In a recent Jacobin article, Thomas Fazi and Bill Mitchell argue that the Left should embrace Brexit. “The Left’s anti-Brexit hysteria is based on a mixture of bad economics, flawed understanding of the European Union, and lack of political imagination,” they write. Although they characterize the Left as suffering from groupthink on Brexit, this is not entirely accurate. “Lexit” has attracted the support of a significant minority on the Left. Nonetheless, it’s fair to say that the majority of progressives in the UK took the Remain side during the referendum.

The authors make two main points to support their position. The first is that many of the UK’s current economic problems have domestic causes and are not the result of the Brexit referendum. The second is that the benefits of EU membership have been overstated. We are sympathetic to both points. But it takes a major leap of faith to get from here to the conclusion that a hard Brexit provides a clear opportunity for implementation of progressive policy. The authors provide no real evidence to support their claim and fail to address any of the major obstacles.

The authors start by noting the inaccuracy of predictions that the UK economy would immediately enter recession in the wake of the Brexit vote. It is true that both sides of the argument have been too willing to make unfounded claims about the economic costs and benefits of leaving. Predictions of immediate recession following the referendum are in a case in point: these were clearly overstated to make political gains. But these short-run projections all originated with politicians or banks; academics confined their attention to the long-run effects.

The authors conflate these politically driven Brexit predictions with the problems of mainstream macro modelling and economists’ failure to predict the 2008 crisis. This is sleight of hand. There are undoubtedly problems with mainstream macroeconomics, but the failures of DSGE macro do not prove that Brexit will be without costs. The authors also conflate projections of the long-term damage of Brexit, based on gravity models, with the highly politicized short-term forecasts of the immediate effects of the referendum. As the Cambridge economists quoted by Fazi and Mitchell explain, the UK Treasury and private financial institutions got the short-term forecasting so wrong because (as ever with forecasting) the assumptions driving the results turned out to be false. It was assumed that the private sector would immediately cut back on spending because of increased “uncertainty” and adjustment to the new long-run trend, and that sterling would fall. Of these, only the prediction about the exchange rate was correct.

The assumption that spending would take an immediate hit never appeared plausible — and was not endorsed by academic economists. Then, as now, the primary engine of growth in the UK was debt-fueled household spending. Unless one believes in the ultrarational, future-predicting fantasy consumer of standard macro models, there was no reason to believe this trend would immediately cease (as one of us has been arguing for some time). It is also the case that politically driven pro-Leave short-term forecasts were equally faulty. “Economists for Brexit” produced the only positive short-term forecast of the effects of the referendum. It predicted that the UK economy would grow at 2.7 percent in real terms in 2017 and 2018 following a Leave vote. This turned to be just as wrong as the Treasury forecasts.

Surprisingly, the authors undermine their point about the structural problems of the UK economy by trying to paint the post-referendum period as a boom. In reality, after eight years of (home-baked) austerity, real wages are still below their pre-crisis peak and not set to recover for years, productivity is flat-lining and the only engine of growth — credit-fueled consumer spending — is now running out of steam. Unemployment figures likely disguise substantial underemployment and precarity. The statement that “total investment spending the UK … was the highest of any G7 country during 2017” is simply incorrect. Investment growth was high, but from a low base. By historical standards, investment spending remains low, and is in fact the lowest among G7 countries as a percentage of GDP.

The authors are on somewhat stronger ground when questioning the extent to which the UK has benefited from membership of the European Union. They draw on the Cambridge Centre for Business Research to argue that the economic benefits of single-market membership have been overstated. This is plausible. It is also reasonable to argue that long-run gravity model projections should be taken with a grain of salt. However, the authors overemphasize the significance of a graph showing export volumes in the EU and eurozone: it shows fairly constant intra-EU trade during the single-market period (with a decline following the 2008 crisis), while the intra-eurozone trade share has been declining. This suggests that non-eurozone intra-EU trade may have been growing to compensate for declining eurozone trade. Even if intra-EU trade hasn’t expanded since during the single-market period, this does not imply “a negligible advantage to the UK of being a member of the EU”, but only that non-European trade has been growing at a faster pace.

The authors also note that UK exports to the EU have remained stable over the period, while UK exports to the rest of the world have increased. Again, this does not demonstrate, as the authors suggest, that membership of EU has not been beneficial. On the relatively narrow indicator of trade alone, we have no counterfactual: we do not know what would have happened if the UK had remained outside the EU. It is perfectly plausible that trade would have declined. The authors also take a somewhat self-contradictory stance: they argue that the UK hasn’t seen large increases in trade with the EU, and therefore hasn’t benefited from EU membership. They go on to argue that trade is not beneficial, appealing to a general case against “neoliberalism” and “globalization” and quoting economist Ha-Joon Chang. This quote is misplaced because it refers to protection of industry in the early stages of development rather than trade between rich countries.

What is also missing from the discussion is the fact that we do not live in a Ricardian world, where countries trade goods produced entirely within domestic borders — but in a world where production takes place through global networks and intra-industry trade dominates. The UK is embedded in these networks: the structure of its production and trade has evolved as part of the European system, reliant on and compliant with its regulations and standards. The primary barriers to trade are therefore not tariffs but “non-tariff barriers” such as regulatory standards. Had the UK remained outside the single market, either it would have complied with the ongoing development of EU regulatory structures in order to maintain and develop its pan-European and global production chains, or faced gradual disintegration with the European economy.

The issue of value chains and regulatory standards enters into conventional long-term gravity-modelling forecasting through inclusion of “non-tariff barriers” into the model specifications. The Cambridge authors carefully analyze the assumptions made in these models and conclude that the negative results are overstated — but not absent. More recent analysis, using input-output tables to account for the effects of global value chains, predicts a hard Brexit would be substantially more costly than a soft Brexit. We believe these projections are plausible, but the uncertainty associated with the forecasts is high. Leaving the Customs Union or the Single Market will inevitably require substantial structural changes and we find it hard to imagine a scenario in which there is no economic cost. Indeed, in 2017 direct investment to the UK declined to $15 billion, the lowest level since 1994, suggesting restructuring may already be underway. It is reasonable to argue that this is a price worth paying, but those supporting hard Brexit should be explicit about this.