|
The causes and consequences of unemployment remain among the most controversial issues in economics, and the interwar experience is the best natural experiment available to test competing theories. Yet the relationship between unemployment and money wages in the 1920s and 1930s in the United Kingdom has never been fully explained: neither the Phillips Curve nor the later 'real wage' approach of Sargan seem to provide a satisfactory account. In Discussion Paper No. 147, Research Fellow Nick Crafts argues that the unemployment rate is, by itself, an inadequate measure of excess demand in the labour market in this period. Estimates of the wage equation can be improved by including trade union membership, the level of unemployment benefit, and in particular the extent of long-term unemployment. Crafts first examines the role of the long-term unemployed, defined as those out of work for over a year. The long-term unemployed rose from 10.7% to 27.1% of all unemployed workers between 1929 and 1937. This increase in long-term unemployment was concentrated in the depressed regions of 'Outer Britain', especially among elderly workers in those areas, and was associated with the permanent decline of older staple industries. Survey data from this period indicate that only 29% of the long- term unemployed were still searching seriously for work, and that lengthy spells of unemployment led to depression, apathy and declining hope of re-employment. Many of the long-term unemployed effectively dropped out of the labour market. Crafts's findings suggest that the large numbers of long-term unemployed did not restrain the growth of money wages, whereas the short-term unemployed did (a finding similar to that of Layard and Nickell for the postwar period). Using quarterly data for the period 1925-39, Crafts finds that the simple wage equation models of Phillips and of Sargan are inadequate. He then introduces variables representing the long- term unemployment rate, trade union membership, the employers' national insurance contribution rate and replacement ratios (the ratio between net income while unemployed and while working). Inclusion of the long-term unemployment rate as a separate term considerably improves the ability of the equation to explain wages and increases the coefficient on unemployment, which becomes statistically significant. The measures of trade union organization and of replacement ratios are also generally statistically significant. These results also suggest that the rate of measured unemployment consistent with non-accelerating inflation (the NAIRU) rose during the 1930s. This finding may justify Keynes's caution in not advocating further demand expansion in 1937 as he had in 1929, Crafts argues, even though unemployment was slightly higher. Although estimates of the NAIRU can be imprecise, the failure of long-term unemployment to restrain wage inflation suggests that the adverse demand shock of the early 1930s permanently lowered the equilibrium level of employment in the United Kingdom. Long-Term Unemployment, Excess Demand and the Wage Equation in Britain, 1925-1939 Nick Crafts Discussion Paper No. 147, December 1986 (HR) |
|