September 4, 2020, Ayse İmrohoroğlu ve Samed Küçükikiz
Unemployment rates have become an incomplete measure of the health of an economy during the COVID-19 pandemic. In the following two graphs we summarize the time path of the GDP index (normalized to 100 in 2015) for Turkey, Germany and the U.S. and their unemployment rates. Sharp declines in output in 2020 are evident for all three countries. In fact, between the second quarter of 2019 and the second quarter of 2020, output declined by 11.3% in Germany, 9.9% in Turkey, and 9.1% in the U.S. Meanwhile, the unemployment rate graph tells a very different story. In the U.S., while the unemployment rate almost triples during the same period, it hardly changes in Germany and Turkey. A big reason for the varied response in unemployment rates across countries is due to the differences in polices enacted during the pandemic in different countries. While the U.S. relied mostly on providing additional unemployment insurance to those who lost their jobs, many European countries as well as Germany and Turkey have provided subsidies to companies to maintain their workforces in short term work schemes.
During the pandemic, Turkey enacted several measures including unpaid leave, ban on layoffs, cash payments as well as a short-time work allowance (SWA). SWA, that provides additional income support to compensate for the wage loss due to reduced hours by firms, is the largest component of these measures. Between April and July in 2020, 3.2 to 1.7 million workers, respectively, were covered by SWA. Similar measures are used by many other countries. In Germany, about half a million businesses have implemented SWA. We used the following links for the data on SWAs in Turkey and Germany (Note: Column D of Tab-04 in the excel sheet for August 2020). IMF provides a summary of measures used across countries.
The following graph displays the number of workers receiving short-time allowance in Germany and Turkey in the first seven months of 2020. In Germany in April 2020, the number of workers covered under SWA had reached over 8 million people. In Turkey, more that 3 million workers received SWA support in April. By July the number of people covered under SWA declined to about 254 thousand people in Germany while they remain to be high in Turkey.
In the following two graphs we provide an estimate for the unemployment rate and the employment rate where we classify all workers covered by SWA as unemployed (dotted lines indicate the SWA correction). This correction increases the unemployment rate from 4.3% in April 2020 to 22.5% in Germany and from 12.8% to 23.9% in Turkey. Similarly April employment numbers decrease from 42,090 million to 34,065 million in Germany and from 25,614 million to 22,370 million in Turkey.
Given that only the workers in the formal sector are able to benefit from measures such as SWA, we are likely to see big differences in the labor market response across these sectors. Indeed, there are marked differences in the decline in the number of people employed in the formal and informal sectors during the pandemic. In December 2019, of the 23,088 million workers (not including those in the agricultural sector) in Turkey, 5,061 million were in the informal sector and 18,027 million in the formal sector. By May 2020, employment had declined by 3.8% in the formal sector and by 27.8% in the informal sector.
Overall, reported unemployment or employment rates have not been a useful description of the labor market conditions during the pandemic in many counties. Correcting for only one of the measures taken during this time by classifying the workers covered by the short-term work allowance as unemployed generates significantly higher unemployment rates for Turkey and Germany. Of course, more careful analysis should be done to correctly classify different workers into the right categories.
