Why we all need to know about NGDP

There’s nothing very new about the idea that a central bank could target nominal GDP rather than inflation.

Sir Samuel Brittan has been lobbying for this since before the Bank of England was made independent; Giles Wilkes, now Vince Cable’s special advisor, wrote a compelling paper proposing it a couple of years ago. More recently, a number of American bloggers have been waging a campaign for targeting NGDP – which is very simply real GDP (the growth rate most people report) plus the broadest measure of inflation – on the other side of the Atlantic.

But don’t be surprised if you hear a lot more about it in the coming months and years. For it has long been an intriguing idea for the future Bank of England Governor as well. Mark Carney has spoken openly in support of NGDP targeting a number of times, and did so again last night.

The salient bit of Carney’s speech:

If yet further stimulus were required, the policy framework itself would likely have to be changed. For example, adopting a nominal GDP (NGDP)-level target could in many respects be more powerful than employing thresholds under flexible inflation targeting. This is because doing so would add “history dependence” to monetary policy. Under NGDP targeting, bygones are not bygones and the central bank is compelled to make up for past misses on the path of nominal GDP.

Of course, this speech doesn’t mean an NGDP target will necessarily become policy in the UK. After all, it’s the Chancellor’s right, rather than the Bank Governor’s, to change the Bank’s remit. Carney’s latest comments on the attractiveness of NGDP targeting were directed towards the Bank of Canada rather than the Bank of England. Plus he’s more equivocal on it than the headlines suggest. And despite having him in office for the past four years, the Bank of Canada still hasn’t abandoned its own inflation target in favour of something else.

But ignoring that for a moment, what would NGDP targeting imply for the UK? The answer a few years ago would have been: not very much. Let’s assume that the Bank had a 5% NGDP target, which is more or less the average growth rate over the past few decades (basically, 2.5% GDP plus 2.5% inflation).

UK NGDP in recent years and projected into the future. Source: ONS and OBR

Monetary policy could well have been a touch tighter towards the end of the boom, and it could have been a bit looser around the turn of the millennium, but, presuming the Bank would have been allowed a percentage point of leeway around the target, the Governor wouldn’t have had to write many letters of apology to the Chancellor.

However, the story since the crisis is altogether different. NGDP has been well below this level and, according to the OBR, it will remain below 5% all the way until early 2018. If the UK was following an NGDP targeting policy, the Bank of England should arguably still be stimulating the economy through more quantitative easing. By contrast, the inflation figures, which show prices are still well above target, indicate that the Bank shouldn’t have been carrying out quantitative easing at all in the past year and a half.

CPI inflation since 1991. Source: ONS

So following an NGDP target would have made the Bank’s job of explaining its policy a lot easier – at the risk of driving up inflation even higher. It would also seem to imply more QE in the future.

However, there is a more subtle way NGDP would make a difference, and that’s in conditioning long-term expectations. If people expect that the Bank will do everything in its power to try to get nominal growth back to 5%, they might have more faith that monetary policy will be looser for longer.

A few years ago, well before the crisis, Ben Bernanke gave a speech which has subsequently been dubbed his “printing press speech”. One of the key points was that if a central bank gets to a stage where interest rates are down at zero, the first thing it should do, before getting into quantitative easing or any other radical active policies like that, is to make it absolutely clear to all consumers and businesses that interest rates will stay low not merely for a few months or a few years but, basically, for the foreseeable future.

That way, people can be sure enough that credit will be cheap that they can make long-term investment decisions without the fear that rates will soon be climbing higher. That’s why Bernanke – and for that matter Carney – were been very clear on this front from early on in the crisis that interest rates in the US and Canada respectively would stay low for a prolonged period of time.

And in some senses, having an NGDP target performs a similar function as committing yourself to low interest rates for a long time – at least in times of crisis: that assumption would be a given, based on the central bank’s NGDP remit.

Central banking is, in part, about not merely changing monetary policy but in conditioning expectations. And, as Carney said in his speech last night, those expectations depend, in turn, on not allowing bygones to be bygones: the longer the Bank misses the growth target for, the more it will have to act, cumulatively (although this clearly depends on how any target is constructed).

Anyway, the point is that were NGDP targeting to be adopted here it would amount to the biggest shift in institutional economic policy since the Bank was made independent back in 1997. So don’t expect the issue to go away any time soon.

UPDATE: see Rob’s comment below – which is quite right: the question of precisely how the target is structured (which I glossed over above) is highly significant. If Carney is suggesting that the level of NGDP is targeted that is altogether different, more radical and indicative of potentially more QE, than if it’s only a growth rate being targeted.

Giles Wilkes also tweets to say: if NGDP target 100% credible, LESS QE needed. Velocity would rise.

And Andrew Lilico tweets to say: NGDP-level targeting would be much more similar to price-level targeting than to inflation targeting. See: http://t.co/1cU4qqi4

What today’s jobs numbers really tell us

I was a little bemused this morning to hear the latest unemployment figures being touted by cabinet ministers as the best in a decade.

Granted: they’re pretty solid, and they show that the labour market is continuing to improve. Unemployment is falling, youth joblessness is down, the total number of people in work is still climbing and the claimant count, an alternative measure which offers an even more up-to-date assessment of where things are going, is also falling.

But are these really the most improved unemployment figures in a decade – as many are reporting? Only if you squint your eyes and look at the figures very selectively.

It is indeed true that overall unemployment fell by 82,000 in the past quarter. It’s true that this is the biggest quarterly fall in simple number terms since May 2001 (though in percentage terms, which are more representative of relative performance, the 3.2% fall is the biggest since 2003).

However, if one compares one month to another – as we do for most statistics – there’s a very different picture. In the month to October, unemployment dropped by only 4,000 across the country. That’s a nothing-to-write-home-about improvement of 0.2%. The previous two months were far more impressive, with the jobless total dropping by 64,000 in August and 13,000 in September.* It’s because of those two months that the quarter-on-quarter fall in unemployment is so impressive.

A similar statistical illusion takes place with the youth unemployment figures. Government insiders claim that the quarterly fall in youth unemployment was the biggest on record.

Again, that’s right only if one looks specifically at the quarterly figures in number terms (72,000; the percentage fall of 7.2% is the biggest since 2000 rather than the biggest on record). In the past month, youth unemployment fell by 18,000 – good but hardly a record. Once again, the quarterly improvement is buoyed up by the big fall (64,000) from a couple of months ago.

But in both cases, the reality is that this was a positive, but basically pretty average, set of employment statistics.

Now, of course all of this is terrifically pedantic. But there’s also an important point. The labour market is still improving, but all the evidence from this latest batch of numbers is that the improvement is starting to tail off. Trumpeting these figures and suggesting they bespeak a labour market which is going great guns is simply misleading.

Moreover, a greater concern for most families is the fact that they are still seeing their earnings squeezed by inflation. Today’s labour market figures showed that the annual increase in the average wage (excluding bonuses) dropped to 1.7% in October – a full percentage point below CPI inflation of 2.7%. In other words, in real terms, adjusting for inflation, families are still facing falls in their incomes.

That’s been the case pretty much since the start of the crisis. And although the Bank of England Governor, Sir Mervyn King, said the squeeze was coming to an end, the statistics contradict him. On the contrary, it is continuing, and although real wages may not be falling at the rate they were last year, they are nonetheless shrinking.

* I should clarify that these figures are actually three month rolling figures, in order to ensure the jobless figures are not overly volatile month-by-month.