Leaving the EU: why do the economic forecasts differ?

Published: 7th Jun 2016

What do the economic forecasts say?

The EU referendum debate has seen a variety of claims about the future. Most (but not all) forecasts predict that leaving the EU would cost the UK economy and be bad for public finances. There’s no agreement about exactly how much the cost would be or even why it would happen.

The differences between the forecasts stem from different judgements about uncertainty, the exchange rate, trade, investment, regulation and migration. It also matters what kind of model is used.

Economic models are far from perfect, but they can contribute to a bigger picture incorporating a whole range of factors. It is important to understand what assumptions it makes, how these are weighted, and what data has been used.

We’ve also published a guide to what models can (and can’t) tell us in to the lead-up to the referendum.

Most economic studies predict a negative effect on economic growth and UK public finances

In the run-up to the referendum, most economic research bodies have predicted that leaving the EU would mean slower economic growth. Two exceptions are Economists for Brexit, who predict a net benefit from the UK pursuing free trade policies outside of the EU, and Open Europe, who predict that it could result in lower or higher growth depending on future circumstances.

If you agree that leaving would slow economic growth, then leaving would also have a negative effect on government finances. The idea is that an economy which grows less quickly will generate less tax revenue, whilst also requiring greater government spending. The IFS recently said that deeper cuts or more time would be needed to eliminate the deficit.

Most studies predict that these negative effects will exceed the benefits of not having to pay the EU membership fee.

But economists do not agree how big the effect would be

There are at least twelve studies that predict a negative effect on long-term economic growth, but their estimates range between a potential loss to growth of -0.1% and -7.9%. (Last month’s Treasury Report was toward the more negative end of these estimates).

One study predicts a positive effect from leaving, which will increase growth by 4.0% in the long term. A final study predicts either a positive or negative effect, depending on what kinds of policies are implemented afterwards.

Very few of the studies offer only one prediction. Within each individual study there will usually be a range of ‘optimistic’ and ‘pessimistic’ forecasts, depending on the exit scenario which is being considered. This creates even more confusion in the media, since often only a single figure is reported – perhaps the most extreme number, or an average of all the scenarios.

So why do the negative estimates differ? 

1. It matters how far you think the exchange rate would fall after a vote to leave, and how you judge the impact of uncertainty during the exit process

In the short term, most economists agree that exit process itself would have a negative effect on economic growth. (The Bank of England has said that the debate surrounding the referendum is already slowing UK economic growth).

Businesses and consumers are likely to put off spending decisions until they know more about what will happen next.

Speculation could lead to a fall in the value of sterling, coupled with people’s preference to hold money in companies and currencies outside of the UK. This will make imports (and holidays abroad) more expensive in the short run, and exports cheaper for foreign buyers.

It’s harder to put an exact number on both these figures, or the level of ‘uncertainty’ in the first place.

There are lots of methods economists do use to transform ‘uncertainty’ in to a specific number. One approach is to estimate its effects based on on past shocks such as the 2008 financial crisis, or the 2010-11 Eurozone crisis. Another way is to compare the price of assets which act as insurance. More creative methods include counting newspaper articles or social media that mention policy uncertainty. Any method will be, to some degree, a rough-and-ready estimate.

2. It depends on what sort of deal you think the UK would be able to negotiate after leaving the EU

Most studies consider more than one possible scenario after leaving the EU. This means that most studies produce a range of possible outcomes, rather than a single estimate.

A common range of scenarios include: joining the European Economic Area (like Norway), striking a free trade agreement with the EU (like Switzerland), accepting World Trade Organisation tariffs on imports and placing no tariffs on exports, or facing WTO tariffs on imports and exports.

Looking at the ‘average’ impact given by the study is probably less useful than deciding which kind of trade deal you think is most likely. Then you know whether you should be looking at the most ‘optimistic’ or ‘pessimistic’ prediction.

3. It depends on whether you think trade would be improved inside or outside EU

Economists tend to agree that trade is good for economic growth. Some argue that extra competition from foreign firms makes UK businesses more productive.

The key to the debate is whether trade will be easier inside or outside of the EU.

The majority of studies assume that trade will be easier within the EU. They judge that membership makes it easier to access European markets, and EU regulations make it easier to export goods and services.

NIESR’s estimates lie roughly in the middle of these studies. They predict a -10.5% reduction in the value of trade by 2030, in the most “optimistic” scenario that UK join the European Economic Area (like Norway). They estimate a -29.7% reduction in trade in the most “pessimistic” scenario that the UK makes no new trade deal and faces World Trade Organisation tariffs.

Obviously, the reality of renegotiation would be much more complicated than either of these scenarios and it would be a mistake to treat NIESR’s forecasts as an exact guide to the future. Instead, it’s better to see them as representing a combination of judgments about what could happen, and the possible effect when all the factors combine together.

Two studies argue that the UK could see positive benefits from leaving the EU in the long term. Economists for Brexit think that the EU reduces free trade with the rest of the world, and advocate leaving the EU to pursue free trade policies outside the economic bloc.

Open Europe suggests that, under certain conditions, the economic cost of leaving the EU could possibly be offset by liberal policies which encourage trade and competitiveness. They emphasise that a positive outcome is far from definite for political and economic reasons (important measures would include, for example, policies that allow a reasonable level of immigration).

4. And there are similar debates about the impact of membership on foreign direct investment, regulation and migration

Economists who predict a negative impact from leaving the EU say that EU membership encourages foreign direct investment, reduces non-tariff barriers through regulation, and encourages immigration.

The two studies which predict a (potentially) positive impact from leaving the EU argue that the EU hampers foreign direct investment, imposes regulatory barriers, and enables the wrong kind of immigration for the UK economy (encouraging low-skilled, rather than high-skilled labour to immigrate).

Once again, all the studies largely agree on what they think is good for economic growth (although not all economists do). The differences come from whether they think these things would be easier to achieve outside or inside the EU.

These assumptions are then fed into different types of model

Once economists have chosen and weighted their assumptions, they combine them together to produce an overall prediction.

These models vary in scope. Models which incorporate the effects which would be felt across the whole world tend to predict a stronger negative impact than models focused just on the EU, or solely on the UK. Models which consider how leaving the EU would affect the productivity of the economy also tend to predict stronger impacts.

Often, a single study will provide several different estimates which incorporate wider or narrower ranges of factors.

Different models combine theory and history in different ways 

We promise this is interesting.

Most modern models use some combination of economic theory and patterns spotted in data from the past. NiGEM, for example, is a widely used model that can incorporate both.

Focusing on history has some advantages. Statistical relationships in past data reflect a whole host of factors including the size of different economies, geographical distances, different politics, regulation and culture.

Using economic theory also has advantages, because it helps us to predict how businesses and consumers might behave when lots of factors change at once.

The Liverpool Model used by Economists for Brexit is unusual in that it doesn’t incorporate statistical relationships found in the past, and makes its prediction only based on economic theory (in particular, the advantages of unilateral free trade). This makes it unique among all the models used in this debate, as well as the only study that predicts an unambiguously positive result.

The most important thing to know is what assumptions went into the model

When a model makes a prediction, and produces a number, it is based on a combination of judgments about what might happen.

The important thing is to see what stories lie behind each prediction, and decide whether you find them convincing.

 

Update 9 June 2016

We updated this article to clarify the position of Open Europe. It predicts that leaving the EU could result in higher or lower growth in the long term, depending on the circumstances. Open Europe's full reports can be read here and here.


Featured

Budget forecasts: take them with a pinch of salt

We aim for our factchecks to be as accurate and up-to-date as possible. If you think we've made an error or missed some relevant information, please email team@fullfact.org.

Tweet

Share