Latvia's sticky unemployment puzzle

Anders Aslund has pointed out during the Friday's conference that the only great concern of Latvia's government is unemployment rate steady at about 16%. Many observers have recently pointed to a high and sticky unemployment rate in Latvia as well (Paul Krugman, Morten Hansen, Edward Hugh, Ambrose Evans-Pritchard, to name just a few), thus raising fears of jobless recovery (economic growth which does not bring too many new jobs). In fact, however, Latvia's unemployment is declining (and employment is increasing) quite fast, but there are some reasons why the former has not been reflected in the official statistics yet. So, let me dispel the view of jobless recovery with just three arguments.

1. Temporary public works effect

Expansion of the temporary public works (mainly under the label of the WWS (100-lats) program (see Hazans (2012) for details) was the main active labour market policy measure during the crisis. The WWS was introduced at the end of 2009 and was in effect for over two years. As from the end of 2011, the number of participants in temporary public works has decreased sharply (see Figure 1).

Figure 1. Participants of temporary public works, % of economically active population


Source: Author's calculations based on Central Statistical Bureau of Latvia data

Since public work participants are treated as employed in the Labour Force Survey (LFS, the source of the headline unemployment figure), the implementation of the WWS definitely decreased the unemployment rate. However, the termination of the WWS has mitigated the unemployment decline (see Figure 2).

Figure 2. Unemployment rate, adjusted for public works participation, % of economically active population

Source: Author's calculations based on Central Statistical Bureau of Latvia data

The headline unemployment rate posted a decrease by just 1.0 percentage point over the last four quarters. Receiving almost twice less than the minimum wage, one would not participate in the WWS if he/she were able to find a job. Thus, in the absence of the WWS, the unemployment rate over the last four quarters would have likely decreased by a remarkable 2.8 percentage points. 

2. Participation rate hike

The LFS unemployment definition includes only those people who have actively searched for a job (i.e., the job seekers' rate). There are many other people (just "sitting without a job") that have never been counted as unemployed, but were instead treated as economically inactive. Given the working age population and the number of jobs, more people willing to find a job (among those who are not working) also means a higher headline unemployment rate. Thus, Latvia's unemployment rate would certainly be lower without a rapid increase in the participation rate (see Figure 3).

Figure 3. Participation rate, % of population aged 15-74 

Source: Central Statistical Bureau of Latvia data

The number of discouraged workers fell particularly fast (by 25% y-o-y). Proponents of the unemployment hysteresis view may feel uncomfortable since the number of discouraged workers in Q2 2012 was the lowest over the period of 3.5 years. The 3 percentage point rise in the participation rate may have raised the unemployment rate by a roughly similar amount. Assuming that the employment is driven solely by labour demand, the unemployment rate would be pretty close to its historical average at about 12% (see Figure 4).

Figure 4. Unemployment: headline and assuming constant participation rate, % of economically active population

Source: Author's calculations based on Central Statistical Bureau of Latvia data

The actual impact from these two effects is likely to be smaller than graphed above (in a real world, the number of jobs is not determined solely by the labour demand even in a loose labour market, and some public work participants may have found a job elsewhere or would not search for a job at all). However, the magnitude of these effects is so large that it seems even 50% transmission is able to dispel the "sticky unemployment" puzzle.

3. Employment growth

The unemployment decline would be meaningless if not accompanied by job creation. In Latvia, the rise in employment and unemployment slowdown go hand-in-hand. According to the most recent Eurostat data (for Q2 2012), Latvia holds the third place in the EU regarding employment growth (see Figure 5).

Figure 5. Employment y-o-y growth in 2012 Q2, %

Source: Eurostat data

And it's not because a year ago employment was atypically low: Latvia recorded the third highest employment growth also in Q2 2011. Even if we agree with Morten Hansen that national accounts data (that Eurostat figures are based on) may be debated, the pace of Latvia's employment growth and its relative position compared to other EU countries cannot be disputed. In fact, Eurostat figures posted the lowest employment growth of all available statistical data sources. Here is a list of the employment y-o-y growth from various official sources: 

Table 1. Latvia's employment y-o-y growth in Q2 2012, %

Methodology Annual growth rate, % Source:
National accounts (persons) 1.9 Eurostat
Labour Force Survey (persons) 2.2 Central Statistical Bureau of Latvia
Business Survey (jobs) 2.3 Central Statistical Bureau of Latvia
Employees paying social security contributions (persons) 3.4 State Revenue Service
Business Survey (hours) 4.3 Central Statistical Bureau of Latvia

Latvia's moderate employment rate (just at par with the EU-12 average) is a consequence of a high participation rate combined with a high unemployment rate. The graph below shows the employment rate significantly exceeds the level recorded in South-European countries where high unemployment rates go hand in hand with low participation rates:

Figure 6. Employment rate in Q2 2012, age group 15-74 , %

Source: Eurostat data

Employees are working hard in Latvia (long hours, secondary jobs). Thus, Latvia's moderate employment rate does not mean there are fewer job opportunities than, for example, in Germany. The number of hours worked per capita even exceeds the level posted by countries with a high employment rate, particularly the Netherlands, Denmark, Germany and Sweden (the graph includes only those countries for which Eurostat data were available):

Figure 7. Hours worked per capita in Q2 2012

Source: Author's calculations based on Eurostat data

Given Latvia's labour market performance in the European context, the current employment growth is indeed fast and is hardly a consequence of merely a low base effect.

APA: Krasnopjorovs, O. (2024, 25. jun.). Latvia's sticky unemployment puzzle . Taken from https://www.macroeconomics.lv/node/2736
MLA: Krasnopjorovs, Oļegs. "Latvia's sticky unemployment puzzle " www.macroeconomics.lv. Tīmeklis. 25.06.2024. <https://www.macroeconomics.lv/node/2736>.

Restricted HTML