3. Long-Term Growth Differentials within Europe

International Monetary Fund. European Dept.
Published Date:
October 2011
  • ShareShare
Show Summary Details

In the past decade, growth rates in GDP per capita have differed markedly among European countries, from zero in Italy and Portugal to more than 4 percent in the best performers. To a large extent, the growth differentials reflect convergence. However, a number of countries have grown less than their potential because of poor macroeconomic policies and barriers to growth. The experience of earlier reformers provides useful lessons for current poor performers. Reforms do make a difference, but their implementation takes time, and their impact is felt only with a lag. Reforms would not only speed up convergence within Europe, but also help close the productivity and innovation gaps with the United States.

Growth Differentials in Europe

Across Europe, countries have experienced a wide variation in per capita GDP growth over the past decade (Figure 3.1).6 Growth rates have ranged from close to zero in Italy and Portugal to more than 4 percent in Albania, Estonia, Latvia, Lithuania, Moldova, Russia, and Ukraine.

Figure 3.1European Countries: Change in Real GDP Per Capita, 2000–10

(Annualized percent)

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Convergence explains a large part of these differences…

Poorer European countries have generally grown faster than richer countries, a process called “convergence.” While there is no clear evidence of absolute convergence in the world, convergence is usually observed within more homogeneous groups of economies—a phenomenon called conditional convergence.7,8 It is noteworthy that convergence in Europe has been stronger than in Latin America or Asia—regions that are not as economically integrated (Figure 3.2). Much of the convergence in Europe is due to rapid growth of emerging European countries, as they have adopted institutions similar to those in advanced Europe and benefited from higher investment rates, financed with intra-European capital flows.9

Figure 3.2Convergence in the Three Global Regions, 2000–10

Source: IMF, World Economic Outlook database.

1Data for Montenegro and Malta are for 2000.

Growth theory identifies two factors that drive convergence: diminishing returns in the accumulation of capital and cross-country knowledge spillovers. Poorer countries usually have a lower capital stock and therefore, a higher marginal productivity of capital: increases in capital stock will thus have a large impact on output. Poorer countries can also boost output by imitating technologies already developed in richer and more advanced countries—a process that will raise total factor productivity (TFP).

Developments in the past decade have been in line with what the theory suggests: higher growth in poorer countries was the result of both faster capital accumulation and higher TFP growth. Higher returns on investment attracted strong capital flows, which financed higher investment rates and a more rapid accumulation of capital stock than in richer countries (Figure 3.3). In addition, poorer countries, facilitated by the EU enlargement process, achieved higher TFP growth by adopting new technologies and better institutions (Figure 3.4).

Figure 3.3European Countries: Contribution to GDP Growth of Investment and Capital Flows, 2000–10

Sources: IMF, World Economic Outlook database; IMF staff calculations; and Penn Tables.

1In percent of 2003 GDP; average of 2003–10.

Figure 3.4European Countries: Change in TFP Relative to Per Capita GDP, 2000–09

Sources: Conference Board Total Economy Database, January 2011; and IMF, World Economic Outlook database.

1 Data for Emerging Europe are for 2000–08.

By contrast, the contribution of employment and human capital to average GDP growth has been lower in emerging Europe than in advanced Europe (Figures 3.5 and 3.6).10 The lower contribution of employment growth in emerging Europe during the period was most likely related to the slower growth of its working-age population, which, in turn, was exacerbated by the emigration of workers to advanced Europe.

Figure 3.5Europe: Contribution to Growth of Output Per Hour Worked

(Annualized average rate, 2000–08, percentage points)

Sources: Conference Board Total Economy Database, January 2011; and IMF staff calculations.

1This reflects the change in the composition of labor measured on the basis of weighted measures of different skill-level groups in the labor force.

Figure 3.6Europe: Contribution to Growth of Employment, 2000–08

Sources: Conference Board Total Economy Database, January 2011; and Eurostat.

Note: Data for working age population for Croatia are 2002–10, for Iceland 2003–10, and for Macedonia, FYR and Turkey 2006–10.

…but institutions and policies are equally important

Growth differentials have been more variable than convergence alone can account for. Figure 3.7 shows “adjusted” growth rates—the difference between each country’s actual growth rate and the growth rate that would have been expected given initial income levels. While the precise figures are sensitive to the shape of the expected convergence line, it is clear that considerable differences exist and that some countries have done much better (and others much worse) than what would be expected on the basis of income differentials alone. For instance, Italy and Portugal have grown much slower than expected, while the Slovak Republic and Sweden have grown faster.

Figure 3.7Europe: Growth Experience Beyond What Is Explained by Convergence

(Adjusted change in real GDP per capita, 2000–10, annualized, percent)1

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

1The adjusted growth measures the difference between each country’s actual growth rate and the growth rate that could be expected given initial income levels.

These growth gaps are associated with key differences in factors such as market structures, human capital stocks, institutions, and macroeconomic policies. The economic literature has identified a large number of factors likely to influence economic growth (Box 3.1). Select factors, discussed below, seem particularly relevant in differentiating fast-growing from slow-growing countries in Europe.

The importance of these factors differs across countries. The growth bottlenecks in countries that are catching up differ from those in countries that are at the technology frontier. For instance, policies promoting macroeconomic stability, flexible labor markets, and a well-educated workforce help growth in both sets of countries. Policies strengthening product market competition, better protection of property rights and legal security, and more innovation appear particularly growth enhancing for countries closer to the technology frontier (Aghion and Howitt, 2009). Finally, early economic liberalization policies during the transition process seem more important for imitating countries.11

Good macroeconomic policies matter

The economic literature suggests that macroeconomic volatility is not good for growth. Empirical studies have shown that countries with higher macroeconomic volatility have lower average growth (Ramey and Ramey, 1995). This may be because higher volatility discourages long-term investments that bring substantial returns only over the long term (such as investments in research and development [R&D]), especially among credit-constrained firms.12

Macroeconomic policies that prevent boom-bust cycles may therefore help raise long-term growth. Some of the countries in emerging Europe saw rapid growth in the run-up to the global crisis, as large capital inflows fueled a credit-driven domestic demand boom. That boom was followed by very deep recessions and generally resulted in slower convergence. For instance, if Estonia and Latvia had avoided the boom-bust cycle and maintained the average growth rates they achieved during the 1993–2005 period, their real GDP per capita in 2010 would have been 40 percent higher.

More generally, excessive growth in domestic demand may be detrimental to long-term growth (Figure 3.8). This is because it encourages a transfer of resources from the tradable to the less productive nontradable sector. In slow-growing countries, excessive domestic demand growth led to a surge in unit labor costs, notably in the manufacturing sector, and to less investment in the tradable sector. The loss in competitiveness increased further the current-account deficit, which was already boosted by higher imports that, in turn, were reducing GDP growth.13

Figure 3.8Selected European Economies: Domestic Demand Booms and Their Impact on Long-Term Growth, 2000–10

Sources: Eurostat; and IMF, World Economic Outlook database.

Fiscal policy also matters. For instance, countries with high public debt have seen lower growth (Figure 3.9), although the causality may run both ways. In addition, it is likely that higher corporate tax rates discourage investment by making it less profitable, causing corporations to shift investment to other countries with lower tax rates.14 It is noteworthy that countries with lower corporate tax rates had higher investment-to-GDP ratios and attracted larger capital inflows (Figure 3.10).

Figure 3.9Europe: Public Debt and Adjusted Growth


Source: IMF, World Economic Outlook database.

Figure 3.10Europe: Corporate Tax Rates and Growth, 2000–10


Sources: IMF, World Economic Outlook database; and Organization for Economic Cooperation and Development.

1Nominal corporate income tax rate.

Labor market flexibility matters…

Higher labor market flexibility boosts growth by increasing labor participation and employment and better matching wage and productivity growth. In countries such as Italy, Portugal, and Spain, more rigid employment regulations (regarding dismissal of employees, collective dismissals, and temporary contracts) and hiring and firing practices may have contributed to lower participation and employment rates, particularly among women (Figure 3.11). In addition, they have likely hindered improvements in TFP or the speed of adoption of new technologies, or both, by discouraging workforce adjustments in otherwise high-turnover industries.15 In more efficient labor markets, that is, those with more flexible wage determination and better relationships between employers and employees, wage increases are also more likely to remain in line with productivity growth, preventing undue losses in competitiveness and slower growth.

Figure 3.11Europe: Labor Market Flexibility, Employment, and Labor Participation, 2010

Sources: Organization for Economic Cooperation and Development; and World Economic Forum.

1Higher value means higher employment protection; 2008 is the latest year for which data are available.

…as does a better-educated workforce

Investment in human capital promotes growth in both innovating and imitating countries (Figure 3.12). The economic growth literature shows that both growth and stock of human capital matter for GDP growth. Higher growth in human capital contributes to higher output growth, and higher stock of human capital increases the ability of a country to innovate or catch up with more advanced countries by imitation.16 The type of education that matters for growth depends on the country’s state of technological development. Investment in tertiary education is more growth-enhancing for countries closer to the technology frontier, because it increases their ability to innovate, whereas primary and secondary education are likely to yield relatively more benefits among countries that are technology imitators.17 Evidence for Europe shows that underperforming countries—both those closer to the technology frontier and those further from it—produce relatively fewer tertiary graduates and have a higher share of population with only a primary or lower secondary school degree. Italy and Portugal are performing particularly badly on this front. In addition, there is evidence to suggest that the first stage of tertiary education programs (outside PhD or doctorate programs, level 5 of the International Standard Classification of Education [ISCED]) is particularly important for countries that are technology imitators.

Figure 3.12Europe: Education Levels and Growth, 2000–10


Source: Eurostat.

Box 3.1Stylized Facts from the Economic Growth Literature

The economic growth literature has developed four leading growth model paradigms. In all models, the level of economic output depends on the stock of capital and labor and the state of technological progress. More accumulation of capital and labor will not in itself lead to a permanent increase of the growth rate: technological progress is needed to offset diminishing returns to capital and labor. The models, which differ concerning what determines technological progress, are as follows (Aghion and Howitt, 2009):

  • The neoclassical paradigm, in which technological progress is exogenous. Higher investment increases the level of output but does not affect the growth rate, which is determined by the exogenously determined rate of technological progress. Consequently, the paradigm does not provide long-term growth policy recommendations.
  • The AK paradigm, which endogenizes technological progress by considering it part of capital accumulation.1 This yields constant returns to scale in capital accumulation. Hence, growth can now be boosted by higher investment—in either physical or human capital.
  • The product-variety paradigm, which endogenizes innovation by linking it to product variety. More product variety raises an economy’s production potential, which offsets the negative impact of diminishing returns. Sustained growth is possible only if new varieties, resulting from R&D investments, are created.
  • The Schumpeterian paradigm, which endogenizes innovation by linking it to firm turnover and “creative destruction.” In this paradigm, a higher rate of firm turnover generates faster growth, as “creative destruction” generates the entry of new innovators and the obsolescence of old products. In this model, growth performance will vary with proximity to the technology frontier, and imitators will converge to the frontier at a higher speed until they need to switch to more innovation. Failure to operate the switch can prevent a country from catching up.

These paradigms have been used to explain different factors that account for the growth process.

Financial sector development promotes growth. A large body of evidence suggests that countries with more developed financial systems tend to grow faster—although causality can go both ways. Better functioning financial systems (i) ease real sector external financing constraints, especially in innovative sectors with fewer collateralizable assets and in countries that lie further away from the technology frontier; (ii) provide ex-ante information on viable projects; (iii) provide ex-post monitoring of investment performance and strengthen corporate governance; (iv) facilitate the trading, diversification, and management of risk (including macroeconomic volatility); (v) mobilize and pool savings; and (vi) ease the exchange of goods and services. Policies aimed at developing financial markets would then be indirectly promoting growth. These policies fall under six categories of purpose: (i) to strengthen political and macroeconomic stability; (ii) to strengthen the operation of the legal and information infrastructure; (iii) to strengthen financial system regulatory and supervisory framework; (iv) to promote market contestability and efficiency; (v) to reduce government ownership of financial institutions and promote public investment in infrastructure that facilitates access to finance; and (vi) to promote financial liberalization and sound institutional development.

Competition has a non-linear impact on growth. Both too much and too little competition can inhibit innovation. In addition, market contestability has a more positive impact on growth in countries closer to the technology frontier but a less positive impact on sectors or countries that lie further away from the frontier.

These findings have important policy implications, including the following: (i) the promotion of national “champions” inhibits growth in countries closer to the technology frontier; (ii) countries closer to the technology frontier should promote entry (also with public funding) of innovative firms; and (iii) domestic competition policy should be complemented by policies aimed at facilitating the reallocation of capital and labor from laggard to innovative sectors.

Investment in human capital promotes growth in innovator and imitator countries alike. On the one hand, countries closer to the technology frontier should invest in secondary and tertiary education, since this facilitates the shift from imitation to innovation and avoids low-growth traps. In particular, funding and autonomous universities are strategically complementary; that is, although funding education is not growth enhancing, it enhances growth when universities are autonomous, because this autonomy better aligns research with market needs. On the other hand, countries further away from the technology frontier should invest in primary and secondary education, because this facilitates the adoption of technologies developed by innovating countries.

Macroeconomic volatility has a non-linear impact on growth. Volatility can promote aggregate savings and, therefore, growth when individuals have a strong preference for future, rather than current, consumption. More commonly, volatility hampers growth especially in less financially developed countries, which are less able to diversify macroeconomic shocks, causing lower investment in long-term R&D. In addition, the procyclical nature of R&D expenditures amplifies further the impact of macroeconomic volatility. Consequently, countries further below the technology frontier should prioritize financial sector development, especially in terms of low-end specialized intermediaries such as microfinance institutions. Countries closer to the technology frontier should prioritize access to external finance through capital market institutions such as private equity and venture capital. All countries would benefit from good macroeconomic policies to avoid short-term booms and busts that lower long-term growth rates.

Based on these theories, the empirical literature on growth has identified a number of macroeconomic, microeconomic, and institutional variables that are linked to long-term growth. Key growth determinants (of variable statistical relevance)2 include (i) macroeconomic variables such as indicators of financial development, exchange rate evolution/variability/distortions, current account balance, money growth, government consumption and/or fiscal balance and/or government taxation, investment rate, human capital, trade openness, and volatility of shocks; (ii) institutional variables such as the rule of law, institutional quality and regulatory environment, expenditures and output of R&D, inequality, and political institutions; and (iii) demographic variables such as population growth or dependency ratio. The statistical relevance of these determinants varies widely and depends on many factors including, but not necessarily limited to, differences in country samples, other variables included in the regressions, and the econometric technique used.

Note: The main author of this box is Gregorio Impavido.1 The name “AK model” originates from the mathematical representation of the production function in the model Y = AK, where Y represents the total production in an economy, A represents total factor productivity, and K is capital.2 See, for instance, Chapter 8 of Aghion and Durlauf (2005).

For countries far from the technology frontier, economic liberalization is key…

For countries that are still far from reaching the technology frontier, economic liberalization reforms conducted during the transition process appear to be strongly growth enhancing. Countries that liberalized and reformed their economies earlier—including through privatizations; enterprise restructuring; liberalization of price, trade, foreign exchange, banking, and interest rates; and infrastructure reforms—have generally grown faster. Such reforms have helped them catch up more rapidly by fostering capital accumulation and the adoption or imitation of existing technologies (Figure 3.13).

Figure 3.13Emerging Europe: Economic Transition and Growth, 2000–10

Source: European Bank for Reconstruction and Development.

1Higher value means better score.

…as countries become richer, improving institutions becomes increasingly important for sustaining growth

When countries move closer to the technology frontier, product market efficiency becomes increasingly important. Ensuring a high degree of product market competition appears to enhance growth in technologically advanced countries. As can be seen in Figure 3.14, good performers in advanced countries (such as Germany, the Netherlands, and Sweden) generally score better in terms of aggregate indicators of efficiency in the goods market and barriers to competition than slow-growing countries in advanced Europe (notably, Italy and Portugal).18 This is in line with theoretical and empirical findings that show how competition encourages growth through two channels: by facilitating the entry of firms with quality-improving innovations and by encouraging incumbent firms in industries close to the technology frontier to innovate as the only available avenue to retain market share (Aghion and Howitt, 2009). Productivity and output will therefore be higher the more intense the competition in countries and sectors close to the technology frontier.

Figure 3.14Advanced Europe: Product Markets Efficiency and Growth, 2000–10

Sources: IMF, World Economic Outlook database; Organization for Economic Cooperation and Development; and World Economic Forum.

1Higher value means better score.

2Higher value means lower score. 2008 is the latest year for which data are available.

Quality of institutions is particularly important for growth in richer countries (Figure 3.15). Good protection of property rights and a high level of legal security are associated with higher growth, in line with theoretical predictions and empirical findings (Bouis and others, 2011; Barro and Sala-i-Martin, 2004). The effect seems to run partly through R&D expenditures, which benefit from good legal systems and tend to boost growth by fostering innovation. There is also a strong positive association between the quality of institutions more generally and growth performance, but only for higher-income countries.

Figure 3.15Advanced Europe: Institutional Quality, Legal Structure, and Growth, 2000–10

Sources: IMF, World Economic Outlook database; Frazer; Organization for Economic Cooperation and Development; and World Economic Forum.

1Higher value means better score.

Good performers also have a generally high capacity for innovation (Figure 3.16). As stressed above, this is likely to be the result of their more efficient and competitive labor and product markets, their reliable institutions that foster investment and innovation, and a more educated labor force. Innovation capacity matters a great deal in advanced Europe, which is closer to the technology frontier and therefore needs to grow via further innovation. Poor performers in advanced Europe manage badly on this front when measured by the number of patents granted, by an index of technological readiness (which measures capacity to develop and absorb new technologies), or by innovation capacity (which includes, in addition to R&D spending, the availability of scientists and engineers, university-industry collaboration, and government procurement of advanced technology).

Figure 3.16Advanced Europe: Innovation, Technological Readiness, and Growth, 2000–10

Sources: Eurostat; and World Economic Forum.

1Higher value means better score.

Trade integration: A critical transmission channel from institutions to growth

The relative degree of trade integration seems to strongly differentiate countries with high growth from those with slow growth. It is striking how many of the strong performers have enjoyed high and increasing levels of trade, both in exports and in imports, and how many of the poor performers have had much lower and stagnating levels, with growth driven more by the nontradable sectors. In Austria, Germany, the Netherlands, and Sweden, the share of export and import in GDP rose by about 15 percent to more than 20 percent between 1995 and 2010. The same was true for the Czech and Slovak Republics, and, to a lesser extent, Poland. At the other end of the spectrum, the export-to-GDP and import-to-GDP ratios of Greece, Italy, Portugal, and Spain stagnated over those years (Figure 3.17).

Figure 3.17Selected EU Countries: Trade Openness, 1995–2010

(Percent of GDP)

Source: IMF, World Economic Outlook database.

Differences in the degree of trade integration likely have little to do with trade policies as such. Generally, trade liberalization is a key driver of growth and in Europe, it has certainly played an important role in the increase of trade shares over time as the EU expanded, deepened, and developed closer ties with non-members in Europe and beyond. However, trade policy is identical for all EU countries and therefore cannot explain growth differentials among them. Rather, trade integration reflects both better institutions and market competitiveness and amplifies their impact on growth. As previously mentioned, a more competitive and flexible labor market, as well as better institutions, encourages firms to imitate and innovate, resulting in higher TFP growth and external competitiveness. Countries with these advantages are likely to display higher export growth (Figure 3.18). At the same time, trade integration enhances growth since competition diverts resources from the nontradable to the more productive tradable sectors. Conversely, poor institutions that pose barriers to competition distort resources toward the protected nontradable sectors, such as real estate and construction, which are weaker sources of productivity growth. As can be seen in Figure 3.19, more manufacturing and less real estate and construction activity are associated with higher labor productivity and TFP growth.

Figure 3.18Europe: Trade Openness and Growth, 2000–10

Source: IMF, World Economic Outlook database.

Figure 3.19Europe: The Size of Tradable and Nontradable Sectors Relative to Productivity and Growth, 2000–10


Sources: Conference Board Total Economy Database, January 2011; and Eurostat.

1Data for Emerging Europe are for 2000–08.

Twenty to thirty years ago, Italy, Portugal, and Spain managed to grow relatively quickly, not only because they were benefiting from the catching-up effect but probably also because their institutions at the time were more appropriate for their stage of economic development, which was then based on technology imitation. However, as these countries moved closer to the world technology frontier, they needed to switch toward institutions more suited to innovation-based growth. That did not occur, and their ability to innovate and move up into new industries and technologies suffered. Consequently, their growth was hurt by low-cost competition from emerging Europe and China in their traditional labor-intensive manufacturing sectors.19

Low Growth Traps and How to Get Out of Them

Heavily regulated goods and labor markets, poor institutions, and macroeconomic policies can interact to pull countries into low growth traps (Figure 3.20). Countries with inadequate institutions and less competitive markets are likely to see lower rates of innovation and stronger growth in the nontradable than in the tradable sector, leading to slower TFP growth. This, in turn, discourages investment in human capital, thereby reducing innovative capacities. The uncompetitive firms in these countries are apt to lobby the government to maintain barriers to competition to thwart new entrants. A vicious circle results with economies bound to grow less over the long term. Overly stimulative macroeconomic policies can further encourage investment in protected sectors by inflating domestic asset prices; this can lead to boom-bust cycles that yield private and public debt overhang, depressing growth further.

Figure 3.20Advanced Europe: Market and Institutional Efficiency Relative to Export Growth, 2000–10

Sources: Frazer; IMF, World Economic Outlook database; and World Economic Forum.

1Higher value means better score.

Conversely, more competitive goods and labor markets, a better educated labor force, good institutions, and prudent macroeconomic policies can set a country on a higher growth path. With more efficient institutions and more competitive markets, workers and companies are better positioned to innovate and more flexibly adapt to international competition. In these economies, trade promotes productivity growth as it provides economies of scale in production and more scope for learning-by-doing externalities and knowledge spillovers. In addition, firms not only are forced to innovate by global competition, they also are more inclined to invest in R&D because of bigger ex-post rents that accrue to successful innovators in a larger external market. A vibrant, skill-intensive sector that offers employment opportunity will encourage investment in human capital, which will contribute further to the innovative capacity of the economy. Stronger growth in turn allows government to reduce expenditures (for example, on unemployment benefits) and opens the opportunity to lower taxes and stimulate investment.

Escaping low growth traps: Experience and lessons

The problem of slow growing countries in Europe is not new. In the late 1970s and early 1980s, many countries in Europe suffered from “euro sclerosis”—with high unemployment and low growth. These earlier periods provide important lessons, as they show that a change in policy can turn an economy around.

The experiences of the Netherlands and Sweden in particular show that it is possible to turn poor economic performance around. In the 1980s and 1990s, these two countries undertook sweeping reforms to boost GDP growth after prolonged periods of poor economic performance (Boxes 3.2 and 3.3). Their experiences provide useful insights into how reforms could help other countries significantly reverse their economic fortunes.

Both countries undertook reforms only after a protracted economic malaise that culminated in a crisis. Income per capita was falling relative to that of Germany for about a decade (Figure 3.21). In addition, public finances were deteriorating: spending increased (Figures 3.22), and fiscal deficits were growing. Wage growth was also high in both countries, contributing to a decline in employment in the Netherlands (Figures 3.23 and 3.24). In Sweden, growth was also held back by reliance on relatively low value-added industries, the banking crisis in the early 1990s, and a stifling tax system. Finally, the erosion of competitiveness also contributed to deterioration in current account balances.

Figure 3.21Netherlands and Sweden: GDP per Capita Relative to Germany, 1970–2010

(PPP terms, percent)

Sources: Conference Board Total Economy Database, January 2011; and IMF staff calculations.

Note: Germany’s GDP per capita before 1989 was constructed using the growth rate of West Germany’s GDP per capita.

Figure 3.22Netherlands and Sweden: Government Primary Spending, 1970–2010

(Percent of GDP)

Sources: IMF staff calculations; and Organization for Economic Cooperation and Development.

Figure 3.23Netherlands and Sweden: Real Compensation Rate of the Private Sector, 1970–2010

(Index, 1980 = 100)

Sources: IMF staff calculations; and Organization for Economic Cooperation and Development.

Figure 3.24Netherlands and Sweden: Employment Rate, 1970–2010

(For age group 15–74 in the Netherlands and 15–64 in Sweden; percent)

Sources: Central Planning Bureau (Netherlands); and Organization for Economic Cooperation and Development.

While each country’s reform package differed in details, both included the same mix: measures to correct macroeconomic imbalances and measures to achieve comprehensive structural reforms. Both sets of measures were needed: on the one hand, macroeconomic stabilization policies without structural reforms would only deliver low, albeit stable, growth—as Italy has experienced in recent decades (Box 3.4); on the other hand, structural reforms without good macroeconomic policies could lead to large swings in economic growth and make sustained high growth all but impossible.20 This was made painfully clear by the crises experienced in these two countries before their reforms.

Box 3.2Labor Market Reform: The Experience of the Netherlands in the 1980s–1990s1

Triggered by poor economic performance in the 1970s and early 1980s, the Netherlands undertook a series of labor market reforms that resulted in strikingly rapid employment growth. Excessive wage growth in the 1970s and early 1980s had led to a decline in private sector employment, as investment and job growth slowed. When unemployment shot up sharply from 1979 onward, in a recession that was much deeper than elsewhere, a consensus for reform gradually emerged.

Labor market reforms started in earnest in late 1982 with an agreement between unions and employers to pursue wage moderation in exchange for employment creation (the “Wassenaar agreement”). The agreement abolished automatic price indexation—not only in new wage agreements but also in existing wage agreements.

Subsequent governments implemented a series of labor market and fiscal reforms that complemented and reinforced each other.

  • The level of the real minimum wage was reduced sharply. It was first cut by 3 percent and subsequently frozen in nominal terms for many years. As a result, by 1997, the real minimum wage had declined by 22 percent from its 1979 peak. The youth minimum wage was reduced even more sharply.
  • Civil servants’ salaries were subject to the same cuts and freezes as the minimum wage and declined in real terms by about the same percentage.
  • The social security benefits replacement rate was cut significantly. Wage-related unemployment, sickness, and disability benefits were cut from 80 percent of wages to 70 percent; and the duration of unemployment and disability benefits was shortened. The minimum benefit, which is linked to the minimum wage, fell substantially in real terms.
  • To support wage moderation, taxes and social security contributions paid by employees were cut substantially. As a result, disposable incomes rose substantially even in the absence of real wage increases.
  • To finance the tax cuts, the government cut primary public expenditures by 14 percentage points of GDP. As a result, the government managed to reduce taxes and the budget deficit at the same time. The budget balance changed from a deficit of 6.2 percent of GDP in 1982 to a surplus of 2.2 percent in 2000.

The reforms contributed to a rapid increase in employment. Employment grew from 1984 onward, initially at a moderate rate, and accelerated further with the strong economic performance in the 1990s, helped as well by substantial financial sector and product market reforms. Employment growth largely benefited new entrants to the labor market, including recent graduates and women. The youth unemployment rate dropped sharply, from a peak of 25 percent in 1985 to 6¼ percent in 2008—the lowest rate in the European Union. The labor force participation of women rose sharply. Although most women worked part-time, this phenomenon seems to reflect cultural preferences rather than government policies.

Further reforms were undertaken in the last decade that reduced the generosity of the unemployment insurance and disability insurance programs. Also, tax rules were changed to stimulate working for second earners, and the tax advantages for early retirement were abandoned. During this period of reforms, the number of benefit recipients was reduced substantially and labor force participation rates increased.

Note: The main author of this box is Yan Sun.1 This box is partly based on Chapter III of IMF (2004), “The Netherlands: How the Interaction of Labor Market Reforms and Tax Cuts Led to Strong Employment Growth,” with additional material from Gautier and van der Klaauw (2009).

Box 3.3Sweden: Structural Reforms in the 1990s

The banking and financial crisis of the 1990s in Sweden triggered far-reaching macroeconomic and structural reforms that set the stage for Sweden’s higher output growth in the late 1990s and 2000s. The reforms involved restoring a credible macroeconomic policy framework, and included a battery of structural reforms in the product and labor markets.

The first step in the aftermath of the crisis was to restore a credible, rule-based macroeconomic policy framework. This included: (i) the establishment of an inflation target in 1993 resulting in a drop in inflation (from an average of 7.5 percent in 1980–1990 to about 1.5 percent in 1993–2000); (ii) an impressive and successful fiscal consolidation, with the general government debt-to-GDP ratio falling from 72.5 percent in 1994 to 53 percent in 2000, the government expenditures-to-GDP ratio falling by about 16 percentage points between 1993 and 2000, and the budget balance turning into a surplus in 1998 from a double-digit deficit in 1993; (iii) the introduction of a detailed fiscal framework, including a nominal expenditure ceiling for the central government, a structural budgetary surplus target for the general government, and a balanced budget requirement for local governments, which helped the government to run a budget surplus every year between 1998 and 2008, except for 2002 and 2003; and (iv) a comprehensive pension reform put into effect in 1999. The new and more stable macroeconomic framework greatly improved policy credibility, thereby contributing to more moderate wage agreements. Moreover, the stronger public finances enabled some reduction in the high tax burden.

Successive reforms were implemented to improve labor market outcomes.

  • In 1991, a comprehensive tax reform was implemented to mitigate the negative effects of the growing welfare state on labor supply. The reforms aimed at shifting the tax burden from labor income to consumption and capital income. The measures included lowering the marginal tax rates on earned income, widening the tax base, eliminating tax shelters, and introducing a more uniform taxation of capital. It is estimated that the tax reform led to an increase in labor supply of about 2 percent.1
  • In 1997, a new agreement was reached by industrial labor unions to restrain wage increases. The consensus followed an unprecedented increase in the unemployment rate after the crisis (from 1.7 percent in 1990 to 9.4 percent in 1994). The agreement established explicit rules concerning the regulation of negotiations and the resolution of disputes; reintroduced more coordination in wage bargaining; and re-established the pacesetting role of the sectors exposed to international competition.
  • Other complementary reforms targeted training, work incentive, employment protection, and education. These included a reorientation of active labor market policies toward training programs and/or practical insertion courses; relaxing employment security provisions; and to some extent, reducing the replacement rate in social insurance, and raising the qualification period for unemployment benefits.2 Extensive reforms in the education system (primary to tertiary) were also conducted in the 1990s.

Building on the successes of early deregulations, additional product market reforms further promoted competition and restructuring. Early deregulation and the promotion of competition in the late 1980s fostered rapid restructuring and large productivity gains in the export sector. In particular, deregulation in the telecommunication sector (Sweden being the first country in Europe to deregulate its telecommunications market) helped to spur competition and establish mobile phone access throughout the country.3 Building on these early reforms, a new Competition Act and a new enforcement agency were created in 1993. After the EU accession in 1995, Sweden rapidly implemented all major directives of the internal market program and by 1998 was far ahead of many other EU countries.4 These rapid product market reforms led to efficiency gains and helped the manufacturing industry transform from traditional industries in the 1980s to more knowledge-intensive and less labor-intensive production in the 1990s, leading to higher productivity gains.5

The restored macroeconomic stability and structural reforms, coupled with a strong IT sector, paved the way for growth, which averaged close to 3.5 percent between 1994 and 2007. With a more flexible labor market, a more competitive product market, and a strong IT sector, which Sweden was well positioned to capitalize on, it was able to emerge rapidly from the crisis when the international outlook improved. Wage growth restraint, improved macroeconomic stability, and higher productivity growth led to rapid growth in the export sector (notably the IT sector), which in turn became the main engine of growth. The real growth rate of export and labor productivity both doubled from the 1980s to the 1994–2008 period, rising from about 4 percent to 8 percent and from 3 percent to 6 percent, respectively (see figure).

Sweden-Labor productivity and volume of exports

(Index, 1980 = 100)

Source: IMF, World Economic Outlook database.

Note: The main author of this box is Géraldine Mahieu.1Agell and others (1998).2 The generosity of the social security system, however, remained elevated compared with other European countries, whereas the level of employment protection was relatively low, in line with the concept of “flexicurity” also introduced in Denmark.3 Other factors, such as early investment by Ericson and the public telecommunications monopoly in establishing a mobile network, a high level of expenditures in R&D, a highly skilled labor force, and several public incentives to the adoption of ICT, also contributed to the emergence of a strong high-tech sector in Sweden.4OECD (1998).5 In the 1980s, traditional industries such as steel, iron, and paper represented close to 20 percent of exports, whereas chemicals and telecom accounted for less than 9 percent. By the 1990s, the share of traditional industries had fallen to 13 percent whereas chemicals and telecom represented more than 20 percent.

From the start, structural reforms in these countries focused on clearing up the worst bottlenecks to growth. These bottlenecks manifested differently in different countries, so the initial priorities of reform also differed. In the Netherlands, a strong initial objective was to contain excessive wage growth and boost employment, and thus the centerpiece of the reform package was a wage agreement between employers and unions.21 Sweden addressed its core problems through a combination of fiscal consolidation, tax reform, financial sector clean-up, and an overhaul of the wage bargaining system. In addition, further liberalization of network industries and reduced barriers to competition provided room for new industries to flourish. This followed earlier reforms in the telecom sector, which had already contributed to the emergence of a strong telecom industry. Structural reforms were generally implemented in both the product and labor markets. Coordinated reforms helped countries reap complementary gains—although the exact sequencing reflected different priorities and political constraints.22

Although reforms spanned more than a decade and took time to yield their full benefits, they were successful, as per capita income in both the Netherlands and Sweden is now significantly higher than in Germany (Figure 3.21), driven by strong growth in employment and higher productivity (Figures 3.24 and 3.25). In addition, there is a vibrant tradable sector, with exports increasing by 17 percentage points and 19½ percentage points of GDP, respectively, in the Netherlands and Sweden (Figure 3.26).

Figure 3.25Netherlands and Sweden: Labor Productivity per Worker, 1970–2010

(1990, PPP US dollars)

Sources: Conference Board Total Economy Database, January 2011; and IMF staff calculations.

Figure 3.26Netherlands and Sweden: Exports of Goods and Services, 1970–2010

(Percent of GDP)

Sources: Organization for Economic Cooperation and Development; and IMF staff calculations.

The United Kingdom provides another example of a country that undertook major reforms, which began in the late 1970s (Box 3.5). Those reforms helped reverse its relative decline in economic performance and set in motion a sustained improvement in income relative to Germany until 2005. The reforms contained many elements that were later adopted in the Netherlands and Sweden. In fact, reforms in the United Kingdom preceded most of the structural reform efforts in Europe since the 1980s. The United Kingdom’s macroeconomic policy, however, has been less countercyclical, and that has contributed to severe boom-bust cycles. Denmark also undertook very ambitious labor market reforms and experienced a relatively benign impact of the 2008/09 crisis on unemployment, in part owing to its fairly liberal product market regulation.23,24

Box 3.4Why Has Italy Grown So Poorly in the Last 20 Years?

In the last two decades, Italy’s growth has been disappointing (IMF, 2011d). By the late 1980s, Italy’s GDP was no longer catching up with the U.S. GDP. More recently, Italy suffered one of the largest output contractions in the euro area during the global financial crisis and is now experiencing one of the slowest recoveries. Per capita GDP, as well as productivity, was lower in 2010 than in 2000 (Figure 3.1).

Although Italy undertook various structural reforms during the period, they were not comprehensive enough to lift growth. During the period, manufacturing, banking, and public utilities were all either privatized or liberalized. These reforms contributed to higher productivity and profitability in the banking and manufacturing sector, but they were less successful in the public utilities sector, where reforms were either incomplete or failed to address additional bottlenecks.1 Labor market reforms eliminated full wage indexation, reduced central bargaining, promoted the use of temporary employment, and increased flexibility for new hires. However, failure to eliminate protections granted to workers with permanent contracts created a dual labor market. Successive pension reforms reduced entitlement and fiscal outlays for the future. Recent regulatory reforms have attempted to re-organize public administration to increase its efficiency and transparency, but they are largely experimental and constrained by fiscal cuts.

Largely the result of underlying structural constraints, several interconnected and often endogenous factors explain why Italy’s growth performance lags by international comparison:2

  • Inefficient public expenditure and a complex tax system hinder fiscal consolidation and growth. Italy scores poorly in terms of the quality and efficiency of public expenditure, and it stands out among countries with the highest tax burden and lowest tax compliance (EC, 2009). Overall, progress on improving public expenditure has been limited, although some steps have been taken to improve the budget classification, institutionalize spending reviews, and reorganize public administration. Further fiscal consolidation is constrained by the size of transfers to subnational governments, large entitlement programs, and the sizable interest expenditure. Finally, the tax system is unduly complex and prone to abuse. This limits labor utilization (see below), promotes evasion, and limits the ability to reduce the debt relative to GDP and attract FDI into the country.
  • Labor productivity is low and falling. Limited labor market reforms have not prevented real wage growth from outpacing the modest productivity gains, causing unit labor costs to increase. Wage bargaining remains excessively centralized so that in effect, real wages in all sectors cannot fall and the ability to align them with productivity at the plant level is severely constrained.3 Increasingly, firms are unable to compete with low-cost producers in the global market as Italy’s pattern of export specialization in low-skill labor-intensive goods has contributed to weak productivity growth.4 In addition, the ability to reabsorb the capital and labor released remain constrained by a poor regulatory and business environment. Finally, while low-skill immigrant workers have partly offset the negative impact of low labor participation (see below) and a rapidly aging society, it may also have contributed to a decrease in average productivity.
  • Labor participation is low. For instance, the labor participation of women is constrained by the lack of a family policy and formal child care structures. In addition, old age participation is reduced by a pension system that until recently favored early retirement. In general, demand for labor is constrained owing to (i) skill mismatches between what the education system produces and the labor market demands (see below); (ii) a high tax wedge, especially for low-skilled workers; (iii) complex rules stemming from the aforementioned labor market reforms protecting insiders with permanent contracts at the expense of part-time and younger workers (and in turn encouraging a brain drain among young talent); (iv) lack of competition in the product market; and (v) stagnant growth.
  • Innovation in new dynamic firms is low. A series of factors hinder innovation by Italian SMEs (see figure). Until recently, the barriers to business creation were significant, and the experience with “one-stop shops” (introduced in 2005) to reduce these barriers has been varied (OECD, 2009). In addition, until 2006, bankruptcy legislation also hindered exits, as entrepreneurs were exposed to risky criminal proceedings, putting personal wealth at risk. At present, the equity market is underused and venture capital market is slow to develop. This is partly due to a still embryonic market of institutional investors, lack of regulations encouraging investment in SMEs (Bongini and Impavido, forthcoming), and weak corporate governance practices, despite adopted regulations that follow OECD best-corporate governance practices (OECD, 2009). Indeed, the widespread use of pyramidal and cross-ownership structures gives insiders control that significantly exceeds their share in ownership and limits the effective rights of minority shareholders. In general, the high levels of public ownership, especially at the local level; regulatory barriers to competition; high administrative burden on firms; and constraining regulations for professional services have all been factors contributing to Italy’s inability to attract FDI and limiting access to finance innovation. These, in turn, have hindered the growth of firms beyond the family-control threshold.
  • Civil courts remain inefficient (OECD, 2009). Regulations aimed at protecting the privileges of judges and lawyers (in terms of pay and status) imply that the average duration of cases is the highest in Europe, although with important regional variations. Claims are usually settled out of court through private negotiations. The inefficiency of the civil justice system amplifies the problems caused by the aforementioned underdeveloped capital markets and weak corporate governance, and it represents another bottleneck keeping firms from growing beyond the threshold below which family control is still an effective organizational form for enforcing contracts.
  • Attainments in education are low (OECD, 2009, 2011). Although there are important regional variations, Italy ranked among the five worst OECD performers in the 2006 PISA tests, possibly because of the absence of a national standard test for secondary schools. The share of tertiary graduates in the labor force is just 14 percent, only half the OECD average. Drop-out rates are high with only 45 percent of students entering tertiary education actually completing their studies, well below the 65 percent OECD average. Duration of studies at the university level is significantly higher than the OECD average as well (66 percent of students remain at the university beyond the normal duration of courses). This low performance is partly explained by the fact that universities are generally not allowed to select students; they are not autonomous in defining curricula and hiring faculty; they are poorly governed with inadequate funding; and they lack accountability. In addition, they have poor human resources owing to the lack of performance-based careers and remuneration, and a system of nationwide public competition for academic positions that favors insiders (OECD, 2011; Perotti, 2008). As a result of these shortcomings, Italian universities often generate skill mismatches relative to what is demanded by the market; have a low contribution to human capital formation in general, thereby reducing labor utilization (see above); and worse, contribute little to R&D expenditure, hindering innovation. The new University Reform Act (Law 240/2010) aims to address many of the aforementioned shortcomings, but the reforms are still in their infancy.
  • Finally, regulations at the regional and local levels reduce further the flexibility of domestic markets. In some sectors, such as commercial distribution, pharmaceuticals, and the transport sectors, regional and even municipal regulations add further complexity. For instance, (i) the proliferation of regional statutes creates inefficiencies; (ii) the process through which measures are enacted lacks transparency and statutes are not subject to evaluation to determine their costs to firms, consumers, and the public administrations involved; (iii) commercial distribution is overregulated; and (iv) restrictions on market entry, both general and sectoral, are still rife.5

Italy’s innovation performance lags by international comparison

Source: Organization for Economic Cooperation and Development (2011).

1BERD: Business Enterprise Expenditure on R&D; GERD; Gross Domestic Expenditure on R&D; and HRST: Human Resources in Science and Technogy.

Note: The main author of this box is Gregorio Impavido.1 For instance, corporate governance practices imply that despite privatization, the government maintained de facto control of privatized companies though minority shares.2 See IMF (2011f) for a detailed analysis on the product and labor market’s inefficiencies hampering growth.3 The 2011 reform introduces flexibility in wage negotiations with unions at the plant level. In addition, it allows firms to go outside the national framework. In the future, this should help better align wages and productivity.4 For such a specialization pattern, the power to pass prices to international markets is low, and international market shares are lost in response to a wage growth that has outpaced productivity gains, hence, accelerating the loss of competitiveness.5 However, recent reforms introducing tender requirements for a number of local public services should increase transparency and accountability and thereby encourage entry of qualified service providers.

These reform experiences suggest the following lessons for other poor performing countries that may be embarking on major reforms, in both advanced and emerging Europe.

  • It is critical to correct macroeconomic imbalances. This is particularly crucial for those countries that are facing increasing market pressures to secure fiscal sustainability (such as Hungary and Italy). A credible consolidation plan would reduce vulnerabilities. For Hungary, where government spending accounts for about half of GDP (a much larger share than in any of its regional peers) (Figure 3.27) and where government debt is high, a durable reduction in expenditures is needed (IMF, 2011c).25 For Italy, reforms may include improving the efficiency of fiscal expenditures and rationalizing the tax system to boost compliance and reduce the tax burden.
  • Initial reforms should focus on clearing the most restrictive bottlenecks. Reducing duality in the labor market by lowering preferential protections for those with permanent contracts is a priority for Italy and Spain. Another priority, particularly for Italy, is to overhaul the regulatory framework to uproot a deeply antimeritocratic system that hinders competition by protecting insiders in industry, education, and the services sector.
  • Complementary fiscal reforms boost the success of other structural reforms. Fiscal reforms could include reductions of the tax burden for labor, civil servant wage restraint, simplifications of the tax system, and widening of the tax base, all of which would improve popular buy-in and chances of success of other reforms.
  • Implementing structural reforms is a long process. It is hard to get the right reforms in place in one stroke. Sometimes reform agendas develop gradually, in part because political economy considerations may suggest tackling problems sequentially. Reforms also take time to implement; for example, the privatization process in the United Kingdom took several years to complete. And reforms may need to be refined in the course of their implementation as constraints change.26
  • It takes time before reforms reap their full benefits, and initially conditions can worsen. The Dutch reforms of the early 1980s took 10 years to come to full fruition. In Sweden, it was not until a decade after the reforms of the 1990s that a new peak performance was reached. In the United Kingdom, notwithstanding the labor market reform efforts, unemployment only began to fall sharply in the mid–1990s. Germany began serious wage moderation in 1995, but German GDP growth remained low for 10 years before export growth pulled the country out of its slump in 2005. Research on a wider set of reform cases suggests that the cumulative gains in growth achieved from structural reforms in the trade, product market, and labor market areas are positive. However, they predominantly materialize only over the long term because they involve costly and painful reallocations of resources.27
  • Reforms need to adapt as constraints evolve. As constraints are relaxed gradually and sequentially, new bottlenecks to growth may emerge in complicated and sometimes unpredictable ways. Countries therefore need to be prepared to shift and adapt the focus of their reforms.

Figure 3.27Hungary and Its Peers: Government Spending, 2009

Sources: IMF, World Economic Outlook database; and IMF staff calculations.

Box 3.5United Kingdom: Structural Reforms during the 1980s1

In the late 1970s, poor economic performance in the United Kingdom made the environment conducive to far-reaching reforms. Decades of relative decline had been exacerbated by a major recession after the first oil shock in 1973, high inflation partly owing to wage pressures and increasing government spending, and further deterioration in the economic situation. A severe balance of payments crisis in 1976 exposed further the weaknesses in the existing economic structure and made it clear that something needed to change.

The Thatcher government, which took office in 1979, and the following governments undertook sweeping structural reforms, including the following key elements:

  • Reducing the state’s role in the economy. The state’s role was reduced through large-scale privatization of state-owned enterprises and public housing, cuts in civil service employment, and pension reforms that reduced the relative value of state pension benefits but created incentives to enroll in private pension schemes.
  • Improving work incentives in benefit programs. Net unemployment benefits were reduced by abolishing the earnings-related supplement, suspending benefits’ statutory indexation, and making their taxation less favorable. Eligibility criteria for receiving unemployment and other benefits were tightened. Job-seeking efforts were monitored via the 1986 “Restart Program,” which required biannual counseling for all unemployed.
  • Reforming the tax system. The number of bands for marginal rates of personal income tax was reduced, while rates themselves were lowered. Exemptions were reduced or eliminated and the taxation of capital income was streamlined. The share of indirect taxes was increased, and corporate profit taxes were lowered while their base was broadened.
  • Reforming trade unions to make participation more voluntary. Reforms included extending the grounds for refusing to join a union, introducing limits on picketing, prohibiting actions that force contracts with union employers, and weakening the closed-shop and union immunities.
  • Liberalizing financial markets. Administrative measures curbing bank lending and lending by building societies were removed, and pricing for financial services was liberalized.
  • Promoting entrepreneurship and self-employment. The government introduced measures to foster self-employment, such as offering tax relief, facilitating bank borrowing for small companies, and establishing local agencies to counsel small businesses on planning, marketing, and design.

Unemployment was initially high, but growth and inflation began to improve in the early to mid–1980s, enabling the reforms to continue. Privatization also contributed to the rise in equity and home ownership, which may have created additional support for the reforms.

Although the impact of the reforms on economic performance remains subject to debate and some of the reforms remain controversial, there is consensus that the reforms contributed to halting the previous trend of relative decline in GDP levels per capita (for example, Card and Freeman, 2002; IMF, 2003), as the overall labor market and growth performance improved in the 1980s and 1990s. Microeconomic evidence—examining the impact of specific reform efforts on firm-level productivity—also suggests that structural reforms of the 1980s contributed to the United Kingdom’s improved relative productivity performance (Card and Freeman, 2002).

Note: The main author of this box is Yan Sun.1 This box is based on Box 3.2 in Chapter III of IMF (2004), “Fostering Structural Reforms in Industrial Countries.”

The current reform agenda in Greece and Portugal largely follows the strategies of the early reformers. The response to the global financial crisis has been a consensus on the need for comprehensive reforms in a short timeframe. Under the EU-IMF-supported programs, Greece and Portugal have adopted fiscal consolidation and structural reforms in product, services, and labor markets. In Greece, for example, the fiscal deficit was targeted to shrink by 12¾ percentage points of GDP between 2009 and 2014 through cuts in wages and other spending and overarching tax reforms. Greece has begun to undertake strong labor market reforms; pension reforms; reforms to open up closed professions; reforms to simplify procedures for start-up and licensing; reforms in the approval processes for large investments; and reform of the education system. It also has plans for ambitious privatizations and judicial reforms. Its fiscal reform includes plans to reduce the high tax wedge on labor. In Portugal, growth and job creation take center stage. Portugal’s reforms address the country’s fundamental problem—low growth—with a policy mix based on restoring competitiveness through structural reforms, a balanced fiscal consolidation path, and financial stabilization. In addition to fiscal consolidation, reforms include a fiscally neutral reduction in labor taxes.

Extending the European Growth Frontier

Implementing reforms would both bring more countries closer to the European technology frontier and bring the European frontier itself closer to the United States. In 2010, average GDP per capita (purchasing power parity—PPP) in the EU was only 64 percent of the U.S. level, and only 85 percent for the three richest EU countries (Austria, the Netherlands, and Sweden) (Figure 3.28). The higher income in the United States is largely explained by its higher labor productivity—particularly in the information and communications technology (ICT) sector and services. In 2010, the level of employment of the United States was similar to that of Europe, although the number of hours worked per employee is still higher in the United States (Table 3.1).28,29 Part of the difference in productivity levels can be attributed to lower human capital in ICT-related sectors and services (Figure 3.30). The large productivity gap in services also reflects a generally more regulated service sector in Europe (Figure 3.31).30

Figure 3.28United States and Selected EU Countries: Per Capita GDP, 2000 and 2010

(Index, United States 2010 PPP dollars = 100)

Sources: Conference Board Total Economy Database, January 2011; and IMF staff calculations.

Table 3.1United States and EU15: Comparison of GDP Per Capita and Its Decomposition (2010)
GDP per capita

(thousands of 2010

PPP US dollars)
Labor productivity

per hour worked
Annual hours worked

per worker
Employment as

ratio of labor

Labor force as ratio

of population
United States46.8611,6900.910.50
United Kingdom37.1521,6470.890.49
EU15 Average40.0521,6470.980.48
Sources: Conference Board Total Economy Database, January 2011; EU AMECO database; and IMF staff calculations.

2007 figures.

Sources: Conference Board Total Economy Database, January 2011; EU AMECO database; and IMF staff calculations.

2007 figures.

Figure 3.29United States and Selected EU Countries: Contribution to TFP Growth of Major Sectors, 1995–2007

(Annual average growth rates)

Source: KLEMS database.

Note: ICT production includes manufacturing of electrical and optical equipment, and post and telecommunication services. Goods production includes agriculture, mining, manufacturing (excluding electrical machinery), construction, and utilities.

1From 1995 to 2006.

2From 1996 to 2006.

3Data for EU13 refer to the 13 countries in the figure.

Figure 3.30EU13 and United States: Human Capital Stock Comparisons, 2005 Level1


Sources: EU KLEMS database; IMF, World Economic Outlook database; and Organization for Economic Cooperation and Development Tax database.

1Human capital is proxied by the share of the labor force having completed tertiary education.

2Includes electrical and optical equipment, and post and telecommunication services.

Figure 3.31United States and Selected EU Economies: Services Sector Contribution to TFP Growth (1995–2007) and Regulatory Conditions

Sources: EU KLEMS database; IMF staff calcuation; and Organization for Economic Cooperation and Development Product Market Regulation (PMR) database.

1Simple sum of indicator of regulatory conditions in retail trade and professional services.

Lower productivity may be linked to Europe’s poorer performance in innovation. A few of the innovation leaders in Europe, such as Denmark, Finland, Germany, Sweden, Switzerland, and the United Kingdom, exhibit innovation performance that is similar to that of the United States in some categories (Figure 3.32).31 But the United States excels in areas such as university-industry collaboration on R&D, high quality research institutions, availability of scientists and engineers, and government procurement of advanced technology products. In terms of innovation outcomes, the United States also far exceeds Europe in patents granted per capita.32

Figure 3.32United States, Japan, and Selected European Countries: Innovation Indicators, 2009–10 Weighted Average

(Higher numbers are better, United States = 100)

Sources: Executive Opinion Survey; World Economic Forum; and the United States Patent and Trademark Office.

How could Europe bolster innovation?

  • Remove barriers to cross-border mergers and acquisitions. Corporate restructuring can be a quick way to inject new innovative spirit into the domestic economy, particularly in those sectors where innovation follows a more radical pattern—where the ranking orders of innovators are unstable and entry rates of innovators are high.33 Europe still has more pervasive barriers to corporate merger and control than the United States. Indeed, based on the last available data from the European Commission, in 2006, 77 percent of intra-EU merger and acquisition deals were domestic deals.34 There is a need therefore to quickly implement the EU Directives for Cross-Border Acquisitions and Takeovers to remove the formal and informal national obstacles to the free flow of equity capital.
  • Improve access to risk capital to boost the innovation of new firms. Europe has a relatively less-developed venture capital market. The constraints in access venture capital financing may have limited the growth of new firms, and helped protect incumbent firms. In the United States, where venture capital is more readily available, new firms have been a major source of innovation.35 One—albeit controversial—option to improve access to risk capital would be through a redesign of EU industrial policy (Box 3.6). Recent studies (Aghion and others, 2011) suggest that subsidies would be more successful in promoting growth when they are targeted at sectors that are new, growing, and with intense intra-sector competition, and when they are not concentrated in single firms or “champions.”36

Box 3.6EU State Aid Policy and Industrial Policy

A key objective of EU industrial policy is to alleviate market failures. The policy is motivated by the need to correct market failures, such as (i) externalities in R&D investments; (ii) asymmetries of information in SME development; and (iii) agglomeration and network externalities as well as strategic trade considerations justifying sector-specific, cluster-specific, or “national champion”-specific support.

Current EU industrial policy is largely focused on horizontal measures—measures that target the economy as a whole—rather than on measures that support individual industries or firms. Aid is delivered mainly through horizontal measures for upstream R&D, such as support for R&D, SME development, education, environmental protection, and energy saving.

State aid to individual sectors or firms (“vertical support”) has been decreasing in the EU27 countries, falling from 1.1 percent of GDP to 0.5 percent of GDP between 1992 and 2007.1 These trends are the result of explicit policies limiting state aid in general, which is perceived as a threat to internal competition and integration, and policies limiting vertical support measures in particular, which are more prone to risk of capture by vested interests and rent-seeking, resulting in support to nonproductive “national champions.”2

A recent study (Aghion, Boulanger, and Cohen, 2011) suggests that capital market imperfections and credit constraints limiting the reallocation of firms to new and more productive sectors justify a more active government intervention, with more direct support for sectors. Vertical support measures would have a stronger impact in countries and/or sectors closer to the technology frontier, but where bank credit is still the primary channel for company financing. Vertical measures are particularly effective when subsidies are targeted to highly contestable markets, as they force incumbent firms to innovate to keep up with competition of new entrants. Sectoral aid also works better when it is not concentrated. In other words, aid that enhances competition within a sector by not focusing on single firms or “champions” is more likely to be growth enhancing.

Note: The main author of this box is Gregorio Impavido.1 It then increased to 3.6 percent of GDP in 2009 as a consequence of the support provided to the financial sector during the financial crisis.2 In practice, however, horizontal support measures have not stopped vertical, sector-specific, state aid. On average, 70 percent of aid classified by the EU as horizontal is in fact awarded to the manufacturing sector (EIB, 2006), although a large variation exists among member states. Indeed, there are numerous cases of state aid where the primary objective is horizontal but the aid is limited to a certain industry, subsector or firms. In particular, aid is often directed to key incumbent firms, or clusters of firms, that have established their reputation as innovators, consequently reducing the contestability of the market and, therefore, incentives to innovate (Aiginger and Sieber, 2005).
  • Develop capital markets more generally. The bank-centric nature of the financial system of most European countries may be another factor that is holding back innovation and growth (IMF, 2006). Both supply- and demand-side bottlenecks hamper the development of capital markets in Europe. On the supply side, constraints on capital market financing include corporate governance problems.37 In particular, problems include the weak enforcement of shareholder rights against insider trading;38 the weak protection of creditor rights;39 the concentrated shareholder ownership structure prevalent in most countries;40 and the general absence of institutional investors among large shareholders.41 On the demand side, constraints include weak judicial systems in enforcing private contracts,42 which discourage firms from growing beyond the family-controlled scale.43 Developing capital markets further is particularly important given the continued deleveraging by banks. With constrained access to credit for innovative firms, reforms aimed at diversifying the source of finance, especially for small start-up SMEs, could boost growth. Key areas of reforms include strengthening corporate governance rules and practices to encourage external private equity and venture capital finance; promoting the development of institutional investors to increase the supply of risky capital; and reforming the judicial system to encourage growth of family-owned SMEs.
  • Reduce the corporate tax burden in the context of a broader tax base. This could stimulate innovation in established firms—particularly in countries with less-developed markets for risk capital. Indeed, there is empirical evidence that for Organization for Economic Cooperation and Development (OECD) countries, corporate tax has a disproportionately negative impact on the TFP of established and large firms.44 In addition, reducing corporate taxes could also encourage innovation in the most dynamic and profitable firms.45 Because more profitable firms are better able to finance R&D—particularly in countries with less-developed markets for risk capital—lowering the effective corporate tax rate could allow such firms to invest more in R&D.46 Finally, it could also attract more FDI, which tends to bring more productivity gains to domestic firms. The reduction in the tax burden should be combined with reforms to broaden the tax base and reduce loopholes.
  • Reduce regulation of services. Relative to the United States, some European countries, including Germany, have a more heavily regulated service sector. In these countries, further liberalization could deliver quick gains in productivity growth.
  • Improve higher education and university-industry collaboration. Europe’s ability to innovate relies on the quality of its higher education (for example, human capital in ICT or the number of engineers and scientists) and its scientific research institutions. While more public funding may help, there should also be reforms in the governance of universities and other higher education institutions to increase their ability to attract funding and invest in R&D, as it has been recently attempted in Sweden (Box 3.7).47 Finally, measures to encourage more university-industry collaboration, such as those to encourage the commercialization of university-held patents and technology, could help.

Box 3.7Promoting Research and Innovation: Sweden’s Research and Innovation Bill (2008)

In recent years, the Swedish government has clearly identified innovation as the key competitive factor in a knowledge-based economy, and has sought to maintain its position as one of the innovation leaders in Europe through public policy. In the 2008 Research and Innovation Bill,1 the government introduced several new features that attempt to improve the quality of publicly funded research. They include the following:

  • Linking funding to the quality of academic and research institutions. In a departure from old practices, quality, as measured by publications, references to publications, and external research funding (including industry contracts), determines how much public funding each university or higher education institution receives. This has introduced an element of competition and has tied public funds to those institutions that have proven ability in attracting external funds—an indication of strong research relevance.
  • Promoting the commercialization of research. The government presented an initiative to increase the commercialization of research results. Dedicated funding is allocated for this purpose. Innovation offices have been set up at a number of higher education institutions.
  • Boosting funding for strategic research with active industry participation. The bill planned a permanent, annual increase in funding to research in a number of strategically important areas, such as medicine, new technology (IT), and environment (energy). Sectoral R&D programs are based on calls for proposals in selected areas and involve active industry participation.
Note: The main author of this box is Yan Sun.1 Sweden’s 2008 Research and Innovation Bill sets the framework for central government-funded research for the coming four years.

Note: The main authors of this chapter are Gregorio Impavido, Géraldine Mahieu, and Yan Sun.


The chapter focuses on the period since the introduction of the euro. The limited number of observations and the impact of the global crisis in the latter part of the decade make it difficult to apply advanced econometric methods. Nevertheless, the findings in this chapter are similar to those found in the economics literature.


See, for instance, Barro and Sala-i-Martin (2004) and Aghion and Howitt (2009).


Conditional convergence predicts that countries converge to their own steady state where growth rates are only determined by technological progress. If countries share the same technology and fundamentals, they share the same steady state, so differences in per capita GDP will tend to disappear over time (Aghion and Howitt, 2009).


Three advanced economies, which until 2008 (Slovenia) or 2009 (the Czech and Slovak Republics) were classified as emerging markets, have been included in this chapter among the emerging markets, reflecting their classification during most of the decade.


The lower employment growth in emerging Europe was partly compensated for by longer hours worked.


Catching-up countries will typically not focus on innovations and inventing new technologies, and instead, rely on the adoption and imitation of techniques developed outside.


The literature also suggests that less-developed economies tend to go through a period of low and highly variable growth in the early stages of development. This is because the inability to diversify risks causes agents to invest in safer but inferior technologies in order to reduce risk. As a result, growth tends to be low and more dependent on the random outcome of a few existing activities (Acemoglu and Zilibotti, 1997).


Empirical growth literature also finds a strong negative correlation between inflation, inflation volatility (even after controlling for the level of inflation), excessive credit growth, and growth. For our sample, we find a relatively weak negative relationship of these variables with growth (after adjusting for convergence). The weak correlation between growth and inflation could partly be due to the low variability of inflation among European countries, many of which share a common monetary policy.


Higher tax rates also reduce retained earnings—a major source of financing for investment, particularly in smaller firms.


See Bassanini, Nunziata, and Venn (2009). As suggested in Aghion and Howitt (2009), high labor market flexibility could be particularly beneficial for innovation activities, which often require initiative, risk taking, the selection of good projects and talents, and weeding out projects that are not expected to be profitable or operational. However, a more rigid labor market may favor the accumulation of firm-specific human capital, which could be more important in imitation activities.


See Aghion and others (2005).


The aggregate indicator of efficiency is an aggregate measure of domestic and foreign competition computed by the World Economic Forum in its annual Global Competitiveness Report, including the extent of market dominance, effectiveness of antimonopoly policy, tax and trade tariffs, restrictive rules on FDI, and so on.


Chen, Milesi-Ferretti, and Tressel (forthcoming).


Low growth, in turn, could derail the macroeconomic stabilization attempt, because it reduces fiscal space.


In the Netherlands, labor market reforms were accompanied by efforts to privatize government stakes in high-profile enterprises (steel and airlines) and to allow the bankruptcy of a major loss-making shipbuilder (which had received substantial government support). Both of these efforts signaled a change in industrial policy. Additional product market reforms, such as liberalizing licensing requirements and introducing new competition laws that included anticartel measures, were introduced a few years later. In Sweden, the wage bargaining system was reformed by reintroducing more coordination in the wage bargaining process and by using wages in the tradable sector (exposed to global competition) as a benchmark for wage negotiations. The reform of the wage bargaining system was accompanied by other labor market reforms in 1997 and by a deregulation of product markets.


It has been well documented that there are complementary gains from these reforms. For example, based on an OECD country panel from 1990–2004, Berger and Danninger (2005) find that lower levels of product and labor market regulation foster employment growth, including through sizable interaction effects, and note that unless deregulation costs are asymmetric across markets, optimal deregulation requires some form of coordination. There is also theoretical support for some sequencing of these reforms; for example, see Blanchard and Giavazzi (2003).


Governments in Scandinavian countries (including Denmark and Sweden) promoted the “flexicurity” concept that combines high labor market flexibility in a dynamic economy with high social security for workers. This integrated strategy is becoming widely acknowledged as a way to preserve the European social model. The European Council of June 2009 concluded that flexicurity is an important means to “increase adaptability, employment and social cohesion” (European Council, 2009).


This would include a rationalization of the public wage bill and social benefits as well as a restructuring of public transportation companies. The government has recently started a major program to achieve this objective.


In the Netherlands, for example, an initial wage agreement led to a rapid increase in employment, primarily reflecting the absorption of new entrants, including women. However, the percentage of the working-age population receiving benefits had remained high, because the stock of inactive workers had changed little, and this prompted the need for further reforms, such as tightening of eligibility criteria.


Although Europe is catching up with the United States with slightly higher growth (for the period 1995–2010, EU15 average annual growth was 1.8 percent compared with 1.4 percent in the United States), its TFP growth is still below that of the United States. Compared with the United States, advanced EU economies have higher productivity growth in manufacturing, but lag behind in the ICT and services sectors (Figure 3.29).


Gordon (2011) suggests that lower hours worked in Europe are due to high labor taxes, high minimum wages, and tight product market regulations.


Gordon (2007) notably suggests that regulatory barriers and land-use regulations inhibiting the development of large retailers (such as Wal-Mart) as well as corporatist institutions designed to protect incumbent producers partly explain the lower labor productivity level in Europe compared with that of the United States.


Maastricht Economic and Social Research and Training Centre on Innovation and Technology (2011).


As suggested by Gordon (2007), the environment is more favorable for innovation in the United States than in Europe, notably owing to an openly competitive system of private and public universities, government subsidies to universities based on peer-reviewed research grants rather than unconditional subsidies for free undergraduate tuition, strong patent protection, and a flexible financial infrastructure making available venture capital finance to promising innovations.


Daminai and Pompei (2008) note that takeovers are more frequent in these sectors. External takeovers are less frequent in sectors that feature creative accumulation and a stable core of leading innovators, because they represent a break in the continuity of deepening innovation processes.


In the United States, of the 20 publicly listed companies with the highest market capitalization in 1967, only 11 were still in the top 60 at the beginning of 2004 (Gersemann, 2004).


An example of these policies is the effort recently undertaken by France to support innovation and competitiveness. The tax credit for R&D activities (Crédit Impôt Recherche) provides generous incentives for research whereas the “Grand Emprunt” aims at supporting a variety of university-based and other research efforts in strategic sectors such as green energy, life science, and digital society and at promoting the creation and the financing of innovating (small and medium-sized enterprises—SMEs). Agencies and committees have also been set up to ensure objective allocation and efficient management of the funds. However, although spurring innovation has the potential to increase competitiveness and growth, the cost effectiveness of these measures needs to be evaluated within an appropriate timeframe. Strict adherence to rigorous governance standards is needed to ensure efficient and productive investments. In addition, other obstacles to innovation should be evaluated, such as those possibly arising from the relative paucity of medium-sized firms.


Only Switzerland has a capital market comparable in relative terms to that of the United States, whereas the euro area capital markets tend to be smaller and on average comparable in size with the Japanese capital market.


In particular, in Austria, Greece, Luxembourg, and the Netherlands (Djankov and others, 2007, 2008).


In particular, in France, Greece, Ireland, Portugal, Sweden, and Switzerland (Djankov and others, 2007, 2008).


With the exception of Ireland, Netherlands, Switzerland, and the United Kingdom (Hartmann and others, 2007).


With the exception of countries such as Ireland, Sweden, Switzerland, and the United Kingdom (Hartmann and others, 2007).


Especially in Greece and Italy (Box 3.4).


A family-owned structure does not hinder productivity growth and innovation. The German “Mittelstand” (Germany’s legion of small and medium-sized family firms contributing to 50 percent of GDP) is often seen as a good structure for innovation owing to its enviable distribution network, close partnerships with researchers at universities, highly skilled labor force, efficient supplier clusters around large manufacturers, harmonious employer-employee relationships, and high contribution (about 30 percent) to total exports that have not benefited SMEs in other countries such as France or Italy. However, even German SMEs find the need to hire foreign staff and raise new capital, perhaps via private equity, as a way to compete with firms abroad (Linneman, 2007).


See Johansson and others (2008), which shows that investment is adversely affected by corporate taxation. This is likely to have depressed TFP growth or innovation because of less spending on R&D and less FDI. In addition, these authors show that (i) employer and employee social security contributions have a more negative influence on TFP in industries that are relatively more labor intensive; (ii) top marginal personal income tax rates have a more negative effect on TFP in sectors characterized by high firm entry rates; and (iii) TFP in growing firms that are in the process of catching up with the technological frontier is particularly affected by corporate taxes, as these firms have on average a larger tax base.


Molagoda and Perez (2011) found evidence that, in addition to human capital and the degree of market regulation, corporate tax rate (interacting with profitability) and trade openness affect TFP growth, based on data for EU13 and the United States for the period 1980–2007.


The effect of corporate taxes is possibly negligible for young and small firms (which include a large share of start-ups), as they tend to have low or zero profits, even in highly profitable industries. The higher corporate tax rate in the United States, with the more developed market for risk capital, may explain why a much higher share of innovation is done by new firms.


Aghion and others (2010) find a strong correlation between the governance structure of universities and their research and education output.

    Other Resources Citing This Publication