Labor Market Flows and Stocks over the Business Cycle: The Role of the Participation Margin

ABSTRACT

We investigate the cyclical properties of labor market flows in the Czech Republic and Poland using impulse responses from a VAR model with three net flows and an output gap. We apply the stock-flow framework to derive impulse responses of employment, unemployment and participation rates, and their decomposition into the contributions of net flows. We find that the role of flows from and into inactivity in explaining the cyclical properties of unemployment and employment rates is small, but is not negligible. The participation rate is driven by the countercyclical net flow from inactivity to unemployment.

KEYWORDS:

• Labor market flows
• employment
• unemployment
• participation
• vector autoregression

JEL Classification:

• E17
• E24
• E32
• J21
• J64

Acknowledgments

The authors would like to thank Jan Babecký, Jan Brůha, Robert Dixon, Michal Franta, Jaromír Gec, Martin Gürtler, Petr Král, and two anonymous referees for their comments and suggestions. The views expressed in this paper are those of the authors and not necessarily those of the Czech National Bank or the National Bank of Poland.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1. Elsby, Hobijn, and Sahin (2015) find that flows from and into inactivity account for around one third of unemployment fluctuations.

2. This is a plausible assumption given the data structure, where we match individuals in two adjacent quarterly datasets. We describe the data in the next section.

3. The X-12 seasonal adjustment method is used throughout the paper.

4. The countercyclical nu is also consistent with the added worker hypothesis, which assumes that secondary earners start seeking jobs in those households where household heads lose their jobs in a recession.

5. The same explanation based on composition effects is provided in Elsby, Hobijn, and Sahin (2015).

6. The ADF test rejects the null hypothesis of nonstationarity. In particular, the ADF statistics for the Czech sample are $\overline{ue}$ (−4.28), $\overline{en}$ (−3.83) and $\overline{nu}$ (−4.76). The 1% critical value is −3.52. For the Polish sample, the ADF statistics are $\overline{ue}$ (−3.98), $\overline{en}$ (−5.56) and $\overline{nu}$ (−5.71), while the 1% critical value is −3.53. The KPSS test does not reject the null hypothesis of stationarity in the Polish data, as the statistics are $\overline{ue}$ (0.10), $\overline{en}$ (0.60) and $\overline{nu}$ (0.10) and the 10% (1%) critical values are 0.35 (0.74). In the Czech data, the KPSS statistics are $\overline{ue}$ (0.25), $\overline{en}$ (0.99) and $\overline{nu}$ (0.16). The critical values are 0.35 at 10% and 0.74 at 1%, so the null of stationarity is rejected for $\overline{en}$ and not rejected for $\overline{ue}$ and $\overline{nu}$.

7. We associate cyclicality (countercyclicality) with positive (negative) impulse responses to the output gap shock in the first several periods after the shock. In Poland, the response of the participation rate reaches a maximum after 9 to 11 quarters. It is of similar size to the response immediately after the shock. As we will see in Figure 7, the second peak in the response is due to the net flow from employment to inactivity.

8. Labor market variables respond to a GDP shock with a delay due to rigidities, for example, high firing costs. According to the OECD, in 2019 the Czech Republic ranked first among the 37 OECD countries in terms of the strictness of dismissal regulations for regular workers, while Poland ranked 16^th (OECD 2020).

Additional information

Funding

This work was supported by the Czech National Bank [Research Project No. B4/17.].