Forecasting GDP

The OBR’s downbeat forecasts for the economy have caused Brexiteers to question its competence. For example, Iain Duncan Smith says it has been “pretty much been wrong on everything.” Such attacks miss two important facts.

Fact one is that there is a crucial difference between unconditional and conditional forecasts. A conditional forecast is the sort that says: “if you put your hand in the fire you’ll get burnt.” An unconditional forecast says “you’ll get burnt”. These are obviously two different things.

Personally, I think unconditional forecasts - the sort that say “GDP will grow 1.4% next year” - are pretty much worthless and simply bring economics into disrepute.

But this isn’t the issue here. Instead, Brexiteers’ beef is with the OBR’s conditional forecast – with its claim that:

Over the time horizon of our forecast any likely Brexit outcome would lead to lower trade flows, lower investment and lower net inward migration than we would otherwise have seen, and hence lower potential output.

Is this claim wrong? Attacks upon the OBR’s record at unconditional forecasts are just childish ad hominem arguments that are irrelevant – not least because the OBR’s view is shared by so many others.

Better arguments would be that Brexit would permit freer trade (as Gerard Lyons still believes) and that this, plus sterling’s fall, would boost GDP growth.

Personally, I doubt this. But what’s the evidence?

You could argue that today’s figures showing a better than expected rise in business investment in Q3 is consistent with Brexiteers’ optimism; it might show that industry is indeed tooling up in anticipation of higher demand.

But here comes my second fact. It’s not just pointy-headed experts and elites that are gloomy about the economy. So too are the people.

The thinking here is simple. Households’ decisions on how much to spend should be forward-looking: if we anticipate good times, we’ll spend more than if we expect bad.

And this thinking is correct, to some extent. My chart shows that the ratio of consumer spending to house prices (a rough proxy for spending-wealth ratios) has predicted medium-term GDP growth in the past. Low spending in 1973, 1979, the late 80s and mid-00s all led to slower GDP growth, and high spending in the mid-90s led to decent growth. Since 1971 there’s been a correlation of 0.63 between the consumption-house price ratio and subsequent five-year growth in real GDP.

Which brings me to the punchline. This ratio is currently below its post-1971 average, which points to below-average GDP growth. Not catastrophic, note – merely below average. In fact, if post-1971 relationships hold, this ratio predicts GDP growth of 1.8% per year until 2021 – which, coincidentally, is exactly the same as the OBR forecasts.

Now, you can quibble with this, not least because there is of course a margin of error in this relationship. Brexiteers might argue, I suppose, that because consumer spending is partly shaped by habits it hasn’t yet fully responded to the good news of the Brexit vote. Such a hypothesis implies that spending should rise a lot in coming months. Let’s see.

For now, though, this leaves Brexiteers with a problem. It’s not just the experts they so despise that are downbeat about the economy. So too are those founts of all wisdom, the people.  

Comments

[Magnus]

This is clever, but to be fair you'd need to know how down/upbeat they'd have been under Bremain. I suspect your indicator might have been a little bit better but not massively different.

[joe]

Business investment for 2016 Q3 was probably planned in 2015. There was no reason to cut them on June 24th as it will be years before leaving EU, if ever. 2017 on should be horrendous for brand new investment.

Consumers have had little spending power since 2008 crash, though some small pay rises are coming through. They are aware inflation is going up and the big spend essential items they have put off buying are best bought asap. They are buying on credit cards - hoping for the best to pay the cards off. Smaller retailers are already really struggling. Once these big items are out of the way retail spending will rapidly wane.

I do wish these Brexiters would have the courage to produce a forward financial plan of their own for UK, so that their views could be exposed, once and for, all as totally ridiculous.

[am]

The OBR forecast 2.1 growth for 2016. Two days later, Friday, the annualised rate was 2.3 at end of q3 2016. It doesn't help folks credibility when their forecast, or part of it, looks wrong after only two. You probably haven't read their document but some of their other 2016 forecast are equally bizarre including inflation at 0.8 whereas most would think it would go above the current 0.9.
An important point is what is their cut off date for submission of their projections and how does it relate to time of published hard data. Bit like the OECD today: they seemed to have missed a lot of ONS published hard data.
IF you can't read the ONS website then you just won't have much credibility.
brexitforecastmacro = inability/unwillingness to read ONS website.

[Blissex]

«"lower trade flows, lower investment, lower net inward migration than we would otherwise have seen, and hence lower potential output."

Is this claim wrong?»

What a clever oxonian technique!

Of course any extra worker that gets added to the workforce adds to its "potential output": it is a truism; and the same for investment and imports.

Any increase in available inputs in practice adds to "potential output", particular emphasis on "potential". But any increase in inputs is likely to add to actual outputs.

But this is little more than sophistry. What we should ask about is actual overall impact and in particular the distributional impact, whose real income will be higher or lower "than we would otherwise have seen".