Anatomy of a scare story

The frustrating thing (well one of the many frustrating things) about the EU referendum campaign is that all too often some of the numbers spurted out by both sides simply aren’t interrogated.

There are a few eye-catching exceptions: the Government’s claim we’ll all be £4,300 worse off; Vote Leave’s EU-costs-£350m-a-week. But many of the other claims are just trotted out without much challenge.

Take Boris Johnson’s recent claim that the UK will be landed with a £2.4bn bill as a result of rising EU budget contributions. This is misleading on two counts: first, because UK contributions always rise as the broader economy expands (they are a constant chunk of an increasing pie). Second, because extra budget contributions on top of this can be vetoed by the UK.

Or take another claim I’ve spent the day trying to get my head round. This one:

This isn’t the first time we’ve heard a warning from the Remain camp that mortgage costs could go up in the event of Brexit. But it is the first time they’ve put some numbers behind it. And yet if you go to their website to try to find out how they got to this £900 figure, you’ll find just two paragraphs with no statistics and no real explanation.

Now, they did send out a press release to lucky journalists with the following bullet points:

Much of which you’ll have seen cut and pasted into news stories in the papers over the past few days. Unfortunately none of those stories point out a few critical problems with the £900 figure.

Before we get to those problems, allow me to explain how they get to that number in a few bullet points.

1. They take the calculation in the Treasury’s paper on the short term impact of Brexit that if the UK left household borrowing rates would go up by between 70 basis points and 110 basis points, which is the same thing as saying 0.7-1.1%.
2. They then take the average house price, as of March, which was £292k, they assume that people will be taking out a 76% mortgage on it (in other words a £222k mortgage).
3. They then take the average two-year fixed interest rate on a new mortgage, currently 1.87%, and worked out, using the Money Advice Service’s mortgage calculator, how much a 1.87% repayment (eg not interest-only) mortgage would cost each year (£11,123) and how much that mortgage would cost if the interest rate were 2.57% (£12,045)
4. Voila, they have a figure of a difference of £922 a year.

But there are quite a few problems with this analysis.

1. It is not a given that mortgages would go up by precisely the same amount as the “household borrowing rate” referred to in the Treasury report. Even if you buy the HMT analysis (and some economists do not), unsecured lending rates typically bounce about less than other forms of lending.
2. The average outstanding mortgage in the UK is not £219k but £101k. Use this figure as your starting point for the same sums above and you get an increase in average mortgages of £430 a year, or just £35 a month, which sounds significantly less scary.
3. The fact is that a 0.7 percentage point or even a 1.1pp increase would still leave mortgage rates well below much of their historic levels. You can see that from the following chart:

On top of this, one could also point out that were there to be a post-referendum recession, the Bank of England could plausibly cut interest rates, which is not something accounted for in the above working.

Finally, recall that the Treasury have also projected a fall in house prices in the event of a severe shock. If one is using the same methodology, that would also reduce the implied mortgage cost (lower house price = lower loan = lower mortgage repayments).

None of this is to dispute the likelihood (and it is a likelihood) that there would be at least a short term economic shock if Britain left the EU. But the Remain camp do themselves no favours by trying to depict this as a major mortgage shock, especially given the opaqueness of their sums.