The 200 page report, released by the Treasury a couple of weeks ago, has been used as the basis for the claim, by both the Prime Minister and Chancellor, that there is a consensus amongst economists that Brexit would lead to recession (or even depression).
However over-blown these claims, the Treasury Report has also been used by the TUC as the basis for their claim that withdrawal from the EU would hit the pockets of working people, and for suggestions that the retired will receive smaller pensions.
Consequently, since it has been used a reference point for many of the claims made in the campaign, on the Remain side, it is worth looking in a little more detail about how it reaches its predictions.
The headline figures are that, should the UK leave the EU, there would be substantial and permanent costs totalling between 3.4 and 9.5% of UK GDP, depending upon which type of trade agreement is negotiated following ‘Brexit’.
Chancellor George Osborne claimed that it proved that leaving the EU would be the “most extraordinary self-inflicted wound” and that those supporting ‘Brexit’ were “economically illiterate”.
This knockabout politics is all very well, but it obscures the fact that all reports of this type are based upon a set of simplifying assumptions. As a result, the predictions are only as good or as rigorous as the assumptions upon which they are founded.
Thus, in much the same way as the 30+ other academic and policy studies, published over the last 20 years, which have sought to predict the potential cost of the UK leaving the EU, the Treasury Report makes its best, educated guess.
There is nothing wrong with that, of course, because we need to base our economic decisions on our best forecasts of the future, but the limitations of such studies need to be kept in mind. After all, there have been a number of occasions – i.e. the disastrous ERM experiment in the early 1990s to the very recent problems Osborne had with his own budget figures – when predictions have not been entirely accurate.
1. Because (gravity model) studies have indicated that EU membership may stimulate trade with other members by up to 85%, it is assumed that leaving the EU would automatically and seemingly instantly reverse this position (p. 129). There are three points to be made here.
(i) Once trade has been established between nations, and firms have formed established supply chains, it is not likely that there would be a sharp and immediate change following ‘Brexit’. It is possible that non-tariff barriers could make such existing links more time consuming and costly to continue, and that, over time, reductions in trade may occur, but it unlikely to be instantaneous.
(ii) When considering the impact if tariffs were to be applied to UK exports, in the WTO case, the House of Commons library indicates that the average EU trade-weighted ‘MFN’ tariff is only around 1%, rather than the higher (non-trade weighted) figure that the Treasury Report uses for its calculations.
(iii) The Treasury Report claims that it found only trade creation, and not trade diversion, when examining why trade grows when countries join the EU (pp. 157-164). This is surprising, because economic theory suggests that both are likely to occur. A customs union or single market makes it cheaper to shift trade within the trade zone. This is not trade creation but diversion. To the extent that trade diversion occurs, then if the UK is freed to trade elsewhere in the world and can establish its own free trade agreements, a proportion of that trade which formerly occurred with EU nations would shift elsewhere. Hence, to the extent that trade diversion occurs, the Treasury predictions are exaggerated.
There is also nothing included in the Treasury report seeking to predict how trade with the rest of the world might develop if the UK left the EU and was able to negotiate its own trade agreements. This is unfortunate, because it might be expected that non-EU trade might expand, particularly if the price of Sterling falls (also mentioned in the Treasury Report), thereby making UK goods and services more internationally competitive. However, this is sadly ignored.
A second key assumption is that, because of a shortage of data, the impact on foreign direct investment (FDI) is assumed to occur in proportion to trade effects (Treasury Report, p. 130). This is surprising, partly because there is plenty of data on FDI, but also because any errors in calculating trade effects will be magnified by applying the same logic to FDI.
A third assumption is that net migration will only fall by about 40% following ‘Brexit’ (p.136). If migration is reduced by more than this amount, then predicted costs and benefits will change accordingly.
Fourthly, the Report only views regulation as an integral part of the single market, and therefore ignores that research which suggests that national (UK) regulation has a better benefit : cost ratio than EU regulation (pp. 59, 139). In other words, were the UK to leave the EU and even we introduce regulation on all of the same issues as the EU, the fact that this could be better tailored for the needs of our own economy would produce a net gain to the UK.
Finally, one limitation of this study is that it makes predictions based upon the simplifying assumption that everything else remains the same (ceteris paribus). That makes calculations easier, but conclusions are more dramatic and less useful.
To take one example, if the EU were to impose a 10% tariff on UK car exports, a study of this type would predict a fall in trade would occur, without considering whether, freed from the rules and constraints imposed by EU membership, the UK government might either compensate UK car manufacturers directly, to offset any increase in costs, or else allow the value of the pound to fall by an equivalent amount. If they did either, then the net effect would be closer to zero.
Yet this effectively precludes one of the largest potential benefits that may arise if the UK left the EU, and that relates to the greater range of policy options and instruments that policy makers could use in managing the economy.
Within the EU, active industrial policy is hampered because any measures seem as disproportionately benefitting one’s own industry would be viewed as ‘discriminatory’ within a single European market. Yet, outside, there is no such restriction and hence it would be more likely that the UK might be able to rejuvenate its manufacturing sector and achieve more of the re-balancing of the economy that the Chancellor states is an official objective, but, within the constraints imposed by the EU, has done relatively little to achieve this goal.
The Treasury Report, therefore, is an interesting document, but it has to be judged as unreliable in its predictions because it leaves too much out of its calculations.
It focuses primarily upon trade with the EU and FDI, both of which are more likely to have some sort of extra costs following Brexit, but ignores trade with the rest of the world, migration, regulation and economic policy options, which are more likely to deliver benefits.
If I were to design a study in this way, I would be accused of either bias or having only a partial model – either way, my results would be treated with caution, which is the equivalent of a research ‘health warning’.
It is a pity that the Treasury were not able to evade their organisational imperatives to deliver a very comprehensive report. This may not have made such a splash in the media, because including all of the variables in their model may have produced a more modest net effect (positive or negative), but it would have been far more use to voters who are struggling to trust the evidence presented during the referendum campaign. A missed opportunity.
Philip B. Whyman
Professor of Economics and Director of the Lancashire Institute for Economic and Business Research (LIEBR), University of Central Lancashire
 Thompson, G. and Harari, D. (2013), The Economic Impact of EU Membership on the UK, House of Commons Library Briefing Paper SN/EP/6730, p. 7.