Custom Search

Unemployment, which according to Feinstein, averaged 4.4% between 1890 and 1913, averaged about twice as much in the period 1921-1929. How would you explain this?

The interwar period in Britain saw a very strong downturn in the level of economic activity resulting in a very high unemployment peaking 22% in 1920-21. The unemployment was high compared to both the previous history and the other countries, although other countries were having problems as well. Furthermore, the change in unemployment was not just a cyclical movement, the unemployment level after 1921 recession did not fall to a historical trend level, but remained permanently higher resulting in an "output gap". This must mean that the British economy was not going through just a cycle, there must have been some kind of shocks present to cause a permanent change. In this essay I am going to discuss the two types of shocks that could have been present: the demand and supply side shocks.

However before I start exploring the shocks I must first stress the periodical nature of unemployment in the interwar era. The two major recessions were in 1920-21 and 1928-32, so the years chose for comparison take account approximately half of both cycles, which is fair. However one must still note that substantially different average unemployment rates are obtained for 1920-32 average or 1922-28 average. There are also specific reasons for both of the peaks. I will not go into them in great detail - I will only look at changes that could have had permanent effect. Unemployment in the interwar era showed a pattern like this:

Initially it was argued that supply side shocks caused the peaks in unemployment. Main concern here has been the strength of trade unions after the war. They managed to reduce the number of hours worked substantially, without reducing the wage level, because rather keeping the hourly wage constant employers kept weekly rate. This amounted to approximately 30% real wage increase. It has been calculated that the elasticity between employment and real wage is approximately -3/4. Economy was not experiencing an increase in the productivity at that time, so this was a serious rise in wage costs for the firms. This caused the wage gap to rise (Broadberry). The wage gap is the difference between wage costs and output. This gap only eroded after second world war.

The prices were also falling, because of the return to gold. Nominal wages did adjust, but with a lag and meanwhile people had too high wage. Also workers had to reduce consumption during first world war that caused high accumulated savings. They shifted out the budget constraint. People could afford to consume more goods and more leisure at the same time and thus reduced the hours worked.

However Dimsdale (1984) showed that the real wage movements were small at that time and could not account for any major changes. It should be noted that the real wages were measured by RPI by Broadberry. GDP deflator should have been used instead to get the rise in the producer wage cost. This change will eliminate long term effects of the gap, by 1922 there is no gap anymore. Also the calculation was for all people, the wage was increased only by 18% for the heads of households in full employment.

In long term, however the initial high employment had permanent effects as explained by the hysteris theory. This will mean that as the number of unemployed rises, the number of long-term unemployed rises as well. They will lose the hope of getting a job and will give up trying. They will also have low probability of being re-employed, as employers like recent practice. Insider-outsider theories explain the persistence as well, as the institutionalised wage setting covered 3/4 of the working population. Outsiders, that were unemployed, could not affect the wage level and thus the wage level did not come down when unemployment rose. Unions were also using wage relativities to determine their demands (wanted to have as much as other) and had co-ordination failures. This meant no union wanted to be the first to give in and reduce the wage level first.

There is however not much empirical evidence for long-term unemployment. There was a group of people who had very high turnover in unemployment, that means they were changing jobs very often. High turnover meant less job security and lower pay. It can be explained by the change in the benefits level.

Benefits were increased after 1921 and they were made available to more people. People could now take some time off between looking for new jobs. They were also given unemployment benefit from the first day they became unemployed, so firms were using the benefits for supporting the wage - they employed people for short times, made them redundant and then re-employed. Maki and Spindler(1979) have found 3 ways in which an increase in the benefit level can cause unemployment. First, more people will register themselves as the labour force to claim benefits. This happened with women registering more during the interwar period. Secondly the effective supply of labour decreases due to increase in searching time and finally demand for labour will decrease as firms are more willing to lay off workers or put them to part time work.

Crafts found little evidence of benefits affecting the employment, though. Unemployment Assistance Board showed the hopelessness of long term unemployment and argued that it was unlikely to be voluntary. In 1936 73% of unemployed lived below the poverty line, so they are unlikely to receive too many benefits. And according to national statistics, only about 150 000 workers were in some kind of work sharing system, so the cheating of the system was likely to be insignificant. However Benjamin and Kochin (1979) have found evidence to support that the unemployment was caused by the increase in the replacement ratio (what percentage a person can earn of their wage when they are on benefit). However calculations of this replacement ratio are rendered less valid again when we use data for the heads of households alone not for a man with a wife and two children (as Benjamin and Kochin did). Eichengreen first looked some household evidence in the New Survey of London Life and Labour (1928). This showed that the heads of families never chose the unemployment, although some secondary workers did.

The replacement ratio rose only after 1922, when new Acts were passed, but significant employment existed already in 1920-21. Benjamin and Kochin could also give no explanation for the difference in unemployment levels between different industries. Matthews have argued that the real benefit level, not the replacement ratio, caused unemployment. This view arises from the fact that as the prices fell the value of real benefits rose over the period. and so did unemployment.  But the wages actually fell in line with prices - if people would have become voluntarily unemployed, then the employers would have bid up the wage level to attract them back, but they clearly did not.

Smyth has done a major criticism to Benjamin and Kochin. He said they used an inappropriate ad hoc model. Benefit basically affects supply, but its effect is lost GDP, which is demand. The government could also not alter the benefit to wages ratio, it could only determine the absolute level of benefits. So absolute benefits and not the ratio should be used.

 Huton found no evidence of more search for employment. "The behaviour of the unemployed in searching for employment gives no evidence that the possibility of drawing unemployment insurance benefit has retarded the efforts of the unemployed to get back to work. It has removed the cutting edge that would otherwise attend that search"(Wright). The models of unemployment are also not robust. Worswick showed that if one takes out the recession year 1920, then the result will not hold.

Changes in the age/sex ratio and arising due to the natural increase in unemployment were also permanently affecting unemployment. Especially as older people, once made redundant, were far less likely to get a new job. And the population got older. But these effects were relatively small. Still, women and young people were less likely to be unemployed, and higher social class meant less unemployment too. Unemployment was initially high in these industries that were expanded during the War, but later unemployment rose in coal also and in other staple exports industries. Industries towards home markets have less unemployment. Unemployment is also regional, it was very high in North and West, in some places even 70%. These facts led people to think the unemployment occurred due to structural change to newer industries from the old staple ones.

It was estimated that 25-45% of the unemployment was structural. Alcraft - Richardson took and optimistic view and said that the resources were transferred to new industries. But there is not much evidence for this according to Mathew-Smee. They noted that the output growth was not very high in new industries. Total factor productivity was also not changing much. In new industries TFP should be rising fast. It was also noted that there was a greater change during the actual War period than during the interwar. New industries were also less capital intensive, so employment should have risen. Writers found no correlation between mergers and efficiency gains, too. So the structural change should not be taken as an explanation.

As seen from this widely scattered evidence on supply side, there were shocks present, but none of them was very big and many did not affect the output permanently. I now turn to look at the demand side shocks. Frequently demand side shocks are not permanent and thus tend to be overlooked. However, recent historians have shown there were permanent demand shocks present during the interwar era.

 Main concern has been on the return of the Britain to the gold standard with the pre-war parity. There is some strong comparative evidence supporting this. Scandinavian and British countries used fast transition to gold standard. France, Belgium return to gold at a depreciated rate (over 20 years). Rest of the Europe was not concerned about the transition, they had other problems from the War to sort out (Germany). Britain goes to restrictive monetary policy and sticks with it. France had restrictive policy but it defeated that as soon as problems arose. And evidently, Scandinavia and Britain face the biggest recession, it is mild in France and almost non-existent in Germany.

Flexible exchange rates and restrictive interest rates caused deflationary expectations and decreased exports. This graph shows an equilibrium A with a given exchange rate. If exchange rates rose with interest rate increase appreciation of an exchange rate to B will happen:

The actual appreciation was much bigger than we expect, because of the overshooting of the monetary theory. Different markets have different adjustment rates. Assets markets are quick, but goods markets have a number of rigidities, so they adjust slowly. If the government made the announcement (1919) of restrictive policy, a quick adjustment to B happened:

So initially the exchange and interest overshot and then adjust slowly downwards as IS shifts. That is exactly what we observe, there was a big overvaluation in 1919-21. Overshooting has been observed also in other countries with flexible exchange rates. Overshooting leads to too high exchange rate and contraction of export that should lead to a fall in GDP. High interest also decreased the investment and the national income. But for sure it is only the price of exports and imports that changes, but on the balance of payments account we are concerned with the value. Value will not decrease if the price increases (for imports) when the demand for imports is inelastic. In fact it is quite the opposite. Similarly in order to get a deterioration in  the current account we have to assume elastic exports, but there is no strong empirical evidence for that.

But the exchange rate change was only temporary. If one wants to explain why the UK became locked in to the lower GDP level one has to look at the multiplier effects. Imports are leakages and when they increased, GDP decreased by more and possibly permanently. However writers have assumed that the high exchange rate lead to an automatic increase in investment and decrease in exports.

We are unlikely to find a single explanation for the high unemployment. Perhaps the overshooting and wage gap together made the UK to lose its trading position and competitiveness and changed the whole world trade patterns. Once the competitors had set in the UK they did not leave when the UK’s industry made a comeback. Foreign firms might have had high set-up costs, instead of leaving they lowered their markups. There is some empirical evidence for that, particularly the rise in import ratios from 20% of GDP before the War to 24% after.

There was also lots of optimism and new investment up to 1918. Many debts were taken out with near 0 real interest rate. As the economy deflated real interest rose and real debt burden shot up. Empirical evidence shows that the debt to income ratio rose from 1.6 in 1913 to 2.8 in 1920's. In France, for example, the inflation was eroding some of the debt.

It is thought at present that the combinations of both supply and demand side lead to persistence of low GDP and high unemployment during the interwar period. Although supply side reasons tend to be weaker, there are many of them and they most probably reinforced the big demand shock of returning to the gold standard.

Click here to see more economics,politics and school papers from me