By Léo Aparisi de Lannoy and Xavier Ragot

The return of growth cannot eradicate the memory of how the crisis was mismanaged at the European level economically, but also socially and politically. The divergences between euro area countries in unemployment rates, current account balances and public debts are at levels unprecedented for decades. New steps in European governance must aim for greater economic efficiency in reducing unemployment and inequalities while explaining and justifying the financial and political importance of these measures in order to render them compatible with national policy choices. The establishment of a European unemployment insurance meets these criteria.

The idea of a European mechanism for unemployment compensation is an old idea dating back to at least 1975. The idea is now being extensively debated in Europe, with proposals from Italian and French economists and policymakers and studies conducted by German institutes, with the latest OFCE Policy Brief offering a summary. The possibility is even being mentioned in communications from the European Commission. The Policy Brief describes the European debates, as well as the system in place in the United States.

The European unemployment insurance mechanism presented in this note aims to finance the unemployment benefits of countries experiencing a severe recession and draws on the US experience to do this. A programme like this would constitute a second European level, supplementing the different national levels of unemployment insurance. It would help provide the unemployed support in countries hit by a deep recession, which would also contribute to sustaining aggregate demand and activity while reducing inequality in the recipient countries. It is also consistent with a reduction in the public debt. This mechanism would not lead to permanent transfers to countries that are not carrying out reform, nor to unfair competition or the transfer of political powers that are now covered by subsidiarity. As in the case of the United States, it is consistent with the heterogeneous character of national systems.

To give an order of magnitude, an insurance system that is balanced over the European economic cycle and involves no permanent transfers between countries would have boosted growth in Spain by 1.6% of GDP at the peak of the crisis, while Germany would have received European aid from 1996 to 1998 and from 2003 to 2005. France would have experienced a GDP increase of 0.8% in 2013 thanks to such a system, as shown by the simulations conducted by the European teams.

For the complete study, see: Policy Brief de l’OFCE, no. 28, 30 November 2017.

By Eric Heyer

Six months following the signing of a national industry-wide agreement on unemployment benefits between the social partners, with new rules that normally are to apply until 2016, the French government, which wants to go further in reforming the labour market, is evoking the possibility of once again reforming the unemployment insurance system by reducing the level of benefits and the period they are paid.

It is far from clear that reforming the unemployment insurance system is in keeping with the idea that any reform must improve the “quality of life” of our citizens. This is, in any case, what is indicated by the latest publication of the OECD.

In Chapter 3 of the 2014 edition of the OECD’s Employment Outlook, the international organization has implemented the recommendations of the 2009 Stiglitz-Sen-Fitoussi report by evaluating the quality of employment in the OECD countries. This new indicator supplements conventional measures of the quantity of work and should eventually lead to transforming the content of public policy by imposing new assessment criteria on the public authorities.

The OECD constructs an indicator on the quality of employment on the basis of three factors: the quality of wages, the security of the job market, and the quality of the working environment. According to the OECD, this last dimension is relatively mediocre in France: the high level of professional requirements and insufficient resources to accomplish tasks leads to a high level of on-the-job stress for French employees. As for wages, a review of both their level and distribution places France close to the average of the OECD countries. Finally, while the quality of work in the country is close to average in the developed countries, this is, according to the OECD, due mainly to a high level of job security in France, due to both the extent of social security … and the generosity of unemployment insurance.

The proposals for reforming unemployment insurance would therefore tend to deteriorate rather than improve the “quality of life” for the French, and would thus miss their target from that perspective. But would they lead to improving the quantity of work?

There is some food for thought on this subject in Chapter 1 of the Report, in which the OECD indicates that the structural unemployment rate – i. e. the unemployment rate depending on the impact of rigidities that prevent the labour market from functioning properly – has not increased since the onset of the crisis in France, just as is the case in many other developed countries: for the OECD, the sharp increase in unemployment seen since 2008 has a mainly cyclical component that cannot be combated by reforming unemployment insurance.

As a consequence, given the current situation of the French economy, reforming unemployment insurance along the lines suggested by the government will, if the OECD analysis is to be believed, undermine the quality of employment – and in particular the quality of life of the unemployed – without reducing the level of unemployment!

By Xavier Timbeau

In the latest publication of France’s Treasury Department,  Lettre Trésor-Eco, no. 132, June 2014 (Ministère des Finances et des Comptes publics and Ministère de l’Économie du Redressement productif et du Numérique), Thomas Lellouch and Arthur Sode develop the operating methods and the merits of a common unemployment insurance for the euro zone. They specify the main steps of how it would be applied, which would ensure neutrality between the Member States. They argue for harmonized employment and labour market policies, leading in the long term to a single contribution rate in the euro zone:

– “Harmonization at the euro zone level of an unemployment insurance component would provide the euro zone a new solidarity instrument capable of giving a social Europe real substance while ensuring greater stability of the zone as a whole…

– This common base could compensate e.g. those who are unemployed less than one year (the most cyclical component) at 50% of their past salary, with financing determined on a harmonized base (e.g. payroll). It would be supplemented by national compensation in accordance with the preferences of each state, thus ensuring the continuation of the current level of compensation…

– Modulating the contribution rate of each member according to its unemployment level, with regular updates based on past trends, would ensure ex ante budget neutrality between the Member States…

– In the longer term, and after the unemployment rates of the various Member States converge, a system marking greater solidarity between the Member States could be considered, with financing through a single contribution rate …”.

New solidarity, but posing three problems …

Unemployment insurance functions as an important automatic stabilizer. Having a common system for the euro zone members would have made possible significant transfers during the crisis we have just been through. Based on the scheme proposed by the authors (pooling the most cyclical component), Spain could have benefited from almost 35 billion euros by end 2012, mainly from Germany and France. This would not be sufficient to cancel Spain’s public deficit, but it would have kept down its level.

A system like this could play a major role in avoiding the sovereign debt crises that dry up a State’s credit. It would introduce solidarity and neutral transfers during cycles, but would be responsive to the state of the cycle.

However, this proposal raises three problems: the first is that unemployment insurance systems are the fruit of a national social compromise that has won general acceptance and is consistent with the rest of the country’s labour market policies, whether these are active policies or not. A European unemployment insurance component built on top of national systems could lead to confusion and to questions about the national balance. This could disrupt the social dialogue, since the social partners would have a potential resource for which they are not responsible, in addition to the issue of whether the European authorities or partner countries might also wish to have a say. Furthermore, unemployment insurance is often a sensitive subject, as was seen by the issue of entertainers and artists (intermittents) in France in early summer 2014.

This could be solved by limiting the sharing to macroeconomic transfers, independent of national arrangements. But, and this is the second problem, to ensure that transfers between states do not become permanent, the transfers need to be balanced over the business cycle. This requires a procedure for identification of the cycle that the stakeholders agree on. The recent experiences of the crisis and the calculation of structural deficits show that this is far from the case today. Another option would be to “replenish” the system prior to using it by accumulating contributions over a number of years before a major downturn. It would suffice to limit use to what has been accumulated to resolve discrepancies. But then the system would be bereft of value in the face of a systemic crisis. The day the buffer collapses, the Kings would be as naked as before. At best the crisis is delayed, at worst it is aggravated.

A final option would be to give up balancing the transfers a priori (or by the mechanics of the way it operates), leaving it to polarize gradually one way or another and to ensure an asymptotic convergence. But in this case the system could lead to undesired structural transfers that could very well call it into question.

Spain for instance has high unemployment, well above its structural rate; entering into a transfer system based on the differences between current unemployment and structural unemployment could be done only on an equilibrium basis, or would run the risk of a long-lasting initial transfer.

This then raises the third issue, governance. It is difficult to design such a system without implying, at least potentially, significant transfers between States. How could such transfers be justified without a legitimate common representation? Furthermore, what could be done to avoid these transfers becoming an instrument for control of macroeconomic policy as a whole? The establishment of a banking union is a reminder of how key this problem is. Likewise, Spain’s refusal to submit to the conditions set for a conventional assistance program (EU / IMF) clearly indicates that in the absence of legitimate and sincere solidarity, the beneficiaries of transfers will be as suspicious as the payers.

By Gérard Cornilleau

The Cour des comptes [Court of Auditors] has presented a report on the labour market which proposes that policy should be better “targeted”. With regard to unemployment benefits in particular, it focuses on the non-sustainability of expenditure and suggests certain cost-saving measures. Some of these are familiar and affect the rules on the entertainment industry and compensation for interim employees. We will not go into this here since the subject is well known [1]. But the Cour also proposes cutting unemployment benefits, which it says are (too) generous at the top and the bottom of the pay scale. In particular, it proposes reducing the maximum benefit level and establishing a digressive system, as some unemployed executives now receive benefits of over 6,000 euros per month. The reasoning in support of these proposals seems wrong on two counts.

In the first place, the diagnosis of the system’s lack of sustainability fails to take the crisis into account: if Unedic is now facing a difficult financial situation, this is above all because of falling employment and rising unemployment. It is of course natural that a social protection system designed to support employees’ income in times of crisis is running a deficit at the peak of a crisis. Seeking to rebalance Unedic’s finances today by cutting benefits would abandon the system’s countercyclical role. This would be unfair to the unemployed and economically absurd, as reducing revenues in a period of an economic downturn can only aggravate the situation. In such circumstances, it is also easy to understand that arguments for work incentives are of little value: it is at the top of the cycle, when the economy is approaching full employment, that it makes sense to raise the issue of back-to-work incentives. When the economy is bumping along the bottom, encouraging a more active job search may change the distribution of unemployment, but certainly not its level.

The current deficit in the unemployment insurance system simply reflects the situation of the labour market. A few calculations can help to show that the system’s generosity is fully compatible with financial stability in “normal” times. To establish this, we simply measure the impact of economic growth, employment and unemployment on the system’s deficit since 2009. In 2008, Unedic was running a financial surplus of nearly 5 billion euros [2]. This turned into a deficit of 1.2 billion euros in 2009 and 3 billion in 2010, before recovering somewhat in 2011 with a deficit of only 1.5 billion, which then rose to 2.7 billion in 2012. For 2013, the deficit is expected to reach 5 billion. The Table shows our estimates of the impact of the crisis on the system’s revenues and expenditures since 2009. The estimated revenue lost due to the crisis is based on the assumption of an increase in annual payroll of 3.5% per year (which breaks down into 2.9% for increases in the average wage and 0.6% for rises in employment) if the crisis had not occurred in 2008-2009. On the expenditure side, the estimated increase in benefits due to the crisis is based on the assumption of a stable level of “non-crisis” unemployment, with spending in this case being indexed on the trend in the average wage.

The results of this estimation clearly show that the crisis is solely responsible for the emergence of the substantial deficit run up by the unemployment insurance system. Without rising unemployment and falling employment, the system would have continued with a structural surplus, and the reform of 2009, which allowed compensation for unemployed people with shorter work references (4 months instead of 6 months), would have had only a minimal effect on its financial situation. There was no breakdown of the system, which was in fact perfectly sustainable in the long term … so long as counter-cyclical economic policies are implemented that prevent a surge in unemployment, whose sustainability is now undoubtedly more of a concern than the finances of Unedic [3].

Based on a diagnosis that is thus very questionable, the Cour des comptes has proposed reducing the generosity of unemployment benefits. Since it is difficult to put forward proposals for cutting lower benefit levels, the Cour put more emphasis on the savings that could be achieved by limiting very high benefits, which in France may exceed 6,000 euros per month for executives on high-level salaries that are up to 4 times the maximum social security cap, which in 2013 was 12,344 euros gross per month. In reality, from a strictly accounting perspective, it is not even certain that this will have positive effects on Unedic’s finances. Indeed, few people benefit from these top benefit levels, because executives are much less likely to be unemployed than are other employees. On the other hand, their higher salaries are charged at the same contribution rates, meaning that they make a net positive contribution to financing the scheme. Calculations based on the distribution of wages and of the benefits currently received by unemployed people insured by Unedic show that employees who earn more than 5,000 euros gross per month receive about 7% of unemployment benefits but provide nearly 20% of the contributions. For example, we simulated a reform that would bring French unemployment insurance into line with the German system, which is much more severely capped than the French system. The German ceiling is 5,500 euros gross per month (former Länder), against 12,344 in the French system. By retaining a cap of 5,000 euros gross per month, the maximum net benefit level in France would be around 2,800 euros. Based on this assumption, the benefits received by the unemployed in excess of the ceiling would be reduced by nearly 20%, but the savings would barely amount to more than 1% of total benefits. On the revenue side, the lower limit would result in a reduction in revenue of about 5%. The existence of a high ceiling in the French unemployment insurance system actually allows a significant vertical redistribution because of the differences in unemployment rates. Paradoxically, reducing insurance for the most privileged would lead to reducing this redistribution and undermining the system’s financial stability. Based on the above assumptions, shifting to a ceiling of 5,000 euros would increase the deficit by about 1.2 billion euros (1.6 billion revenue – 400 million expenditure).

This initial calculation does not take into account the potential impact on those whose unemployment benefits would be greatly reduced. To clarify the order of magnitude of this effect, which is, by the way, unlikely, we simulated a situation in which the number of recipients of the highest benefits would be cut in half (e.g. by a reduction in the same proportion of the time they remain unemployed). Between the new ceiling and the highest level of the reference salaries, we estimated that the incentive effect increased linearly (10% fewer unemployed in the first tranche above the ceiling, then 20% fewer, etc., up to -50%). Using this hypothesis of a high impact of benefit levels on unemployment, the additional savings on benefits would be close to 1 billion euros. In this case, the reform of the ceiling would virtually balance (with an added potential cost [not significant] of 200 million euros). But we did not include the fact that the shortening of the duration of unemployment compensation for unemployed people on high benefits could increase the duration of the unemployed on lower benefits. In a situation of near full employment, it is possible to consider that the rationing of employment results from the rationing of the supply of work; in the current situation of a generalized crisis, the more realistic case involves the opposite situation of a rationing of demand for labour. Achieving budget savings by cutting high benefit levels is not credible, at least if we stick to a reform that does not change the very nature of the system.

One could of course obtain a more favourable result by reducing only the cap on benefits and not the cap on contributions. This would be very destabilizing for the system, since it would strongly encourage executives to try to pull out of a unified solidarity system that provides them with reasonable assurances today through the acceptance of a high level of vertical redistribution, while lowering the cap on benefits alone would force them to insure themselves individually while continuing to pay high mandatory fees. This type of change would inevitably call into question the basic principle of social insurance: contributions based on each person’s means in return for benefits based on need.

The general economics in the Cour’s report on unemployment benefits thus seem highly questionable because, by not taking into account the effect of the crisis, it winds up proposing a pro-cyclical policy that puts additional burdens on the unemployed at a time when it is less possible than ever to make them bear the responsibility for underemployment. As for the key measure that challenges the compromise on high level benefits, it would at best be budget neutral and at worst destroy the social contract that today makes possible strong vertical redistribution within the unemployment insurance system.

[1] Unemployment insurance has a special scheme for interim workers in the entertainment industry worth a billion euros per year. It would obviously be sensible for this expenditure to be borne by the general budget and not by Unedic.

[2] Excluding exceptional operations.

[3] On economic policy in Europe and the lack of macroeconomic sustainability, see the initial report of the Independent Annual Growth Survey project (IAGS) .