Brexit and the EU budget Contents

Chapter 3: Potential demands


27.It may seem intuitive that when the UK leaves the EU, it leaves behind both the responsibilities and benefits of membership. However, this does not take account of the complexity of the UK’s participation in the EU, nor of the procedures for agreeing current and future budgets, which involve mutual commitments projected many years into the future. We summarise the views of witnesses without taking a view as to their merits. We do not focus on any single figure for a potential EU demand. However, any withdrawal agreement will need to take account of the status of existing commitments that the UK is a party to (and will be until the moment of Brexit).

Elements of a potential EU claim

28.Professor Iain Begg, Professorial Research Fellow at the European Institute, London School of Economics and Political Science, told us that “what Barnier’s team has been saying”, is that the bill for leaving “could be of the order of €60 billion”.22 Ingeborg Grässle MEP, Chair of the European Parliament’s Budgetary Control Committee, commented that “up to now the Commission has refused to put forward the bill, and it is the only one that knows what the exact bill is every day”.23 A recent report by the Centre for European Reform (CER) think tank, written by Alex Barker, head of the Financial Times’ Brussels Bureau, suggests a figure of €57.4 billion. This seems to align with the amounts reportedly being ascribed to the Commission—though the CER report also concludes that, if the rebate is excluded from calculations, the arithmetic liability is likely to lie between €47.9 and €72.8 billion.24

29.There is, though, nothing ordained about the figure of €60 billion. Ms Grässle suggested an alternative figure of €21.7 billion, which she viewed as “an obligation that will cover the whole procedure”.25 This figure was based on what she deemed to be the UK’s share of the EU’s current and non-current liabilities, including pensions, plus a share of the RAL.26 She qualified this by saying: “I think that €20 billion is at the lower level, but you can imagine it going up to €70 billion.”27

30.Ms Grässle, taking an approach that differed substantially from those of our other witnesses, reached her figure by allocating the liabilities and assets on the EU’s balance sheet in proportion to the UK’s share of the EU’s population (12.5%), while adding 5% (the proportion of the UK’s pre-allocated spending under the MFF) of the RAL (€254 billion). Importantly, the balance sheet (drawn from the annual accounts for 2015)28 measures the financial position of the EU itself (primarily loans and assets), and not the total revenue and spending of the annual budget. In other words, it excludes the UK’s payments and receipts under the EU’s various spending programmes (such as the ESIFs), and looks solely at the EU’s financial position.

31.According to the CER report, Michel Barnier “takes a more expansive view”.29 Ms Grässle’s figure and that attributed to Mr Barnier are based on different assumptions. In order to obtain the larger figure, it is necessary to take a maximalist perspective on the UK’s potential contributions. The CER report details how this might be done. According to their calculations, it is necessary to carve out responsibilities based not merely on the EU’s current financial position, but also to add in an obligation to honour spending under the MFF to 2020. Including this measure contributes at least €17.7 billion30 to their overall computed figure of €57.4 billion (assuming the UK would have a 12% share of ongoing EU funding).

32.Obtaining such a figure also requires minimising any putative share of receipts. The CER report therefore accounts selectively for the EU’s balance sheet assets, by counting only property and assets available for sale, which make up €22.5 billion of the EU’s total listed assets of €154 billion. It also excludes the UK rebate.

33.The range of values in circulation for the UK’s potential ‘exit bill’ indicates that the absolute sum of any posited settlement is hugely speculative. Almost every element is subject to interpretation.

Calculating the value of the UK’s current contribution

34.A key issue in reaching a figure for the EU’s potential demand following Brexit is the UK’s ‘share’ of the EU budget, which is itself not a fixed sum. The amount of the UK contribution varies from year to year, depending on currency fluctuations,31 spending commitments, own resources and VAT receipts, and the relative size of Member States’ GNIs. It can also be measured as a gross or net figure, either taking account of the UK rebate and receipts from the budget, or not. Table 1 breaks down the UK’s contributions in recent years.

Table 1: UK gross contributions to the EU budget (£ million)








Duties and levies







VAT-based contribution







GNI-based contribution







Other income to EU institutions














Source: ONS Pink Book 2016, section 9.9

35.Professor Begg proposed a figure of 12% for the UK share. This represents an approximate amount for the UK’s gross contribution to the budget, once the rebate has been deducted. The CER report used a figure of 12.1% as the average of UK contributions, after the rebate, from 2012 to 2016.32 Ingeborg Grässle MEP proposed a UK share of 12.5%, calculated by reference to the UK’s population as a share of the EU whole. In contrast to the CER report, both Ms Grässle and Richard Ashworth MEP, a member of the European Parliament’s Budget Committee, proposed that other figures be used to calculate any UK share of RAL liabilities (5% and 8%, based on different measures of spending) and pensions (4%, 8%, or 12.5%)—issues that are discussed further below. In sum, our witnesses and other experts did not agree on a single, uncontroversial number that might be considered a measure of the UK’s ‘fair share’ across all liabilities.

36.It would also be possible for the EU to take a maximalist view of the UK’s current contribution by excluding the rebate. According to Professor Begg, this would leave a gross contribution of approximately 15%.33 In the 2015 financial year the rebate accounted for £4.9 billion, against a total gross contribution of £19.5 billion34 (about €25 billion at the time35). The EU budget in 2015 allocated €145.3 billion for commitments and the UK’s contribution less the rebate was 12.57%.36 The rebate therefore forms a significant proportion of the UK’s receipts. It is an accepted part of the EU budgeting process, being incorporated into the ORD, but its inclusion in any calculation is likely to be a point of negotiation.

37.In addition to the rebate, and the UK’s 20% ‘collection fee’ for levying duties at the external border (worth £772 million in 2015), the UK also receives money in the form of public sector and private sector receipts. At present, public sector funds accrue to the UK mostly through EU programmes such as the Common Agricultural Policy (CAP) and the European Structural and Investment Funds (ESIFs). The ONS reports that, collectively, these EU-funded public sector credits accounted for £4.8 billion in 2014.37 For 2015, the most recent data available, the ONS accounts list a return to the UK of £3.6 billion from spending programmes, plus £0.8 billion accrued from the administration of duties and levies.

Table 2: UK receipts from the budget (£ million)38








Duties (collection costs)





















Other receipts





















Source: ONS Pink Book 2016, section 9.9

38.This left the UK’s net contribution to the EU institutions—contributions less the rebate and receipts to the UK public sector—at £10.4 billion in 2015, compared to an average balance since 2010 of £9.1 billion per year. However, the ONS figures only count public sector revenue, and not private sector grants (for instance research funding under Horizon 2020, for which €1.4 billion had been awarded to UK organisations by November 2015). The European Commission, by contrast, includes spending disbursed directly to the private sector39 in its accounts. These numbers are not directly comparable to the ONS data, because they include adjustments to the budget between years.40 But on this measure, the five-year average of UK net contributions was £7.1 billion per annum.41 This figure amounts to approximately 5–6 % of the EU budget.42

Table 3: UK Net contributions to the EU budget, including and excluding private sector receipts (£ million)









Contribution including private sector receipts






No figure available


Contribution excluding private sector receipts








Source: First row––European Commission, cited in HM Treasury analysis: the long-term economic impact of EU membership and the alternatives, Annex B; second row––ONS Pink Book 2016, section 9.9

39.It is worth underlining one final point: EU spending is not hypothecated, which is to say that revenue from particular taxes is not ring-fenced for application to associated areas of spending. This means that determining an amount that the UK could potentially ‘pay in’ to maintain access to particular programmes such as Horizon 2020—the approach suggested by the Prime Minister in her speech in January—is not straightforward.

40.The total UK contributions to and receipts from the EU budget are variable, difficult to calculate, and subject to interpretation. It is therefore difficult to reach an unequivocal figure for the UK’s current commitments. This has ramifications for attempts to determine the costs or benefits of disentangling from current financial commitments. Nonetheless, on average, the UK’s net contribution in recent years has been around £7.1 billion when receipts to the private sector are taken into account.

41.Various figures have been suggested for the UK’s ‘share’ of the EU budget that could be used to determine a share of EU liabilities. This could be based on the UK’s average gross contribution (around 15% of EU revenues), that same contribution minus the rebate (12%), the net contribution taking into account public sector receipts (8%), or taking into account private sector receipts (5–6%). It has also been suggested that the ‘share’ be calculated on the basis of the UK’s population, as a proportion of the EU total (12.5%). The percentage to be used is likely itself to be a matter for negotiation, and different figures could be adopted depending on the precise liabilities (or assets) under discussion.

Currently committed spending


42.The UK is currently scheduled to make further payments to the EU, in accordance with the provisions of the ORD, in order to meet spending set out under the MFF. Were the UK to continue to pay into the MFF up to the end of 2020, it would remain a net contributor, but might also expect to receive continuing benefits. The EU might also make the case that the UK should continue to fulfil the commitments that have been made up until 2020, regardless of whether it enjoys the benefits of membership.

43.The UK currently receives significant disbursements, and Professor Begg commented that “if we were still paying in, I would expect the counterpart to that to be continuing to receive on the same basis”.43 As we have discussed, the net figure of the UK’s contribution is contingent, but the five-year average cited in Table 3 suggests a net figure of about £7.1 billion per year. Assuming that receipts from the EU to the public and private sector continued until the end of 2020, alongside UK payments into the EU budget, the net cost would be around £12.4 billion, for the 21 months between 1 April 2019 and 31 December 2020.44 This also assumes that that the UK will meet liabilities that fall due during this period in the normal way: it does not take into account any extra liabilities that might be claimed.

44.The Government’s White Paper repeats commitments to certain UK beneficiaries made by the Chancellor of the Exchequer, to maintain equivalent levels of spending in particular funding categories, such as ESIF and CAP funding, through to 2020.45 This guarantee applies to all UK ESIF projects agreed before the 2016 Autumn Statement (and some agreed thereafter). HM Treasury has also guaranteed that funding under the CAP will be matched until the end of the MFF in 2020. In the years 2018/2019 and 2019/2020 HM Treasury predicts that this will amount to £3.3 billion and £3.5 billion respectively.46

45.It is impossible to calculate what would be the net cost of remaining in the MFF until 2020 without an accurate comparison of expected receipts from the EU budget and those that the Government would match domestically if the UK withdrew from the budget at the point of Brexit. The figure of £12.4 billion discussed above, based on the UK’s average net contribution in recent years, would represent the net cost only if the Government were to fund every penny of the expected receipts in this second scenario in which it did not contribute to the EU budget. The Government has provided guarantees for most, but not all, of these receipts. The gap between a possible gross payment to the EU, and payments that will be made regardless is therefore likely to be higher than the £12.4 billion figure. Without knowing precisely how much the Government will guarantee domestically it is impossible to predict the size of this gap.

46.Jens Geier MEP, Vice-Chair of the European Parliament’s Budget Committee, suggested that the other EU Member States might be anticipating that the UK would not continue paying towards the MFF post-Brexit, and referred to a proposal by the German Finance Minister, Wolfgang Schäuble: “The German proposal is that we would have a new MFF. That would make it necessary to start negotiating a new MFF in 2018, which will be in place at the moment the UK stops its payments.” This would involve renegotiating payment commitments in line with an anticipated drop in resources. As he presented it, the alternative was “to cope with the MFF as it is until it is finished on 31 December 2020. That would be the more uncomfortable situation for the 27 Member States left.”47 Mr Geier’s comments imply that, in Germany at least, there is an acknowledgement that the commitments made in the MFF may not be legally or politically due.

47.The EU might demand that the UK continues to contribute to the 2014–20 MFF until its natural conclusion at the end of 2020, regardless of the date of Brexit. Were this demand to be met, this suggests a net cost to the UK, based on its average net contribution, of at least £12.4 billion, assuming other liabilities are not accrued.

48.In the event that the UK does not remain part of the MFF the Government has guaranteed some, but not all, receipts from the EU. The Chancellor has stated that “structural and investment funds projects signed before the Autumn Statement and Horizon research funding granted before we leave the EU will be guaranteed by the Treasury after we leave. The government will also match the current level of agricultural funding until 2020, providing certainty to our agricultural community.”48 However, without knowing precisely how much the Government will guarantee domestically it is impossible to arrive at a precise figure.

Reste à liquider (RAL)

49.RAL contributes a large proportion of the speculative EU demand being discussed. Ingeborg Grässle MEP argued that the UK had “an obligation to pay as-yet-unpaid commitments for the RAL at least until [it leaves] the European Union”.49 In her view there was “also an obligation that covers the rest of the MFF period”.50 The Commission forecasts that RAL will amount to €254 billion51 in total by the end of 2020.

50.A claim based on RAL may be particularly attractive to European negotiators, because it represents future payments for commitments already made, to which the UK has been a party. Nevertheless, accounting for the total amount of RAL, let alone any potential share that could be attributed to the UK, is not straightforward.

51.The payments of RAL sit within the spending limit set over the term of the MFF, allowing the Member States to carry forward spending commitments ahead of their disbursement as payments.52 However, some commitments are settled beyond the term of the relevant MFF, as Professor Begg noted:

“In particular, cohesion policy is allowed a rule of what is called N+3. For example, Poland53 has yet to start most of its cohesion policy programmes, even though they are for 2014 to 2020. It can still present bills up to 2023, using the N+3 rule.”54

52.Some of the current RAL commitments date back further even than the N+3 rule would suggest, from the period before the 2007–2013 MFF.55 Published EU data only go back nine years, but indicate that in 2015 (when total RAL was €217.1 billion) the pre-2007 share of the RAL’s maturation structure was €2.1 billion.56 Thus even by 2023, not all of the current RAL may be cleared. Dr Giacomo Benedetto, Senior Lecturer in the Department of Politics and International Relations at Royal Holloway, University of London, reflecting on comments made by the German Finance Minister, stated: “There is this lag [in liabilities] that, apparently, if we are to believe Wolfgang Schäuble, could go on until 2030.”57

53.The amount of RAL varies somewhat unpredictably over time. Richard Ashworth MEP cited the current RAL liability as €237.5 billion,58 but Ms Grässle noted the Commission’s forecast that it was due to rise to €254 billion by 2020.59 For 2017, the gap between commitment appropriations and payments was €22.7 billion, with payment appropriations totalling €134 billion.60 Over the term of an MFF, RAL is accumulated slowly before reaching a maximum in the last stages of a programming period. It is then disbursed over the N+3 period as commitments are paid out.

54.Mr Ashworth stated that the amount of RAL was difficult to predict, because it was “a projection budget … dependent on the Member States wishing to sign up to the project that has been undertaken, wishing to fulfil it, wishing to make their own financial part-contribution towards it”.61 In consequence, some anticipated commitments may never be converted into payments, being instead ‘decommitted’. The decommitment amount was €3.8 billion for 2015, and the Commission anticipates an overall rate of about 2% based on the previous period. This means that the current projection for the size of the RAL to 2020—€254 billion—may never fully materialise as a liability, although most of it will.

55.Determining what any UK share of RAL might be is equally challenging, although the UK’s average post-rebate contribution to the EU budget is around 12%. Most ESIFs, and almost all CAP spending, are pre-allocated to Member States at the beginning of the MFF period (other forms of spending—in particular on research—are by contrast allocated on a competitive basis over the term of the MFF, with the allocations to individual Member States unknown in advance).62 The spending pre-committed to the UK was €39.6 billion over the term of the MFF (out of a total of €767.3 billion for the EU-28). This produces a UK share of the allocations of 5.1%.

56.As the RAL is largely comprised of these pre-commitments, Ms Grässle felt that it would be fair for the UK to contribute around 5%, in line with its expected pre-committed receipts, although she also mentioned that “12.5% is the maximum share of this obligation … there is room for negotiation”.63

57.Mr Ashworth calculated the UK’s potential share of the RAL at approximately 8%.64 This figure represents a very rough estimate of the UK’s “net cost of contribution” to the budget as a whole (less both the rebate and an approximation of receipts), based on HM Treasury data for 2014.65 This net contribution for 2014 was £9.8 billion, minus receipts to the private sector of approximately £1.4 billion.66

58.Mr Ashworth, however, suggested that even his own approximation of the UK’s share was problematic: “It is far too simplistic to divide €237.5 billion [the current stock of RAL] by 8% and say that is the figure you owe.” This was because “it is almost impossible to determine how much of that would have come back to the United Kingdom”, rather than being returned to other international contractors.67

59.Any liability for RAL needs to be seen in the context of forecast UK receipts. In particular, Ms Grässle’s proposition overlaps with the Chancellor’s commitment to fund domestically a large proportion of the receipts expected from the EU if the UK’s receipts from the EU budget cease before the end of 2020. In total, 70.1% of the MFF’s commitment appropriations are pre-committed—€767 billion—with 5.1% of the total going to the UK (€39.6 billion). However, the UK would face an additional contribution if it were to fund the RAL in line with its gross share of total budgetary contributions, (15%, or 12% less the rebate) rather than in accordance with pre-allocated receipts (5%). This could therefore amount to a difference of between 7% and 10% depending on how the share of any UK liability were calculated.

60.Assessing any UK liability for reste à liquider (RAL) amounts is complicated by uncertainty over the extent of RAL. The Commission’s current forecast is for RAL to amount to 254 billion by the end of 2020, but this assumes that none of that amount will be decommitted in the meantime.

61.If a UK liability for RAL were to be agreed within the Brexit negotiations, it may be liable for a share of RAL up to 2023 or beyond. The precise point at which any UK RAL liability is calculated would have an impact on the amount, as the extent of RAL depends on the sequencing of commitments and payments. The EU’s starting point for negotiation could be the total accumulated RAL at the point at which the UK ceases to make payments to the budget.

62.If it were argued that the UK were liable for a share of RAL, using the Commission’s forecast for 2020 and the maximum assessment of the UK’s share as 15%, the total liability could amount to some 38.1 billion. But this is based on the largest possible assessment of RAL and does not take into account the rebate or any expected receipts from RAL. Given the difficulties in establishing a figure for RAL and the various options for assessing any UK ‘share’, this figure is extremely tentative.


63.Pension entitlements for retired EU staff are a liability of the EU institutions to their employees, and are guaranteed by Member States. They are paid out of the annual budget, while existing staff contribute a sufficient proportion of their salary (currently 9.25%)68 to fund one third of the costs of the scheme. These contributions are not invested in a separate fund, but are ploughed back into the annual budget. This has been referred to as a ‘pay as you go’ system. A number of our witnesses proposed various ways to approximate a potential UK share of pensions, which do not necessarily tally with proposed methods of calculating its share of the rest of the budget. Others questioned the assumption that the UK should be liable for a share of pension liabilities beyond the date when the UK ceases to be a member of the EU.

64.The UK’s potential liability for EU pensions has garnered some attention in recent months—though Mr Ashworth argued that pensions “will have a disproportionate amount of press to the actual size of the liability, and I suspect at the end of the day there are quite a lot of grey areas”.69 He argued that pensions should be “split off” as a separate budgetary issue.70

65.Mr Ashworth gave us figures for the UK share of employment in the institutions over the years. He noted that, as the EU has enlarged, this share has changed: “Historically, the [UK] contribution to the European Union general budget was of the order of 15% … at the highest level of employment of UK nationals, [the UK was] at 8%. Currently, [the UK is] at 4%.”71 Therefore, in his view, “There would be very good grounds for the United Kingdom to say, ‘Hold on a minute. We have over-contributed over the years.’” On the other hand, he cautioned:

“The European Union institutions have a very strong case for saying, ‘Come on, these are your people. There are about 1,800 people, going right back to the coal and steel days, who are dependent for their pensions on the European Union. You always contributed through the European Union budget, so you need to continue to meet that or take those people back under your own wing.’”72

66.Ms Grässle believed that a figure either of 4% or of 12.5% could potentially be used to derive a pensions liability for the UK. These numbers refer to the 4% of EU officials who are UK nationals, and the UK’s 12.5% share of the EU’s population: “It depends on how you attack the question, but this means that there is room for negotiation.”73 In her view, the UK had a moral obligation to contribute towards pensions: as officials had paid into the budget as part of their pension entitlement, during its membership the UK “had plenty of money from their pension fund … It would be fair to pay at least this money back, plus, perhaps, the money for the UK officials.” She put the number of these British former employees who receive a pension at around 1,000.74

67.Professor Begg noted that the EU accounts included a figure for the capitalisation of the long-term pension commitments;75 Ms Grässle made a similar point, noting that “there is accrual accounting for all pension liabilities”.76 The current figure for this total EU liability is €63.8 billion;77 according to Ms Grässle’s calculations, possible UK shares of this might be between €2.5 billion and €7.9 billion.

68.However, one of our witnesses, Dr Zsolt Darvas, Senior Fellow at Bruegel, has co-authored a report that questions the validity of the calculation producing the €63.8 billion capitalisation.78 He and his co-authors argue that the UK should omit the one-third contributed by officials, and furthermore that the “reasonable” value of the pension liabilities is €43.1 billion instead of €63.8 billion. Two thirds of this lower liability is approximately €29 billion.79 Although Professor Begg thought that “the liability could be capitalised” he suggested that “the cash flow on it could be spread over several years. It could be a lingering bill for 20 years.” As we suggest later, there may not be a legal liability for pensions at all.

69.The current total capitalised EU pension liability is recorded as 63.8 billion. If the UK were to accept a liability for any of this, our witnesses suggested that its share could be calculated by reference to the UK’s average contribution to the EU budget (between 8% and 15%), the proportion of those currently serving in EU institutions who are UK nationals (4%), or the proportion of those in receipt of an EU pension who are UK nationals (8%). The total liability on this basis would be likely to come to between 2.5 billion and 9.6 billion. However, an accurate assessment of any potential liability could involve a complex actuarial calculation of pension rights accrued by those serving in the EU institutions while the UK was a member, whether UK nationals or not. Negotiations may also encompass a figure for capitalised pension liabilities that is in practice substantially lower than 63.8 billion.

EU assets

70.The EU has a range of assets. They total €153.7 billion, and include property, equipment, loans and investments, and cash and other fungible assets. The biggest elements of the EU’s assets are loans (37%) pre-financing and other advances to Member States held in anticipation of issuing payments under the MFF (32%), and cash and equivalents (14%).80 Any agreement on apportioning assets would need to take account of their varied composition.

71.An article co-authored for Bruegel by Dr Darvas, attempts to account for the divisibility of the EU’s assets. The article states that €41 billion of balance sheet assets constitute ‘accumulated wealth’, which could potentially be apportioned, including cash (€21.7 billion), property (€8.7 billion), available-for-sale financial assets (€9.6 billion), and other assets (€1 billion).81 Dividing financial assets is largely an accounting exercise, which may make it potentially easier for the UK to include asset offsets in any negotiations over payments. The CER report, however, envisages only the property, and an expanded category of assets available for sale (€13.6 billion), being divisible.82

72.In the case of financial assets, contingent liabilities—primarily in the form of loans, which amount to €57 billion83—are particularly problematic, in that they are counted on the EU’s balance sheet as assets, but are also potential liabilities. As the Bruegel article notes, such loans do not constitute ‘net wealth’, because they are matched by EU borrowing.84 Nonetheless, the CER report comments that the UK may be asked to provide some kind of capital backstop in relation to its budget share, in anticipation of eventual (but not guaranteed) returns. It considers that this scenario “is perhaps the most implausible, but may nonetheless be the Commission’s starting point in talks”.85 The Bruegel article argues that if EU borrowing is considered a liability that should be apportioned, then the corresponding loans should be considered an asset.

73.The Bruegel article notes that pre-financing (€45 billion) is likewise not a divisible asset, but argues that it should be considered in any negotiations over offsetting.86 In the case of pre-financed projects, commitments are already matched by payments and are therefore not included in the RAL. Pre-financing is considered an asset, as a proportion of it will be returned if unused. The UK has, by definition, already provided the required resources for its share of pre-financed commitments. Therefore, if it were divided on an equivalent basis “EU pre-financing would offset a small part of the UK’s share of future commitments.”87

74.Only 6% of the EU’s assets are held as property. These property assets are recorded at historic cost value, minus accumulated depreciation (with the exception of land and artworks, which are deemed to have an indefinite useful life).88 As a result, the current market value of the EU’s property holdings is probably significantly larger than their book value.

75.There was some disagreement among our witnesses about what proportion of property was owned or rented: Mr Ashworth stated that “the assets in terms of buildings in Luxembourg, Brussels and Strasbourg are worth €9 billion. The policy is to own those buildings”.89 By contrast, Dr María-Luisa Sánchez-Barrueco, Senior Lecturer in EU Law at the Deusto Law School, University of Deusto in Bilbao, Spain, said that most buildings were rented or leased, and the recent trend was towards purchasing institutional buildings through emphyteusis (a lease with an option to purchase).90 Jonathan Arnott MEP, a member of the European Parliament’s Budget and Budgetary Control Committees, noted that while EU assets had been “described in the tabloid press under headlines such as ‘Give us wine, art and property’”,91 in reality these formed a relatively small proportion of the total.

76.Our witnesses were split on the question of whether, or how, such assets might be divided. For accounting purposes, assets and liabilities are considered together. In this vein, Ms Grässle’s calculations net the UK’s potential share of EU assets against its share of EU liabilities in suggesting a possible contribution. These calculations fully incorporate the EU’s balance sheet assets (€153.7 billion), regardless of their composition.

77.On a political level, some witnesses felt that assets should be considered as part of any settlement. Dr Darvas stated:

“Since the UK paid in for more than 40 years and the UK was a net payer to the European Union budget, I expect that some of these assets will be [apportioned]. The big question is what the guiding principle should be”.92

He argued that the UK differed from net recipient states, such as Greece, which would not “have the same claim”, and concluded that a potential UK share of assets could be higher than 12%.

78.On the other side of the argument, Dr Sánchez-Barrueco commented that “no additional contribution was requested from the UK to cover a share of the assets already owned by the European Union”.93 In other words, at no point did the UK (or any other acceding Member State) explicitly ‘buy into’ the assets of the EU. The corollary of this point, she stated, was that “no acceding member so far has benefited from a reduction in its budget contribution”, to take account of accrued liabilities such as pension costs.94

79.Jonathan Arnott MEP ultimately concluded that invoking a UK claim on the EU’s assets might be futile, if not counterproductive. He observed that the EU’s liabilities totalled €226 billion, significantly more than the sum of its assets. Therefore, any initial attempt to claim assets could undermine the UK’s attempt to minimise or obviate liabilities. Nonetheless, he believed that it could be a valid strategy, if the EU were to ask the UK take on a share of EU liabilities: “The first thing that we would mention in a response is the assets, I would hope.”95

80.The UK may or may not have a claim against EU assets. However, the EU’s assets are less than its liabilities, and therefore are likely only to come into play in the event that the UK is willing to accept responsibility for contributing to the budget post-Brexit. The EU’s assets total 154 billion: the theoretical maximum the UK could claim would be 23.1 billion, using 15% as the relevant ‘share’. This share is likely to be hotly contested by the EU.

The European Investment Bank

81.The UK’s position with respect to the European Investment Bank (EIB) gives rise to related but separate issues, as it involves a stakeholder share. It might be assumed that if the UK leaves the EIB, it will be paid at least for its stake.

82.Professor Begg stated that “the overall subscribed capital of the EIB is €232 billion … of which 16.1% is British”.96 Shareholdings were calculated according to relative economic size at the time of accession, and the UK is one of the four biggest shareholders (along with Germany, Italy, and France). Professor Begg went on to state that “only about 5% or 6% of that is called up”. He therefore thought that “16.1% of 6% of €232 billion” would “represent the British amount”.97

83.The data published by the EIB in its Financial Report 201598 list overall subscribed capital of €243.3 billion, of which €21.7 billion was called up as of December 2015. The UK’s contribution to the called up capital is €3.5 billion.99 It is likely that the UK would claim its called up capital in the event that it ceased to be a shareholder of the EIB. However, a more useful measure of the EIB’s assets may be the listing of own funds (its equity), which includes reserves, and profits for the financial year. This amounted to €63.3 billion in 2015. A 16.1% stake of this sum, were it to be put into play, amounts to €10.1 billion.

84.The EIB Statute, echoing the words of the Treaties, declares that “the Bank’s members shall be the Member States” (Article 3), and therefore the default position would be for the UK to leave the EIB upon Brexit, unless the other Member States decided (by unanimity) to amend the Statute and the Treaties to allow the UK to retain membership. Dr Jorge Núñez Ferrer, Senior Research Fellow at the Centre for European Policy Studies (CEPS), mused: “Could we open the door to a more regional model?  We could to some extent, also allowing Iceland or Norway to join would be good for the EIB. Why not?”100 Retaining the UK stake could be the more financially straightforward option, although it would be politically and legally more complex.

85.Professor Begg stated that, from the UK perspective, “it may be of interest simply to leave [the UK’s stake] as an investment that generates a rate of return”,101 and in order to help retain access to EIB financing for infrastructure projects. The UK is a significant recipient of EIB funds: in the last three years, £16 billion was invested directly in the UK by the EIB.102 While the EIB does lend to non-EU countries, 90% of its funds are spent within the EU.103

86.The Statute of the European Investment Bank, and the Treaties, state that its members are limited to EU Member States. Unless the 27 remaining Member States were to agree unanimously to amend this rule the UK would have to leave the EIB upon Brexit, losing access to around £5 billion per year in financing. The UK might expect its 3.5 billion in called up capital to be returned if it ceased to be a shareholder. Based on the current net worth of the EIB, the UK may be due a share of equity in the region of 10 billion.

The figures

87.The discussion throughout this chapter illustrates the fact that, even if it were to be accepted that the UK had any financial liability on leaving the EU, no single figure can incontrovertibly represent an amount that the UK might be requested to pay. As Professor Begg stated: “It can be done only on the back of an envelope.”104 Nonetheless, our witnesses suggested a range of numbers across several headings that indicate the possibilities. We have calculated and summarised the absolute amounts associated with their evidence. Where this evidence was insufficient to reach a determination, we have made reference to publicly available reports, such as those published by CER and Bruegel.

88.On the side of liabilities, we heard from a variety of witnesses with different views. Some felt that the UK might be asked to pay towards the MFF after leaving: at rates that reflected measures of the current UK contribution (gross, gross minus the rebate, net of public sector receipts, and net of private sector receipts, or 15%, 12%, 8% and 5% respectively). Under the MFF headings, there are €296 billion in outstanding payment appropriations for the 21 months between April 2019 and the end of 2020.105 Dividing this produces values of €44.4 billion, €35.5 billion, €23.7 billion and €14.8 billion.

89.The upper range of these estimations is unlikely to be demanded or conceded, not least because it relies on the UK’s gross contribution to apportion liabilities. Furthermore, Alex Barker, the author of the CER report, took the view that the whole MFF would not in any case be included, but rather only expenditure covered by the Common Provisions Regulation (principally ESIFs), which would imply that any share could be instead be determined as a proportion of pre-allocated funds (€767 billion for 2014–20, or approximately €191.8 billion over 21 months). Ingeborg Grässle MEP omitted the MFF from her calculations entirely, giving weight to any claims that the potential liabilities under this heading could be zero.

90.For RAL, we heard from witnesses that the share could be 5% (the UK proportion of allocated pre-financing under the current MFF), 8% or 12% (the UK contribution less the rebate, and less public sector receipts), 12.5% (the UK’s share of the EU population) or 15% (the gross contribution). The EU’s estimation of the stock of RAL at the end of 2020 is €254 billion. Using our witnesses’ suggested shares gives numbers of €12.7 billion, €20.3 billion, €30.5 billion, €31.8 billion and €38.1 billion. Again, the largest of these approximations is unlikely; the smallest is worked out on the basis of an apportionment that is unusual with respect to the other financial headings.

91.For pensions, UK shares were mooted in line with the current proportion of the EU’s staff who are UK nationals (4%), the proportion of UK claimants amongst current pensioners (8%), the UK’s share of the population (12.5%) or the UK’s gross contribution (15%). We used the total pensions capitalisation included in the 2015 Annual Accounts of €63.8 billion, while noting Dr Darvas’ opinion that a “reasonable” number could be significantly lower.106 Using the EU’s total produces numbers of €2.5 billion, €5.1 billion, €8.0 and €9.6 billion. Taking Dr Darvas’ total “reasonable” sum of €29 billion as the numerator, by contrast, gives a range from €1.2–4.4 billion. As pensions are paid out from the EU budget, any attempt to apportion them nationally will be the subject of negotiations.

92.In the case of assets, few witnesses made explicit statements about what a UK share might entail, or which assets could be included for accounting purposes. €153.7 billion is the total sum of balance sheet assets. This was included by Ingeborg Grässle MEP in her calculations as an offset against total budget liabilities (€162.3 billion excluding pensions), but as explained above, she was the only witness to include the balance sheet liabilities in this way, as many of the items are not obviously divisible. On this basis, we conclude that the full sum of assets may only be a valid consideration only where all balance sheet liabilities were also included (leaving a share of negative €8.6 billion). Dividing this figure in line with Ms Grässle’s proposed share (12.5%) gives a net figure of negative €1.1 billion.

93.However, not all assets are necessarily divisible, and nor—for legal and political reasons—are they likely to be included en bloc in negotiations. Therefore, we have identified a range of estimates from external sources: a sum of €22.5 billion in divisible assets is quoted in the CER report, consisting of property and assets available for sale; €41 billion is quoted in the Bruegel article by Dr Darvas and his co-authors, representing ‘accumulated wealth’. Similarly a figure of €86.2 billion may be derived from the Bruegel article by adding to the ‘accumulated wealth’ their figure for pre-financing (€45.2 billion). Bruegel notes, though, that the status of this is “ambiguous. Some of it may essentially ‘pre-cover’ part of the UK’s liabilities for future expenditures agreed while it was still a member”.107 This would suggest that any incorporation of this amount would be as an offset towards liabilities.

94.We calculated the proportion of each of these categories of EU assets according to the UK’s current gross contribution and contribution less the rebate, in the absence of any specific suggestion by witnesses as to an appropriate share. This produces a minimum estimate of €2.7 billion (12% of €22.5 billion) and a maximum of €12.9 billion (15% of €86.2 billion). These calculations are purely arithmetic, and may not reflect the nature of political and legal negotiations over assets.

95.In sum, depending on which financial headings are included and on what basis, the range of possible demands is wide. Producing what might be thought of as a ‘best case scenario’ for the UK on this basis involves omitting all contested areas of spending (namely the MFF and balance sheet liabilities). Such an estimation produces a figure of €15.2 billion for RAL and pensions at the lowest shares suggested by our witnesses (5% and 4% respectively). It may also be possible to offset this yet further with a share of assets, although if minimalist figures are used for liabilities, it is not likely that a 15% share would be the determinant of such a number, nor that a large proportion of assets would be considered.

96.At the opposite end of the scale, generating a maximalist demand would imply taking an expansive view of the UK’s current contributions and therefore dividing by the gross figure (15%). It should be noted that Professor Begg was the only one of our witnesses explicitly to identify 15% as a possible denominator;108 most other witnesses used some variety of net figure. Depending on which liabilities and assets are included on this basis, it is possible to produce numbers that match, or indeed exceed, the €60 billion attributed to Mr Barnier.

97.It is possible to arrive at various, widely ranging, figures for any EU claim against the UK. This is not least the case because our witnesses disagreed over which categories of assets and liabilities might be included in any potential demands. If only RAL and pensions were included, at the minimum suggested level of contribution, the total sum would be approximately 15 billion; if gross contributions were used as the determinant and both the MFF to 2020 and budget liabilities were included, the total that is produced by the EU could be even larger than the oft-quoted figure of 60 billion.

24 Alex Barker, Centre for European Reform, The €60 billion Brexit bill: How to disentangle Britain from the EU budget, (February 2017), p 10: [accessed 21 February 2017]. The range is a function of whether or not contingent liabilities are included.

28 European Commission, Consolidated annual accounts of the European Union, Financial Year 2015, (11 July 2016), p 32: [accessed 21 February 2017]

29 Alex Barker, Centre for European Reform, The €60 billion Brexit bill: How to disentangle Britain from the EU budget, (February 2017), p 5: [accessed 21 February 2017]

30 Share of ESIF funds due in 2019 and 2020. Committee’s own calculations based on CER report.

31 The exchange rate between the UK and EU is set by the Commission for the coming year on the 31 December of the previous year. Where figures have not been taken directly from evidence or from third sources, we have used these exchange rates to calculate the difference between the two currencies for the relevant year. These exchange rates are published in the Annual accounts of the European Union: and in equivalent previous editions. European Commission, Consolidated annual accounts of the European Union, Financial Year 2015, (11 July 2016), p 40: [accessed 27 February 2017]

32 Alex Barker, Centre for European Reform, The €60 billion Brexit bill: How to disentangle Britain from the EU budget’, (February 2017), p 9: [accessed 21 February 2017]

34 Office for National Statistics, UK Balance of Payments, The Pink Book: (29 July 2016), Section 9.9: [accessed 21 February 2017]

35 Conversions, where necessary, use the Commission’s exchange rates to the euro calculated at the year end. These can be found in the Annual Accounts of the European Union 2015, p.40. European Commission, Consolidated annual accounts of the European Union, Financial Year 2015, (11 July 2016), p 40: [accessed 27 February 2017]

36 HM Treasury, European Union Finances 2015, Cm 9167, (December 2015), p. 13: [accessed 21 February 2017]

37 Office for National Statistics, ‘UK Perspectives 2016: The UK contribution to the EU budget’, (25 May 2016): [accessed 21 February 2017]. This figure differs slightly from that implied by Table 2, also sourced from ONS figures.

38 This table excludes private sector receipts.

39 Although, as Professor Begg commented, “Bizarrely, that includes some of the public sector, such as universities” (Q 1). The distinction is that such funds are awarded on a competitive basis rather than being enveloped. One of the most important sources of private sector receipts is the Framework Programme for Research and Innovation (known in its 2014–2020 iteration as Horizon 2020), which has an overall budget of €74.8 billion (£58.3 billion) according to the HM Treasury report, European Union Finances 2015, Cm 9167, (December 2015), p 14: [accessed 28 February 2017]

40 Letter to The Rt Hon Norman Lamb MP from Sir Andrew Dilnot CBE (21 April 2016) and Letter to Sir Andrew Dilnot CBE from Jonathan Athow (21 April 2016) on UK contributions to the European Union: [accessed 27 February 2017]

41 Letter to The Rt Hon Norman Lamb MP from Sir Andrew Dilnot CBE (21 April 2016) and Letter to Sir Andrew Dilnot CBE from Jonathan Athow (21 April 2016) on UK contributions to the European Union: [accessed 27 February 2017]

42 European Commission, EU Budget 2014: Financial report (2015): [accessed 27 February 2017]. Committee’s own calculations, based on the EU’s Financial Report 2014) using the total euro-denominated contributions for 2014 less the rebate and public and private sector spending produces a figure of approximately €7 billion (UK payments after the rebate being €14.1 billion) against total payment appropriations of €139 billion, or 5%. However, this figure fluctuates to a large degree from year to year. 2014 was a year of unusually high receipts from Horizon 2020 in particular, meaning that the net contribution is usually larger.

44 This is calculated from the expected date of Brexit—the end of March 2019—to the end of 2020. The UK’s average annual net contribution of £7.1 billion, calculated over 21 months, comes to £12.4 billion.

45 HM Government, The United Kingdom’s exit from and new partnership with the European Union, Cm 9417, (February 2017), p 12: [accessed 21 February 2017]

46 HM Treasury, European Union Finances 2015, Cm 9167, (December 2015), p 18: [accessed 27 February 2017]

48 HM Treasury, Chancellor Philip Hammond guarantees EU funding beyond date UK leaves the EU, (13 August 2016): [accessed 28 February 2017]

50 Ibid.

51 Commission staff working document accompanying Mid-term review/revision of the multiannual financial framework 2014–2020, An EU budget focused on results, COM (2016) 603 final

52 HC Deb 2 June 2016, House of Commons, Session 2016–17, Written Answer 38264

53 The largest recipient of commitment allocations: European Parliamentary Research Service, ‘Structural and Cohesion Funds in the Member States: an overview’, (7 April 2014), [accessed 21 February 2017]

55 Communication from Michel Barnier to the European Commission, Application of the “n+2” rule under article 31.2 of Regulation 1260/1999: [accessed 27 February 2017] This is partly because the initial decommitment rule (‘N+2’) was only introduced in 2002

56 European Commission, Report on budgetary and financial management: Financial Year 2015, p 25: [accessed 27 February 2017]

60 Draft General budget of the European Union for the financial year 2017: General statement of revenue, COM(2016) 300

62 Institute for Fiscal Studies, The budget of the European Union: a guide, (April 2016): [accessed 21 February 2017]

65 HM Treasury, European Union Finances 2015, Cm 9167, (December 2015): [accessed 21 February 2017]

66 Ibid., pp 13–14. These figures differ somewhat from those included in Table 3 due to different sources and calculation methods.

68 European Commission, Staff Regulations of Officials of the European Communities, (1 May 2004): [accessed 21 February 2017]

70 Ibid.

71 Q 3. Professor Begg also suggested a figure of 8%, although based on the proportion of those in receipt of an EU pension who were UK nationals.

77 European Commission, Consolidated annual accounts of the European Union, Financial Year 2015, (11 July 2016), p 32: [accessed 27 February 2017]

78 Zsolt Darvas et al, ‘The Brexit bill: uncertainties in the estimate of EU pension and sickness insurance liabilities’, Bruegel (17 February 2017): [accessed 27 February 2017]

79 The lower ‘reasonable’ number is a function of the Commission’s use of different discount rates in the employee contribution rate calculations and the EU balance sheet calculations, which in the view of the Bruegel article overstates the actuarial cost.

80 European Commission, Annual consolidated accounts of the European Union, Financial Year 2015, (11 July 2016), p 14: [accessed 21 February 2017]

81 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 21 February 2017]

82 Alex Barker, Centre for European Reform, The €60 billion Brexit bill: How to disentangle Britain from the EU budget, (February 2017), p 9: [accessed 28 February 2017]

83 European Commission, Annual consolidated accounts of the European Union, Financial Year 2015, (11 July 2016), p 32: [accessed 21 February 2017]

84 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 21 February 2017]

85 Alex Barker, Centre for European Reform, The €60 billion Brexit bill: How to disentangle Britain from the EU budget, (February 2017), p 8: [accessed 21 February 2017]

86 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 21 February 2017]

87 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 27 February 2017]

88 European Commission, Consolidated annual accounts of the European Union, Financial Year 2014, (23 July 2015), p 40: [accessed 27 February 2017]

97 Ibid.

98 European Investment Bank, 2015 Financial Report, (29 April 2016): [accessed 28 February 2017]

99 Ibid.p 47

103 House of Commons Energy and Climate Change Committee, The Energy Revolution and future challenges for UK energy and climate change policy (Third Report, Session 2016–17, HC 705), p 41

105 European Commission: ‘Budget; Multiannual Financial Framework’ (September 2016): [accessed 28 February 2017]

106 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 21 February 2017]

107 Zsolt Darvas et al., ‘The UK’s Brexit bill: could EU assets partially offset liabilities?’, Bruegel (14 February 2017): [accessed 21 February 2017]

© Parliamentary copyright 2017