July 2019 newsletter: Royal Economic Society – 2019 Annual Conference

The Society’s Annual Conference took place at the University of Warwick, 15-18 April. This report comes from Rachana Shanbhogue of the Economist.

Even by the standards of 2019, the days immediately preceding the conference were eventful. Britain’s deadline for exiting the EU had been extended, granting us a stay of execution for six months. A group of activist climate-change protesters were attempting to bring London to its knees. Further afield, Donald Trump seemed intent on waging trade war not just with China but with the US’s traditional allies in Europe.

There was a certain appeal to entering a rarefied academic bubble. And arriving at the Warwick University campus — a perpetual building site, goes the joke, but at least on this fine spring day adorned by a profusion of daffodils — I was briefly lulled. But it was not to be. This was a conference very much intertwined with the real world, with researchers making good use of the ample fodder provided to them by current events. It was not the otherwordly retreat that I had hoped for. But it was very much to the credit of the economics profession.

We jumped straight into the question of why capitalism leaves some people behind — a phenomenon thought to have fuelled populist movements around the world.

In the first session of the conference, the Hahn Lecture, Anne Case gave a powerful talk about her work with Angus Deaton on ‘deaths of despair’. (Andrew Oswald went on to laud it as the best lecture he had ever heard.)

Life expectancy in America had begun, extraordinarily, to fall, thanks to a sinister increase in drug overdose, suicide and alcoholic liver disease. America, Anne showed, is in a different league when it comes to death rates from alcohol compared with the rest of the West, and is more in line with former Soviet republics. These deaths of despair had been concentrated solely among non-graduates, but in a sense they are symptomatic of a rise in chronic pain and mental-health problems that are affecting a much bigger group of nongraduates, perhaps associated with decreasing job quality, stagnant wage growth, and a diminished sense of community.

In a deadly cocktail, the rise in illness and pain coincided with the increasing availability of fentanyl and oxycontin on the market, and a greater willingness of doctors to prescribe the drugs. The tragedy of it all was that better policy might well have helped avert these deaths, by strengthening stronger safety nets and discouraging rent-seeking in the medical and pharmaceutical industry.

The next session put the process of development in a poor country under a microscope, assessing how lives and livelihoods change. For fifty years now economists have been tracking the fortunes of the residents of Palanpur, a village in northern India. Nicholas Stern and Nick Lanjouw presented findings from the study’s most recent wave.

The study conducts a census of the village, which allows for a neat distributional analysis. The fortunes of some castes change over time as they seize non-farm opportunities. But there is also a degree of stickiness: some castes’ fortune stays the same; and fathers’ income becomes a better predictor of their sons' fate in the second phase.

Rather than activity moving from the agricultural to the industrial sector — as Arthur Lewis might have predicted — the past two decades or so have been characterised by ‘pluriactivity’. Those who do well are those who seize non-farm opportunities, doing more than one job or occupation.

But this is where the drawbacks of using the village as a unit of analysis become evident: it excludes inflows and outflows. As Anne Case and Pramila Krishnan, both respondents to the talk, commented, understanding why people migrate could also help us understand why some people strike it rich while others do not.

Refugees leave home not just in search of a better life but because to stay is intolerable. At a panel on the second day, Christian Dustmann, Rossella Pagliuchi-Lor and Gerald Knaus discussed the facts and rhetoric around refugees. The facts suggest an urgent need to work out how to process, resettle and integrate large numbers of refugees. As the panel's chair, Mark Easton, pointed out, 400 people had already died trying to get to Europe this year. The stocks of refugees in countries such as Turkey and Lebanon dwarfed the flows into Europe and were sizable shares of the population. Refugees are young and malleable, but even so, pointed out Christian, take longer to integrate than immigrants. And rapid population growth in the world's fragile suggested that the number of refugees could rise further.

Policy ideas were largely aimed at Europe: Gerald suggested a ‘new Schengen’ agreement that excludes countries that take a hard line on accepting refugees, such as Italy and Hungary. Another idea was to offer home countries a regular quota for migrants as long as they took back any extra arrivals. But everyone agreed more research was needed: to work out how to integrate refugees, how to speed up the pace of resettlement decisions, how to design agreements with home countries.

Helping hands

If the first day established that growth and development can be uneven, the second day focused on policies to remedy the unevenness.

One welfare programme that is often cited as evidence for early childhood interventions in America is the Perry preschool programme, which ran in the early 1960s and was aimed at 3-4 year old disadvantaged black children. A criticism of it, though, has been its small sample size and concerns about how treatment and control groups were created.

In his Sargan Lecture, Nobel Laureate Jim Heckman used the Perry programme to illustrate how experiments can be saved even when there are doubts about their randomisation. Some children, for example, appeared to have been assigned to control groups because their homes could not be visited, rather than through true randomisation. But in work done with Ganesh Karapakula, Jim showed that even if a higher bar were applied to the results and ‘worst-case’ p-values computed, many of its positive findings still hold.

On top of that, a new survey suggests that the intervention has had impressively long-lasting effects, even spilling over on to the next generation. Those who were in the treatment group as children went on to spend less time, on average, in prison. Thanks to that, they had longer employment records, higher earnings and were healthier. Their children appear less likely to be arrested and suspended from school. The effects on aggregate inequality from such interventions alone are small, Jim reckoned: but every little must help.

Giving the Economic Journal lecture the next day, Eliana Ferrara presented work done with Fernanda Brollo and Katja Kaufmann. It was a nicely worked example of how the structure of a scheme and the incentives it sets up can have unintended effects. They analyse Brazil’s Bolsa Familia, the world’s largest conditional-cash transfer programme which requires that recipient families send their children to school, to assess the impact of punishing those who fail to meet the scheme’s conditions. Financial penalties do encourage compliance, they find, not just by those directly affected, but also by neighbours and children’s classmates.

But that is not where the ripple effects end. Those punished also try to punish in turn at the ballot box, voting against incumbent mayors. And in a further knock-on effect, mayors appear to massage attendance figures around election time, in the hope of not angering their voters.

All this, even in a scheme had overall had very positive results. Even the best thought-out policies, have wide-ranging knock-on consequences. In fact, as Eliana said, the more sophisticated the scheme, the more sophisticated the changes in behaviour it provokes.

Is the solution to rising inequality more university? Anne Case’s lecture certainly suggested it might be. But a word of caution came from Anna Vignoles and Jack Britten, who showed that the monetary returns, at least, are not always positive in the UK. The two studies were the outcome of a project that links school and university records with HMRC data on earnings.

The taxpayer offers university students a rather large subsidy: around half of tuition-fee loans probably go unpaid. Anna presented work on extent to which a university degree improves valued added, using graduate earnings at age 29 to proxy for productivity. Controlling for socioeconomic and regional variables, she computes the returns to studying a particular subject at a particular university.

University attendance appears to boost earnings by 26 per cent for women and 8 per cent for men. In a boost for the profession, she finds that those studying economics earn the biggest wage premium, particularly for women. Medicine and physics have high returns too; the creative arts and social care are less well-compensated for. The Russell Group universities had the highest premium; for some universities the wage premium was negative, particularly for men.

Jack Britten looked at the question from the point of view of a prospective student, and concluded that it might not be optimal for as many to go to university as they do. Men with a lower educational attainment have a lower return to going to university, of below 4 per cent, with the returns on some subjects negative.

Of course, none of this provides a complete answer to the question of whether to subsidise or not. University attendance has significant non-monetary benefits, some of which might even help to avert despair.

As Jim Heckman observed, economists are like detectives, hot on the trail of the causal effect. Many papers used intriguing natural experiments to uncover causality. Outbursts of anti-refugee sentiment on Facebook are correlated with real-life hate crime in Germany; outages on the website are associated with less hate crime, suggesting the causal link runs from the virtual to the real. The establishment of the Metropolitan Police and bobbies on the beat had crime-fighting effects; the introduction of crack cocaine in California had corruption-boosting ones, as drug profits were funnelled into weakening enforcement.

But history is not inevitable: take the impact of the partition of Poland in the 19th century on human capital. Poland had been divvied up between Russia, Austria and Prussia, each of which had education systems of varying quality. By a couple of decades after its unification and independence in 1918, those differences in educational attainment had faded away.

Causal links are less easily established in macroeconomics where everything seems to be endogenous. But there were illuminating session on the drivers of low interest rates (monetary regimes, consumption to wealth ratios, and demographics) and of low inflation (menu costs, frequency of price changes and in the future, though not yet today, cloud computing). In an interdisciplinary session organised by Rebuilding Macroeconomics, John Kay mulled over the implications of radical uncertainty for modelmakers, based on his work with Lord Mervyn King. In summary: treat model estimates with care.

Geography lessons

The vote to leave the European Union may well have been a response to rising inequality and capitalism’s failings. Given the shadow the subject of Brexit has cast over the country for the past three years, it was only right that the profession discuss its economic implications.

In his past president’s lecture, Peter Neary took on the Brexiteers’ argument that distance is dead, pointing out very amiably that geography, unfortunately for them, was still very much a powerful force when it came to trading patterns.

Gravity models find that trade falls off more or less one for one with distance. And that has not changed over time: just as reduced trade barriers and lower transport costs have made trade with distant countries cheap, it has also made trade with nearby ones easy. But the data suggest that demand is subconvex: as the value of trade rises, its elasticity to distance is lower. So the introduction of trade barriers with neighbours is unlikely to have, in Peter’s words, a ‘catastrophic’ effect.

Michael Wickens had either not attended Peter’s speech, or disagreed with it; at the public ‘RES presents’ event, he declared himself no fan of gravity models. Vicky Pryce argued forcefully that the effects of Brexit were already being felt, pointing to its effects on foreign direct investment and civil-service morale. Alan Winters suggested keeping an eye on services firms. Theresa May’s deal (whatever that is now worth) ignores them and they can all too easily up sticks and set up abroad. Nick Macpherson counselled ministers to focus now on policies that make Britain an attractive place to invest and work, such as innovation and tax policy and further education.(They do not yet appear to have taken the advice.)

But why had economists’ views on the effects of Brexit fallen on deaf ears during the referendum campaign? Some panellists put that down to the profession's failure to spot and explain local congestion effects from increased migration.


Is the profession really incentivising the best research? The INET special session also involved self-reflection.  (See below, p.9). Jim Heckman showed that publications in the top 5 journals are associated with better promotion prospects at universities. But such laser-like focus on the top 5 might be costing the profession. The journals are not perfect measures of quality: papers published in the top 5 are not always the most cited, for example, and journals affiliated with a certain university appear to have a bias towards researchers from the same university.

And it might mean that unusual questions or books, are underweighted when it comes to thinking about promotion. Giulia Zacchia, also on the panel, noted that research on the US and the UK tended to be most rewarded, even in Italy. Those hoping for change might take some comfort that Jim Heckman — a Nobel laureate and an editor of the Journal of Political Economy, one of the top 5 — has taken up the cause.

One of Jim’s findings was that the bar for promotion of women is higher than it is for men. (Though small samples and maternity leave make it difficult to know how much weight to put on the finding). Across the Atlantic, the question of discrimination in profession had dominated the AEA congress in January, after a climate survey of women that harassment and discrimination were not uncommon in the profession. (The RES is drawing up a code of conduct, and is mulling a survey of its own.)

At the women’s special session, Rachel Griffith, Eliana La Ferrara, Carol Propper, Sarah Smith and Rain Newton-Smith spoke about their own experiences. Several noted the importance of role models, and observed that men and women were still viewed differently in the context of childcare. The advantage of academia is its flexibility, says Sarah Smith. But perhaps that could be a disadvantage too, if the expectation is meetings and calls can take place at any time of day. Perhaps, the panellists suggested, it might fall to colleagues to be sensitive when thinking about arranging meetings or calls. 

More introspection came at another ‘RES Presents’ session on the communication of economics. Nick Stern, the departing president, took comfort from the research presented at the conference that profession was indeed paying attention to real-world problems. But why was the message not getting through? That must be down to economists.

Perhaps, as Nick suggested, economics could take a few lessons out of geography’s book. The subject is more popular, perhaps because ‘everything is seen as geography’, says Wendy Carlin. Economics, by contrast, is perceived very narrowly as to do with money.

Arun Advani pointed out that economics was either less accessible or less appealing to school kids. Independent and grammar schools were three times more likely to teach it than state schools. Boys were three times more likely to study the subject than girls. The relative exclusivity of the subject could affect the way the public engages with economists’ claims: if they have either studied the subject for a while, or are being spoken to by people of a similar background, they might feel more inclined to engage with the subject.

Others argued that the profession needed more humility. Rachel Griffith, the society’s newly anointed president, suggested a Brian Cox approach: being honest about what we don’t know, and how that was actually an exciting avenue for exploration rather than a failure.

What about engaging with politics? Political interest in evidence-based policy making seems to be at an all-time low. That was dispiriting. But for economists to withdraw now would surely be a mistake. There might be no easy way to maximise the profession’s impact. But as I returned to London, I had no doubt that economists have plenty to contribute.