FACT-CHECK: Are 820,000 jobs at risk as Remain claim?

A friend of mine asked me to look into George Osbourne and the Treasury’s claim that 820,000 jobs could be lost in Brexit, which he described as “blatant lies”. It does sound like a pretty extreme claim — so does it pass a fact check?

It’s worth noting here that these are economic forecasts, so obviously this is more of a “prediction check”. Is there any basis at all for this claim? If so, is it sound, does it have any huge holes in how it was made, and does it tally with other predictions from different sources?

Does The Prediction Hold Up?

Mostly, yes. The methodology is broadly sound, and it tallies with predictive studies from other sources. The Financial Times, the International Monetary Fund, the BBC, and the Bank of England broadly agree on the predictions. I’ve read through all of the 90-page report from which Osbourne draws these figures (barring appendices) and I only found one obvious problem in one section of the document. Against that, the report deliberately ignores several elements which could have made their predictions even more dire, because they couldn’t accurately model them.

(Personal note — I would very much like to find a Remain claim that is genuinely suspect, to prove that these analyses are impartial. I’m also not the biggest fan of Mr Osbourne, to put it mildly. But my read after a detailed investigation is that these figures are plausible.)

The Details

The first thing to do, of course, was to download the official report containing these predictions. It’s a 90-page brute of a report going into considerable detail, using a modelling methodology called “ vector auto-regression”, which I’m not familiar with but a bit of Googling reveals is a pretty widely used and accepted model.

I was on train wifi, so whilst I waited for the download bar to crawl along, I moved to Phase 2: the giggle test. Are the numbers presented obviously ridiculous?

The UK has 31.58 million people in work according to the latest official government statistics. In the last year that number has gone up by 400,000. Given the lack of a major economic shock in the last year on the scale of Brexit, that implies that a change of double that during a major economic change isn’t impossible.

The report passes the giggle test.

So, part 3: do those figures tally with other predictions about Brexit? 820,000 jobs are approximately 2.6% of the UK’s jobs. The London School Of Economics’ predictions (not affiliated with the government) are that Brexit under the Leave campaign’s preferred option of Free Trade Agreement or WTO membership will cost the UK about 8% of our Gross Domestic Product (total money earned by the country).

It’s fairly obvious that if UK businesses are earning less money, that will be reflected in job losses. Even if we assume that half that number will be absorbed in other ways (deferred investment and wage decreases, for example), a 2.6% loss in jobs seems reasonable given that half of the 8% prediction is higher than that.

Part 4 — what do independent analysts think of the Treasury’s paper?

The BBC’s Economics Editor wrote a detailed critique of the paper, concluding that “ it is very much worth the paper it is written on”.

The Financial Times said “ you should remain very sceptical about the report’s details”, but also saying “ These figures are not outlandish and many would say they are cautious.”. They highlight many of the same issues that I was concerned about, namely predictions of household reactions to Brexit.

The Treasury also employed an outside expert, Sir Charlie Bean (former deputy governor of the Bank of England), to vet the report. The Financial Times refers to him as “highly respected”. He found that its predictions were “reasonable”.

Finally, the predictions that are made in the report are in line with those made by the Bank of England and the International Monetary Fund.

Part 5 — The details of the paper itself.

Finally the paper had crawled down the Wifi, so I dived in. I’m not a Charlie Bean-level economic expert, but as someone with considerable experience in both entrepreneurship and online marketing (which uses a lot of sophisticated statistical modeling, at least if you’re doing it right) I can understand most of the paper even if I’m not familiar with the specific modeling technique used.

(Other people are, and apparently it’s a very well-understood and sound approach.)

The paper’s arguments fall into three categories: the Uncertainty Effect, the Transition Effect, and the Financial Conditions Effect.

I’ll take them in reverse order.

The Financial Conditions Effect argues that the financial markets are not going to love Brexit, and points to rises in volatility and falls in the value of UK-connected financial instruments that have already happened. This section is very well worked out, with a lot of detailed numbers from existing, not predicted figures. I checked one of these with independent sources, and they seem to be consistent (for example, here’s the FT on the increased Sterling volatility we’re seeing in the run-up to Brexit).

This section appears to be fairly rock-solid to me, although I’m not a financial markets expert so I might have missed something. On the other hand, the fact that most financial markets experts appear to be screaming and running around with their hair on fire about Brexit implies that I didn’t.

The Uncertainty Effect points out that there are a lot of very complicated processes that would need to be undertaken in the event of a Brexit, we have no idea how long they would take, they’ll almost certainly take longer than the time allowed under Article 50 of the Treaty on European Union, and they’ll have all sorts of interesting side-effects.

I found this part of the paper fascinating, and it highlighted a lot of elements of Brexit I’d not considered. For example, a lot of our existing legal and regulatory structure rests on EU laws, and we’d have to overhaul all that. It’s unclear what this would do to, for example, customer protection laws — would we just be missing them until new laws were passed? (I’d be fascinated to hear a lawyer’s take on this.)

The section also includes break-out segments on the impact of Brexit on individual sectors of UK trade. Obviously these are the worst-case examples, but they’re still very interesting. The UK’s agricultural sector, for example, relies heavily on EU subsidies — would they be replaced? The financial services industry relies on “passports” to offer services in the EU, and might have to urgently relocate jobs to EU nations in order to get new passports. And so on.

Again, a very convincing and interesting section.

Finally, we come to “The Transition Effect”, and this is where the wheels, at least, start to creak a bit.

They include some sensible statements about businesses delaying investment whilst they find out what’s going to happen, and households cutting back temporarily, but they also claim that “households would start to adjust to being permanently poorer”.

That seems like a specious claim to me: they offer no evidence that the British public is convinced we’d be poorer out of the EU. Obviously Remain voters are convinced of that, by and large, and they make up around half the population, but what about the Leave voters? Might they not increase their spending on the assumption that outside the EU we’ll be more prosperous, as the Leave campaign claims?

However, they also claim that households would reduce spending because of, again, uncertainty over the future. That seems more plausible.

So, one weak point in an otherwise very strong 90 page document. And overall, a pass on the fact check from almost all angles.

Sources

Sources are linked inline in the document above.