Sunday, January 26, 2014
Another fine forecasting mess you've got me into
Posted by David Smith at 09:00 AM
Category: David Smith's other articles

carney-new.png

My regular column is available to subscribers on www.thesundaytimes.co.uk This is an excerpt.

When Mark Carney took up his job as Bank of England governor in July last year, he impressed everybody with how well he had been briefed, or briefed himself, on the British economy and financial system.

In two important respects, however, he could have done with a bit more briefing. The first was that he should have been more aware of the Bank’s recent very poor forecasting record.

The second, which is related, is that he should have been more cautious about using the unemployment rate as the basis for his forward guidance strategy.

As I wrote in August, when the policy was announced: “Is it the best-designed policy in the world? No. Frankly, it looks as if it was designed by committee, which it was. The unemployment rate threshold of 7%, at which point the monetary policy committee (MPC) will start thinking of higher rates, suffers from familiar flaws.”

Those chickens are coming home to roost, of which more in the moment. But let me first put in a good word for forward guidance, which I genuinely think helped the recovery along by reducing interest-rate uncertainty, if in the end only briefly, for households and firms. Carney said on Friday that the MPC will consider a range of options to update its guidance, which I think is the right thing to do.

The appeal of using the unemployment rate was its simplicity. The question about the updated guidance, which may have to feature a combination of indicators, is whether it will be possible to convey the Bank’s message – rates will not be raised until the recovery has had plenty of time to breathe – in as straightforward a way.

Let me also speak up for the labour market. It is a pretty rum state of affairs when a big fall in unemployment, and the strongest rise in employment on record (280,000 in the September-November period) is interpreted as bad news. The job market is strong across the board, from sharply rising working hours to a falling jobless total on all measures.

This is something to be celebrated and, to answer a question often put to me, is not mainly a reflection of government measures to clamp down on bogus benefit claimants. Such actions do not affect the wider, and fast-falling Labour Force Survey measure of unemployment.

We also appear to have reached the point where real wages are no longer falling. In fact, it appears to have happened at about the time when the Labour party switched from attacking the coalition’s austerity to focusing on the cost of living.

I base this not on the government’s analysis showing a rise in real take-home pay in 2012-13 for all but the top 10% but on the wages and salaries data in the most recent gross domestic product figures. They showed a rise in total wages and salaries compared with a year earlier of 3.7% in the second quarter of last year, and 3.3% in the third. Adjusted for rising employment they suggest earnings per employee were rising at an annual rate of around 2.6% during the two quarters, broadly in line with inflation. The big squeeze is over.

But what about the Bank’s terrible unemployment forecast? Carney was right to say that it is better to get it wrong in this direction than if the jobless total was going up. That does not, however, redeem it. The Bank’s forecast has given him exactly what he did not want, intense speculation about an imminent rate rise, which he has had to deal with, less than six months into his forward guidance strategy.

This is an interesting time to talk about economic forecasting. This year is the 50th anniversary of the launch of the Sunday Times Business section. One of the earliest innovations in the section was to introduce the regular London Business School economic forecast, prepared by such luminaries as James Ball, Terry Burns and Alan Budd.

As a very young reader, I remember being fascinated by these forecasts. How could anybody look into the future? There was a mystique about it which implied that forecasters were endowed with magical qualities. The Treasury, remember, did not publish its forecast until the mid-1970s, and the Bank did not do so until the 1990s.

Why have forecasts not got better? The old joke used to be that economists produced forecasts to make weather forecasters look good. Weather forecasts, however, have become significantly more accurate, thanks to satellite and computer technology. Why has there been no similar improvement in economic forecasts? The answer may be that there are too many aspects of economic behaviour which are simply unpredictable.

In the case of the Bank, Charlie Bean, one of its two deputy governors, took the issue of its forecasting record on the chin in a recent speech. The Bank, like most other forecasters, failed to spot the severity of the downturn in 2008-9. In 2010, it was guilty of being over-optimistic about both growth and inflation.

Growth, it predicted in 2010, would be supported by improved credit conditions and a reduction in uncertainty, boosting consumer spending and business investment, while exports would provide a “significant stimulus” as the pound’s big 2007-8 fall “worked its magic”. As for inflation, it would be boosted temporarily by George Osborne’s January 2011 VAT hike, but swiftly return to the 2% target.

It did not, as Bean ruefully admitted turn out like that. Growth was much weaker than it expected, and business investment has only just started to turn up. Net exports contributed to growth in the first half of last year, but have taken an age to do so. Inflation only belatedly returned to the target last month.

Why did the Bank get it so wrong? Bean’s explanation was a variation of Harold Macmillan’s “Events, dear boy, events.” Things intervened, like the eurozone crisis and soaring commodity prices to scupper the forecast. But forecasters are employed to at least keep an eye out for these things, and the Bank should have had more intelligence on them than it did.

You could argue that the Bank’s errors did not do great harm. Its optimistic forecast did not result in damaging interest rate hikes. Had it known what was really going to happen to inflation, maybe it would have done so.

Is there any point to forecasting? Yes. But forecasters have to be smarter and more aware than the Bank’s (and most other forecasters) have been in recent years. Providing alternative scenarios is one way of demonstrating the uncertainty.

As for Carney, wiping a little egg off his face for believing too much in the Bank’s unemployment forecast, he now has to ensure that his new forward guidance rules are as error-proof as they can be. Once- bitten, twice-shy.