Oz Blog News Commentary

A response to Judith Sloan on monopsony

July 14, 2014 - 07:46 -- Admin

Ross Gittins’ column in Saturday’s Sydney Morning Herald was devoted to the topic of minimum wages. Among other things, he explained the ‘dynamic monopsony’ model of labour markets. These models are associated in particular with Alan Manning of the London School of Economics, who set out the theory at length in his 2003 book Monopsony in Motion: Imperfect Competition in Labor Markets.

It appears that Judith Sloan isn’t a fan of dynamic monopsony. She says:

Gittins also quotes the laughable dated work of Manning et al about monopsony bei[n]g associated with a link between higher employment and higher wages.  The fatal problem for Manning’s work (which is now ignored by the labour economics profession) is that the prevalence of monopsonies (single employer, many workers) is extremely low.  (Perhaps some regional towns and one big employer, but even here, employment tends to be spread across a number of employers).

And as for the argument about transaction costs, this is just theory.  In the case of Australia, there are no information problems because low paid workers have regulated pay and information is readily available.

I was surprised to hear that the dynamic monopsony model developed by Manning and others “is now ignored by the labour economics profession.” The latest edition of the Handbook of Labor Economics (released in 2011) includes a 68-page article by Manning titled “Imperfect Competition in the Labor Market” that sets out the theory in detail. There’s a working paper version at the LSE’s website. The paper concludes:

One’s views of the likely effects of labour market regulation should be substantially altered once one recognizes the existence of imperfect competition. All labor economists should take imperfect competition seriously.

I would not expect a theory ignored by the labour economics profession to be featured in the Handbook of Labor Economics. I also would not expect to see a whole issue of the Journal of Labor Economics devoted to an ignored theory, yet that is what occurred in 2010. The introduction to that issue of the journal, by Orley Ashenfelter and Henry Farber (both of Princeton) and Michael R. Ransom (from Brigham Young University), opens by saying:

There has been renewed interest in monopsony in labour markets that includes both the traditional static approach to monopsony… and the ‘new’ approach to monopsony with more attention paid to dynamic issues, developed in detail by Manning (2003). 

The introductory piece concludes by saying:

The papers in this volume provide remarkable evidence that labor markets are far from competitive. The evidence comes from a variety of countries and labor markets using a variety of econometric techniques and models. 

The notion that search frictions in labour markets can render them imperfectly competitive isn’t, in my understanding, a fringe theory in the economics profession. Diamond, Mortensen and Pissarides won the 2010 Nobel Prize in Economics for their work in developing models of markets with search frictions. Here’s an extract from a summary of the work that earned them the prize:

Burdett and Mortensen (1998)… developed a model with monopsonistic wage competition in an economy with search frictions… Reservation wage heterogeneity creates a tradeoff for firms between “volume” and “margin”: high-wage firms are able to attract and retain more workers than low-wage firms are, but the rent per worker that high-wage firms can extract is relatively low. As in traditional models of monopsony, an appropriately set minimum wage can increase employment and welfare.

Are these models of monopsonistic labour markets relevant to Australia? Alison Booth and Pamela Katic of ANU used data from the HILDA survey to try to answer that question. They conclude:

The essence of monopsonistically competitive or oligopsonistic labour markets is that labour supply to a firm is imperfectly elastic with respect to the wage rate. The intuition is that, where workers have heterogeneous preferences or face mobility costs, firms can offer lower wages without immediately losing their workforce. This is in contrast to the perfectly competitive extreme, in which the elasticity is infinite. Therefore a simple test of whether labour markets are perfectly or imperfectly competitive involves estimating the wage elasticity of the labour supply to a firm. We found that the Australian wage elasticity of labour supply is around 0.71, only slightly smaller than the figure of 0.75 reported by Manning (2003 for the UK)… 

These estimates of the wage elasticity of labour supply to a firm are so far from the perfectly competitive prediction of an infinite elasticity that it would be difficult to make a case that labour markets are perfectly competitive.

Alison Booth has another paper in the forthcoming 20th anniversary edition of the Labour Economics journal (working paper version here)  in which she discusses the evolution of labour economists’ view of wage setting. She notes:

It would be fair to say that even a quarter of a century ago many economists viewed the labour market as intrinsically perfectly competitive…

Perhaps the most interest development in wage determination theories of the past decade or so has been the realisation that employers have some market power in wage setting. This is not only a plausible and reasonably tractable characterisation of the labour market, but it can also help explain certain labour market phenomena. 

Professor Sloan says dynamic monopsony theory is “laughable,” “dated” and “ignored.” I am not a professor, but my reading of the literature referred to above does not support these assertions.