Structural reforms: learning the right lessons from the crisis
Keynote speech by Benoît Cœuré, Member of the Executive Board of the ECB,
Economic conference, Latvijas Banka,
Riga, 17 October 2014
One lesson that we learned after the financial crisis is that achieving sustainable economic growth is always about both demand and supply. Moreover, policies that support demand empower policies that support supply, and vice versa.
In the post crisis environment of a large debt overhang, partly structural unemployment and weak productivity levels, supply-side policies can increase the effectiveness of demand-side policies and help empower demand. By raising potential growth they create more space for spending (governments, firms and households). Additionally, by reactivating/reskilling the workforce, they allow employment to respond faster to higher demand. Finally, they ensure that higher demand is channelled towards the right sectors.
While their long-term positive effects are undisputed, structural reforms, especially in times when monetary policy is at the lower bound, are seen to have negative short-term effects through increasing real interest rates. However, structural reforms can be effective in the short run as they raise expectations of higher incomes and cause firms to increase investment today and bring forward the economic recovery.
Achieving a positive short-term impact from structural reforms depends on both the pace and composition of reforms. First, instead of the traditionally gradual approach, reforms need to be frontloaded (as e.g. in Latvia). Second, they must focus not only on competitiveness, but also on increasing the rate of productivity growth in the economy.
When thinking about the composition of reforms, one should have in mind innovation (see firms’ market entry and turnover), adoption (e.g. information and communication technology) and reallocation (labour market reforms).
But while the full benefits of these efforts are still emerging, aggregate demand should be supported to counteract the potentially destabilising effect of lower prices and higher real interest rates. This can be done by fiscal policy, when it is available without questioning long-term debt sustainability. And it can be done by monetary policy, even when interest rates have reached the lower bound, by expanding the central bank’s balance sheet and channelling liquidity to the real economy.
Dear Governor Rimšēvičs, ladies and gentlemen,
The early classical economist Jean-Baptiste Say argued that the real purchasing power in the economy is about real production . This claim is often summarised, as "supply creates its own demand" , although that phrase does not appear in Say's writings. While Say’s Law has its limitations, it nonetheless initiated a fruitful academic discussion about how supply and demand interact in the economy.
There is a tendency nowadays to ignore this discussion and look for simple answers to what is a complex economic problem. Some argue that demand side policies are the only way out of the current economic malaise. Others claim that the crisis can be solved by supply side policies only.
But one lesson that we learned after the financial crisis is that achieving sustainable economic growth is always about both demand and supply. Moreover, policies that support demand empower policies that support supply, and vice versa.
The ways in which demand side policies can lift supply are relatively well-established from previous recessions.  A period of weak demand creates a shortfall of investment and a risk that workers are unemployed for too long, lose their skills and becoming structurally unemployed – i.e. hysteresis – both of which then lead to lower potential growth. Thus, the quicker demand policies can return the economy to potential, the lesser the risk that our living standards and employment opportunities will be permanently lower.
The relationship in the other direction, however, is less well established. In most macroeconomic models the effectiveness of demand stimulus does not depend on supply-side measures. But what is missing from such analyses is our specific context in the euro area – namely, the initial conditions we face after the crisis.
Those conditions include, in several countries, a large debt overhang across all sectors due to the pre-crisis credit boom. They include a labour market with very high unemployment, part of which is structural due to the downsizing of previously overblown sectors, especially construction.  They include excessive rent seeking in sectors which have been shielded from international competition. And they include weak productivity trends, emanating from the pre-crisis period and reflecting in part the misallocation of resources into construction and other low productivity sectors. 
In this environment, stabilisation policies become less effective in boosting demand and an increase in demand becomes less effective in stimulating economic activity.
Fiscal space is squeezed by high public debt, and the impact of monetary policy on investment and consumption is blunted by high private debt. The positive impact of higher demand on jobs is constrained by the partially structural nature of unemployment, such as the skill mismatch between job seekers and vacancies, low mobility of workers and the priority given to insiders. Moreover, it is also constrained by capital market fragmentation, i.e. the inability to channel savings where investment is needed to create jobs.
And absent other policies, stimulating demand may in fact only delay the necessary restructuring of the economy towards higher productivity sectors.
This is where supply-side policies – by which I mean structural reforms – can help empower demand. By raising potential growth and hence future income, structural reforms create more space for governments, firms and households to spend today. By reactivating and reskilling the workforce, they allow employment respond faster to that higher demand. And by helping the economy to restructure, they ensure that higher demand is channelled towards the right sectors, and thus accelerates rather than retards the creative destruction process.
What type of supply?
This relationship between supply and demand, however, rests on the assumption that structural reforms will have mainly positive short-term effects. But this needs some more qualification. Though the long-term benefits of structural reforms are by and large undisputed , there are different views about the short-term effects in the academic literature.
Reforms can be shown to produce two, opposing sets of forces in the short-term. One is contractionary, as reforms lead to lower prices and higher real interest rates. If monetary policy is at the zero lower bound and unable to respond and fiscal space has been exhausted, higher real rates cause the private sector to postpone consumption and investment decisions and GDP to contract. 
The other set of forces is expansionary. As reforms raise expectations of permanent income, firms bring forward investment and households bring forward consumption. Moreover, for households the prospect of higher potential growth increases the likelihood of future employment, thus reducing precautionary savings and boosting consumption today. These effects can be achieved at the zero lower bound  and in conditions of a debt overhang .
The key issue is therefore which set of forces dominates. And what matters here is not whether structural reforms are implemented, but how. This is where two further lessons of the crisis come in, which relate to our experience in Europe during the crisis: both the pace and the composition of reforms matter crucially for achieving positive effects in the short-term.
Why the pace of reforms matters
Before the crisis there was an active debate about the appropriate pace of reform implementation, which had its origins in the liberalisation of Central and Eastern European economies in the 1990s. Some argued that reform was best conducted through a “big bang” approach, while others argued for more gradual reform implementation. 
More recently, studies of reform implementation had suggested that outcomes might be better if reforms were sequenced, with product market reforms preceding labour market reforms, so that lower mark-ups can support the purchasing power of households.  There was also a political economy argument in favour of gradualism: as governments have a fixed amount of political capital, they would be most effective if they allocated their scarce resources to one set of reforms at a time.
These arguments were however not always applicable to the euro area as, again, they lacked our specific context. Several euro area economies had experienced a sharp rise in unit labour costs (ULC) associated with the pre-crisis credit boom, which required a period of internal devaluation to correct. And they were starting out with rigid economic structures.
In this context, there was a much stronger case to reform fast so that internal devaluation could take place more quickly. A one-off price adjustment has distributional effects ex post, but it does not affect the perception of real rates going forward – insofar as it boosts future income, it may indeed raise inflation expectations and lower real rates. By contrast, if adjustment is gradual, it creates disinflationary expectations and real rates increase. In other words, a slower adjustment makes the contractionary effect of reforms more prolonged.
This seems to have been borne out from our experience with adjustment during the crisis. Countries that have enacted a more frontloaded reform strategy have, on the whole, seen better outcomes than those that have applied a more staggered approach. Today’s low inflation expectations in the region as a whole may indeed be telling us that the approach was on average too staggered, and that it is time to accelerate.
In this country, Latvia, for instance, a frontloaded programme of fiscal consolidation and structural reforms led to a fast adjustment of wages and prices and a fast rebound in growth and job creation. The Latvian economy began growing again six quarters after it entered recession, and has been on an upward path since. A similar, if less dramatic picture was visible in Ireland, where the economy stabilised after eight quarters, while unemployment levelled off and then fall from the first quarter of 2012. In both countries, prices took around two years to fully adjust and have been rising since.
By contrast, in Greece, Portugal and Spain reforms were less fast and comprehensive. In Spain, for instance, slow implementation of labour market reforms meant that wages continued to rise until the third quarter of 2011, despite output falling by almost 5% in that time. In Greece, labour market reforms were enacted early on but without in parallel opening up product markets, meaning that when wages adjusted, prices did not: the price level only began falling in October 2012. The result was that these countries experienced a quantity adjustment – higher unemployment – which triggered a protracted disinflationary process that is still ongoing. 
In sum, while both options have been costly in terms of output and employment, and notwithstanding differences in labour market flexibility, in a situation where countries had to undertake an internal devaluation, accelerating the pace of reforms seems to have been the better approach. It has led more quickly to higher growth prospects and higher current and expected inflation – i.e. conditions where the positive income effects of reforms can outweigh the negative real interest rate effects.
This contrasting experience also, in my view, disproves the view that reforms should be staggered because political capital is scarce. We have seen that this type of thinking tends to lead to authorities avoiding taking on organised vested interests – such a protected sectors – and instead prioritising measures that affect a more diffuse group – such as lowering minimum wages.
Indeed, this is why, despite the EU-IMF programmes proposing a comprehensive approach to reforms, product markets reforms have tended to lag labour market reforms. And the consequence is not only the mismatch between different measures that I just described. It is that reforms have often been perceived as unfair, which in turn undermines popular support for the reform process.
So for those that still need to press ahead structural measures, including the large euro area economies, there are two key points to take from this.
First, talking vaguely about structural reforms, but not doing them, is the worst of all worlds. It creates uncertainty over the path of real interest rates, without in tandem raising expectations of future growth.
Second, while in “normal” times it might be acceptable to reform one sector at a time, in crisis times it is not. Fairness must be a priority. And the best way to align vested interests is to reform them all at once.
Why the composition of reforms matters
The aim of structural reforms should however be more than internal devaluation. It should above all be to raise productivity growth. As I said earlier, the key to structural reforms having positive short-term effects is firms and households expecting higher permanent income, and altering their behaviour today. And to achieve this, productivity growth is essential.
Remember that in a basic Solow growth model, investment should grow at the growth rate of productivity plus the growth rate of hours worked along the steady state path of the economy. For many European countries, there is scope to increase labour participation rates over time. But given demographic trends, a major part of raising investment demand must be engineering an upward shock to productivity.
Besides, internal devaluation can be the backbone of reform in a small, open economy, but not in a larger and more closed one – and even less in the euro area as a whole, where it is by and large a zero-sum game.
This is why the composition of reforms also matters – we need reforms that both boost competitiveness and raise productivity.
The reforms undertaken so far, despite proceeding at different speeds, have achieved a significant internal devaluation in some euro area countries. Since 2009 Greece has almost entirely reversed the ULC growth it experienced in the pre-crisis period (-19.4% relative to the euro area), while good progress has also been made in Spain (-11.3%), Ireland (-12.3%) and to a lesser extent Portugal (-8.2%).
As a result, most of these countries have been able to improve their competitiveness and export. Spain and Portugal, for example, have persistently outperformed the euro area between 2011 and 2013, with both countries gaining more than 10 percentage points of export market shares relative to their competitors. This seems to be feeding, with about a two years delay from the start of reforms, into a gradual recovery in employment.
But these achievements are not yet showing any signs of feeding into productivity growth. Total factor productivity has actually fallen by 0.8% in the euro area since 2008, whereas in the US it has grown by almost 7% in the same period. Higher competitiveness may well contribute to raising productivity, but we know that cheap labour alone will not produce innovative firms.
Thus, I see a need today to rebalance our focus: to focus less on achieving internal devaluation, and more on raising productivity. And this entails a broader set of reforms than countries have adopted thus far.
Creating a productivity-enhancing environment
For advanced countries in today’s globalised economy, productivity is increasingly linked to investment in knowledge-based capital and moving closer to the technological frontier, especially in high value-added sectors such as the digital economy. While this is ultimately about the effort and ideas of individuals, not governments, public policy plays a crucial role in creating the right environment.
That is, an environment where competition increases the supply of new ideas and pushes frontier firms to innovate and produce new technology; where well-functioning markets diffuse that technology, causing it to be adopted more widely; and where resources are quickly reallocated from failing to growing firms, so that the economy as a whole can reap the returns of innovation, and entrepreneurs have the financial incentives to undertake risky projects.
So, when thinking about how to raise productivity, it is helpful to think in terms of these three categories: innovation, adoption and reallocation. And in each case, our policy frameworks are at present not as effective as they could be, especially in comparison with the US.
In terms of innovation and competition, recent research shows that size of entering and exiting firms tends to be smaller in the US than in Europe, and successful young firms tend to expand relatively more quickly in the US than elsewhere. In other words, successful firms in the US grow faster and unsuccessful firms shrink faster. 
These differences are in a large part due to labour and, in particular, product market regulations that affect firm entry and turnover. Despite recent improvements, according to the OECD index of product market regulation the euro area still lags some way behind the US. The obstacles in the euro area are in part structural – that is, legally protected professions. But they also stem from a generally onerous business environment. For example, if an entrepreneur wants to start a new business in Spain, he or she has to go through 10 separate procedures, while doing so in Slovenia requires only two.  In Latvia, incidentally, the figure is 4 procedures.
Our economic structures in Europe also tend to be less conducive to technology adoption than in the US, a channel which has been found to be highly favourable for productivity growth.  Indeed, an important explanation why, from the 1990s onwards, Europe has lagged behind the US in terms of productivity growth is the slower adoption of information and communication technology, in particular in the services sector. 
The reasons for this are several, but two themes are worth highlighting for the current policy debate.
One is the relatively closed nature of services markets within Europe, which slows down the diffusion of new technologies. Given the relative importance of services in our economies, this weighs on aggregate productivity growth, as well trickling down to other sectors.  This calls for the completion of the single market in services.
A second theme is the relative rigidity of national labour markets. Reaping the productivity gains from adopting new technology tends to require fundamental organisational restructuring – for instance, due to the automation of certain tasks. This means that firms need to have the flexibility to reallocate workers to different tasks, which in turn requires them to have access to adequate initial and vocational training – something which on the whole US firms seem to have been able to do better than European ones. 
As part of this process, reallocation of resources is both unavoidable – as firms need to restructure – and essential – to concentrate resources where they are most productive.
There are however important differences in the extent to which capital and labour flow to innovative firms. For example, researchers at the OECD have found that a 10% increase in the patent stock is associated with an increase in the typical firm’s capital stock of about 3% in Sweden and the United States; 1.5% in the United Kingdom and Germany; and a 0.5% in Italy and Spain. Importantly, these cross-country differences tend be driven by younger firms, which are those that are most likely to supply radically innovative ideas. 
The same product and labour market reforms that would spur innovation and adoption would also support better reallocation. Employment protection legislation in particular is thought to significantly reduce the ability of innovative firms to attract resources.  This is not an argument to lessen workers’ insurance against labour market risks, rather to shift the burden of that insurance away from firms and towards society more widely. This is where “flexicurity” type policies – active labour market policies, unemployment insurances – could play a greater role.
Reallocation is also linked to several other policies, which cannot be covered in detail now. They include aligning insolvency regimes with best practice so that resources can be released more quickly: for a firm to be resolved takes less than six months in Ireland, for example, while in Slovakia it takes about four years.  And they include channelling financing more effectively to innovative firms, where we also lag behind the US due to the fragmentation of venture capital within Europe.  Indeed, this is now part of the wider agenda of the European Commission of building a capital market union. 
Why reforms are a necessary but not a sufficient condition for growth
Let me finally say a word on the macroeconomic environment needed for structural reform to succeed.
As I explained earlier, reforms produce both expansionary and contractionary forces in the short-term. Designing reforms in the right way and creating a productivity-enhancing environment can maximise the former, and let me add: the faster, the better.
But while the full benefits of these efforts are still emerging, aggregate demand should counteract the potentially destabilising effect of lower prices and higher real interest rates. This can be done by fiscal policy, when it is available without questioning long-term debt sustainability. And it can be done by monetary policy, even when interest rates have reached the lower bound, by expanding the central bank’s balance sheet and channelling liquidity to the real economy.  This is what the ECB has been doing and will continue to do.
It is sometimes argued that monetary policy can weaken the incentives to reform. I do not share this view. Price stability, defined as the inflation rate being below but close to 2%, is the anchor of the adjustment process and it is our duty to deliver it within our mandate.
Very low government bond yields can indeed weaken the incentives for sound fiscal policies, and this is why we need strong and credible fiscal rules. But the suffering of young and long-term unemployed in our countries should suffice as an incentive to reform.
To conclude, the main lesson of the crisis is that a major financial shock leaves a set of initial conditions that can only be worked through by policies on both the supply and demand sides of the economy – that is, structural reforms and monetary and fiscal support. Key to the effectiveness of those reforms, however, is that they are fast – to avoid drawing out the disinflationary process – and that they focus on productivity as well as competitiveness.
Today the reform agenda facing European countries is largely about productivity, and this means that pursuing reforms aggressively is less likely to have negative short-term effects. Many of the reforms that lead to downward price pressures and higher real interest rates have already been done, and their effects are working their way through the economy now. The remaining reforms are more about boosting investment demand and productivity and so raising growth today.
Thus, it makes little sense for European countries to stop reforming now. This would mean taking the pain but missing out on the gain. And it would only make getting out the crisis harder.
Jean-Baptiste Say (1834): “A Treatise on Political Economy”.
John Maynard Keynes (1936), “The General Theory of Employment, Interest and Money”.
The seminal paper on this issue began is Blanchard, O., and L. Summers (1986), “Hysteresis and the European unemployment problem”, NBER Macroeconomics Annual 1986, Volume 1.
For more details see speech by Draghi, M. (2014), “Unemployment in the euro area”, annual central bank symposium in Jackson Hole, 22 August 2014.
For details on pre-crisis productivity developments see European Commission (2013), “Catching-up processes in the euro area”, in Quarterly Report on the Euro Area, Vol. 12, Issue 1, March 2013.
See Gomes, S., P. Jacquinot, and M. Pisani (2011), “Structural reforms and macroeconomic performance in the euro area countries: a model-based assessment”, ECB working paper no. 1323; 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. and S. Scarpetta (2005), “Product Market Reforms and Employment in OECD Countries”, OECD working paper no. 472.
See Eggertsson, G., A. Ferrero, and A. Raffo (2013), “Can Structural Reforms Help Europe?”, Journal of Monetary Economics, forthcoming.
See Gerali, A., A. Notapietro, and M. Pisani (2014), “Structural reforms at the zero lower bound”, mimeo.
See Andrés J., O. Arce, and C. Thomas (2014), “Structural reforms in a debt overhang, Banco de Espana, working paper n. 1421.
For a discussion see Blanchard, O., K. Froot, and J. Sachs (1994), The Transition in Eastern Europe, University of Chicago Press.
See Blanchard, O., and F. Giavazzi (2003), “Macroeconomic Effects of Regulation and Deregulation in Goods and Labor Markets”, Quarterly Journal of Economics, Vol. 118.
For a more in-depth discussion on the link between wage developments and labour market institutions see the aforementioned speech by Mario Draghi at Jackson Hole.
Bartelsman, E.J., J. Haltiwanger and S. Scarpetta (2013), “Cross-Country Differences in Productivity: The Role of Allocation and Selection”, American Economic Review, vol. 103(1).
World Bank (2013), Doing Business 2014: Understanding Regulations for Small and Medium-Size Enterprises.
Foster, N., Poschl, J., Rincon-Aznar, A., Stehrer, R., Vecchi, M. and Venturini, F. (2013), “Reducing productivity and efficiency gaps: The role of knowledge assets, absorptive capacity and institutions”, in European Competitiveness Report, DG Enterprise and Industry, European Commission.
See for example Gomez-Salvador, R., A. Musso, M. Stocker and J. Turunen (2006), “Labour productivity developments in the euro area”, ECB Occasional Paper No. 53, October 2006.
Bourlés, R., Cette G., Lopez J., Mairesse, J. and Nicoletti, G. (2012), “Do product market regulations in upstream sectors curb productivity growth? Panel data evidence for OECD countries”, Review of Economics and Statistics, MIT Press, vol. 95(5).
For an explanation of this phenomenon in the US context see Bernanke, B (2005), “Productivity”, remarks at the C. Peter McColough Roundtable Series on International Economics, January 2005.
Andrews, D., C. Criscuolo and C. Menon (2013), “Do Resources Flow to Innovative Firms? Cross-Country Evidence from Firm-Level Data”, OECD Economics Department Working Papers, forthcoming.
Andrews, D., C. Criscuolo (2013), “Knowledge-based capital, innovation and resource allocation”, OECD Economics Department Working Papers No. 146.
World Bank (2013).
Veugelers, R. (2011), “Mind Europe’s early-stage equity gap”, Bruegel Policy Contribution 2011/18, December 2011.
See also speech by Cœuré, B. (2013), “Completing the single market in capital”, ICMA Capital Market Lecture Series 2014, Paris, 19 May 2014.
See Eggertsson, G., and P. Krugman (2012), Debt, deleveraging and the liquidity trap: A Fisher-Minsky-Koo approach, Quarterly Journal of Economics, Vol. 127, No. 3 (August), 1469-1513.