Financial Times (London, England), August 26, 2005, Friday

Copyright 2005 The Financial Times Limited
Financial Times (London, England)

August 26, 2005 Friday
London Edition 1

SECTION: LETTERS TO THE EDITOR; Pg. 14

HEADLINE: No escaping it: unions seem to be a kind of anti-fertiliser

BYLINE: By ANDREW OSWALD

BODY:
From Prof Andrew Oswald.
Sir, Ian Brinkley of the Trades Union Congress (Letters, August 25) has written a brilliant reply to my letter of August 23. Because it cannot dispose of the key evidence, however, it is ultimately not persuasive.
Research stretching over 20 years, done by many independent researchers, using data on thousands of randomly chosen workplaces, in countries such as Britain, the US, Canada and Australia, and published in peer-reviewed journals, seems to be producing essentially a common finding. Once we factor out other influences, workplaces with unions grow less quickly. Unions appear to be a kind of anti-fertiliser.
First, Mr Brinkley asks whether my summary isSee, for instance, the new 2006 edition of the best-selling Modern Labor Economics by Ronald Ehrenberg and Robert Smith of Cornell University. Roughly speaking, the authors argue that unions reduce job growth by between 2 and 4 per cent a year.
accurate. In fact, it is a statement of the latest textbooks. Second, is the relationship causal? Unfortunately we cannot do giant laboratory experiments, and, although the pages of this newspaper are not the place, there is certainly room for technical debate. Nevertheless, given existing knowledge in 2005, it appears to be. And the referees of the journal papers apparently believed so.
Third, is there other evidence? There is indeed a literature on the deleterious macroeconomic effects of unionism - say, upon the level of unemployment. Much of this has emerged close by, from the London School of Economics, as Mr Brinkley can check.
Fourth, does it predict the real world? Some may feel we do not need equations to see that it is union-sector activities that steadily disappear around us.
Fifth, I agree that trade unions lead, in many instances, to greater equality inside workplaces, and that this can be viewed as an important plus from unionism. Yet the main reason we know this is because of research based on the same statistical methods as suggest that unions harm job growth.
My own doctorate, in the late 1970s, was a study of trade unions, when less was known; I was open-minded about their effects. But, of course, we must move with the winds of evidence.
Andrew Oswald,
Professor of Economics,
University of Warwick,
Coventry CV3 7AL