Can unemployment rate go down if jobs are lost? Explore this data story to learn more about labor market trends in Kosovo over the past 7 years.
The topic of unemployment rate is very often discussed in political discourse and media. There seems to be a vague understanding in the general public about how the rate of unemployment is calculated. Therefore, the purpose of this data story is to clarify how unemployment rate is calculated and how this magical number is tightly linked to the labor force participation.
Using the data obtained from Kosovo Agency of Statistics Labor Force Survey this data story shows the unemployment rate trends in Kosovo and relationships between labor force participation, number of jobs created or lost and unemployment rate.
In the first chart named “Jobs created or lost” it is shown how the rate of unemployment has changed in Kosovo for the period 2012-2018. Since 2012, Kosovo Agency of Statistics started using the Eurostat methodology in conducting its labor force survey. From 2012 until 2015 the labor force survey was conducted in annual basis then from Q1 2016 it was conducted in quarterly basis. This is the reason why the data are provided in yearly basis for the period 2012-2015 and on quarterly basis for the period 2016-2018.
The bar charts indicate the number of jobs lost or created for each respective period (year or quarter). The red line which is superimposed over the bar chart shows the trend of unemployment rate for the same period. If we focus on the year 2015 we notice that there were 26,568 jobs lost during this period (this means 26,568 more jobs existed in 2014 as compared to 2015); however, unemployment rate was reduced from 35.3% in 2014 to 32.9% in 2015. This is a counter intuitive results since if jobs were lost then we would be inclined to believe that unemployment rate would be increased.
In order to understand better this counter intuitive result please open the second chart named “Labor Force vs Unemployment” where I have added the data about the change in the labor force participation (change in number of persons that participate in labor force, who are actively looking for a job).
It can be noticed that during this period when unemployment rate dropped from 35.3% to 32.9% the number of people in labor force was also decreased drastically (57,535 persons left the labor force during 2014) Therefore, since the formula for calculating the unemployment rate is dependent on number of people in the labor force it means that the effect of 26,568 lost jobs was overshadowed by the effect of decrease in labor force by 57,535 persons which was 2.1 times higher than the number of jobs lost.
The opposite trend occurred during period 2015 to Q1 2016. In the first quarter (Q1) of 2016, 3733 new jobs were added to the economy and intuitively the unemployment rate fell from 32.9% in 2015 to 27.7% in Q1 2016, this is an expected result indeed. However, the decline in unemployment rate during this period was very sharp (by 5.2 percentage points) even though there were only 3733 new jobs created this during Q1 2016, the unemployment rate fell much faster by 5.2 percentage points because the labor force also decreased by 26,953 persons during this period which contributed to an even sharper decline in unemployment rate.
The aim of this graph is to show that unemployment rate is very sensitive to changes in labor force participation and just because we see in the news that unemployment rate has declined doesn’t mean that new jobs were created in fact it can be the opposite, jobs can be lost and unemployment rate may still go down if labor force participation also changes. That is why the media should focus more on the reported number of jobs created or lost during each quarter rather than unemployment rate alone.