Economy & trade Project Fear Civil service Blog Featured

A new prime minister must end the Treasury’s anti-Brexit propaganda campaign

HM Treasury
Written by Harry Western

The UK Conservative Party has now begun a search for a new leader, and thus a new British Prime Minister. Who may ultimately land this job is far from clear, but one thing is very clear: any new PM serious about delivering a genuine Brexit must end the anti-Brexit campaign that the UK Treasury has been waging since 2016 – and repudiate the analyses the Treasury has produced to date.

The UK Conservative Party has now begun a search for a new leader, and thus a new British Prime Minister. Who may ultimately land this job is far from clear, but one thing is very clear: any new PM serious about delivering a genuine Brexit must end the anti-Brexit campaign that the UK Treasury has been waging since 2016 – and repudiate the analyses the Treasury has produced to date.

These analyses have been characterised by circular reasoning, double counting, the selective use – and often misuse – of academic evidence, the deliberate calibration of models to produce very negative results, and wildly skewed assumptions. The intention has been, in every case, to produce the biggest negative effects possible for ‘clean’ Brexit scenarios (i.e. WTO- or free trade agreement-based Brexits) by exaggerating the downsides and minimising or assuming away entirely any upsides.

At least one of these analyses has already been totally discredited by events. Ahead of the 2016 referendum, the Treasury produced an analysis of the short-term economic effects of a vote to leave the EU. This analysis claimed that such a vote would ‘…cause an immediate and profound economic shock’ plunging the UK into a recession and leading to unemployment rising steeply.

In fact, nothing of the sort has happened and it is worth comparing the key predictions of the Treasury report with the actual outcomes:

  • The UK would fall into a recession, and over two years GDP would fall by 4-6% below a ‘baseline’ level (i.e. the level if the UK voted to stay in). In fact, there was no recession and the UK economy grew by 1.8% in both 2016 and 2017. At most, what was seen was a very moderate economic slowdown and even this may have been unconnected to the vote.
  • Unemployment would rise by 500,000 to 800,000. In fact, unemployment fell by 280,000 in the two years between Q2 2016 and Q2 2018.
  • House prices would crash 10-18%. In fact, house prices rose by 8%.

As well as noting that the Treasury was totally wrong in their forecasts, it is worth noting why. It is because the whole analysis was reverse engineered, with assumptions deliberately chosen that would create, when plugged into the Treasury’s model, very big negative effects on the economy. In particular, the Treasury assumed the UK would suffer a rise in ‘uncertainty’ on a scale only generally seen in recessions. This assumption, unsurprisingly, generated a recession. It was a classic example of circular reasoning and about as far from rigorous economic analysis as you can get.

Similarly questionable techniques and assumptions are visible in the Treasury’s other analyses, in particular its analysis of the long-term effects of Brexit and its analysis of the costs of customs processes outside the EU customs union.

2016 also saw the Treasury produce estimates of the long-term effects of Brexit on the UK economy. Once again, the headline figures were lurid: GDP would be 6-8% lower in the long-term under a ‘clean’ Brexit compared to a ‘Remain’ baseline, UK exports to the EU would halve.

But once again, these numbers can be exposed as greatly exaggerated. As Gudgin, Coutts et al. have shown, this study had serious methodological flaws. In particular, the ‘gravity model’ the Treasury used estimated the impact of EU membership on exports across all EU members and then applied this to the UK, despite the UK’s very different trade structure. This and other flaws hugely biased the results, so that the long-term fall in UK exports to the EU estimated by the Treasury after a ‘clean’ Brexit was at least three to four times too high. Strikingly, the Treasury itself had suggested just a few years earlier that the effect of EU membership on UK exports to the EU was less than one-tenth of the size they claimed in 2016. Oddly enough, the Treasury no longer refers to this earlier study.

Moreover, the Treasury’s long-term study also tacked on a huge additional effect from weaker UK productivity growth. This was based on the notion of a direct causal link between trade and productivity. But this notion has no solid empirical foundation: the economic literature the Treasury drew upon to justify it is inconclusive, with much of it suggesting either no clear link or that any link operates in reverse, i.e. from productivity to trade. The Treasury must have known this yet took the decision to effectively double their estimates of UK GDP losses under ‘clean’ Brexit scenarios using this highly dubious trade-productivity link.

Fast forward to early 2018 and another version of the long-term analysis materialised, dubbed the ‘Cross-Whitehall Briefing’. Interestingly, this analysis saw the Treasury abandon the ‘gravity model’ approach previously used, which had come in for considerable criticism. But remarkably, despite using a different model structure (this time a general equilibrium type, more common for trade studies) the new analysis managed to come up with almost exactly the same numbers: UK GDP was again 5-8% lower in the long term under ‘clean’ Brexit scenarios.

Yet again, this new analysis featured a mass of wildly pessimistic assumptions and the new model was clearly specifically designed to produce big negative numbers. The Treasury’s estimates of the size of ‘non-tariff barriers’ to trade that UK exporters to the EU would face outside the EU customs union and single market were absurd: at up to 30% of trade values, these were up to ten times higher than the 3-8% range of estimates seen in Kee & Nicita, and other recent model-based work. The big (and unjustified) productivity effect was also there again, this time partly disguised within the structure of the model.

As well as exaggerating the possible negatives of Brexit, the ‘Cross-Whitehall Briefing’ also minimised any positives. For example, it claimed free trade deals with the US and other fast-growing economies might add at most a few tenths of a percent to UK GDP. The Treasury had previously totally assumed away any gains from new UK trade deals, so this might be seen as an improvement. But the EU’s own estimates for such deals suggest long-term gains of 1-2% of EU GDP, up to ten times higher than the cross-Whitehall briefing. Similarly, possible gains from deregulation were also assumed by the Treasury to be minimal, despite the EU’s own estimates suggesting that the worst EU regulations cost the EU up to 4-6% of GDP.

Arguably, the Treasury’s propaganda campaign against Brexit reached its nadir in May 2018, when the head of HMRC (part of the Treasury) claimed leaving the EU customs union would cost UK firms £20 billion per year, or 1% of GDP. Careful analysis of the HMRC calculations revealed them to be shoddy. They featured double-counting, dubious assumptions, misreading of the evidence and exaggerated claims about rules of origin costs. Recent scholarship and examples from real-world firms, customs brokers and port authorities reveal the costs of rules of origin compliance and customs processes to be relatively low and a sensible reading of the available evidence in this area suggests HMRC’s estimates of customs-related costs were ten to twenty times too high.

Overall, what we see since 2016 is a clear pattern of the Treasury systematically exaggerating the costs of ‘clean’ versions of Brexit while denying the upsides. The result has been estimates of economic losses from Brexit that defy all common sense. As we have previously pointed out, the long-term losses the Treasury have come up with are equivalent to the worst estimate you can possibly make for long-term UK GDP losses resulting from the Great Depression of the 1930s – the most enormous financial and economic shock of the last century.

That the Treasury might be an aggressive supporter of the economic status quo is no real surprise. It has been so for most of the last century, and this has often put it on the wrong side of many of the great economic issues of the day. It was wrong about the gold standard in the 1920s and 1930s, (dogmatically) wrong about fiscal expansion during the depression, wrong to try to keep sterling at an overvalued level in the 1960s, and wrong about the ERM in the 1980s and 1990s.

But the Treasury acting in a nakedly partisan political manner and thus departing entirely from its supposed job of providing unbiased advice to ministers, is novel and unwelcome. It is quite extraordinary that the government allowed the Treasury, after the 2016 referendum, to effectively continue to campaign for Remain and against Brexit – or at least any Brexit worthy of the name. The fact that it did so is, unfortunately, rather strong evidence that the May administration was never serious about delivering the kind of Brexit outlined at the Lancaster House speech in 2016.

By the same token, any new UK prime minister who claims to be planning to deliver a genuine Brexit must repudiate the Treasury’s analyses since 2016 and put together a new economic vision for Brexit, recognising and planning for both the potential upsides and downsides – in a balanced way. We await any such efforts with interest.

About the author

Harry Western