Inflation and Unemployment
Is Europe so different?

Excessive real wages are not the explanation for European unemployment, argued Research Fellow Robert Gordon at a CEPR lunchtime meeting on June 11. The European problem of declining employment and rising unemployment is centred in the manufacturing sector, yet the wage-gap index for European manufacturing has fallen steadily since the late 1970s and is now well below the value of the same index for US or Japanese manufacturing. Gordon also claimed that differences between the economies of Europe and the United States had received too much emphasis in research explaining wage and price behaviour: the response of labour input and labour productivity to changes in the real wage is similar in Europe, the US, and Japan. Gordon cautioned against one-sided arguments in favour of lower real wages. A stimulus to aggregate demand would not only raise output and employment, he argued, but would also raise the real wage and induce a productivity-boosting substitution away from labour.

Robert Gordon is Professor of Economics at Northwestern University, Illinois, a German Marshall Fund Fellow, and a Research Fellow in the Centre's International Macroeconomics programme. He has published extensively on a wide range of macroeconomic policy-related issues including inflation, wages, employment and monetary policy. The lunchtime meeting at which he spoke was one of a series sponsored by the German Marshall Fund of the United States, focussing on the international economy. The opinions expressed were his own, however, and not those of the German Marshall Fund or of CEPR, which takes no institutional policy positions.

Gordon began by noting that the major economic policy issue of the 1980s in Europe was its persistently high unemployment. Policy-makers were apparently unwilling to reduce unemployment by expanding aggregate demand, and attention had been focussed instead on what Gordon described as 'a litany of European supply- side maladies'. To construct this list of ills, Europeans cast envious glances toward the US and Japan to discover those aspects of European economic institutions that are different and hence 'worse'.

This was too simplistic, Gordon argued. In their envy of US employment creation, Europeans neglected the dismal productivity record of the US economy. In the 70% of the US economy outside of manufacturing, productivity growth had been a pitiful 0.5% per year since 1968, and since 1972 aggregate US productivity growth had been just 0.8% per annum, in contrast to 2.4% in the United Kingdom, 2.8% in Europe as a whole, and 3.6% in Japan. The pessimistic tone of UK policy debates, for example, did not recognize that in the productivity 'league table', the United States was further behind the UK than the UK was behind Japan. Gordon's empirical analysis, which he outlined at the lunchtime meeting, cast doubt on some of the contrasts between the United States and Europe that have received such heavy emphasis in previous research. His research did confirm that labour input responds to changes in the real wage, a relationship stressed by UK economists Richard Layard and Steve Nickell. But the response of labour input and labour productivity to changes in the real wage is similar in Europe, the United States and Japan.

Explanations of European unemployment often centre around a distinction between 'Keynesian' and 'classical' unemployment, in which the real wage plays a central role. Arguments that unemployment is classical are often based on a demonstration that employment responds negatively to an increase in the real wage and that growth in European real wages has been 'excessive'. The 'wage gap', a concept popularized by Bruno and Sachs, is frequently used to assess whether real wages are in fact 'excessive'. In their analysis of European unemployment and stagflation, the wage-gap indices constructed by Bruno and Sachs increase much more in Europe than in the United States. But the explanation of European unemployment by excessive real wages has been carried too far, according to Gordon. He outlined some major difficulties with the wage-gap explanation of unemployment.

In 1984 the European wage gap was lower in manufacturing but higher in non-manufacturing, although most European job losses have occurred in manufacturing. The wage-gap index for European manufacturing has fallen steadily since the late 1970s and is now well below the value of the same index for US manufacturing. The Japanese wage gap dwarfs anything experienced in Europe, yet Japan seems free of European unemployment problems.

Much of the increase in the European (and Japanese) wage gap was a statistical illusion, Gordon argued: the wage gap concept is nothing more than the share of employee compensation in national income. But the earnings of farmers and individual shopkeepers are not included in employee compensation. Once the share of household earnings from farms and small businesses was added to the share of employee compensation, the increase in the Japanese wage gap since the mid-1960s was eliminated, and the increase in the European wage gap was reduced by more than two thirds. Empirical analyses of production functions suggested that the elasticity of substitution in production was less than one. Given this, an increase in labour's share of output was only to be expected as labour became more scarce relative to capital.

Too much had been said about the evil of higher real wages: the benefits of higher real wages seem to have been forgotten. An increase in the real wage could reduce hours worked, but increase output per hour. Such substitution away from labour in response to an increase in the real wage has been at the heart of the economic growth process for centuries. In fact, Gordon's analysis demonstrated that it was exactly this 'wage gap' which accounted for a substantial component of productivity growth in Europe, Japan and even in the United States. It also helped explain the mysterious slowdown in productivity growth in the United States after 1972 and in Europe and Japan after 1979. Gordon noted that Japan's productivity growth outside of manufacturing had actually been zero between 1979 and 1984.

Gordon suggested that the situation of the US economy in 1939 offered a useful precedent for European policy-makers today. In that year in the United States there was 17.5% unemployment (mainly long-term) and no output growth, yet prices refused to decline. The approach of Bruno and Sachs would point to the large wage gap as the culprit, since the share of employee compensation had risen by 11% from 1929 to 1939. Yet from this dismal and unpromising starting point, the United States enjoyed an explosion of growth in 1939-42. The fiscal stimulus brought about an increase in real GNP of 50% and a fall in unemployment to 4%. 75% of this expansion of nominal aggregate demand represented a real output increase, while the remainder was inflation. Gordon did not recommend that such a radical fiscal stimulus be implemented today in Europe. But he used the example to argue that a modest acceleration of nominal demand growth, for instance by 3 percentage points per annum, would plausibly be accompanied by 2% faster real output growth, with an increase in the rate of inflation of no more than 1%.

Gordon's lunchtime talk was based on his econometric analysis of the relationships between productivity, wages and prices in the United States, Japan and Europe, available in a forthcoming CEPR Discussion Paper. His study focussed on three major issues: (1) the response of employment and productivity to changes in the real wage and in the wage gap; (2) the 'Phillips-curve' response of real wages to economic slack; (3) the division of a change in nominal aggregate demand between inflation and real output growth.

He based his analysis on a consistent set of time-series data for 14 countries over the period 1961-84. The data were developed for each economy as a whole, as well as for its manufacturing and non-manufacturing sectors separately. Gordon argued that this distinction between sectors is crucial: the productivity growth slowdown, as well as constructed 'wage gap' measures, display quite different behaviour inside and outside of manufacturing.

For each of the 14 countries, Gordon estimated equations describing the behaviour of productivity, wage and price changes for the aggregate economy and for each sector separately. He then compared the results for Europe as a whole to those for the United States and Japan. His estimated productivity growth equations allowed him to identify the effects of the real wage, cyclical changes in utilization, and the underlying productivity trend. Real wages played a similar role in Europe and in the United States.

Gordon found that wage gaps did not emerge after 1973 in Europe and Japan out of a failure of real wages to decelerate in response to the post-1973 productivity slowdown. These wage gaps originated instead from episodes of autonomous 'wage push' in Europe in the late 1960s and in Japan during 1973-4. In this sense real wages in Europe (and Japan) were too flexible, at least in the upward direction, rather than too rigid.

Gordon also reported results for equations which explained the division of nominal GNP growth between inflation and real growth. In terms of the aggregate economy, this division was the same for Europe and the United States. Thus the US experience of demand expansion in 1982-86 did indicate the scope for European action. Gordon argued that the adverse side-effects of the US expansion on its exchange rate and the trade balance could have been avoided. These side-effects could also be avoided in Europe by pursuing balanced monetary and fiscal expansion rather than by using monetary or fiscal expansion alone. Such policies were also more likely to be successful if coordinated among all the major European nations rather than if carried out by a single nation acting alone.

The discussion which followed touched on Gordon's empirical analysis, as well as his advocacy of a balanced European expansion. Some questioned whether the United States in 1939 did offer a useful precedent. A demand expansion, it was suggested, might merely increase imports with little or no benefit to domestic production. This, Gordon responded, suggested that a lower exchange rate might be desirable.