Unemployment and COVID-19: an analysis of change in persistence

ABSTRACT

This paper examines the degree of persistence in monthly unemployment rates for a group of 24 European countries along with the global rate for the Euro area, the European Union, the G7 and the OECD countries. For this purpose, fractionally integrated methods are employed. Using data from January 2010 to November 2020, our results indicate that fractional integration is present in all countries examined, with the orders of integration of the series ranging in the (0, 1) interval. Comparing the data before COVID-19 with those including it, the significant time trend coefficient and the mean reverting property disappear in most cases when COVID-19 data are considered. This implies that governments should consider that, after the pandemic, shocks on the labour market will have permanent effects. Thus, policies should address unemployment accordingly. Our work, however, does not focus on the analysis of nonlinearities, what could provide a more complete understanding of the series behaviour.

KEYWORDS:

• Fractional integration
• long memory
• persistence
• unemployment rate

JEL CLASSIFICATION:

• C22
• E24
• J64
• O57

Acknowledgment

Luis A. Gil-Alana gratefully acknowledges financial support from the Ministerio de Economía, Industria y Competitividad’ (MINEIC), the Agencia Estatal de Investigación’ (AEI), Spain and `Fondo Europeo de Desarrollo Regional’ (FEDER).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ If u(t) in (1) is ARMA(p, q), x(t) is said to be a fractionally integrated ARMA (ARFIMA(p, d, q)) model.

² The sum of the AR coefficients in an AR(p) process is another well-known measure of persistence.

³ See Gil-Alana and Robinson (1997), Gil-Alana and Moreno (2012) and Abbritti et al. (2016) for applications involving Robinson’s (1994) tests.