## Philipp M Hildebrand: The EU economic recovery - why is it so slow?
Speech by Dr Philipp M Hildebrand, Member of the Governing Board of the Swiss National Bank, at the Brussels Economic Forum, Charlemagne, 21-22 April 2005.
* * *
## Introduction
Let me begin by thanking the Commission and Klaus Regling in particular for inviting me to join this outstanding panel. As an EU outsider, it is a special privilege to join the Brussels Economic Forum. The question I was asked to address today - why is the EU economic recovery so slow - is crucial for many reasons. Let me highlight two.
The asymmetric nature of the global growth dynamic during the last decade is a worrying development for the European Union. Together with Japan, Europe has failed to make a growth contribution commensurate with its share of the world economy. The U.S. economy has had to step in and assume the role of what some have referred to as a 'global consumer of last resort'. One of the consequences of this dynamic is a potentially vulnerable world economy beset with historically unprecedented economic imbalances. How these imbalances will adjust over time is a complex question that goes beyond the topic of today's panel. 1 Suffice it to say, Europe will not be an island of prosperity in the hopefully unlikely event of a financial market crisis triggered by these imbalances.
The EU also has an internally motivated incentive to address its economic weaknesses if it aims to preserve many of the core social values which lie at the heart of the European integration project. In light of the prevailing European demographics, welfare reforms are imperative. In the end, however, the core values of the European social model can only be maintained for future generations if Europe's growth potential is augmented.
I would like to begin by briefly addressing a number of measurement issues that need to be kept in mind when comparing Europe's economic performance with that of the United States. I will then address the underlying cause of Europe's economic weakness, namely its potential growth rate with a specific focus on the all important measure of productivity growth. Finally, I will venture outside of the traditional domain of a central banker's portfolio and offer some tentative conclusions as to where the main EU economic policy challenges will likely lie in the years ahead.
## GDP growth differences EU - U.S.
As you can see in Graph 1, real GDP growth in the U.S. has consistently outperformed growth in the European Union since the early 1990s. Between 1995 and 2003, real annual GDP growth in the U.S. has averaged 3.3% compared to 2.2% in the EU-13. This sustained growth gap suggests that there is a structural dimension to the seemingly cyclical question of why the EU recovery is so weak.
Measurement discrepancies account for some of the gap between official growth statistics in Europe and the United States. These discrepancies emanate from different sources of data, from differences in the scope of measurement and from different ways in which prices and hence price deflators are measured. Price measurement differences exist in the quality adjustment of prices (hedonic pricing) and in the pricing of services in trade and finance as well as in non-market sectors like education and health. In addition, the measurement scope itself can vary. An example is the treatment of military expenses which are categorized as investments in the U.S. and as current expenses in the EU. Finally, data source issues are particularly relevant in the technology sector. In the EU, software investments are derived from typically conservative corporate accounts. In the U.S., software investments are estimated directly from software expenses.
These measurement issues are well known amongst economists on both sides of the Atlantic. Nonetheless, I am struck by how little attention they receive in comparative economic discussions. With increasing focus on data measurement synchronization, measurement differences are fortunately on the decline in cross-country comparisons of national income accounts. Nonetheless, such
1 See e.g. IMF World Economic Outlook April 2005, Chapter III.
BIS Review 27/2005
measurement issues currently explain between 0.5 and at most 1 percentage point of the growth differential between the EU and the United States. 2 We should consider the upper limit of this range as a litmus test to the question of whether there is a structural growth differential between the EU and the United States.
In addition to various measurement differences, about one percentage point of the difference in real GDP growth rates between the EU and the U.S. can be attributed to the higher growth rate of labour input in the U.S. for the period between 1980 and 1995. This is attributable to a stronger increase in the working population in the U.S. and a marked relative decline of effective working hours in the EU.
Factors causing this decline in individual labour input can be divided into two categories: First, numerous rigidities in European labour markets were introduced over the past decades: working hour restrictions, minimum wage agreements, and other restrictions on the conditions of hiring and firing employees. These restrictions are often valid over a broad range of sectors and geographic areas irrespective of economic fundamentals. Second, changes in taxing labour evolved over time with significant tax cuts realized in the United States. 3
In total, overall labour input for the period 1990 to 1995 increased in the U.S. by an average of 1.2% per annum against a decrease of 0.8% per annum in the EU-13. This is pictured in Graph 2. From a European employment perspective, the good news is that Europe is catching up in line with the goals of the Lisbon agenda. From 1995 to 2003, the respective growth rates of overall labour input are still 0.9% for the U.S. against a now positive growth rate of 0.7% per annum for the EU-13.
Thus, while the relative growth of labour input and measurement differences can explain the GDP growth difference up to 1995, this is no longer true since. The data indicates that instead, GDP growth divergence since 1995 has been influenced much more by changes in productivity growth. Productivity growth as change in GDP per work hour is crucial as it determines wages and - together with labour input growth - the potential growth rate.
## Productivity growth differential EU - U.S.
As you can see from Graph 3, labour productivity growth rates since 1995 have been markedly less in the European Union than in the United States. Previously, the reverse was true for several decades. Graph 4 shows that between 1995 and 2000, Europe lagged behind the United States. Labour productivity grew on average 0.6 percentage points per annum less in the EU-11 than in the United States. After 2000, the situation worsened. Between 2001 and 2003, EU-11 lagged the United States by an average of 1.7 percentage points of labour productivity growth per year. In total, the EU-11's relative position in annual productivity growth worsened by 2.9 percentage points for the period of 2000 to 2003 or by 1.8 percentage points for the period of 1995 to 2000 against the period of 1990 to 1995.
The GDP measurement differences I outlined earlier explain at best less than half of this productivity growth gap between the European Union and the United States during the last ten years. Admittedly, we need to be aware of the fact that Europe's economic performance over this period has been significantly and adversely affected by the terms of German unification. The residual productivity growth gap lies at the heart of the question of why European potential growth has been so weak. To understand the gap, we need to focus on different relative conditions in input as well as output markets.
Strongly diverging labour market conditions between the EU and the United States over the decades up to 1995 have led to a massive reduction in EU overall labour input. Rigid labour regulations in Europe restricted labour input and thereby increased the capital-labour ratio which helps explain the higher productivity growth rates in Europe until 1995. From an employment perspective, labour market reforms deserve the high priority they now get by refocusing the Lisbon agenda on 'growth and jobs'.
2 See e.g. Nadim Ahmad, François Lequiller et al., 'Comparing Labour Productivity Growth in the OECD Area: The Role of Measurement', OECD Statistics Working Paper Nr. 5, Paris, 2003.; Philipp M. Hildebrand, 'Wo bleibt das Produktivitätswachstum in Europa?', Speech at the 'Vereinigung Basler Ökonomen', 28 october 2004.
3 Edward C. Prescott, 'Why Do Americans Work so Much More than Europeans?', NBER Working Paper 10316, Cambridge, 2004.
BIS Review 27/2005
Beside the labour market reforms, other reforms should not be neglected. Different conditions in other input markets, for example capital markets, have probably contributed to lower potential growth in Europe. Frictions between national EU capital markets and payment systems render financing more costly. This pushes the capital-labour ratio down, resulting in relative undercapitalization and lower labour productivity. The same line of argument of inefficient allocation of capital applies to non-economic restrictions on corporate mergers and acquisitions as well as on profit transfers. If firms are not allowed to transfer profits from investments ex post the same disincentive to invest ex ante leads to undercapitalization. As in labour markets, the same dynamic is relevant here: Exit barriers are entry barriers.
Different regulations in the product markets represent another important obstacle to an augmented growth performance in Europe. Modern information and communication technologies (ICT) in essence provide leverage to an economy. Product market restrictions on the scope of the use of the technologies lead to undercapitalization and underinnovation ex ante. The potential of these growth dynamics may be grossly underestimated as the costs of the regulations are not easily visible.
In the United States, the largest productivity gains were not realized in the high-tech sector or from the IT investments itself but from the widespread use of information and communication technologies in traditional sectors like wholesale trade, financial intermediation and construction. The new technologies led to a complete reorganization of the value chain thereby enhancing total factor productivity, albeit with a lag. One cannot entirely exclude that this leverage effect will simply occur with a lag in Europe. After all, productivity growth is one of the least well understood phenomena in economics. While I do not want to exclude a productivity lag effect in favour of Europe, I suspect the story is a more complicated and a more challenging one. Structural conditions must be in place in Europe to allow maximum flexibility so that the full potential of the new technologies can be unleashed throughout the real economy.
## Conclusions
The question of why the EU recovery is so slow implies a cyclical answer. The European track record of the last ten years, however, strongly suggests that there is a crucial structural dimension to the problem of weak European growth. The real challenge for the EU is to boost productivity growth and augment its potential growth rate. In the words of Mervyn King: 'Raising productivity growth is the key to improving the prosperity of future generations.' 4 It is therefore not particularly meaningful to analyze the growth weakness in Europe in purely cyclical terms. In other words, expanding fiscal and monetary policy will be no panacea to Europe's growth challenge. Longer-term fiscal consolidation is imperative in light of a rapidly aging European population. On the monetary side, there is no indication that policy is too restrictive in light of real interest rates at or near zero for the Eurozone. Indeed, based on a standard Taylor rule framework, the ECB's monetary policy stance is equally or in the recent past even more expansionary than that of the Federal Reserve. At any rate, neither fiscal nor monetary policy can ultimately address structural impediments to higher growth dynamics in the Euro area.
In an effort to augment the EU's growth potential, the focus must therefore be on breaking down labour market rigidities for additional labour input on the one hand and on accompanying policy measures to increase productivity on the other hand. Despite the subdued nature of the current growth outlook which is adversely affected by a strong Euro and rising oil prices, there are some promising signs that structural change is under way. Both the gradual dismantling of the 35 hour week in France and the shift to part-time and temporary employment in Germany are boosting labour input. In line with this positive albeit hesitant trend of bringing more workers back into the economy, productivity growth rates have begun to recede in Europe. The primary challenge for the European economy is to compensate this natural initial decline in productivity as a result of higher labour input. Product markets must be further liberalized in order to unleash the full productivity potential of the new information and communication technologies. This will in turn further enhance ICT enabled capital deepening by European businesses. Such a virtuous cycle may well be enhanced by an insufficiently understood productivity lag effect from ICT investments which could further benefit European businesses in the years to come.
4 Mervyn King, 'Foreword', in: James Proudman und Stephen Redding (ed.), 'Openness and Growth', Bank of England, 1998.
BIS Review 27/2005
This virtuous process can be initiated. Indeed, many European countries are well on their way. Germany, for example, has steadily increased its competitiveness and is the only G7 country with a growing global export share over the last 5 years. By the same token, we must recognize that the process creates uncertainty for many European citizens. Such heightened levels of uncertainty may impede consumer confidence. Increasing overall labour input decreases labour productivity at least in the short-run and leads to downward pressure on wages. It seems to me of vital political and economic importance to counteract this pressure along two lines.
First, as I have already indicated, we have to ensure that by way of product market reforms these productivity gains find their way through to consumers in the form of lower prices thereby dampening the downward pressure on real wages. In Switzerland, where we face similar issues in the framework of the bilateral treaties, we have introduced an instrument package called 'accompanying measures' to ensure an orderly transition. The officially fixed transition periods offer a time window within which the corresponding reforms on product markets can offset an eventual decline in nominal wages.
Second, European policy makers should focus on educational efforts to efficiently integrate the workers that are being brought back into productive activities through labour market reforms. We must not forget that many of these men and women have been unproductive for many years due to exclusionary labour market practices. Their skills are unlikely to be state of the art when they return to work. Continuing education and on the job training will therefore be crucial elements in bringing up their skills to a level commensurate with the new technologies they are likely to encounter in their newly found jobs.
Let me conclude on an optimistic note. In 1994, Paul Krugman published 'The Age of Diminished Expectations', a beautifully written book that emphasized much of what seemed wrong with America at the time. We now know that 1994 can in many ways be viewed as the starting point of the great American boom of the 1990s. In his foreword, Paul Krugman expresses the hope that 'America will eventually be roused from its slumber and once again begin to face up to problems instead of letting them slide.' It seems to me these words are propitious for the challenges that Europe has begun to tackle. Much as it did in 1994 in the U.S., the hard part lies ahead.
## Appendix
Graph 1
## Average Annual GDP Growth Rates
Source : OECD, Productivity Database, February 2005.
Graph 2
## Average Annual Growth Rate of Labour Input: EU-13 vs. U.S.
Source : OECD, Productivity Database, February 2005.
Graph 3 Growth of Labour Productivity: EU vs. U.S.
Source : Datastream (OECD Economic Outlook).
Graph 4
## Average Annual Growth Rates of Labour Productivity: EU vs. U.S.
Source : OECD, Productivity Database, February 2005.