The financial crisis and real labour earnings in Denmark, Finland and Sweden.
Did the choices of exchange rate regimes determine how the three countries were impacted by the Global Financial Crisis?
The three countries all experienced high growth from the beginning of the Millennium until the financial crises. GDP increased until 2007 in all three countries and EU28. Following that year, it decreased two years in a row in Denmark and Sweden. Denmark did not experience “the double dip” as the three other regions, especially Finland, did c.f. Figure 1.
Splitting the developments shown in Figure 1 into different periods of time shows that post-crisis growth has been significantly lower in all three countries (and in EU28 although not shown here). In Finland was even the annual average growth negative between 2008 and 2017, c.f. Table 1.
The lower growth since the Financial crisis, or even earlier, means that incomes of ordinary people will be lower or even stagnate. Since most of the incomes of ordinary people is in terms of wage income, this entry looks at growth of labour income measured as real labour earnings per capita between 2000 and 2017 and further decomposes the changes in real labour earnings per capita by following the approach by Hall (2017).[1]
Real labour earnings per capita grew steadily in all three countries until 2008 when the Financial Crisis broke out. While Swedish real labour earnings reached the pre-crisis level already in 2011, Denmark reached it as late as 2017 and real labour earnings per capita remains lower in Finland, c.f. Figure 2.
Figure 2. Real earnings per capita 2000-2017. Log Index 1970=1.
According to Figure 2, developments in real labour earnings were similar until the outbreak of Financial Crisis. Pre-crisis developments were similar in the three countries. The only qualitative differences between them was that the employment rate in Sweden contributed negatively while the working age population rate contributed positively.
It’s therefore more interesting to discuss differences between countries from the outbreak of the Financial Crisis and onwards. As mentioned above, this will be done by breaking down real labour earnings into different parts.
Real labour earnings are defined as:
Real labour earnings per capita = the labour share * real output per hour * hours per worker * employment rate * labour force participation rate * working-age population rate.[2] Note that the employment rate is 1 – the unemployment rate. See the Annex for a more precise definition.
Focusing on the time period since the Financial Crisis, shows that while the labour share increased in Sweden, it fell in both Denmark and Finland. Also, declining hours worked per employee contribute largely to the developments of real labour earnings in Denmark and Finland while playing a minor role for Sweden. The same applies for the employment rate, c.f. Table 2.
Table 2. Average annual growth rates of real earnings per capita and its decomposition for different periods of time.
The developments shown in Figure 1 and Table 1 show that the effects of the Financial Crisis were less severe for the Swedish economy and that the Swedish economy recovered faster than the Danish and Finnish. The decomposition showed that the main difference between Sweden on the one hand and the two other countries on the other hand, are to be found in the variables “Hours per worker” and the “Employment rate”. These variables are cyclical and reflect variations in aggregate demand.
It appears that the Swedish economy was both more resilient and more able to recover from the slump. The three countries all have different exchange rate regimes which means that there also are differences in monetary policies. Sweden is outside the euro area and has a flexible exchange rate which means that the Riksbank can pursue its own monetary policy. Monetary policy in Denmark and Finland is more restricted. Finland has the same monetary policy as the euro area. Denmark conducts a fixed exchange rate against the euro. The primary aim of Danish monetary policy is to keep the value of the Danish currency stable against the euro.
The extent to which the differences in exchange rate regimes between the three countries determined the differences in impacts and resiliencies across Denmark, Finland and Sweden, is a complex question to analyse. An analysis of this issue requires that other factors should be considered. That is beyond the ambitions of this blog. The ramblings in this entry only scratch on the surface and suggest that the above mentioned differences in policy could play a role for the recoveries, c.f. Figure 3.
Figure 3. Years to recovery from the Financial Crisis
A final remark on an observation from Table 2 is that even though demographics do not appear to play a big part in explaining the differences between the three countries, those kinds of developments may have an impact on the economies’ potential output in the future.
References
Hall, R. E., (2017). "Sources and Mechanisms of Stagnation and Impaired Growth in Advanced Economies" In European Central Bank, ECB Forum on Central Banking - Investment and growth in advanced economies, conference proceedings. https://www.ecb.europa.eu/pub/pdf/other/ecb.ecbforumcentralbanking2017.en.pdf
Annex
Where w = W/P, Y = Y/P , and P is the GDP deflator. wH/Population is total labour earnings. wH/Y is the labour share, Y/H is output per hours worked, H/L is hours worked per employee, L/LF is the employment rate (the number of employed relative to the number of people in the labour force, LF), LF/Pop15-64 is labour force rate (the percent of the population in the working ages that are ready to work) and Pop15-64/Population is the working age population rate (number of people in the ages 15-64 over the total number of people in the population).
[1] As Hall (2017) explains, the decomposition is definitional and not intended to be interpreted as that variations in real labour earnings were caused by variations in one of its components, e.g. hours worked per employee. It is more appropriate to say that factors that caused hours worked to change also changed real labour earnings. [2] Hall (2017) further decomposes output per hour in into TFP and the capital-output ratio.