The sharp and deep recession that followed the global economic crisis looked very much like the Great Depression to start with, but so far it has not turned out that way (see Eichengreen and O’Rourke 2010).
While some observers such as Almunia et al. (2010) point to monetary policy and fiscal stimulus as the decisive difference, here I argue that some attention should also be given to the supply side in the Great Depression. In particular, developments in interwar labour markets slowed recovery from the Great Depression.
It is illuminating to compare the labour market experiences of the UK and the US and to consider the recession of the early 1920s as well as that of the early 1930s. As Figure 1 shows, the average unemployment rates before WWI averaged about 6%. Between 1920 and 1921 there was a sharp recession that pushed the unemployment rate above 10% in both countries. But then the histories diverge. While the US labour market bounces back to something resembling the pre-war unemployment rate, the UK does not. Yet after 1929, the US unemployment rate rapidly overtakes the UK and, despite some recovery, it seems to settle at a permanently higher level.
Figure 1. Unemployment rates, 1890-1940
The visual impression from Figure 1 is supported by estimates of the Non-Accelerating inflation Rate of Unemployment (NAIRU). The UK NAIRU shifted up decisively in the 1920s, with no further change in the 1930s; by contrast the US NAIRU shifted dramatically up in the 1930s but not before. The message from these comparisons is clear. If we want to identify key shifts in labour market performance then we should look at the 1920s in the UK and the 1930s in the US. In a new CEPR Discussion Paper (Hatton and Thomas 2010), Mark Thomas and I re-examined the historical evidence for the two countries across the two decades.
Shocks and institutions
One way of thinking about persistently high (or low) unemployment is to consider the combination of “shocks” and “institutions”. “Institutions” is the constellation of labour market structures, rules, and customs that determine how flexibly the labour market responds to demand or supply shocks. Certain institutions (and combinations of institutions) can magnify the impact of shocks and cause their effects to persist. This approach has been used to explain the stubbornly high European unemployment in the 1980s and 1990s as compared with the US (e.g. Blanchard and Wolfers 2000), and it is just as relevant to the 1920s and 1930s.
The second decade of the 20th century saw dramatic changes in UK labour market institutions. Union density tripled, from 14% in 1910 to a peak of 44% in 1920, and with it industrial militancy dramatically increased. Wartime pressures accelerated the trend towards industry-level collective bargaining and, in industries where there was little pre-existing organisation, minimum wages were set by Trade Boards. Taken together, these forms of centralised wage setting machinery covered about half the UK labour force by 1920. Also important, the unemployment insurance scheme, established in a small way in 1911, was expanded to cover the majority of workers; benefits were increased and made available for longer unemployment durations.
The sharp negative demand shock of 1920-1921 was compounded in the UK by a supply shock. In 1919 average weekly hours were cut by 13% with no cut in the weekly wage. During the 1920s the British economy recovered slowly from these shocks while contemporary observers, including Keynes, Pigou, and many others, worried about the lack of labour market adjustment. Looking across the Atlantic they would have seen a very different picture. In the US unionism rose modestly during the WWI and then fell back to the pre-war level while the structure of industrial relations remained largely unchanged. No national system of unemployment emerged. And the more gradual decline in working hours was easily absorbed by stronger productivity growth.
Now to the Great Depression
The demand shock that initiated the Great Depression was more severe in the US than in the UK and it created pressure for action. Roosevelt’s New Deal was once seen as a Keynesian-style recovery package, but historical research has downgraded the positive demand-side effects and upgraded the negative supply-side effects. Foremost among these is the National industrial Recovery Act introduced in 1933. It involved raising wage rates and cutting hours but, above all, it spawned industry agreements that some have seen as a form of cartelisation. Its successor, the National Labour Relations Act (1935), further strengthened the hand of labour organisations. As a result both nominal and real wages failed to adjust downwards despite unemployment remaining historically high.
In both the UK and the US, labour markets were much more fluid than they are today, labour turnover was higher and unemployment durations were shorter. But as workers queued for jobs in the depression, unemployment durations lengthened. In 1936, 27% of unemployed men in Britain had been unemployed for more than a year. By 1940, the figure in the US exceeded one third (even excluding those intermittently employed on government relief work). The long-term unemployed became less and less able to compete for work as their skills and motivation atrophied. As a result wage pressure was not as strong during the recovery as the persistently high unemployment rates would suggest.
The experience of the Great Depression has cast a long shadow. The early and aggressive use of fiscal and monetary policy in the recent recession attests to that. In the 1930s policy activism meant leaving the gold standard, and abandoning the policy dogma that went with it. But these policy shifts did not come into effect until two years into the recession in the UK and nearly four years into the recession in the US, and they were insufficient to promote a vigorous recovery given the developments in the labour market. By that time the NAIRU had increased to around 10% in the UK and even higher in the US.
In the interwar years, as today, governments were put under severe pressure to do something to cope with the depression. Much of their effort went into maintaining high wages and ameliorating the plight of the unemployed in order to stave off unrest. These policy packages contained interacting elements that magnified the shocks and caused their effects to persist. In the current recession active demand-side policies have averted such pressures so far, although that might change with a return to austerity as the case of Greece shows only too clearly.
An important but neglected lesson from the Great Depression is that labour market policies should be firmly focused on fostering labour market flexibility and maintaining employability, and avoiding policies that cause unemployment to persist.
It is worth remembering that the Great Depression lasted for ten years and mass unemployment was not brought to an end by equilibrating adjustments in the labour market but by the massive boost to aggregate demand occasioned by WWII. Without that the depression would have lasted well into its second decade.
Almunia, M, A Bénétrix, B Eichengreen, KH O’Rourke, and G Rua (2010), “From Great Depression to Great Credit Crisis: Similarities, Differences and Lessons”, Economic Policy, 25:219-265.
Eichengreen, Barry and Kevin H O’Rourke (2010), “A tale of two depressions: What do the new data tell us?”, VoxEU.org, 8 March.
Blanchard, O and J Wolfers (2000), “The Role of Shocks and Institutions in the Rise of European Unemployment: The Aggregate Evidence”, Economic Journal, 110:C1-C33.
Hatton, TJ and M Thomas (2010), “Labour Markets in the Interwar Period and Economic Recovery in the UK and the USA”, CEPR Discussion Paper 7983.