What the gravity model of trade can tell us about the implications of Brexit and Trump
Consider a hypothetical Robinson Crusoe thinking of selling coconuts to the inhabitants of one of the neighbouring islands. Which island should he target first? The obvious answer is – whichever island is closest and richest. A rich island will be more attractive than a poor one since it offers a bigger market for his coconuts. But so too will a closer one – any time and money spent transporting his coconuts will eat into any potential profits. This basic insight is embodied in one of economics’ most respectable theories – the gravity model of trade. So-called, because it mirrors Newton’s gravitational equations, it says that the volume of trade between two nations is proportional to their GDP, and inversely proportional to the distance between them. Britain and the Netherlands trade a lot because they are both rich and close.
The model boasts some impressive empirical credentials. But now that countries such as the US and UK are seeking to recalibrate their trading relationships, it has come into focus as a potential source of guidance about the evolution of world trade. In the US, Trump is seeking to renegotiate aspects of NAFTA to better serve American interests, as he sees them – in particular, to reduce the country’s trade with Mexico. And for Britain post-Brexit a crucial question is whether it can forge new trading relationships – for instance with the US or China – that could compensate for any loss of trade with the EU that may come as and when arrangements with the EU deteriorate. Are these countries simply trying to defy (trade) gravity or does distance no longer matter?
Tyler Cowen, of George Mason University, has recently argued that despite technological advances which have accelerated of globalisation, the influence of distance on trade has not diminished:
“The most surprising result from the research on trade and distance is that the ability of trade patterns to surmount barriers of distance has not in fact increased over time. You might think that with the internet, highly efficient ports and powerful multinationals, geographic distance would predict trade patterns less well over time, but that has not been the case.”
Why is this? His tentative answer:
“…individuals form network connections with those who live nearby, and those connections boost subsequent trade. Many top Mexican executives, for instance, went to American business schools, where they developed contacts more useful for trading in the U.S. than for trading with, say, China or Russia.”
Cowen suggests that the effect is to tether countries like the UK or US more closely to their current trading partners. If that’s right it suggests that, regardless of political imperatives, trading patterns will not be easy for either Trump or May to change.
Adam Tooze, a historian and economist, has also examined some of the literature to try and solve the puzzle of why, despite globalisation, distance still seems to matter so much. Citing a paper by Yoto V Yotov, of Drexel University in Pennsylvania, he suggests that this is really an artefact of how distance has been measured. Empirical studies of the relevance of distance to trade have (incorrectly) relied on measures of the relative cost of transport between nations, which will remain unaffected by advances (such as containerisation) which affect all countries and so are neutral.
To put it another way – if Robinson Crusoe is given a speedboat instead of a rowing boat, he would probably still choose the closest, richest island. So distance would appear to matter to trade just as much as it did before. But if, as the researchers did, the costs of trade between countries are compared with the cost of trade within them a different result emerges. Then, the researchers do indeed find that distance between countries matters much less to trade than it used to – 37% less. When Robinson Crusoe got his speedboat, he did indeed do more trade with more distant islands – even if he still traded most with his closest, richest, neighbour. Putting effort into that more distant trade became more worthwhile compared with the other alternative, doing more business inside his own island. As Tooze puts it:
“Yes, geography matters. Yes, trade with those closest to us is most important. But our eyes have not been deceiving us, the distance-annihilating effect of globalization is not an optical illusion.”
So much for the theory. But what of its implications for policy-makers? Samuel Lowe and Grant Lewis, two British researchers, have examined its potential significance for the UK economy after Brexit. They offer a striking chart (see below) which shows quite clearly the impact of distance on UK trade. Trade is by far the greatest with the EU (which is rich and close), less so with the US (which is rich and further away) and minimal with New Zealand (rich but extremely far away).
The implications of this are potentially serious. Any economic model based on replacing lost EU trade with more trade with the US and China– as some Brexiteers such as Liam Fox like to posit – is, like pushing water uphill, an attempt to defy gravity. Using Treasury estimates of the long-term impact of Brexit as well as some generous estimates of the potential gains from new FTAs with other trading partners, they find that the long-term negative impact of leaving the EU will still outweigh the potential trade gains with other partners. So a greater effort should be devoted to limiting that negative impact than on forging new trade arrangements with distant nations.
The gravity model of trade suggests that countries like Britain and the US will continue to trade a lot with their closest neighbours. It also suggests that both countries will do rising trade with China (which despite being far away has a large and growing GDP) and that there is a strong economic logic to regional trading blocs such as NAFTA and the EU. That is a useful corrective to some of the more hare-brained ideas of protectionists in the governments of Trump or May. But it needs careful interpretation. Globalisation has lowered the overall costs of trade between nations, and the absolute importance of distance in driving trade has (probably) declined. But distance’s importance in determining which nations we trade with may be as significant as ever.
Edited by Bill Emmott