Desemprego e segurança no emprego

Employment protection and jobs, por Chris Dillow:

(...) Empirical work has struggled to find a link between employment protection laws and unemployment. For example, the OECD has found (pdf) that:

There appears to be little or no association between employment protection legislation strictness and overall unemployment

although it did find that such laws raise long-term unemployment.

And Stephen Nickell found (pdf) that employment protection laws were "negligible and completely insignificant” as a cause of cross-country differences in unemployment in 1980s and early 90s.

So, why is Tim’s intuition at odds with the facts?

It’s because there are a couple of mechanisms operating in the other direction.

If there were no employment protection at all, workers would gain less from staying with the same firm for a long time. Ceteris paribus, they would therefore have more incentive to leave. This would have two effects:

1. To retain staff, employers might have to pay more. This would tend to reduce employment. To put this in a way Tim would prefer, the incidence of employment protection laws falls upon workers in the form of lower wages.

2. Workers would have less incentive to invest in firm-specific skills. This would tend to reduce the firm’s efficiency and hence prevent it growing. (...)

Because of these offsetting mechanisms, some research (pdf) has found “an inverse U relationship between employment protection and economic growth”:

From the point of view of per capita growth of GDP…employment protection in the early 1990s was too low in countries like Australia, Canada, Denmark, Ireland, Switzerland, the UK and the US

Diga-se que o último estudo referido por Dillow (Welfare improving employment protection, por Michele Belot, Jan Booneyand e Jan van Oursz) não faz qualquer referência a Portugal (de forma que não sabemos se eles nos incluiriam nos que têm protecção no emprego de mais ou de menos).