2 Comments

Thanks Matt, a very useful analysis. If I am reading your post correctly, then wage scarring on average is driven by those failing to re-attach to the labour market within 1 year. Presumably, those subject to a mass layoff during the GFC faced a labour market with high unemployment. My question: would you expect the same average scarring in response to a mass layoff if the labour market was buoyant (i.e. unemployment was low)?

Expand full comment

Good question. Evidence from Germany suggests that it is worse during downturns: https://www.aeaweb.org/articles?id=10.1257/aer.20200252 - but when it comes to Australia and New Zealand I think there is nothing available (yet).

It could go either way. Becoming unemployed during a downturn could lead to worse scars because of a "congestion externality" in labour markets - everyone is trying to find a new job while jobs are relatively scarce. This could work through match efficiency or bargaining power, but either way it is a clear way that scarring would be cyclically worse.

However, this does ignore the "information" embedded in job loss. If job loss during a downturn is seen as a weaker signal of "poor productivity", firms may be more willing to hire someone who was laid off during a recession than someone who was laid off in good times. As a result, the scar could end up being procyclical!

So theoretically it is ambiguous, while international evidence suggestes that scars are very countercyclical. Lets see what the Australiasian evidence ends up saying!

Expand full comment