21 June 2011

Expectations, Employment and Prices

Roger Farmer
2010, Oxford University Press, 189 pages, £25.00
ISBN 9780195397901

Reviewer: Diane Coyle, Enlightenment Economics

This is a more technical book than is usually reviewed in this journal, indeed more technical than I usually read these days. The reason for this is that some months ago I read the popular version of Farmer’s work, How the Economy Works (OUP, £14.99), much of which seemed very sensible. But it ended with the argument that monetary authorities should target a stock-market index, which seemed very much not sensible. Would that not just lead to overly-volatile policy, given the overshooting of rational asset-markets, never mind the scope for non-rational bubbles? Taking account of asset prices as a supplementary indicator for monetary policy is one thing; a target is quite another.

That review is online at


Still, there was enough there to persuade me to take a look at the theoretical work underpinning the argument. For those interested in the state of our understanding of the macroeconomy and monetary policy, this book is well worth working through. (The equations are pretty straightforward differential calculus of the kind familiar from masters and PhD courses.) There are two important contributions. One is to incorporate in a macro framework the kind of search model of the labour market for which Chris Pissarides along with Peter Diamond and Dale Mortenson won the economics Nobel Prize in 2010. The other is to incorporate wealth effects (which is the source of the recommendation to target asset prices).

Both of these seem giant strides on the way to a more coherent framework for macromodelling. After all, it is plain that conventional DSGE models have failed utterly in the past decade and we do not yet have a new consensus about what to put in their place. Farmer is at great pains to place himself in the tradition of Keynes’s General Theory and the spiritual debt is obvious, but I think his approach is acceptable to all but the most committed real business-cycle, Ricardian-equivalence merchants, given the emphasis he places on microfoundations.

Having said all this, I do have some quibbles. Mostly these concern the chapter on business-cycle data, which plunges into presenting real GDP in terms of wage units without ever explaining what this means or why it is preferable to conventional statistical approaches. In my PhD work I specialized in time-series econometrics and although this was a long time ago, I thought this confrontation of theory with facts was weak and difficult to follow.

In addition, this second book did not persuade me either about the sense of targeting the stock market. When I worked in the Treasury in the mid-1980s (the era of monetary targeting), the monthly briefing for the interest rate meeting included many of the indicators the Monetary Policy Committee looks at now, but added a range of asset prices and quantities. For example, agricultural land prices, house prices, the stock market, and activity in the corporate debt market, were all included. As inflationary pressure can emerge in any of a number of asset markets, I think this dashboard approach has much to recommend it, rather than a formal target for a single index.