The spatial analysis of the employment effect of the minimum wage in a recession : case of the UK 1999-2010

We assess the impact of the National Minimum Wage (NMW) on employment in the UK over the 1999-2010 period explicitly modelling the effect of the 2008 to 2010 recession. Identification is facilitated by using variation in the bite of the NMW across local labour markets with the use of the 'incremental differences-in-differences' estimator on the base of a panel data set. We explicitly take account of the spatial nature of local labour markets by using commuting patterns to augment our empirical model by spatial lags of the endogenous variable or spatial-autoregressive disturbances. LM tests indicate that spatial dependencies of the dependent variable and/or the error terms cannot be ruled out with a few exceptions. This is especially true for the full specified models. The results lead to the conclusion that the SEMP model should be preferred in the majority. Furthermore, we examine whether the definition of the geographical unit used for the analysis matters. Since the definition of what constitutes a 'local labour market' in Great Britain is still open to discussion, the analysis is undertaken at three different levels of geographical aggregation (140 Unitary Authorities and Counties; 406 Unitary Authorities and Districts; 138 "travel to work area", TTWA). We find slight differences, but overall there are very small positive or zero incremental employment effects of up ratings to the NMW in a year on year context. The years where there is a small positive effect historically the years when the up rating exceeded the Retail Price Index (RPI) rise in the cost of living and hence the up rating of the NMW was relatively generous. In contrast the underlying effect of the presence of the NMW is reflected in the Kaitz Index coefficient. This coefficient is nearly always negative and significant suggesting that the effective implementation of the NMW has an underlying negative impact on employment but that this is offset in a year on year context by the small positive effects of the NMW up ratings - especially when the up rating is above the level of the RPI. However, our measured marginal effects were consistently made smaller when we condition out for the presence and severity of the recession in the regional context.