Measuring the effect of matching problems on unemployment

• Author: Ante FARM
• Date: 01 June 2020
• Published date: 01 June 2020
• DOI: http://doi.org/10.1111/ilr.12095

Copyright © The author 2020

Journal compilation © International Labour Organization 2020

International Labour Review, Vol. 159 (2020), No. 2

Measuring the effect

of matching problems

on unemployment

Ante FARM*

Abstract. This article shows how matching problems reduce employment gures

– and hence also raise those for unemployment – by creating a gap between labour

demand and employment. It also shows how this gap can be measured by unlled

jobs (unmet demand) as distinct from job vacancies (recruitment processes) and

reports results from the Swedish vacancy survey which measures both. In fact,

while a shift of the matching function indicating longer recruitment times suggests

increasing matching problems, this can only be veried by measuring unlled jobs,

which also quanties the eect on unemployment.

Keywords: job vacancy, unemployment rate, Beveridge curve, matching function.

1. Introduction

Information on job vacancies has increasingly been used in economics to meas-

ure the inuence of friction in job matching on unemployment.1 But exactly how

can vacancy statistics be used to dene and measure friction and its impact on

unemployment?

If hires are instantaneous, there are no job vacancies as measured in va-

cancy surveys. Even in these cases, though, there can be friction in job matching

according to the dynamic theory of labour demand. This predicates that, when

hiring is instantaneous, a rm’s employment decisions depend on adjustment

costs and expectations of market conditions in the future.2 If, for example, a rm

* Swedish Institute for Social Research (SOFI), Stockholm University, email: ante.farm@so.

su.se. The author gratefully acknowledges the useful comments made by Jim Albrecht, Mahmood

Arai, Stig Blomskog, Per-Anders Edin, Nils Gottfries, Jan Johannesson, Matti Pohjola, Åsa Rosén,

Lena Schröder, Susan Vroman, and, in particular, Gerard van den Berg, Anders Björklund, Peter

Fredriksson, Eivind Homann, Bertil Holmlund, Tomas Korpi, Walter Korpi, Matthew Lindquist,

Martin Nybom, Mathias Schumann, Michael Tåhlin and Eskil Wadensjö.

Responsibility for opinions expressed in signed articles rests solely with their authors, and

publication does not constitute an endorsement by the ILO.

1 Seminal works on this subject include Abraham (198 3 and 1987), Pissarides (1986), Blanchard

and Diamond (1989), Jackman, Layard and Pissarides (1989), and Layard, Nickell and Jackman (1991).

2 See, for example, Nickell (19 86) or Hamermesh (1993, Chapter 6).

International Labour Review

244

expects a rise in the demand for its products to be only temporary, it will not

necessarily increase its employment due to costs of hiring and ring.

Not all hires are instantaneous, however. Accordingly, job vacancies (time-

consuming recruitment) occur and a large outward shift of the UV (or Beveridge)

curve – relating unemployment (U) to vacancies (V) over time – has often been

interpreted as a decline in search eectiveness with strong eects on unemploy-

ment.3 The stocks of vacancies and unemployment have even been interpreted

as inputs in a production process described by a matching function with the

ow of hires as its output.4 This hiring rate is assumed to aect unemployment

in terms of the probability of the unemployed nding a job.

A matching problem exists when rms struggle to nd workers at the same

time as workers struggle to nd jobs. The main focus of this article is on rms’

problems in nding workers, noting that deviations from instantaneous hires

can reduce employment – and hence also increase unemployment – by giving

rise to unlled jobs. We shall also see that hiring is much simpler than the con-

cept of a matching function suggests.

Firms create one type of vacancy (recruitment processes) in order to avoid

another type of vacancy (unlled jobs). Because of this ambiguity I will refer to

recruitment process vacancies as “job openings” (or openings) and unmet labour

demand vacancies as “unlled jobs”. The distinction is important in that only

unlled jobs add directly to unemployment by bringing the number of employ-

ees below the number of jobs.

In contrast, there are three types of hires, namely instantaneous hires

(e.g. recalls of former employees), hires preceded by job openings (recruitment

processes), and hires preceded by unlled jobs (unmet labour demand). How-

ever, in spite of recent advances in vacancy statistics,5 the basic facts regarding

these types of hires are still unknown in most countries, in particular the share

of instantaneous hires and the relation between job openings and unlled

jobs. This article aims to show not only how this information can be collected,

but also how it can be used to characterize the hiring process and estimate the

eect of matching problems on unemployment. For example, if the rate of un-

lled jobs is 2 per cent of employment, then mismatch adds to unemployment

by reducing employment by 2 per cent. The rate of unlled jobs is a measure of

the gap between labour demand (jobs) and employment at every point in time

and not only in a steady state.

3 For example: “In a world where economists have little certain knowledge, the shift of the

U/V curve provides us with vital clues to the sources of the rise in unemployment. Large shifts in-

dicate that a major part of the rise is due to changed behaviour of workers and employers in the

lling of vacancies.” (Layard, Nickell and Jackman, 1991, p. 220).

4 See Petrongolo and Pissarides (2001) for a precise denition of the matching function and

a survey of its role in equilibrium models of unemployment since the 1970s.

5 Today there are quarterly (or monthly) business surveys of vacancies not only in Australia

(since 198 3) and the Netherlands (since 1988), but also in, for example, Sweden (since July 20 00),

the United States (since December 2 000) and the United Kingdom (since September 2002). A pro-

cess of developing business surveys on vacancies in all European Union (EU) countries was initi-

ated by Eurostat in 2 003 and reinforced by Regulation 453/2008 in 200 8.