"Minimum Wages and Firm Profitability"

With all of the recent discussion about the minimum wage (e.g.), I thought this paper was worth noting. It finds evidence that the minimum wage transfers income from owners to workers, i.e. that it reduces profit and increases wages, but it does not change the probability of a firm going out of business, and it does not reduce employment. Thus, this paper raises the possibility that an increase in the minimum wage reduces inequality without having much of an impact on aggregate activity or employment:

Minimum Wages and Firm Profitability, by Mirko Draca, Stephen Machin, and John Van Reenen, NBER WP 13996, May 2008[Open Link]: I. Introduction In debates on the economic impact of labour market regulation, much work has focused on minimum wages. Although standard economic theory unambiguously implies that wage floors raise the wages of the low paid and have a negative impact on employment (Borjas, 2004; Brown, 1999), the existing empirical literature is not so clear. Whilst many studies have shown that minimum wages significantly affect the structure of wages by increasing the relative wages of the low paid (e.g. DiNardo et al, 1996), empirical evidence on the effect on jobs is considerably more mixed (see the recent comprehensive review by Neumark and Wascher, 2007). Some studies have found the expected negative impact on employment[1], yet others have found no impact or, in occasional cases, a positive effect of minimum wages on jobs.[2]

In the light of this, one may wonder how firms are able to sustain the higher wage costs induced by the minimum wage. One possibility is that firms simply pass on higher wage costs to consumers in the form of price increases. However, there is scant evidence on this score (exceptions are Aaronson, 2001, and Aaronson and French, 2007).[3] An alternative is that the higher wage costs are not fully passed on to consumers and the minimum wage eats directly into profit margins.[4] Since there is a complete absence of any study directly examining the impact of minimum wages on firm profitability, this is the focus of this paper.

Our identification strategy uses variations in wages induced by the introduction of the national minimum wage (NMW) in the UK as a quasi-experiment to examine the impact of wage floors on firm profitability. The introduction occurred in 1999 after the election of the Labour government that ended seventeen years of Conservative administration. There is evidence that the NMW increased wages for the low paid, but had little impact on employment[5] and so this provides a ripe testing ground for looking at whether profitability changed. We use the fact that the intensity (or “bite”) of the NMW is higher for firms with many low paid workers relative to firms with fewer low paid workers in order to construct treatment and comparison groups. We then compare outcomes in terms of wages, profitability and firm exit and entry using difference in differences methods.

Our work does uncover a significant negative association between the minimum wage introduction and firm profitability. This association is robust across two very different panel data sources, namely a specialized UK data source on workers in residential care homes (a very low wage sector) and an economy-wide firm level database FAME (Financial Analysis Made Easy) that covers all registered firms in the UK.[6] In both data sets, firm profit margins fall in relatively low wage firms following the introduction of the minimum wage. These effects correspond to about a fifteen percent fall in profit margins for the average care home and an eight to eleven percent reduction in profit margins for the average affected firms in FAME.  ... Finally, we could not find any evidence that low wage firms were forced out of business by the higher wage costs resulting from the minimum wage. Our analysis of an industry level panel dataset suggested that there was some fall in net entry rates following the minimum wage, hinting at a longer run negative effect on the number of firms. These results were rather imprecise, however, and not significant at conventional levels.