially seen as breaching precisely the subsidiary principle. Stephens (1999) has noted that "Arguably, the current asymmetry between allowing European funds to be used to attract physical capital, but excluding housing as a key aspect of enhancing human capital, is itself a breach of subsidiarity."[19]

2.3 An obvious example is Cornwall—the only region in the UK that will receive Objective One funding throughout the 2007-13 financial perspective. One of the main problems in Cornwall is of course the disproportionately high property prices, due to the number of houses sold as second homes to wealthy outsiders (estimates vary—some argue levels are as high as up to 50% in some districts). In some areas the average house price is 17 times the average annual income in the region—while average income, in turn, is 25% below the national average. As a consequence, Cornwall has seen what MP Matthew Taylor calls a "shocking" rise in the number of people seeking social housing.[20]

2.4 In this scenario it would be sensible to at least have the flexibility to direct structural funds money towards social housing—as that is clearly a need on the ground.

2.5 Conversely, while it is clear that EU management of the structural funds prevents the UK Government from doing things it wants to do, it is difficult to argue that the EU's involvement in running UK regional policies allows anything to happen which could not be achieved by the UK Government. It is clear that EU management of UK regional policy is a violation of the subsidiarity principle.

3. What impact has the enlargement had on Structural Funds, and are any changes necessary to meet the challenges of further enlargement?

3.1 One of the more apparent effects is that the SCF since enlargement have become bigger and therefore even harder to manage and audit.

3.2 As for the UK, the significance of the SCF has obviously decreased due to the "statistical effect". For the UK as a whole, the annual allocation of the structural funds will amount to about 0.1 -0.15% of national GDP for the 2007-13 financial perspective. Across the regions, the share of GDP varies between approximately 0.2% (Northern Ireland) and 0.02% (South East).[21] This is of course not enough to have any significant macro-economic effect. Despite the further decreased significance of EU funds, the complex regulations and the bureaucratic structure will stay in place, and hence so will the limitations on how the UK can run its regional policy.

3.3 In an enlarged Union, it therefore makes even less sense for the EU-15—particularly countries such as the UK—to be enrolled in the SCF. Allowing for the option of opting out of the policy, could be a major step towards focussing the SCF on the MS and regions that actually are in most need—as well as sizing down the programme and thus making it more manageable. In addition, the targeting of the SCF must be radically improved (see above and below) if the SCF are going to have a real impact on growth and jobs in the EU-27 and beyond.

4. What criteria should guide the decisions on the proportion of the EU budget to be allocated to Structural Funds?

4.1 The key is to move away from bargain-driven deals and towards development-focussed strategies. As long as the richer MS are in the game, political domestic constraints will inevitably force them to strike deals, and the budget will continue to balloon, without the funds necessarily having their desired effect. Both for the purpose of fiscal discipline and better targeting of funds, the UK Government's proposal to have all MS with a GDP above 90% of the EU average managing and paying for their own regional policy, should be one of the main criteria for deciding how much money to inject into the SCF. If this indeed is the main criteria, the EU's overall regional budget could be cut by as much as 48% (given that 48% of the funds were allocated to the EU-15 for the 2007-13 period).[22]

4.2 There must also be an honest discussion about the SCF's track record. As with the CAP, the failed aspects of the SCF cannot be allowed to simply stay in place. For example, the absence of evidence in support of the SCF's added value must be considered also in the budget negotiations.[23] Where MS consider the value added to be absent, funds should not be allocated.

4.3 In addition, focus must be put on impact. While it is widely agreed that the evidence is inconclusive as to the precise effectiveness of the SCF—and that the counterfactual is virtually impossible to determine—several indicators seem to suggest that the SCF are simply not delivering where they should. Most generally, convergence between regions is simply not happening.

4.4 A recent OECD report noted that regional disparities in Europe are not falling, or at best are declining very slowly. At the current rate of convergence, the report stated, it would take 170 years to halve divergence across the regions in the EU.[24] The report argued that this should put a big question mark next to the SCF.

Source: OECD 2007.

4.5 A paper by Gardiner et al concludes that both when looking at regional GDP per capita and regional productivity, regional convergence in the EU is at best a slow process of only 1-2% per annum. Meaningful convergence would take decades to achieve.[25]

Source: Gardiner et al 2004.

4.6 And as the Sapir report has argued, the available empirical data on economic convergence in the EU gives a very different picture depending on whether one looks at the Member States, or its NUTS 1 regions across the Union.[26] Although convergence of GDP per capita has taken place between member states, inequality within each country accounted for roughly half of total EU regional inequality in the early 1980s, but this rose to about two thirds by the mid-1990s (while inequality between countries fell by about a third during that period).

4.7 Even the Commission's report admits the continuing divergence across the regions (in passing), stating that "in spite of progress, absolute disparities remain large. This is partly as a result of recent enlargement and partly as growth tends to concentrate—during the initial phases of development—in the most dynamic areas within countries".[27]

4.8 Elsewhere, the report also points out that in the period 2000-04, real GDP per head fell in 27 regions and in 24 regions it grew by less than 0.5% a year. In five regions, GDP per head slipped below 75% of the EU average.

4.9 The Commission seems to argue that such discrepancies reflect different stages of development and will naturally disappear in due course. However, first, as we saw, convergence is very slow—170 years to halve welfare gaps can hardly be an acceptable rate. Secondly, Objective One regions—those with an average GDP under 75% of the EU average—have remained remarkably stable. Of the 44 regions originally granted Objective One status in 1989, 43 were still eligible for such funding in 2003.

4.10 Meanwhile, households in the EU's poorer regions have increased their dependency on grants and social transfers for their disposable income.[28]

4.11 In the UK, when comparing EU funding with the GVA per indices there seems to be little evidence that the structural funds have had a net effect on wealth and job creation. If the structural funds were efficient, we would expect to detect some sort of boost in a regions' GVA as EU funding increases. This is not the case, however. Although not a hugely meaningful comparison, it still undermines the Commission's case, as the very idea behind the SCF is to raise GVA per head, as poorer regions are granted aid precisely on the basis of their low levels of GVA per capita relative to the EU average.

4.12 Also more meaningful measures show that economic opportunities remain unevenly spread across the UK, and are unaffected by the SCF. The one-fifth of the UK's population that live in the poorest local areas still only accounts for about 13% of national income—with few signs of any significant upward mobility taking place in the last decade.[29]

North East[30] Wales

Sources: ONS 2005; DTI/BERR allocation of the structural funds, 89-93, 94-99, 2000-06.

4.13 Like in the case of value added, if there is no solid evidence that the SCF lead to convergence—evidence usually tend to tip the other way—nor any measurable impact of funds, what is the justification for continuing with the programme in its current size and shape—particularly with the deadweight losses in mind. These are all issues that have to inform the decision on how much money to allocate to the regional budget post-2013.

5. Are the current eligibility tests for regions to receive support under the EU's Structural Funds relevant, fair and appropriate? Should they remain in place after 2013? Is it appropriate that they are discussed simultaneously with wider agreements on allocating EU budget spending?

5.1 First, the EU's criterion for special status regions—75% of average EU GDP—is outdated. It may have made sense when the regions were either well above or well below the threshold. Since enlargement, however, the regions are now much more evenly scattered, with several areas being around the threshold. This opens up for regions with similar wealth levels, receiving very different amounts of funding.

5.2 Secondly, the Commission's preferred measure for wealth—the Gross Value Added (GVA) measure—is open to substantial criticism.[31] GVA measures the contribution to the economy of each individual producer, industry or sector. GDP is derived from GVA by adding taxes and subtracting subsidies on products.[32] Essentially, GVA measures output divided by population.[33]

5.3 However, as often noted, since GVA does not take commuting into account, it tends to produce an upwards bias for regions that have large levels of inward commuting, and vice versa. Thus, for example allocation of Objective One (now Convergence) spending can be distorted by the fact that some NUTS 2 regions, for example, are urban areas or cities (ie Hamburg, Bremen or even Inner and Outer London), whereas others are coastal or rural areas. The urban areas naturally have higher GVA per head, due to concentration of financial and business services, more value-added activities and so forth. Areas with a large share of inwards commuters will have artificially high GVA per head. The reverse is then true for areas where commuters tend to live. Some estimates put the Greater London's upwards bias in the GVA calculation in the area of 15%.[34]

5.4 This is precisely the case with the infamous example of Lüneburg outside Hamburg, Germany. The district serves as residential area for Hamburg's professionals, and is considered one of the wealthiest commuting areas in Germany, growing by 2,000 residents each year. However, since GVA focuses on workplace, the high income of Lüneburg's residents counts towards Hamburg wealth statistics.[35] Consequently, Eurostat has identified Lüneburg as an Objective One region, marginally poorer than Merseyside. Lüneburg was granted a staggering €900 million from the SCF for the 2000-06 period—only slightly less than what was given to Merseyside for the same time period. This, of course, adds to the criticism of the insufficient targeting of the SCF.

5.5 In addition, GVA does not adequately account for demographic trends, productivity factors and population structure. Neither does it reflect lifestyle choices. For instance, an area that has a large number of pensioners, students or part-time workers will have low GVA per head since that type of income is omitted in the GVA measure. This, without the area necessarily experiencing a shortage of economic opportunities.[36] Many have pointed to Cornwall and Cumbria as being on opposite ends of this problem.

5.6. The ONS' recommendation for measuring regional prosperity and hence for deciding how to allocate the SCF, strike us as fairer, more appropriate and more relevant:

(a) regional GVA per full-time employee, which they argue is the best way to measure regional differences in productivity as it is not affected by the number of non-working residents; and

5.7 But even beyond the measure per se, the appropriateness of the EU's distinctly region-based funding system is questionable. Due to the set-up, the SCF often fail to target smaller pockets of poverty that can exist within a region. The flip side is that more affluent areas can end up with substantial amounts of money for no apparent reason.

5.8 This is a major shortcoming in a country such as the UK, where local pockets of poverty often exist within affluent larger areas. In urban areas, for example, discrepancies can come down to as small units as a block of council estates. The SCF are way too inflexible to address such gaps. This problem is also exacerbated by what many observers consider the artificial nature of the UK's regions.

5.9 The South East is a good case study. In 2004, the South East was the 13th wealthiest region in the EU[38] with a household disposable income 12% above the national average. The region was unsurprisingly given Objective Three status—which reflected its wealth levels—and was granted close to £300 million for the 2000-06 financial perspective.[39]

5.10 However, within the South East there are smaller economically deprived areas such as Hastings, where 27% of all children come from families living on out of work benefits.[40] This and other indicators make it one of the poorest areas in England. But Hastings is situated within the NUTS 3 region of East Sussex, which has a GDHI above the national average. NUTS 3 is the smallest unit in the EU's classification system. Consequently Hastings is completely omitted. For the 2000-06 financial period, it has only been given ESF grants worth £1.1 million. This translates into 0.4% of the total ESF amount given to the South East.[41]

5.11 Likewise, some parts of Thanet belong to the 1% of the most deprived areas in the UK.[42] Its NUTS 3 region, however, is Kent, whose GDHI is above the national average. Thanet received some £1.8 million from the ESF over a six year period. Although both towns get substantially more money from the ERDF, it still seems rather odd that they would end up with so little money from the ESF.

5.12 In comparison, Canterbury—a wealthy area by all measures—has been given some £2.6 million from the ESF.

5.13 Examples like these seem to suggest that a substantial share of the SCF ends up with projects and individuals that, comparatively speaking, are not in need of the money, and that the reverse is true.

5.14 We used the ACORN system for UK postcodes—which categorises the UK population according to income, lifestyle choices and a whole range of demographic statistics—to get an idea of where the SCF actually end up. In the South East, we found that postcode areas whose populations are classified as "wealthy achievers" received 5.5% of the funds. The upper half of the population, from the "wealthy achievers" to "secure families" in the "comfortably off" category got almost 30% of the money granted. Only 10% of the funds had gone to the bottom one fifth of households that fall under the category "hard pressed". For the upper categories, we excluded grants that have been given to councils under general headlines, as opposed to specific projects. It is fair to assume that lump sums given to councils could be redistributed elsewhere.

5.15 As said, one of the major problems with evaluating the SCF is the poor quality of the available data. The South East is the only region that provides detailed information on where its SCF money has gone in terms of smaller units. In virtually every other region, data is only available for the NUTS 2 level—at best. As discussed, this is not very useful if we want to investigate whether the funds in reality are targeted at areas where they can do most good.

5.16 If we assume that the South East is indicative for the UK as a whole and using the SCF allocated to the UK for the 2000-06 financial period, which is around £10 billion, more than £3 billion has gone to the "top" 50% of the population in terms of wealth. The wealthiest 20% have received about £550 million from the SCF. Meanwhile, as little as £1 billion has gone to the bottom 20% of the UK's population. Even assuming that all the grants that have been given to Councils in the wealthiest areas have been redistributed to the bottom 20% (something that seem very unlikely), the bottom one fifth has still only been granted £3 billion out of the allocated £10 billion.

5.17 This is a very rough estimate, but it does give an idea of how off target and unfair the SCF can be. Furthermore, it provides a strong argument as to why the EU's criteria and region-based spending structure should not stay in place after 2013.

5.18 As the flaws are so fundamental and cut to the very heart of the distribution/impact of the SCF, it seems highly appropriate to consider them simultaneously with the wider discussions on spending allocation.

6. What would be the effect of linking the availability of the Structural Funds with compliance to Broad Economic Policy Guidelines?

6.1 It would not be appropriate for member states which are not members of the euro to face financial penalties (in the form of lost structural funding) for non compliance with the Broad Economic Policy Guidelines.

6.2 The euro member states, where macroeconomic policy is arguably a matter of greater mutual concern, already have in place rules on fiscal policy in the form of the stability and growth pact. Creating a parallel process to enforce the BEPG would be likely to create confusion.

9 January 2008