Since as far back as 2016, the Treasury and other elements of the UK Government have been pushing for the UK to remain in some form of customs union with the EU after Brexit. Recent weeks have seen this idea again coming to the fore, with talk of a ‘temporary’ customs union to be created after the ‘transition period’. Those proposing this idea put it forward as a solution to what they see as the problems of Brexit. In reality, it would be in no way a solution, and it would create a whole new raft of problems. The main arguments for a customs union are that it will guarantee tariff- and quota-free access for UK exports to EU markets and that it will avoid UK firms having to bear customs and ‘rules of origin’ costs that they would face in a free trade agreement (the latter involve the costs of ensuring your product has enough ‘local’ content to qualify for zero tariffs). On top of this, it is claimed that a customs union solves the problem of the Irish border. In my view, the purely economic arguments in favour of UK customs union membership with the EU are weak: There is not much evidence that a customs union would be more beneficial for UK-EU trade than a standard free trade agreement (FTA). The study by Cippolina and Salvactici (2006), based on a large number of estimates of the trade-creating effects of FTAs and customs unions, finds no evidence that customs unions outperform FTAs. The similarly large-scale study by Head and Mayer (2013) found membership of the EU customs union had modest trade-boosting impacts (15-20%) but that these were often smaller than the trade-creating effects of FTAs such as NAFTA. Rules of origin costs are often hugely overstated. Claims that rules of origin costs for UK businesses in case of a UK-EU FTA could be as high as 7-8% of trade values are far too high. A careful study by the WTO points to the cost of compliance with rules of origin being less than 1% of traded values, and often negligible. Costs of customs processing are also massively exaggerated. Claims by HMRC earlier this year that customs costs could total 1% of UK GDP or 6% of trade values are anything from five to twenty times too high, being based on a mixture of double-counting (now admitted) and dubious claims about the future costs and numbers of customs declarations. It is not even clear that a ‘new’ UK-EU customs union would entirely remove customs-related costs. The EU’s customs union with Turkey has not led to ‘frictionless’ border trade – queues at the border are often lengthy. Formal customs checks within the EU only ended in the early 1990s due to the Single Market Programme. The UK’s foreign trade structure is not suited to a customs union. Customs union arrangements have some logic where one economy does a very large share of its trade with another. But the EU now represents only around 45% of UK goods exports. And this share has been dropping rapidly as the EU grows slowly compared to the rest of the world. Twenty years from now it is likely that the EU will take only around a third of UK goods exports. The UK would remain locked into the EU’s highly protectionist agricultural trade system. While average EU tariffs are not very high, they are often steep on agricultural products. This represents a heavy effective ‘tax’ on UK consumers, especially given the UK’s status as a large net food importer. UK consumers are denied the choice of cheap food from outside the EU and pushed towards consuming expensive products from within it. This cost is high at 0.5-1% of GDP – almost certainly higher than possible rules of origin costs for manufacturers under an FTA. Moreover, the strategic/political arguments in favour of a customs union are even less compelling: Entering a customs union would make meaningful trade deals with other economies impossible. While there might be scope for very limited deals on trade facilitation or deals on services the scope even for these would be very small. Why would India or the US be interested in a deal on services (potentially benefitting the UK) when the UK had nothing to offer on the goods side? The EU would be effectively able to ‘sell’ access to UK markets with no reciprocal benefits for the UK. The EU-Turkey customs union is a good example of this. When the EU does trade deals with third parties, these third countries gain tariff-free access to Turkish markets but Turkish exporters do not gain automatic reciprocal access to these third countries and Turkey has to try to negotiate parallel arrangements (not always successfully). Notably, Turkey came close to cancelling its customs with the EU when the EU was negotiating the TTIP trade deal with the US. Britain in a customs union would be in the same position as Turkey. A future customs union would jeopardise the UK rolling over existing EU FTAs. Under a Turkey-style arrangement, the UK would be reduced to trying to replicate any new EU FTAs with third countries to get access to the markets of the EU’s FTA partners – but these partner countries would have little incentive to agree having already got access to UK markets. For the same reason, the prospect of an open-ended future UK-EU customs union could undermine attempts the UK is currently making to transform existing EU FTAs with countries like Korea into UK-only FTAs. Current EU FTA partners would continue to benefit from free access to UK markets under a new UK-EU customs union – so they would have no incentive to negotiate new UK-only FTAs. Britain would have no voice at future WTO discussions about global tariffs. It would simply have to accept whatever the EU agreed. The EU would be able to damage UK business using anti-dumping actions. Under a new UK-EU customs union the EU would be likely to be in charge of the UK’s ‘trade defence’ measures such as ‘anti-dumping’ actions (where large tariffs are levied on countries deemed to be ‘dumping’ their goods on the EU market). Again, the EU would make anti-dumping decisions without a UK say and such decisions could damage UK business and consumers. So, engineering and vehicles manufacturers could be hurt if punitive tariffs were imposed on some steel imports, and consumers hurt if food or other imports were raised in price by anti-dumping duties. Worse still, the EU might well insist on being able to impose anti-dumping duties on the UK as well – as is the case with Turkey. A customs union would not simply cover tariffs and quotas, i.e. a ‘bare bones’ arrangement. The EU would also require the UK to follow EU rules in a broad swathe of policy areas including competition policy, environmental policy and social and labour standards – without any say at all in how these rules were set. This would not only be a huge loss of UK sovereignty but also dramatically narrow the UK government’s freedom of action in key economic policy areas. The ‘temporary’ customs union would be unlikely to be temporary. The EU has made it quite clear in recent days that it would require such an arrangement to be permanent. The proposed withdrawal agreement currently being negotiated between the UK and EU (and which would potentially contain proposals for a ‘temporary’ customs union) looks unlikely to have a unilateral exit clause, leaving the UK tied to the customs union indefinitely. A customs union does not solve the Irish border ‘problem’. Customs checks only represent a small element of potential border checks at EU borders today. A bigger issue is generally product conformity and other single market rules. This is another reason why any customs union would require either effective UK single market membership (see above) or border checks between Britain and Northern Ireland and/or Britain and the rest of the EU. In sum, a customs union arrangement whereby the UK contracted out huge areas of trade and economic policy-making to the EU would be totally unsuitable for an economy like Britain’s. Customs unions arrangements may work well for small economies that do an overwhelming share of their trade with a large neighbour (Liechtenstein and Switzerland for example). But the UK is the world’s fifth largest economy, with a diverse pattern of foreign trade and with business and consumer interests that will often diverge from those of the EU. It is no accident that Canada and Mexico are not interested in joining a customs union with the US, despite their strong trade orientation towards the US. They know that the loss of economic independence involved would be far too great to justify what would probably be quite a modest reduction in border frictions. The calculation should be the same for the UK. Supporters of a customs union have suggested the UK could somehow retain some influence over decision making in such a new UK-EU arrangement. But this looks like a fantasy. It would be legally and politically difficult for the EU to grant any significant decision-making power to the UK. The best the UK could hope for would be some kind of observer status (again as proposed for Turkey). But the arrangement would remain a thoroughly one-sided one where, at the end of the day, the UK would have no power either to veto potentially damaging agreements or push for deals that benefitted it. Entering a new customs union with the EU would be a backward-looking step for the UK, with a massive loss of policy independence and flexibility and businesses and consumers at risk of having damaging decisions imposed on them with no say in how those decisions were taken. It would be a significant downgrade from the UK’s current position whereby it retains some authority over EU trade policy. It would also give the UK minimal additional policy freedom in the trade and economic policy area. Meanwhile the benefits would be small and mostly accrue to a handful of industries (the car industry, for instance, accounts for less than 1% of UK GDP). Overall, it is hard to imagine a more sub-optimal policy.