What sort of trade policy should the UK be pursuing after Brexit?

With all the sound and fury of recent discussion of whether we should be in or out of the EU customs union, one might think this was a new question. However for me and for Economists for Free Trade, it has always been one of the key central issues to which we gave a clear answer: the EU is a highly protectionist body, and it is in the UK’s interest to get rid of this protection against the rest of the world.

Our work calculated the extent of this protection against other OECD countries at 20% for food (mainly via tariffs), and 20% for manufactured goods (mainly via non-tariff barriers).

In fact the evidence suggests it is much higher against non-OECD countries: we did a calculation based on wage costs that suggested it was more than 60% against China! (see page 197 of our book, Should Britain leave the EU?)

On the trade front there are major gains for the UK to be made by eliminating this protection, whether unilaterally or (politically more likely) through free trade agreements (FTAs), with the rest of the world.

These gains come by lowering consumer prices and so generating competition with world producers that raises UK productivity. Our calculations, based on a cautious 10% trade barrier abolition, were that GDP and welfare would rise 4% from leaving the EU customs union and abolishing these barriers.

What then to do about trade with the EU? Plainly what is best all round is continued free trade between the UK and the EU, while maintaining existing free trade in services based on mutual recognition of standards. Let us call this Canada+.

The EU would require ‘Rules of Origin’ checks on products with inputs from the rest of the world, on which we would still have to pay EU tariffs; indeed, this would be the case even if no FTA was signed.

Some argue this is costly. However, modern customs procedures are computerised and ‘virtual’; indeed this is mandated by WTO laws. When one looks at customs procedures in developed countries that have the necessary resources one finds that this computerisation means 98% of traffic goes through borders without any physical inspection while the remaining 2% takes an average of a day.

What this means is that products are entered into databases, and their nature established well before arrival, including all matters related to content and standards.

Once we are out of the single market, UK firms will only have to apply EU product standards on goods they export to the EU. Instead of general EU-based standards across everything they do, they will now be free to follow home regulations in general, and simply obey EU standards on their exports to the EU, just as they must on their exports to the US or any other country. Again this will be logged in the customs file, and provided the UK product meets the EU standard, it will pass the border unchecked.

It is therefore somewhat baffling to be told by some trade economists that there are serious costs associated with these procedures. One could imagine that there are some on-off set-up costs of registering the product on the customs system, defining its nature as above, and ensuring that the data on each load is available to be filed. But in the nature of computers, their repeated use in a known pattern is costless.

It is true that the EU has often been defended as a way of achieving ‘integration’ that gets rid of transactions costs of trade. Back in the days before the internet this was no doubt true. However, today it would seem that technology has done the job of ‘integration’ for us in a way that makes this state role redundant.

When it comes to calculating the effects of these trade arrangements these details can matter. ‘Border costs’ are similar to trade barriers in their operation. If they are essentially nil in an EU FTA, what if there is no trade deal at all – the ‘No Deal’ scenario?

Again there would seem to be no ‘border’ cost as such, for the same reasons as above: traffic from non-EU countries is inspected barely at all now. What there would be most likely are tariffs.

Non-tariff barriers between the UK and EU would be illegal since there are no grounds for discrimination, given that all products currently meet product standards and would by construction continue to do so. It is hard to see where grounds for ‘anti-dumping’ might occur.

As for tariffs, these average around 3-4% on manufactures; on food they are closer to 20%, but food is a small proportion of UK-EU trade. Plainly though we have definite barriers here and we must ask what they would do to trade.

For this we need a World Trade Model. We have one we have built in Cardiff and tested against UK trade facts by indirect inference, which it passes so we know it is fairly accurate for the UK.

We can assess the effects of ‘No Deal’ in the long term when the UK FTAs with the rest of the world have driven down UK food and manufactured prices to world levels.

Under these conditions we find that EU exporters to the UK market must match world prices to sell anything; so any UK tariff will have to be absorbed by the EU seller out of its margins.

UK exporters to the EU are driven by internal UK competition to sell at world prices everywhere, including to the EU; if any firm did get more in the UK its rivals would undercut it immediately. Hence any EU tariff will be passed on to EU consumers.

The unfortunate implication of this for the EU is that ‘No Deal’ makes it pay the UK a flow of tariff revenue, while the UK pays none because any EU tariff revenue is paid by its own consumers. Consequently the EU loses from ‘No Deal’ while the UK gains.

Furthermore ‘No Deal’ means the UK leaves without a transition, including the financial settlement associated with it. This would bring forward the effective Brexit date, including the end of the customs union, by two years.

This means the EU loses the UK financial contribution to end-2020, and also it loses two years earlier the gains it makes under the customs union from higher prices paid on its large current account surplus in goods with the UK.

The net effect of all these ‘No Deal’ effects we have calculated in present value using a 3% discount rate is a loss to the EU of some £500 billion and a gain to the UK of some £650 billion.

Because ‘No Deal’ is so bad for the EU we think there will be an EU FTA under which all such costs between the UK and EU would simply not appear. To summarise our findings for that we calculate that together with FTAs across the world the gains to the UK would be around 4% of GDP.

It is true that in recent weeks the Buzzfeed-leaked Treasury figures show a negative for this on GDP on a general world trade model from GTAP. But we know from published work that to get such a result on GTAP would require huge border costs, which the Treasury have not explained, and which we argue are not there at all.

When we put our assumptions into GTAP we get a gain of 2% of GDP. As noted our model has been tested on UK data and GTAP has not; so we would back our model results against GTAP.  But the key point of note is that whichever model you use you get a gain from the Government’s policies.

By Patrick Minford, Professor of Applied Economics at Cardiff Business School.

The views expressed in this analysis post are those of the authors and not necessarily those of the UK in a Changing Europe initiative.

View all analyses

Sign up to our newsletter

View our latest newsletter