Disingenuous economics

Today a new paper in the World Bank Research Observer by Banerjee, Hanna, Kreindler and Olken announced that there is “no systematic evidence” to support the view that social transfers discourage work. Under different circumstances I might be pleased by this, since I have long believed that claims about welfare-induced voluntary unemployment are mostly ideology-driven garbage. But I am not happy about this paper.

The reason is that the way in which it is written implies that the ‘welfare-induced laziness’ thesis is something conjured up by policymakers and members of the public from thin air. To quote:

despite these proven gains, policy-makers and even the public at large often express concerns about whether transfer programs discourage work.

The authors position themselves, as economists rigorously assessing evidence, against the puzzlingly misguided ‘policymakers’ and ‘public at large’. From this, one would think that economists do not hold such beliefs and have not contributed to them. But you’d be wrong. In fact, a paper published in the World Bank Economic Review in 2003 by Bertrand, Mullainathan and Miller (BMM) claimed that the South African old age pension led young, black men in pension-receiving households to reduce their labour supply. This had a significant impact in South African policy circles, leading a number of local social scientists to challenge the findings of the paper.

Posel, Fairburn and Lund (2006) argued that BMM didn’t adequately account for migration, which changed the conclusions. When I was a Master’s student at UCT in 2006 I wrote a term paper – using data kindly provided by Posel – arguing, further, that BMM’s result using cross-sectional data couldn’t possibly be robust to household formation dynamics:

The primary aim of this paper is demonstrating that, based on existing evidence, one can make at least as strong a case that the pension influences household formation, as that it affects labour supply.

I referred to the BMM paper as a case of ‘perverse priors’. A different approach to the same point by Ranchod (2006) examined the effect of losing a pensioner and his findings called into question the conclusion about labour supply of young men. Ardington, Case and Hosegood (2009) later used panel data to show that differences between households along with migration could explain the finding, rather than an actual reduction in labour supply. To quote them:

Specifically, we quantify the labor supply responses of prime-aged individuals to changes in the presence of pensioners, using longitudinal data collected in KwaZulu-Natal. Our ability to compare households and individuals before and after pension receipt, and pension loss, allows us to control for a host of unobservable household and individual characteristics that may determine labor market behavior. We find that large cash transfers to elderly South Africans lead to increased employment among prime-aged members of their households, a result that is masked in cross-sectional analysis by differences between pension and non-pension households. Pension receipt also influences where this employment takes place. We find large, significant effects on labor migration upon pension arrival.

So, in the South African case, it was well-regarded, US-based economists publishing in a World Bank journal in 2003 who gave credence to the idea that grant receipt led to idleness. Much effort was expended debunking those claims and preventing them from having harmful effects on progressive social policy. It is, therefore, rather outrageous that in 2017 another set of well-regarded, US-based economists publishing in a World Bank journal can pretend as if this never happened.

In my view, this kind of behaviour is characteristic of the failure of critical self-reflection, and taking of responsibility, in economics as a discipline. And it should serve as a cautionary tale to non-economists and economists alike.

Author: peripheralecon

Public sector economist, extra-mural academic

Leave a Reply

Your email address will not be published. Required fields are marked *