Of UK trade, about 55% was with the Single Market in 2005, about 45% is with the Single Market now and about 33% will be with the Single Market by 2030. The fall is for various reasons, but the most important is that the rest of the world (China, India, etc) is growing faster than the non-UK EU economies and will continue to do so for much of the next 15 years.
UK exports are currently about 28% of GDP. Let’s suppose, to make the maths easy, that by 2030 they will be 30% of GDP. Then the one third or so of those exports that are to EU countries will be around 10% of GDP.
It doesn’t follow that if our exports to EU countries vanished, GDP would be 10% lower. After all, we would be likely to import a fairly similar amount, and presumably if exports to EU countries were disrupted, then imports would be disrupted, also. At a first iteration, if imports and exports fell by the same amount, GDP would be unchanged (though of course there would be other more complex negative impacts that I’ll come to in a moment). Even if exports to EU countries were impaired without any corresponding harm to imports (say if, post-Brexit, the EU raised tariffs on UK exports but the UK did not raise any tariffs on EU exports to the UK in retaliation), since the UK has a floating exchange rate, sterling would depreciate against the euro so imports to the UK would become more expensive (making them fall) and exports to EU countries would become more expensive. If those suggesting UK GDP would fall by 6% really meant “in dollar terms” (i.e. instead of one pound being worth, say, $1.50 in 2030 it would be worth $1.41) but with GDP in sterling terms being more-or-less unchanged, I might well find that plausible, but I can’t see it being much of a referendum-winning claim.
Thus, in themselves, those 10% of GDP of exports to EU countries don’t per se add anything like 10% to GDP – if indeed they add anything at all. Their main value lies in supporting imports, and the main value of those is not extra GDP (and certainly not extra jobs) but, rather, extra welfare (enjoyment, use value, usefulness) for consumers. But they are associated with extra GDP in more subtle ways. Because we trade at all (and trade with EU countries is part of that), domestic businesses face additional competitive threats, making them more efficient and making their products cheaper, higher quality and more innovative. Because the UK is part of the market for some products, firms can exploit more economies of scale, making products cheaper for our consumers. Because our firms see what is possible from foreign imports, they are driven on to do better themselves.
These benefits are important and will have an impact on GDP. In fact, they are worth a lot of GDP. It is true that much of those sorts of gains arise from the UK being exposed to trade at all, rather than its being exposed to trade with EU countries per se. And it is also possible that some such gains top out – that, for example, at some point a market is as close to perfectly competitive as human technology allows, so being exposed to even more trade won’t add anything further.
But, even so, the EU is a big chunk of our trade and by 2030, one third of our trade will still be a big chunk. So even though the direct GDP gains from trade with EU countries are likely to be small, and even though the main benefits aren’t GDP at all, it would not be unreasonable to guesstimate that that 10% of GDP of exports to the EU might be associated with 10% extra GDP.
So if trading with EU countries (note, I say, “trading with EU countries”, not “being in the EU”) is worth 10% of GDP, what proportion of that is it reasonable to believe could possibly be lost if we left? The Treasury says GDP would be 6% lower by 2030 if we were outside. Let’s imagine that were achieved purely through harming the GDP contribution that comes from trading with EU countries (it doesn’t, of course; nothing like it – which is precisely the point I want to help you to see). That would mean that leaving the EU would mean the UK’s losing some 60% of the entire economic contribution made from trading with EU countries. Even if that entire economic contribution were roughly equivalent to the sum of imports and exports – so 20% of GDP not 10% – the notion is that leaving the EU would cost of 30% of the entire value of trading with EU countries.
Would that make any sense at all as an estimate? Think about it. The average trade-weighted tariff applied by the EU is 1%. WTO tariffs average 4.4%. So if the UK imposed no tariffs on imports from the EU and imposed an extra UK-wide tax to offset tariffs imposed by the EU (yeah, yeah, WTO rules, blah-blah – forget that for now and concentrate on my thought experiment), if exports to the EU would be 10% of GDP that would cost between 0.1% and 0.44% of GDP (since 10% x 1% = 0.1% and 10% x 4.4% = 0.44%). Not 6%. 0.1-0.44%.
“OK, but what about non-tariff barriers?”, I hear you cry. Well, let’s think about those. Some of the most extreme non-tariff barriers imposed by the EU on friendly countries are those imposed on US car exports to the EU. The US auto sector pays 10% tariffs and about 25% tariff-equivalent after taking account of non-tariff barriers. Let’s imagine a near-total breakdown in diplomatic relations with the EU post-Brexit, with all UK sectors ending up facing as bad tariffs and non-tariff barriers as US auto exporters. That would be 25% tariff equivalents on that whole 10% of exports. So to offset that, the UK government would need to impose taxes equivalent to 2.5% of GDP. Let’s imagine those taxes created huge additional distortions, making their total negative impact half as much again as their scale, so GDP was actually 3.75% lower. More modest variants (a merely disastrous breakdown in relations resulting in 10%-15% tariff-equivalent of non-tariff barriers) might be only half or so of that – perhaps 2% of GDP.
[UPDATE: I note that the European Commission estimates that the gains to EU GDP from the Single Market have been around 2% of GDP (see here, p13) – precisely in line with my figure here. It is, of course, highly implausible that by leaving the Single Market one could lose the entire gains from being a member. Paul Krugman’s estimate of the trade losses from Brexit is, likewise, 2 per cent – in that case based on the notion that Brexit will lead to a drop in UK exports from 30 per cent of GDP to 25 per cent, apparently based on the idea that Brexit would reverse most of the trade creation gains the UK made from joining the EU. So, Krugman says that the UK’s proportion of its trade with the EU rose from around one third before it joined the EU to around 50 per cent today, which he believes is all trade creation. So if non-EU trade was 67 units and EU trade was 33 units, and joining raised EU trade to 67 units, total trade is now 134 units. If EU trade dropped back to 33 units, that would be a fall of 34 units or 34/134 = 25 per cent. If total trade is 30 per cent of GDP, a fall in that trade of 25 per cent would be a drop to 22.4 per cent of GDP. Krugman’s drop to 25 per cent of GDP therefore assumes Brexit means the loss of around two thirds of the trade creation gains from EU membership.]
So you see that in even disastrous and wildly implausible scenarios where there is a near-total breakdown in relations with the EU post-Brexit, we can’t get anywhere near the Treasury’s 6% GDP loss from Brexit by 2030. We can barely get half way. And that’s before we even broach the really big questions of whether there would be any negative consequences of Remaining (there would) and whether there would be any, even partially offsetting benefits from leaving (there obviously, obviously would and it’s incredible nonsense to deny it).
The lesson I want you to draw from all this is that it simply isn’t plausible that leaving the EU could cost us 30%-60% of the total GDP value the UK gets from trading with EU countries. And of course that isn’t how the Treasury gets to its 6% of GDP loss either.
What the Treasury assumes is that if the UK leaves the EU, we will make what the Treasury regards as economically destructive decisions, such as raising tariffs on imports to the rest of the world and not entering into any new trade deals with new geopolitical and economic partners. The allegedly bad economic consequences of Brexit are created, in significant part, by us. That’s the only way Brexit could get anywhere near causing a 6% loss of GDP. Without us making stupid economic mistakes following a Brexit, it just isn’t plausible that, in terms of impaired trade with EU countries, it could cost us more than a percentage point or two of GDP (even before one started to consider offsetting gains).
Well, if these economic decisions the Treasury assumes we would make are so bad, might I humbly suggest we don’t make them?
The Treasury assumes that the UK’s economic openness would decline as a consequence of Brexit. Over the long-term, a country’s economic openness can be seen as the product of five key drivers:
- geography relative to transport technology – i.e. how easy it is to get stuff to and from the country, given the transport technologies of the day;
- comparative advantage – ie how tradable the stuff is that the countries is best at making, relative to everyone else;
- the general appetite of other countries to trade with that country;
- whether geopolitics means special sanctions are applied upon trading with that country;
- that country’s appetite to trade with everyone else.
The only one of these options that could plausibly be changed much by Brexit is the last – our appetite to trade with others. Many commentators do assume that Brexit would be an act of isolationism, a choice to cut ourselves off. I don’t see it that way at all – indeed, I see no reason to believe it would be likely to lead to that. I believe Brexit is about cooperating with non-EU partners more intimately and with the EU a little less intimately – i.e. a change in the mix of our partnerships, not ceasing to have partners. And I think it’s pretty obvious that the dominant political forces in the UK would see things my way post-Brexit and that isolationist forces would have marginal, if any, impact.
The great likelihood is thus, in my view, that the UK would have about the same openness over the long-term if we left the EU as we have now. Modelling exercises which attribute large economic costs to our choosing to cut ourselves off, post-Brexit, are nothing more than mathematical tilting at windmills. The monster they fear simply doesn’t exist other than in the fevered minds of Remain campaigners.
Meantime, the option these models assume of the UK being able to continue as we are doesn’t exist either. It’s important to grasp that the difference between mainstream pro-Brexit economists such as myself and the Treasury or OECD is not that we have different economic theories. In, say, the austerity debate, many people were probably dimly aware that there were Keynesian and non-Keynesian theories about the impacts of deficits. There is no equivalent of that theoretical debate in the Brexit discussion. I use gravity models all the time. The difference between myself and the Treasury is nothing to do with economics, per se, at all.
Where we differ is in what we believe to be plausible political outcomes from Remaining and plausible political outcomes from Brexit.
Here’s a big difference: Remain assumes that the EU can carry on for the next 15 years pretty much as it has done for the past 15 years. I do not. I believe that, because it is what the EU has done for 60 years, because it is what the EU says it will do (e.g. in the Five Presidents’ Report), because it is regarded as an existential necessity in the light of the euro crisis and migration crisis, and because it is such a cool idea in its own terms, the Eurozone will politically integrate much more deeply over the next 15 years. As a non-member of the euro and Schengen area, the UK will become increasingly peripheral to EU decision-making, with the Eurozone caucasing to force through Single Market rule-changes that serve the needs of the Eurozone and Schengen areas and in which the UK has relatively little interest but which will affect us nonetheless. Those who said that if we didn’t join the euro our interest and influence would decline were right. It will.
The consequence will be that within 15 years the UK’s position as an EU member (if indeed – which I don’t believe will happen, in fact – it were permitted for us to stay formally an EU member for that long without joining the euro), would be that we will have the rules set for us by others. Our status will be as per the Treasury’s “Norway” or “EEA” option – bound by the Single Market rules but not able to influence them. (For technical reasons that won’t concern us here, if in fact the UK left the EU to join the EU it would not have the Single Market rules set by others without being able to influence them, but we don’t have space to explore why.)
Thus, much of – perhaps as much as half – the Treasury’s modelled loss of 4% or so of GDP if we took up the EEA option would, actually, in my view, be what happened to us if we remained in the EU, while if we left our net position would (for the reasons explained above) be fairly close to today’s.
More generally, I believe that outside the EU we would enter into new trade agreements. There would be two forms of these.
First, there would be new trade deals that the EU finds it difficult to do because, although its size give it weight in trade agreements, the number of players (e.g. the 28 members) creates complexity in sealing the deal. We have seen recent discussion that the TTIP deal with the US might be vetoed by the Austrians. The agreement with Japan is bogged down by disagreements between Japan and Germany. The Australians argue with the Italians and Poles. The Canadians argue with the Greeks and the French.
It’s worth observing that the European Commission estimates that seven trade deals the EU has under negotiation (with the US, Japan, Canada, ASEAN, India, Mercusor and China) would be worth about 2% of GDP – approximately equal to our conservative estimate for loss on EU trade from Brexit. Several of those are unlikely to happen with the EU – the Japan and US ones are probably gone for now, and the Canada one seems to be in trouble. (The India one is also intrinsically unlikely but irrelevant for our purpose since the UK probably wouldn’t get one with India either.)
Outside the EU, the UK would make quicker deals. It might finalise two generations of deals in the time the EU takes to make one.
A second set of new deals would be new preferential agreements – e.g. a new customs union – with new economic partners. I would like to see a new deal with Canada and Australia, but if we didn’t get that there would be something else.
I think it unlikely that, at least in the short-term, these new free trade deals or new preferential partnership deals would be worth as much, per unit of trade, as the trade arrangements we have with our EU partners. But since there would be twice as much trade with the rest of the world as with the EU, we wouldn’t need them to be worth as much per unit of trade. Even if they were worth only half as much per unit of trade, that would offset the entire loss, in trade terms, of leaving the EU.
Given that EU tariffs are very low and that the EU has a regulatory philosophy greatly influenced by us and similar to ours, and thus the burden of complying with EU non-tariffs barriers would be low for us (they’d be requiring our businesses to do what our own government would probably want those businesses to do anyway), the notion that gains on non-EU trade could be half as much as losses on EU trade seems eminently plausible. Perhaps in the end we’d get nothing, but there’s every reason to think it plausible that what we’d get might be worth as much as what have now. I’ll assume we might get 0%, but we might also offset the whole of that 2% or so plausible maximum loss we set out earlier from reduced trade with the rest of the EU.
If I’m right that such new deals could place the UK in roughly our current position, whereas by Remaining we would suffer a “Norway”-type fate inside, losing perhaps half the Treasury’s 4% estimated GDP losses, that would mean Brexit would gain us around 2% of GDP in trade effects. I suspect, however, that the Treasury model over-estimates that, so I’ll content myself with around 1% – half that half the Treasury’s Norway option effect. Perhaps things wouldn’t be that bad. Perhaps we’d lose only half that much again. Let’s take a range of 0.5% to 1% loss from Remaining from this factor.
There would be other Brexit gains, as well. I’ve explained previously that by getting us out of the way, Brexit would allow the Eurozone to take control of EU institutions to address its governance issues, allowing it to grow faster – to our benefit. I’ve estimated that by 2030 that could be worth 0.3% of GDP.
Although we might lose something in regulatory terms from facing additional non-tariff barriers to trade with the EU, the flip side is that we would be able to set our laws for ourselves. That could allow us to make our laws more focused on our own needs rather than on the average needs of the EU. Perhaps more importantly, it would allow us to devise new regulations in the future, to address tomorrow’s regulatory challenges, with the possibility of experimenting and then u-turning if we get it wrong.
If what you want to achieve in regulatory terms is pretty clear – if you are trying to get tariffs and non-tariff barriers down, to develop free trade, to liberalise markets, to oppose state aid, to create competition – there are policy economy advantages in having a ratchet process whereby once measures are in place it’s very difficult to undo them. The EU’s decision-making is designed like that – with good historic reason. That’s been effective for much of its history, but the challenges of tomorrow are different – the sharing economy, e-cigarettes, climate change – and the best regulatory solutions are not obvious in advance. The EU is likely either to regulate too early, get it wrong, and be unable to back-track, or, for fear of binding itself in and being unable to back-track, failing to regulate early enough.
Better regulation can add decimal points to annual GDP growth rates. If it’s really true that by leaving the EU we could regulate better for tomorrow’s challenges, an estimate that that might add 0.05% to annual growth for a decade – or around 0.5% higher GDP by 2030 if we left the EU in 2020 (which is when I believe we would leave), would be conservative. If we didn’t get the hang of the new regulatory possibilities for the first five years post-Brexit, we’d only get half of that – so, 0.25% extra – by 2030. We might get nothing, but let’s take a range of 0% to 0.25% by 2030 as our gains from this source.
Additional to all of this would be the around 0.5% of GDP, per annum, net contribution the UK makes to the EU. I have every confidence that UK governments would find ways to spend that domestically that I would disapprove of and thus I don’t usually like to talk about the repatriation of our EU contribution as a significant Brexit benefit. But if we are on the topic of GDP, it’s pretty clear that if more of that were spent domestically that should count – at least as a source for some of those EU tariff-offsetting subsidies I mentioned earlier. It might even have a multiplier effect (i.e. 0.5% of GDP spent here might produce more than 0.5% in extra GDP). Let’s stick with assuming that, after some of it ended up leaking abroad and some of it was spent in ways that would reduce GDP rather than enhance it, the multipliers applied to what was left would take the net impact to between 0.25% and 0.5% of GDP – i.e. that somewhere between half and all the contribution would be added to GDP.
Thus, overall, we have the following GDP benefits from Brexit by 2030:
- +0.5% to +1% of GDP from avoiding being in the position of being over-ruled on Single Market regulations if we Remain.
- -2% of GDP from having tariffs and non-tariff barriers equivalent to 10%-15% tariffs imposed on our exports to the EU, post-Brexit.
- 0% to +2% of GDP from new free trade deals and new preferential trade agreements (e.g. a new customs union) with new geopolitical partners.
- +0.3% of GDP from the Eurozone growing faster once we were out of the way.
- 0% to +0.25% of GDP from regulation more focused on our needs and being better able to experiment and u-turn when we face new challenges.
- +0.25%-0.5% of GDP from the repatriation of the UK’s EU contribution.
So, our estimate is that by 2030 the net impact of Brexit would be between a loss of 0.95% of GDP (i.e. a small but non-trivial loss) and a gain of around 2% of GDP, depending mainly on how we do with new trading arrangements with new international partners and on how badly our position would deteriorate within the EU if we Remain.
This seems to me to be the sensible order of magnitude for this discussion. The outer end of the loss scale sits at us losing around 10% of the total GDP gain we make through trading with our EU partners (irrespective of whether we do so as EU members). The upper end of the gain scale assumes the politics works well for us, in terms of new trade deals, that the Eurozone seizes the opportunity Brexit would afford to integrate more and take control of the EU institutions, that our repatriated EU contribution is spent reasonably fruitfully and that we make wise regulatory choices with our new flexibility.
(There are other factors as well. I don’t believe there would be much enduring change to immigration from Brexit, other than some change in the mix of who comes, so that won’t have much economic significance. I believe Scotland would be less likely to leave the UK if we Brexit – indeed, I fear that if we Remain Scotland is more likely than not to leave. I could have added some GDP loss to the Remain scorecard for this, but I have excluded that effect here.)
None of this is guaranteed. We could horlicks the whole thing up – elect communists or isolationists or engage in some other madness post-Brexit.
It’s also worth noting that in my view even the upper end of the gains scale – the +2% scenario – would not by itself justify Brexit, especially as I expect we might have temporarily lost growth of 2% or so of GDP in the short-term in around the 2019-2022 period. The economic case for Brexit should not be that there would be large economic gains. The EU is not mainly an economic project. The main good things about it have not been economic, and the main bad things aren’t economic either. The reason for leaving isn’t economic – it’s the large constitutional, self-determination and geopolitical gains we could make by leaving that are the real prize from Brexit.
The economic case for Brexit is not that we would make large economic gains. It is that there would not be such large economic losses that it would not be feasible to seize those big constitutional, self-determination and geopolitical gains without disaster. I hope I have demonstrated here that by 2030 one should expect the economic consequences from Brexit to be roughly balanced – somewhere between a loss of 1% of GDP and a gain of 2% of GDP. That means we can and should seize those – hugely important – other gains by a Leave vote on June 23rd.