Euro-pessimism is back. . . .  between 1990 and 2000 GDP increased at an average annual rate of 3.2% and 2.1% in the US and in the EU respectively. . . .  

Adding to the concerns of the Euro-pessimists is the steady erosion of the EU exports share in world markets. Between 1990 and 2001 Germany's export share has fallen from 12.1% to 8.1%. Italy and France have not fared better, with their share declining respectively from 5.1% and 6.1% in 1990 to 3.9% and 4.8% in 2000. Again, the contrast with the US is striking: over the same period, the US share in world export markets rose from 13.1% to 14%.

Labor productivity is also mentioned as a further, and perhaps more fundamental, indicator of Europe's long term decline . . . between 1997 and 2002, the US has definitely over-performed the EU, with labor productivity growth rising to 2.2% in the former and falling to 1% in the latter.

. . . we take a closer look at the European performance. We take issues with the most of the indicators - GDP growth, world export share, productivity growth - that have been used to demonstrate the facts of Europe's decline:

His conclusions:

• Once corrected for demographic changes and accounting differences, the growth performance of the Euro nations since 1990 is no worse than that of the United States;

• The income gap that still exists between Europe and the US is mostly accounted for by differences in hours worked, suggesting that higher American GDP results mainly from different preferences for leisure and not from European inefficiency;

• There has been no systematic decline in the competitiveness of European economies; changes in world export share can be explained by changes in the nominal euro-dollar exchange rate;

• The real problem with the European economy seems to be its inability to sustain rapid productivity growth and employment growth simultaneously.

By and large . . . old Europe suffers first and foremost from a demographic decline. During the eighties population increased at an average annual rate of 1% in the US and only 0.3% in the Euro area. During the nineties, population growth in the US even accelerated to 1.15% and the gap with the euro area rose to more than 0.8%. Clearly, under these circumstances, comparing aggregate GDP growth is simply misleading. . . . When corrected for population growth, Europe's performance has been better than the US in the eighties, has worsened markedly during the first half of the nineties and, in the more recent years, has been at par with that of the US.

Average Annual Per Capita GDP Growth, US and Euro Area               
                    1981-90    1991-02    1991-96    1996-02
United States    1.9%         1.7%        1.4%      2.1%
Euro Area         2.1%         1.6%        1.2%       2.0%

Economic Growth and Accounting Conventions

The harmonization of national accounts definitions has been a decisive factor in allowing meaningful cross country analyses. Yet, substantial differences persist, even among industrial countries:

• Military Equipment: In 1996 . . . when US national accounts were duly amended, the extension of coverage of investment was more extensive than recommended by international conventions, as it included also those assets that are used exclusively for military purposes. The effect on the level of GDP may not have been sizeable, 0.6% according to the OECD. Its impact on growth may be even smaller, 0.03% over the past decade. Both effects however are likely to grow larger given the recent military build up undertaken by the current US administration.

• Financial intermediation services to households are a further item that is treated differently in the US, where it is included since the late eighties, and in Europe and Japan, where contrary to international conventions it is excluded mainly for measurement difficulties. The impact on GDP level is sizeable, 2.3%; that on GDP growth is not negligible, 0.1%.

• Software Investment: Most European countries use a demand approach, relying on the way businesses record investment. An alternative approach, used in the US, is to measure the total supply of software services and then estimate the portion of such services with assets characteristics. The impact on the level of GDP is substantial. Given the disproportionate growth of software expenditure, differences in the measurement of software have also a sizeable impact on growth. According to the OECD, had the US used the European demand-based approach, growth in 1997-98 would have fallen by a substantial amount, up to 0.2%.

• Hedonic Prices: Last but not least, Europe and the US differ in their use of hedonic prices. This is perhaps the most notorious correction. Perhaps surprisingly, its GDP effects are not particularly sizeable. Moreover, even the sign of such correction is not unambiguous. For instance, if hedonic prices are used for intermediate imported goods, the volume of imports will grow with a corresponding reduction, ceteris paribus, in the volume of net output. Overall, according to the OECD, the impact on US growth is relatively small, 0.1%. More crucially, it is offset by a parallel correction in US national accounts. Following international recommended practices, the US combines the use of hedonic prices with Laspeyres, instead of Fischer indices, the latter being used instead by European statistical offices. Laspeyres indices lead to a fall in measured growth rates of approximately the same order of magnitude as the increase associate with the reliance on hedonic prices.

Overall, allowing for different accounting definitions leads to a reduction in the US growth rate of around 0.2-0.3 %. As emphasized by the OECD (2003), this results only in a fractional reduction of the aggregate growth differential, 1% during the nineties, between the US and the Euro area.

However, the impact on the difference in per capita GDP growth is quite significant. Moreover, there is a further and quite crucial effect on the dynamics of the convergence process between Europe and the US. Recall that most of these accounting corrections have little or no impact on measured growth during the eighties, their main effect being to lower US growth during the nineties, particularly in the second half of that decade. This has a number of substantive implications:

1. First, the decline during the nineties in the relative performance of the Euro area with respect to the US was less pronounced than previously thought, as it reflects to some significant extent different accounting procedures.

2. Second, it is no longer true that the US economy grew at a faster rate than the Euro area during the first half of the nineties.

3. Finally, and perhaps more crucially, in the second half of the decade US per capita growth would be again below that in the Euro area.

To sum up, if the convergence [catch-up of Euro per-capita GDP with that of the US] process ever came to a halt in the first half of the nineties - and even this conclusion is less than fully warranted - it was back on track during the second part of the decade. . . . once different accounting conventions are allowed for, per capita growth in the Euro area has not fallen behind the US and, most likely, has surpassed it in the most recent years.

Composition of Growth

To measure the proportional contribution of the different factors to the US - Euro area income gap:

• Productivity differential were quite important at the beginning of the period, as they accounted for 57% of the gap.

• Faster productivity growth in Europe meant however that this factor played a lesser role over time. Indeed, in 1997, productivity per hour was actually higher in the Euro area than in the US. More recent trends have however favored the US and, in 2001, productivity differential gave once again a positive contribution to the income gap between Euro and the US. At the same time, however, its role was much less reduced.

• Indeed, in 2001, the key factor in explaining the income gap between the two areas was the number of hours worked per employed person. By and large, therefore, the preference for leisure . . . is the main factor behind Europe's short working time and, hence, its remaining income gap with respect to the US.

Competitiveness

[Although] the loss of market shares in world exports is often lamented as an unambiguous indication of the decline in Europe's competitiveness in the international arena. . . . there are no obvious indication of a structural loss in competitiveness for euro area countries. The evolution of their export share in world markets is well explained by the behavior of the nominal euro-dollar exchange rate. The recent recovery on world export share for Germany, France, and Italy also points to the paramount role of valuation effects.

The Real Problem with the European Economy

In a nutshell, while the US economy was able to combine fast employment growth with an extraordinary productivity performance, the Euro area economy appears to be facing a difficult trade-off between fast productivity growth, and hence sustained wage growth, and rapid employment growth but stagnant real wages.

Why might this be the case? Typically inefficient labor and product market regulations are cited, and Faini does not dispute this. Others have suggested a lag in the diffusion of information technology in Europe. But there may be another reason:

New entrants into employment were mostly young and quite inexperienced. The recorded fall in TFP growth may therefore be a statistical artifact, to the extent that no (or inadequate) allowance is made for experience. Over time, as the new entrants acquire more experience, Europe's productivity growth would rebound, softening the trade off between productivity and employment growth. . . . Whatever the answer, it remains true that the opening up of the productivity gap with respect to the US offers new prospects for Europe to catch up.