European Views: Missing wage growth and ECB policy

Euro area wage growth has been unusually weak in recent years.Despite falls in the unemployment rate, wage growth has not picked up as its historical link with unemployment  would have suggested.

A structural change in the labor market. 

Missing wage growth has been especially marked in Germany – where the fall in unemployment has been pronounced since 2012. An unemployment rate of 3.6% in 2017Q4 would have implied annual pay growth of 5.0%yoy based on the historical  link between wage growth and unemployment. With German wage growth currently running at around 2.5%yoy, this implies around 2½pp of ‘missing’ annual wage growth.

Labour market developments have been highlighted in recent ECB communication.

Speaking last month, ECB President Draghi noted that increased supply capacity could become available as the Euro area recovery continued. ECB Executive Board member Praet recently noted: “We are continuously surprised by the response of the labor market to increasing aggregate demand because rising demand has been leading to continuous positive supply reactions in the labour market… there were reservoirs of labor. It could be migration.” Bundesbank President Weidmann expressed a similar view in an important speech in January.

We illustrate how missing wage growth could be explained by a relatively persistent reduction in worker bargaining power. We believe that our model simulation captures several key features of what we label Europe’s more ‘contestable’ labor market. The model frames the rationale for our thinking and our call on ECB policy, which remains slightly dovish relatively to market pricing. Our results here reinforce that existing view.

High dismissal costs and levels of employment protection, quite significant levels of unemployment insurance and influential labor unions remain key features of Europe’s labor market. These structural features mean that job search costs and the costs of matching the unemployed with job vacancies are significant; wage-setting is quite rigid. A realistic model of the Euro area labor market needs to take account of these features.

An important feature of the model is that it embodies a ‘wage channel’ whereby weak wage growth lowers businesses’ marginal cost and inflation directly. This important feature contrasts with most search and matching models of the labour market where wage bargaining divides the potential profit between employees and their employers, rather than by affecting marginal cost and inflation directly.

The stronger link between wages and inflation (and output) that results is consistent with the ECB paying close attention to wage data in assessing the outlook for inflation.

We intended to capture the broad macro effects of a persistent reduction in worker bargaining power. We therefore compare the effects of a standardised and persistent shock to worker bargaining power with a less persistent reduction in worker bargaining power, typical of the past.

Rather than simply assess the impact of a ‘bigger’ shock, we instead allow the structural changes to have had a more persistent effect. We believe this is consistent with the ending of transitional migration controls (and our broader ‘Mitteleuropa’ view of deeper integration between Germany and CEE) being a structural rather than a cyclical change.

The results. We believe underlying changes in European labor markets have raised the persistence of a shock to worker bargaining power. We therefore compare the macro effects of a reduction in worker bargaining power (specifically, a 1 standard deviation reduction) under standard persistence with their effects when the reduction in worker bargaining power persists for longer.

For the given shock we consider, the increased persistence implies a fall in real wages that is between two and three times larger than the same shock with normal persistence, despite the labor market rigidities . What of inflation? Annualised inflation also declines around twice as much when the reduction in worker bargaining power persists .This captures the weaker ‘wage channel’ when workers’ bargaining power is persistently lower.

The material effect on inflation coming from a more persistent drop in worker bargaining power leads to lower policy interest rates which further boost output and account for the higher level of employment. The model includes a policy rule for interest rates so that lower inflation induces easier policy than otherwise and that supports GDP growth.

The simple change we make in our model simulation has quite important effects, including for ECB policy. Assuming that ECB policy rates follow a Taylor-rule , then the policy rate is ¼ pp lower than otherwise for 2 years. This owes to the ECB being confronted with a more subdued inflation outlook. Our simulation is largely illustrative. Yet, in practice we expect the ECB to base its policy decisions on the inflation outlook as its forward guidance currently emphasises. We believe our existing ECB calltakes these effects into account.

The message to take from our model-based exercise is that a more persistent negative shock to wage bargaining lowers the inflation profile significantly relative to previous experience. Associated with a persistent reduction in worker bargaining power would be weaker wage growth and inflation, despite  lower unemployment. Wage growth would therefore appear subdued relative to unemployment. The motivation for our analysis – a structural change that has made the Euro area labour market more contestable – is consistent with recent emphasis on such issues among ECB policy-makers. As such, our analysis illustrates some of the intuition expressed by Mr Draghi, Mr Praet and Mr Weidmann in their recent discussions of how ECB policy could depend on key labour market development