Churning and Labor Productivity in Economic Crisis, Differences between Foreign and Domestic Firms

ABSTRACT

Our analysis of matched employee-employer data from Estonian firms (years 2006–2013) shows that an increase in labor churning is related to a positive change in labor productivity during an economic crisis. During boom years, churning is related to a negative change in labor productivity. Only in services during the crisis did foreign firms have a stronger positive relationship between labor churning and labor productivity changes than domestic firms. However, our analysis at the individual level does not confirm that, during a crisis, foreign firms in services hire more employees with characteristics that have been found to be related to productivity increases.

KEYWORDS:

• Labor productivity
• labor churning
• economic recession/crisis
• Estonia

JEL CLASSIFICATION:

• J24
• J63
• L60
• L80
• M51

Acknowledge

The authors acknowledge Statistics Estonia for providing the data. The authors also acknowledge support for the compilation of the datasets used in the paper from the Estonian Research Infrastructures Roadmap project “Infotechnological Mobility Observatory (IMO).” Jaan Masso and Urmas Varblane acknowledge support from the Estonian Research Council’s research project PRG791 Innovation Complementarities and Productivity Growth. Jaan Masso also acknowledges financial support from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 822781 GROWINPRO – Growth Welfare Innovation Productivity. The authors are also thankful to Priit Vahter, Karsten Staehr and Ari Kokko as well as seminar participants in Tartu and Viljandi for their invaluable comments.

Disclosure Statement

No potential conflict of interest was reported by the author.

Notes

1. See section 2 for elaborate definition of churning.

2. Other topics related to churning have been covered. For example, the role of R&D worker mobility on innovation has been investigated with churning as a measure for worker mobility Müller and Peters (2010). In Norway, in firms with many workers from the new EU countries, excess churning has been found to partly be responsible for the overrepresentation of labor migrants from Eastern and Central European countries in the unemployment insurance programme (Bratsberg, Raaum, and Røed 2020). Ilmakunnas and Maliranta (2005) have used plant and worker characteristics to explain churning.

3. The rather large share of foreign-owned firms in Estonia makes the crisis years in Estonia a perfect example for analyzing the differences between foreign and domestic firms. In 2013, the stock of FDI in Estonia reached 85% of GDP, one of the three highest shares in CEE countries (Jude and Silaghi 2016). In 2017, according to the OECD (2017), foreign-owned firms directly supported 38% of private-sector jobs in Estonia and 41% of value-added.

4. Foreign firms in this paper are defined as firms where the sum of shares held by Estonian shareholders is less than 50% to ensure that the foreign owners fully control the company.

5. In the current paper, we have chosen to use labor productivity growth as a productivity measure instead of total factor productivity growth. According to Sargent and Rodriguez (2001), labor productivity is the better indicator of the two if the analysis includes shorter periods (less than ten years), and we have seven years in the whole sample. The authors also indicate that the changes in the capital capacity utilization usually cannot be adjusted for and this results in a procyclical bias in TFP. The current paper aims to compare different business cycle phases and such a bias might lead us to the wrong conclusions.

6. Strict labor market regulations may, to an important extent, negatively affect the efficiency of labor allocations (Haltiwanger, Scarpetta, and Schweiger 2008). In Estonia, the deregulation of the employment protection legislation in 2009 made its strictness formally comparable to other Central European countries (Brixiova 2009). However, the legislation was often also disregarded before 2009 (Eamets and Masso 2004). In addition, comparisons of legislation of different countries usually fail to account for the differences in other factors e.g. the strength of trade unions and the labor market in Estonia is actually considered flexible at company level (Eamets 2013).

7. As, based on the decline in GDP, the years 2008–⁠2009 already represent a crisis in Estonia, the choice of periods refers to a churning variable that is a lagged variable.

8. We note that the actual regional allocation of the economic activities may differ from that as companies may operate in a region different from their place of registration and some companies may operate in more than one region, especially in certain industries like retail.

9. More detailed results on matching are available from the authors upon request.

Additional information

Funding

This work was supported by the Estonian Research Council’s research project Innovation Complementarities and Productivity Growth. [PRG791]; European Union’s Horizon 2020 research and innovation programme under grant agreement GROWINPRO [822781]; the Estonian Research Infrastructures Roadmap project “Infotechnological Mobility Observatory (IMO).”

Notes on contributors

Liis Roosaar

Liis Roosaar is currently a junior research fellow at School of Economics and Business Administration, University of Tartu.

Urmas Varblane

Urmas Varblane is currently a professor at School of Economics and Business Administration, University of Tartu.

Jaan Masso

Jaan Masso is currently an associate professor at School of Economics and Business Administration, University of Tartu.