VoxEU Column Frontiers of economic research

Economic policy and the New Century public discourse

Recent advances in economics have opened a gap between research and real-world applications. This essay – written for LaVoce’s 5th anniversary – discusses how sites like LaVoce are bridging the research-reality gap.

Economics is more relevant to policy-making than ever before, but you would never know it from the public discourse. The internet is helping bridge the gap, and www.laVoce.info has led the charge. But how will the internet’s role in the public discourse on economics change in the future?

More realistic economics

Major advances in theoretical and empirical economics now allow economists to address policy issues with much more realistic tools.

On the theory side, new areas such as economics and psychology, and behavioural economics, experimental economics and the new institutional economics have taken the profession light years beyond the simplistic view of human society that dominated thinking just 15 years ago. When it comes to firms, contract theory has allowed explicit modelling of the firm’s organisation choices and their reactions to changing economic environments. Mechanism design has allowed us to think more clearly about how government rules affect equilibrium outcomes in realistic settings.1

On the empirical side, the emergence of enormous new data sets and powerful statistical tools have greatly advanced our understanding of exactly how real-world markets work as well as how people, groups and firms react to incentives. In the 1980s, a ‘large sample’ was one with more than a hundred observations. Now researchers routinely work with tens or hundreds of thousands of observations; sample sizes in the millions are increasingly common. One key aspect of all this new information is that it involves ‘panel data,’ i.e. data that track, over time, workers, firms or whatever.2 Data that vary both across individuals and over time allow us to filter out much of the heterogeneity that marks individuals and time periods. What emerges is a much clearer picture of how things like tax policy or education actually affect economic outcomes.

The new tools don’t mean that we now have all the answers. We don’t. We still don’t know, for example, how economic development really works, or what determines productivity growth. But the new tools mean that we no longer are forced to rely on theory with crude foundations, e.g. firms as profit functions, consumers as utility functions (although there are still many problems where such assumptions are useful abstractions that permit the research to focus on the key trade-offs).

With all these excellent tools at hand, one might have expected the newspapers and Parliamentary debates to be filled with new insights, new results and new approaches. Alas, with few exceptions, the public debate has not moved much beyond the simplistic pro- vs anti-market exchanges that have dominated Europe since the post-war rise of welfare states. Case in point? The 2007 French Presidential election debate.

In the 1980s, brilliant young economists like Paul Krugman, Larry Summers, Jeff Sachs and Joe Stiglitz felt obliged to write Brookings or Economic Policy articles, to sit on government panels, to write policy reports, and to send Op-Ed pieces to the Financial Times. At the time, it was part of the definition of a being a leading scholar. It helped you get tenure at Harvard. It also bridged the gap between cutting-edge research and the public debate on trade policy, exchange rates, current account dynamics, etc.

Today’s brilliant young economists are much less interested in participating in the public debate in these ways. I have no empirical evidence to back up this opinion, but I think it is shared by many economists involved in economic policy issues and I had first-hand experience of it during my five years as a Managing Editor of Economic Policy. Young people need publications in good anonymously-reviewed journals; everything else is a luxury.

Hypotheses on why abound. My belief is that a major intensification of competition among the top US economic departments has occurred over the past two decades, and this led to an overwhelming emphasis on scholarly output that can be easily quantified – journal articles in particular. Things like sitting on Select Committees, testifying to Parliament, or writing policy-oriented books and reports have impacts that are too hard to measure, so they get excluded from the rankings. Oversimplifying to make the point, what does not matter for the rankings doesn’t matter for academic promotion, prestige or pay. While things have not yet gone this far in most of Europe, the trend is starting. Some European economics departments pay a bonus of thousands of euros to faculty whose work gets published in top journals. Writing an influential analysis of a legislative policy proposal gets a pat on the back.

To outsiders – people in the policy world, but also our fellow social scientists – this trend may appear to be a classic case of ‘blinded by science.’ Like the protagonist in 1983’s pop song “She blinded me with science,” the use of mathematics, Greek letters and econometrics seems to be an attempt to deliberately confuse by giving the impression of highly complex knowledge. That is surely overstating the case, but it captures the flavour of the world’s reactions to the increasing divide between today’s best economic research and real-world policy issues.

Why not do it like the doctors?

Medical science faces a similar disconnect between recent advances in hard science – where the big breakthrough may be identification of a particular molecule – and real-world implications, say the concerns of the average family doctor. Medical journals address the problem in a very different way – each article includes the hard science, but it also includes a less formal discussion of what the results mean for medicine and how they fit into the bigger picture – the “Discussion Section.” The International Committee of Medical Journal Editors has called this the “IMRAD” structure (Introduction, Methods, Results, and Discussion) and state that the structure is “not simply an arbitrary publication format, but rather a direct reflection of the process of scientific discovery.3

In the top economic journals, “Policy Implication” sections have fallen out of favour. Including such conjectures in a manuscript is unlikely to raise the probability of publication. Given the natural conservatism of the leading economic journals, there is probably no hope that the journals themselves will encourage authors to draw out the policy implications of their work. In any case, there is a widespread perception that policy analysis is not really the business of scientists. For example, the NBER Working Papers explicitly prohibit policy recommendations. The discussion of research results that does not take place in the journals has spilled over into cyberspace.

The internet as a very public Discussion Section

On the bright side, the internet has two features that make it an excellent vehicle for bridging the research-reality gap. Feature No.1: its technology makes publishing very cheap. Feature No.2: its global span makes it possible to find an audience that is both large and homogeneous in terms of background knowledge. On the dark side, these two features have produced a cacophony. Bloggers – embracing Feature No.1 and hoping for Feature No.2 – have set up a shocking number of sites. A bewildering mixture of insight and nonsense is spread over literally thousands of sites (technocrati.com lists 1,745 blogs about economics with 281,823 posts). One can spend some pleasant hours browsing the various blogs – and even learn a lot from the big blogs, like “Economist’s View”, “New Economist”, ”Marginal Revolution”, and the sites of Brad DeLong, Greg Mankiw, and Nouriel Roubini. But this is not the profession’s response to the Discussion Sections of medical journals. It is more like the collegial coffee-room discussions we used to have when there was time for such things.

What Tito Boeri, the founder of www.laVoce.info, discovered 5 years ago, was that there is a large demand for high-level public discussion of economic policy, higher than the level in newspapers or blogs. Something much more like the Discussion Section of the British Journal of Medicine: discussion of real-world, policy-relevant issues by researchers that is screened by researchers. Tito also found that there was a large supply of researchers willing to devote time – for free – to such discussion. We would probably need psychology-and-economics experts to understand the motives, but whatever they are, these economists are doing Italy a public service.

The Consortium

Tito’s idea – gathering research-based policy analysis that is written by researchers and screened by researchers – is spreading fast. A French site – www.telos-eu.info – was set up a couple of years ago and an English-language site – VoxEU.org – was set up four weeks ago. The three sites have formed a consortium for sharing of columns, translating the best ones into the local language. This way the best research-based policy analysis – regardless of its original language – can reach much deeper into the policy-making world (language still matters). A Spanish site is being set up by Juanjo Dolado and a Dutch and a German site are scheduled to launch later this year. The Consortium has opened discussions on a Swedish partner, a Japanese partner and most recently a South African partner. There may be a dozen more by the end of next year.

Happy birthday LaVoce; it has been an inspiring 5 years.

1 My favourite reference is not on-line: Patrick Bolton and Mathias Dewatripont, "Contract Theory", Cambridge and London: MIT Press, 2005.


2 See Manuel Arellano’s 2003 book, "Panel Data Econometrics", Oxford University Press.



1,470 Reads