Benoît Cœuré: The structural aspects of euro area adjustment
The views expressed in this speech are those of the speaker and not the view of the BIS.
I would like to thank Beatrice Pierluigi, Sebastian Schmidt and Oreste Tristani for their contributions. I remain solely responsible for the opinions contained herein.
Introduction
Starting with the Lehman collapse, the financial crisis has now persisted for more than five years. For the stressed countries, this period has brought severe social and economic hardships. The delayed, slow economic recovery has given rise to a renewed debate about the effectiveness of structural adjustment policies.
Against that background, I would like to share with you today my views on the adjustment process in the euro area. In doing so, I will organise my talk in three parts. First, I will recall how the imbalances came about. Then I will outline the factors that have made the adjustment a lengthy and costly process. Finally, I will conclude on a cautiously optimistic tone regarding the future.
Pre-crisis imbalances
At the start of Economic and Monetary Union (EMU), the per capita income levels across the joining countries differed significantly. Among the countries that joined the euro area in 1999 and 2001, three countries - Spain, Greece and Portugal - were considered "cohesion countries", i.e. countries that qualified for considerable financial support in the form of EU structural funds to speed up their convergence to the European average per capita income levels.
The textbook neoclassical growth model would suggest that countries lagging in terms of per capita income should experience faster economic growth until convergence. Along the adjustment path, a temporary increase in indebtedness would be warranted.
On the surface, this is indeed what happened in these countries. With the disappearance of the exchange rate premium, the expected convergence process was facilitated by capital flows originating from wealthier euro area countries with higher GDP and capital endowments. Between 1999 and 2008 Greece and Spain witnessed indeed a significant "apparent" convergence of their per capita income to the euro area average. Other countries went through similar experiences, although with some timing differences. In Portugal, for example, this convergence came to a halt as early as 2003. In Ireland, by contrast, a tremendous increase in per capita income continued until 2007.
Identifying the precise time when legitimate expectations of future growth turned overly optimistic, and led to the accumulation of imbalances, is a task which will occupy researchers and policy-makers for a long time. But the outcome of this process is well known. It resulted in buoyant credit growth, high indebtedness by the private sector, excessive wage increases, and increased risk-taking by banks.
Contrary to expectations, aggregate supply did not catch up with demand. This was due to the misallocation of capital in non-productive sectors, and due to important structural rigidities, which policy-makers neglected to tackle in the boom years.
Concerning the public sector, good times were not used to build up the necessary fiscal leeway and asset booms made fiscal positions appear sounder than they were. In some booming economies (e.g. Ireland and Spain) government debt ratios declined, but given the extent to which ample fiscal revenues had been linked to unsustainable asset market developments, structural fiscal balances remained fundamentally weak.
Finally, prices in the non-tradable sector also ran ahead of actual increases in productivity during the boom years, leading to real effective exchange rate misalignments. The ensuing, persistent losses of competitiveness translated into an accumulation of external imbalances.
The facts of the Lehman collapse in late 2008 and the subsequent sovereign crisis in
mid-2010 onwards are well known. As financial markets realised that the combination of high private indebtedness, external imbalances and structurally weak fiscal positions was unsustainable, a "sudden stop" occurred. Private cross-border financial flows dried up, bank lending collapsed and spreads increased sharply in the countries most affected by the crisis.
Crisis management
The policy response of euro area national and supranational institutions was driven by this analysis of pre-crisis developments. Pre-crisis imbalances had to be redressed. Excessive leverage in both the public and private sectors had to fall.
Countries with large external debts ultimately needed to service and repay their obligations in the form of real goods and services exported abroad. An "expenditure switching" was necessary, i.e. an increase in exports and a rotation of internal demand towards domestically produced goods, implying relative price adjustment. This raises three questions. First, which adjustment polices would in theory avoid large employment costs? Second, why were adjustment costs in fact so high for stressed countries? Third, could we have done things differently? Let me elaborate on each question in turn.
Macroeconomic adjustment policies
There is one obvious mechanism which delivers the required macroeconomic adjustment without unnecessary costs: a swift change in relative prices. Exports should become cheaper and imports dearer. Internally, the price of non-tradable goods should fall more to stimulate demand for them. This mechanism could smoothly lead to the necessary "expenditure switching" effect.
However, a slower movement in relative prices makes this process more costly. The rotation of internal demand and the improvement in export performance is slowed down. And economic agents, foreseeing a prolonged period of relative price adjustment, form expectations of persistently low inflation. This, in turn, pushes up the real interest rate faced by local households and firms and thus weighs down on the domestic economy.
For this reason, the key area of policy intervention during the crisis has been in the realm of structural policies aimed at enhancing competition and reducing economic distortions in goods and labour markets. Greater competition in product markets fosters lower prices through a reduction in firms' mark-ups over their marginal costs. The deregulation of the labour market eliminates unfair treatment of outsiders - the unemployed - vis-? -vis insiders - those with a permanent job. The ensuing decline in the bargaining power of insiders is also conducive to lower prices through a reduction in firms' marginal costs.
This approach could be referred to as a once-and-for-all "structural devaluation", to distinguish it from the idea of "fiscal devaluation", which aims at similar objectives through a combination of changes in tax instruments, say, a cut in payroll taxes combined with higher taxes on consumption.1 Structural devaluation reduces external imbalances and also produces substantial, beneficial effects in the long run. These effects operate through higher productivity and competitiveness, as confirmed by model-based and empirical analysis.2 For instance, several empirical studies of OECD member countries find that labour and product market deregulations, by leading to lower markups, have sizeable positive effects on employment and total factor productivity.3
The success of structural reform also has a strong political economy dimension, and the key issue here is that reform is comprehensive. A comprehensive reform process contributes to increasing the perception of a fair distribution of the burden of adjustment across society. This, in turn, helps to diffuse the risk of social tensions and political crises. The structural reforms agenda initiated in the crisis countries - and in particular in the programme countries - has therefore been wide-ranging.4
Challenges for stressed countries
Given the observed benefits of structural reforms, why has implementing them therefore been so costly for some stressed countries? Indeed, from 2009 to 2013 real GDP fell cumulatively by about 7 percentage points in Portugal and Spain, 8 percentage points in Italy, 11 percentage points here in Slovenia and 26 percentage points in Greece. At the same time, the unemployment rate rose to very high levels in all these countries, with the rate of youth unemployment being more than double the overall one.
The first contributing factor is that most of the countries involved entered the crisis with rigid institutions and structures. An instantaneous adjustment of relative prices was not feasible due to nominal wages and prices being "sticky".5 As a result, both the duration and the real cost of adjustment have increased sharply, as did the burden borne by the more vulnerable members of the society, affecting the political acceptability of reforms.
The cost of high rigidities can be well understood when comparing the adjustment in Ireland with those in the other programme countries. In the first case, the correction in relative wages and prices occurred immediately after the 2008-09 recession and basically all the nominal adjustment had already been made before Ireland entered the financial assistance programme in late 2010. This is very different from what happened in Greece, Portugal and Spain, where the nominal adjustment has lasted many years after the recession. As a result, in Ireland, an export-driven recovery started back in 2011, while the unemployment rate started to decline in 2012. For the other three countries, the recovery story had to wait two or three more years.
Second, when reforms were undertaken, there were implementation obstacles. Ex ante the approach to structural reforms was broad-based and aimed at all critical weaknesses in a comprehensive manner. But ex post it turned out that it was easier to change institutions and rules in the labour market than in product and services markets. This meant that marginal cost improvements did not automatically translate into price improvements. For example, labour costs fell by around by 18% in Greece from 2009-13, but price adjustments did not follow until very recently in Q1 2013. This not only limited the potential for the external sector to generate growth, but also lowered citizens' real incomes.
The third contributing factor was that, while this supply-side adjustment taking place, there were also constraints on the demand side. To begin with, the reform process had to start in an environment where fiscal policies had, by necessity, to become contractionary. Clearly, contractionary policies were needed to restore debt sustainability and market access. Indeed, structural fiscal balances - i.e. fiscal balances excluding the impact of the cycle and one-off costs - improved significantly in the crisis countries. From 2010 to 2013, this improvement was equal to about 15 percentage points in Greece, 6.5 percentage points in Portugal, 6 percentage points in Spain, 5.5 percentage points in Ireland, 4 percentage points in Italy and 3 percentage points in Slovenia. This gives us a notion of the discretionary fiscal adjustment.
That said, there were also implementation obstacles that may have worsened the contractionary impact of consolidation. For example, it is always easier to raise taxes than to reduce spending, and to cut capital expenditure rather than consolidate current expenditure, both of which lowered the quality of fiscal consolidation.
Moreover, the credit transmission channel was impaired. In countries where the banking sector has experienced large balance sheet shocks related to deteriorating sovereign credit, access to bank credit has tightened substantially. Elevated risk premia in the stressed countries have in turn had repercussions on access to finance, in particular for small and medium-sized enterprises (SMEs). Their restricted access to finance in a number of euro area countries has clearly been a threat to economic recovery. The ECB has partly alleviated the funding constraints of banks extending loans to SMEs by accepting such loans as well as asset-backed securities (ABS) backed by pools of loans as collateral in its monetary policy operations. But the main lesson of the crisis has been that the restructuring and strong capitalisation of banks is an important condition for restoring the smooth provision of credit to euro area corporates and households. This is why the comprehensive assessment of the banking system is important so that the banks can free up balance sheet capacity to extend new loans.
Assessment
In light of this experience, it is worth asking whether the reforms could have been implemented differently. For example, did it make sense for euro area countries to reform while fiscal policy was contractionary? Or when monetary policy was close to the zero lower bound and therefore could not react to adverse cyclical developments? I have three points to make here.
First, in my view, the option of delaying reforms was simply not there for euro area countries. Political economy reasons suggest that reforms are often implemented under the pressure of economic crises.6 Delivering a credible commitment to implement reforms in the future is therefore very difficult for any country - and especially so for European countries that do not have a good track record in this field.
At the onset of EMU, many observers took for granted that structural reforms would be a necessary choice for countries that had given up monetary independence, or in other words, for the Eurozone to be an optimum currency area.7 If reforms had not been carried out over the past decade when conditions were very favourable, why should anyone expect them to be carried out when the economy improves again?8 I therefore remain convinced that delaying the reform process would have been counterproductive.
Second, one can exaggerate the effect of the zero lower bound. The ECB has employed various standard and non-standard measures that have helped to smooth the adjustment process for stressed countries. The provision of liquidity to banks through fixed-rate tender procedures with full allotment, and against an expanded list of eligible collateral, has been crucial to prevent further catastrophic effects arising from the sudden stop that some countries experienced.9 The expansionary monetary policy stance has ensured that inflation expectations remained well anchored over the medium term, thereby ensuring that short-run price adjustments do not translate into longer-term deflationary expectations. Price stability has been also crucial in avoiding increases in the real values of nominal debts - increases which would be unwelcome throughout the euro area, but especially pernicious in countries under stress. There are monetary policy instruments that could be used in the event of downward risks to medium-term price stability, even if the nominal interest rate is constrained by the zero lower bound.
Third, besides monetary policy, we should also not forget the additional support for governments that came from external assistance, which also helped to offset the effects of a sudden stop. Between 2010 and 2013 Greece, Ireland, Portugal and Cyprus requested financial assistance from the EU and the IMF. In 2012, Spain requested financial assistance from the EU for the recapitalisation of its banks. At the same time, a permanent solidarity framework - the European Stability Mechanism - was put in place, to provide financial assistance to Member States in difficulty, thereby reducing threats to financial stability in the euro area.
Conclusions
Let me conclude.
The crisis has been long and painful for many people in Europe. Nevertheless, and in spite of persistent challenges, much of the macroeconomic adjustment has now taken place. Looking back, 2013 was certainly the year of the turnaround.
Since 2008, the current account deficits of stressed countries have narrowed substantially. Greece, Spain and Portugal moved from double-digit deficits and in 2013 achieved slight current account surpluses. Over the same period, the current account balance of Ireland went from a deficit of around 5% of GDP to a surplus of around 5% of GDP. Some of the adjustment certainly reflects a decline in imports due to the cyclical compression of domestic demand. But in Spain, Portugal and Ireland, the larger part of the adjustment is attributable to robust export performance. Structural improvements have played an important role in the adjustment of current account imbalances and, in fact, cyclical factors are estimated to account for less than half of the adjustment observed so far.10
Export growth has also been the engine of the positive GDP growth rates since the second quarter of 2013 in some crisis countries. With the return to positive growth rates last year, the upward trend in the unemployment rate came to an end in most euro area countries. In Portugal unemployment has started to decline. Let me stress that this is not enough and unemployment rates, particularly among young people, remain unacceptably high in many crisis countries. However, the labour market reforms undertaken give us confidence that unemployment will start falling faster once the recovery picks up.
It is, therefore, crucial to continue with the reform process in all euro area countries, including those unaffected by the crisis. Undertaking reforms simultaneously can boost GDP more than in a situation where each country acts alone.11 And importantly, a simultaneous approach will also help to make the reforms more socially acceptable. It will underscore the simple truth that in the euro area we are all in the same boat, and we are all heading in the same direction, across countries as well as within countries.
1 See e.g. the IMF Fiscal Monitor on Addressing Fiscal Challenges to Reduce Economic Risks, September 2011, and E. Farhi, G. Gopinath and O. Itskhoki (2011), "Fiscal Devaluations", NBER Working Paper No. 17662.
2 See Blanchard, O., and F. Giavazzi (2003), "Macroeconomic Effects Of Regulation And Deregulation In Goods And Labor Markets," The Quarterly Journal of Economics, MIT Press, vol. 118(3), pages 879-907, August.
3 See Berger, H., and S. Danninger (2007), "The Employment Effects of Labor and Product Market Deregulation and Their Implications for Structural Reform", IMF Staff Papers Vol. 54, No. 3; and Nicoletti, G., S. Scarpetta (2005), Product Market Reforms and Employment in OECD Countries, OECD working paper No. 472.
4 It included, to different extents in different countries, public administration, health and pension systems, education, judicial systems, competition frameworks, industrial relations, labour markets, energy markets, network industries, services sectors and regulated professions.
5 See Álvarez et al. (2006). "Sticky Prices in the Euro Area: A Summary of New Micro-Evidence". Journal of the European Economic Association 4 (2): 575-584.
6 See e.g. Drazen, A., W. Easterly (2001), "Do Crises Induce Reform: Simple Empirical Tests of Conventional Wisdom?", Economics and Politics, Vol. 13; and J. Høj, V. Galasso, G. Nicoletti and T. Dang (2006), "An empirical investigation of political economy factors behind structural reforms in OECD countries", OECD Economic Studies No. 42, 2006/1.
7 See e.g. C. Bean (1998), "Monetary policy under EMU" Oxford Review of Economic Policy 14, pp. 41-53.
8 EMU@10, (2008), "Successes and challenges after ten years of Economic and Monetary Union". European Economy, Nr. 2/2008.
9 See S. Fahr, R. Motto, M. Rostagno, F. Smets and O. Tristani (2013), "A monetary policy strategy in good and bad times: lessons from the recent past," Economic Policy, CEPR & CES & MSH, vol. 28(74), pages 243-288, 04.
10 See the box "To what extent has current account adjustment in the stressed euro area countries been
cyclical or structural?", Monthly Bulletin, ECB, January 2014, http://www.ecb.europa.eu/pub/pdf/mobu/mb201401en.pdf.
11 See Varga J., J. in't Veld (2013), "The growth impact of structural reforms", Quarterly Report on the Euro Area, European Commission Vol. 12/4, page 17-27.