July 2017 newsletter – Conference Report 2017

The Society’s Annual Conference was held at the University of Bristol, 10-12 April. This report was prepared by Soumaya Keynes of the Economist.

Brexit means Brexit 
In March of 2016 I asked an audience of economists at the Royal Economic Society’s annual conference in Brighton to raise a hand if they thought Britons should vote to remain within the European Union. A sea of hands went up. I then asked for a forecast. Did the audience think that Britons would vote to leave the EU? Back then, the opinion polls suggested a victory for remain. Only a smattering of the economists predicted otherwise.

A year later, this time at the University of Bristol, and the assembled economists were feeling decidedly less optimistic about Britain’s future. A panel discussion on the first day of the conference tackled post-Brexit economics. Paul Johnson, director of the Institute for Fiscal Studies, declined to talk about the impact of Brexit — there wasn’t a great deal to add beyond the obvious. Instead, he asked the audience what they should learn from the past year. Economists knew that Brexit would make Britons poorer, but failed to get their message across. What went wrong?

Divided we fall 
Perhaps the public knew that Brexit would drain their wallets, but voted for it anyway. Sometimes it’s not the economy, stupid. But Simon Wren-Lewis, of Oxford University, rubbished this idea, pointing out that those who voted for Brexit said they were unwilling to pay to reduce immigration. Instead, people seemed misinformed. Not only did those who voted to leave the EU think that they would be no worse off as a result, they thought that lower immigration would improve their access to public services.

The problem, Mr Wren-Lewis argued, was with the media. Dismissing a large fraction as producers of propoganda, he reserved most disappointment for the BBC. It failed to communicate the consensus among economists, he said, treating it as opinion rather than knowledge. Swati Dhingra, of the London School of Economics, agreed, saying that the BBC’s quest to generate balance gave the false impression that there was a meaningful debate between economists. (Depressingly, she noted that this false balance had oozed into policymaking, as select committees are being stacked with pro-Brexit voices.)

Mr Johnson did some finger-pointing of his own. He ticked off the treasury for publishing its own analysis before independent researchers had their chance, and for its ‘ludicrous’ statement about an emergency budget. The chancellor should have stood back, rather than adding to public confusion over the distinction between academic economists and those toeing the treasury line. Part of the problem, he added, was in the nature of the debate. Brexit had political and economic dimensions, and whereas there was a broad consensus among economists that Brexit would make Britons poorer in the long-run, there were political arguments to be made on both sides. Where the economics is clear but non-economic factors are important, it becomes difficult to appear neutral and objective.

The first comment from the audience came from Gervas Huxley of the University of Bristol. ‘The panel came across as frighteningly complacent’, he said. ‘We still sound to me terribly pleased with ourselves … we must ask ourselves how we can do better.’ The panel did come up with prescriptions. Simon Wren-Lewis called for economists to organise, as the ‘there’s only one group of people who are going to stand up for economists and that’s economists themselves.’ Jagjit Chadha of the National Institute of Economic and Social Research called for an economics analogue of a chief scientific officer, who would speak for the profession. 

There was perhaps stronger evidence of complacency among the audience. To Paul Johnson’s astonishment, when he asked for a show of hands from all those who had changed their research agenda in response to Brexit, not a single hand went up. His response: ‘in the context of the biggest economic shock that has hit this country in decades, we need to reflect on how and why that is the case.’ (It may have been some consolation to Paul Johnson that businesses are only a little more responsive than academics. My other dose of Brexit over the next few days came from a paper by Jumana Saleheen, Iren Levina, Marko Melolinna and Srdan Tatomir of the Bank of England, which found that in November of 2016 only 23 per cent of firms they surveyed had reviewed their investment targets since the referendum.)

Trade: all we need is love? 
With so much politicking and so few details on Britain’s post-Brexit deal, perhaps it wasn't surprising that Brexit-related research was scarce. Amid the three-day feast of economic research there was, however, plenty of research on trade more generally. 

There were even a (very) few good jokes. The gag of the conference went to Martin Cripps, of University College London, who after a lengthy pause, announced the winner of the Austin Robinson prize as… La La Land. The real winner, Sergey Nigai of ETH Zurich, used a little more algebra than Ryan Gosling, in his quest to analyse the effects of trade when one relaxes the assumption that all consumers are alike. He finds that lowering trade barriers has more of an effect on the price of tradeable goods than non-tradeable ones. And as richer households both have relatively strong preferences for tradeable goods and consume more of them, they will tend to benefit more from trade. Without accounting for this, one could overstate the progressivity of the consumer benefits of trade, he warned.

Another example came from Katrin Huber of the University of Passau and Erwin Winkler of the University of Wuerzburg. Most analysis of the impact of import competition on people looks at individuals. When Ms Huber and Mr Winkler looked at the effect of rising imports from China on Germany, they explored how the pattern of winners and losers changed after accounting for partners. They found that partners did tend to cushion the blow of a trade-induced shock, and called for policymakers to take note. As single-headed households become more common, the insurance provided by family structures may well be weakening.

These two papers prodded the standard ways of studying trade. In her keynote speech, Gita Gopinath of Harvard University aimed for a more fundamental change, to our framework for thinking about currency devaluations.

When the pound devalued, Brexiteers cheered that this would boost British exports. In this, Mundell-Fleming, view of the world, prices are sticky in the exporting country's currency. So a devaluation of the pound would mean that although import prices would rise, Europeans should be tempted towards cheap British products. This should injecting the British economy with extra foreign demand, and improve the British balance of trade.

Ms Gopinath also outlined the competing theory, developed by Caroline Betts and Michael Devereux, which is that prices are sticky in the importing country’s currency. If so, a sterling devaluation is less likely to be a boon for the British economy, as with no change in the euro sticker price, european customers will face no price incentive to switch towards British goods. British exporters’ profit margins might rise temporarily as their euro prices would return more sterling, but they would not enjoy any more demand from european consumers chasing cheaper stuff.

Unfortunately, said Gita Gopinath, the world is a little more complicated than that. The symmetry embedded in these models is convenient for making neat models, but in reality trading countries differ in the way that they invoice their trade. Specifically, one currency, the dollar, dominates global trade invoicing. Since 1999, America have bought just under 10 per cent of world imports, but 55 per cent of world invoices are in American dollars. And whereas 91 per cent of American exports are priced in dollars, 51 per cent of British trade is priced in sterling and just 5 per cent of Chinese exports are priced in yuan.

She therefore introduced a third generation of exchange rate assumptions, known as the ‘dominant currency paradigm’. Here prices are set in neither the importer’s nor the exporter’s currency, but in a third currency, which dominates the dynamics of trade. Her research with Emine Boz of the IMF and Mikkel Plagborg-Møller has confirmed this empirically. For understanding how prices will change in response to a currency shock, they find that the dollar exchange rate is more important than that between the trading partners.

In this alternative paradigm, currency devaluations work differently. A devaluation of the Brazilian real against the Japanese yen, for example, may not boost Brazilian exports, depending on what the dollar does. If orders are priced in dollars and the yen depreciates against the dollar, then they could even fall. A currency devaluation should still have an effect on trade, but most through making imports more expensive and pushing consumers away from them. This leaves small open countries vulnerable to shifts in American monetary policy, insofar as they strengthen the dollar, making imports more expensive without having much of a stimulative effect on exports. At a global level, a strong dollar could even crimp global trade.

Question your assumptions 
Sometimes critics charicature economics as a discipline stuck in its ways, clinging desperately to useless theory. But as Gita Gopinath showed, the very best research comes from ricocheting between theory and evidence, overturning old assumptions and exploring all implications. There were more examples of this on show at the conference. 

Philippe Bracke of the Bank of England and Silvana Tenreyro of the London School of Economics, for example, tried to understand the causes of history dependence in the British housing market, or why the probability of selling one’s house seems to depend on the purchase price. This phenomenon, which is tough to reconcile with standard models of the housing market, helps to explain why in Britain transaction volumes fell so much more than house prices during the financial crisis. As prices fell, a disproportionate number of people declined to move home.

Of two identical houses, one bought in 2007 at the peak of the house price boom and the other bought in 2001, the one bought in the heady days of 2007 will sell at on average a 10 per cent premium over the one bought in 2001, they found, but the chances of it being sold will be 15 per cent less. Of the 4.4 percentage point decline in the chances of selling one’s property between 2000-07 and 2008-14, around 0.4 percentage points (or, in journalist speak, a whopping 10 per cent of the fall) can be explained by this backward-looking behaviour.

The authors test two possible explanations: might the buyers may be so saddled with debt that they cannot move? Or perhaps a behavioural bias is at play, and sellers anchor prices to the price they bought at? They compare cash buyers, who are unlikely to be credit-constrained, with home owners, and find evidence of anchoring behaviour in both, though the effect is accentuated for highly-leveraged buyers. A large part of the effect is not, therefore, gummed up credit markets, but rather because of human psychology.

Moore or less 
In his keynote lecture Oliver Hart, billed by John Moore, the outgoing president of the RES, as ‘the smartest economist on the planet’ told the audience how he had not so much reformed an old framework as built a new one. Delivering a version of his Nobel lecture, he remembered musing on why one firm would ever buy another, rather than operate at arm’s length using a contact. Pinning down the answer in a precise, mathematical way, took ten days of intense work — ten days that shook my world’.

Understanding the difference between contracts and mergers required two ideas, he thought. The first was the idea that in real life, contracts are complete. (When I agreed to cover the Royal Economic Society’s annual conference, it would have been difficult to incorporate a clause relating to the quality of John Moore’s jokes.) The second was that ownership conveys ‘residual control rights’. When reality strays from the letter of a contract, the owning party has the final say over what is what.

Using this framework, Mr Hart explained the trade-offs involved when choosing between ownership and a contract. Incomplete contracts mean that sometimes situations arise that fall outside of what was already agreed. In those situations, whoever has residual control rights has huge power, and can ‘hold up’ the other. In anticipation of that imbalance, the party without control might be reluctant to invest too much in the relationship, and invest less than is efficient. A power plant, for example, might be reluctant to locate too near to a coalmine, for fear of the mine manager abusing its market power later. 

Ownership comes with drawbacks too. Buying the coal mine, and seizing residual control rights from its manager, the power plant owner can be sure not to get suckered by an opportunistic coal-mine manager. But staff at the coal mine might then have less of an incentive to innovate, as profits as well as control would have been wrested from them. 

To take another, dragon’s den scenario, an inventive entrepreneur is strapped for cash. In a perfect world, the investor would fork out the money in exchange for equity in the company, and set out various rules to make sure the entrepreneur doesn’t squander the investment. In reality, writing a complete contract is impossible. Then, the investor faces a tricky trade off. Demanding a very high stake in return for a risky investment comes with costs-if the entrepreneur's baby is wrested from her, she might not work so hard to make it a success.

The solution is to make control state-contingent. The entrepreneur could simply borrow from the investor, promising to hand over her assets in the event that she cannot pay an instalment of the repayment. Here, the ability of the investor to seize assets, or their residual control rights when things go wrong (effectively ownership transfers to them) forces the entrepreneur to pay up, and makes the investor feel comfortable advancing the loan. This framework suggests that when it is harder to seize assets, it will be harder to raise money. It also explains why their real-world version, debt contracts, are so popular.

Since Hart’s early innovations, other economists have built on his insights to show in ‘mind-boggling’ detail how particular clauses can solve theoretical incentive problems. But disappointingly there is often a gulf between theory and practice. ‘Take it or leave it’ clauses, for example, can align incentives, but unfortunately for the theoreticians, in reality it seems that real people shun them.

After having walked the audience though the evolution of contract theory, first from woolly assertions, to mathematised theory, and then a blossoming to plug theoretical gaps and refine the theory where needed, he accepted that there may never be a ‘tractable, widely agreed upon, theory of incomplete contracts’. Modelling reality beyond a certain level of complexity means departing from the classical notion of rational man, and including things like feelings of fairness and loss aversion. As an audience member, this was an unexpectedly satisfying conclusion. Mr Hart had tugged complete contracts towards the reality of incomplete ones, and in doing so had created an infinite number of new questions.

The conference indulged its participants in grand theories, plenty of debate over valid identification and a healthy dose of self-criticism. As if to set up the debate, attendants were handed turquoise water bottles decorated with a quote from Alfred Marshall defining economics as ‘a study of mankind in the ordinary business of life.’ Mr Marshall thought that economics was not a subject for women. And while his views are no longer socially acceptable, men still dominate the discipline.

One paper, presented by Erin Hengel of the University of Liverpool, opened with a line addressing her fellow economists: ‘ladies, there aren’t that many of us.’ Her paper uncovered evidence suggestive of sexism in the peer review process, in that they seem to be held to higher standards than their male peers. Abstracts written by women are clearer than those written by men, even conditional on their classification within the Journal of Economic Literature system, with the gap widening as the peer review process continues. And at Econometrica, papers written by women take on average six months longer to complete the process.

Sarah Smith, the conference convenor, was concerned with gender inequality earlier on in an economist’s career. At every stage of education, girls shun the subject. Only 25 per cent of those studying A-level economics are girls, a lower fraction than maths, though still higher than computer science, engineering and physics. Rather than economics, girls are picking psychology and law.

Given the influence of economists, and the fact that women have views that are systematically different to men, this imbalance matters, said Ms Smith. In the panel on Brexit earlier in the conference, Mr Johnson wondered if the disaffection from women was evidence of something wrong with the discipline. 

In a packed lunchtime panel, Kimberley Scharf reported that the imbalance could have become self-fulfilling. The girls she had surveyed informally reported being reluctant to share a classroom with so many boys, calling them "massive idiots". My own survey before the panel of friends who had dropped economics between A-level and university revealed some who did so because they needed maths A-level, and others who thought it would be too hard to get into the university of their choice.

Sarah Smith’s diagnosis was that economics is misperceived as a subject. As the conference illustrated, economics is a broad subject, and fundamentally is about helping people do their best with scarce resources, improving policy and studying how people behave. This should appeal to both genders, but perhaps especially the higher fraction of girls than boys who reported that their A-level choices were motivated by a desire to help people. For some reason, though, students think that the subject is all about finance or money. Mentoring and engaging with schools were both suggested as solutions. 

Breakfast means breakfast
Amid the pastries, the presentations and the self-flagellation, it might have been easy to forget what the point of it all. An article in The Independent, previewing the conference, and highlighted by Paul Johnson on the first day of the conference, provided a helpful reminder. ‘Economics research can really improve people’s lives’, wrote Hamish McRae. While the public associates economics with GDP and abstract equations, much of the research presented at the conference was focused on how to make people’s lives better.

Examples I saw included a paper presented by Christine Farquharson of the IFS, which suggested that free school breakfasts are a cheap way to help children do better in school. A panel discussion on re-skilling the UK between Steve Machin, Kirabo Jackson, Richard Burgess and Sandra McNally tossed around tax credits for investment in skills and training, a plea for more thinking about teacher quality, and from Kirabo Jackson, to think about the education system as a whole, rather than separate, substitutable stages. Football scheduling was on the list too: boys perform worse in exams when they coincide with international football tournaments.

keynote lecture by Hilary Hoynes of the University of California at Berkeley looked directly at American poverty-reduction programmes, including health insurance, food stamps, tax credits and housing vouchers. These programmes are often hard to assess, due to the heavy data requirements and non-random application. Despite these hurdles, economists knew rather a lot. Mothers given food stamps in the third trimester experience a six per cent drop in the chances of their child falling into a low birth weight category. The work incentives tied up in the American earned income tax credit had as much of an impact on poverty reduction as the cash value of the benefit. And housing vouchers to help families move can improve children’s earnings and test scores — but the benefits only appear for children who move when young. There were still unanswered questions, like which of benefits delivered in cash, vouchers or in-kind were best, but Ms Hoynes was confident that for the most part, the American social safety net had broader benefits than people had previously thought. 

Take back control 
For three days at the University of Bristol, I feasted on economics geekery, indulging in tables of standard errors and questions about identification. Risk and uncertainty were a given. Constructive criticism was taken as it was intended. The speakers were sleuths, piecing together what they could from awkward data sets and non-random interventions. The economists had, sort of, taken back control.