Matching Workers with Jobs: How Well Is the New Zealand Labour Market Doing?
Craigie, Rebecca, Gillmore, David, Groshenny, Nicolas, The Reserve Bank of New Zealand Bulletin
New Zealand's unemployment rate rose sharply from the beginning of 2008 to mid-2009 and has remained high ever since. A persistently high unemployment rate in isolation suggests substantial slack in the labour market and in the economy as a whole. However, that looks inconsistent with some other labour market indicators. Job advertisements, reported skill shortages, and wage growth suggest that the excess capacity that built up during the recession has dissipated gradually over the subsequent three years. Those indicators suggest less downward pressure on inflation than might be implied by the unemployment rate alone.
How to reconcile the high number of people looking for work with the other labour market indicators is the focus of this article. We use a couple of analytical techniques to try to shed further light on the issue. We look specifically at how the labour market is doing in matching workers with available jobs, and how that may have changed in recent years. Finally, we offer some thoughts on what may have accounted for the apparent change.
2 Looking at the labour market data
We first look briefly at some of these other labour market indicators. For example, the Quarterly Survey of Business Opinion's (QSBO) measure of how easy it is to And labour has fallen sharply since 2009, and is now around long-run average levels (figure 1). (1) This is consistent with feedback received during recent Reserve Bank business visits, suggesting that employers are finding it more difficult to fill vacancies for skilled staff. Figure 1 shows how this QSBO measure has historically been moving closely with our estimate of the unemployment rate gap (the deviation of the unemployment rate from its estimated trend rate). However, since 2010 the fall in the QSBO measure has been inconsistent with the persistently high unemployment rate.
[FIGURE 1 OMITTED]
Additionally, at least on some measures, wage inflation in the private sector appears at odds with a view of considerable excess capacity. After rising to quite high levels during the expansion of the previous decade, wage inflation fell very sharply during the 2008/09 recession (figure 2). The recovery in LCI wage inflation since then appears consistent with at least some reduction in the degree of excess capacity in the labour market.
[FIGURE 2 OMITTED]
So what explains this divergence between the signals from the unemployment rate itself and those from some of the other labour market indicators? One theory that might help explain this divergence is a decline in how well the labour market is doing in matching job seekers with vacant jobs--that is, a decline in matching efficiency.
The possible importance of such a mismatch issue is currently being discussed in several advanced economies, particularly the United States. Several policy makers have related the persistently high rate of unemployment in the United States to an increase in both sectoral mismatch (a shortage of workers in some industries at the same time as unemployment of workers with different skills from other industries) and geographical mismatch (shortages and unemployment across different regions or states). (2) A series of studies suggests that matching efficiency in the United States has declined since the 2008/09 recession. (3) Some studies have estimated that mismatch accounts for 1 to 1.5 percentage points of the increase in the unemployment rate. (4)
We use two methods to infer the evolution of matching efficiency in New Zealand: firstly, the Beveridge curve; and secondly, an estimated measure of matching efficiency based on a standard model of labour market flows.
3 The Beveridge Curve
Looking at the relationship through time between the number of vacant jobs and the number of people unemployed is one way into the issue. Even doing that has its challenges. Job vacancies are generally proxied by the number of jobs advertised. …