Unemployment & inflation: the non-trade-off

Bank of England Governor Mervyn King said today that inflation will “probably fall below target as the substantial degree of spare capacity pushes down on prices.” This runs into a problem, shown in my first chart. In recent years, spare capacity - as measured by the ILO unemployment rate - has not pushed down on prices*. Quite the opposite. Higher unemployment has led to higher inflation 12 months later, though the relationship is not significant if you exclude the last few months. The Phillips curve has been the wrong shape.
To see how curious this is, contrast it to the US. My second chart shows that, here, there was a strong and significant properly-shaped Phillips curve between 1998 and 2009. (It seems to have shifted in recent months, but that‘s another story.)
Why the difference? One strong possibility is that, being a smaller, more open economy, UK inflation is determined more by global factors than by domestic ones. From this background, the Phillips curve might slope upwards, if - as happened in 2008 - higher unemployment weakens the pound and so raises import prices.
This raises a problem. Imagine the following scenario. The coming fiscal tightening raises unemployment. Given our Phillips curve, this doesn’t reduce UK inflation. Worse still, inflation actually rises either because sterling falls or because global inflation picks up. What does the Bank do?
It could raise interest rates, thus exacerbating the downturn. Or it could stick to its spare capacity story, in which case inflation would overshoot its target.
Which leads me to a cynical thought. Could it be that Mr King’s belief in the effect of spare capacity upon inflation is useful, even though it’s mistaken? It will allow him to respond to a tight fiscal policy in a way that supports the economy, even at the expense of higher inflation.
In other words, it helps make monetary policy behave as if the Bank were targeting nominal GDP growth rather than inflation.
* Note that I’m using the CPI excluding energy and food. This reduces the impact of oil shocks, which should help to create a more normal-shaped Phillips curve. Despite this, though, no such curve exists.

Comments

[Ralph Musgrave]

I never understand why higher unemployment normally leads to a country’s currency weakening. Strikes me that higher unemployment means reduced demand, which in turn means reduced imports, which should lead to the currency strengthening (though of course this effect is mitigated by out of work benefits, which means that demand does not drop by all that much). In particular instances, higher unemployment could also derive from reduced exports which would have the opposite effect. So it strikes me forex traders should look carefully at the reasons for unemployment rising in any particular country before jumping to conclusions about the effect on the currency. Do forex traders actually do this, or do they just “jump to conclusions”?

[Laura]

I don't think trading (in anything) is too much about what's actually happening. It's more about how you think other people (i.e. "the market") are going to react to what's happening than the events themselves. Even more than that, for people who work for others, it's about looking like your own reaction was the right one to those who employ you, so that you don't get sacked.

[chris]

@ Ralph - relationships between exchange rates and anything are rarely stable for long, but there are at least two mechanisms whereby higher unemployment might reduce sterling:
1. The weak aggregate demand betokened by higher unemployment often means reduced demand for money.
2. High unemployment leads markets to expect either lower interest rates or some form of higher money supply, eg QE.

[Fred Kapoor]

I agree with @Chris, no business rate is stable for a long period of time. @Rlaph I don't really understand what you mean when you say:"Strikes me that higher unemployment means reduced demand..." I mean, isn't it the opposite. If people are employed, then they have money to spend and the demand would go up, right?

[Ralph Musgrave]

Chris: good point about higher unemployment presaging an interest rate cut. They taught me that when I did basic economics, but I forgot that point. The Alzheimer’s must be getting worse.

[AJ]

I suggested the Bank was up to something similar at Freethinking Economist last week, though without data to back me up. The argument is simple - the Bank would love to have a surprise inflation as it would kill various birds with the one stone. This post helpfully explains the mechanics of the surprise - it results because people are expecting a Phillips curve relationship that doesn't actually hold.

The unkind response from FTE was

'I have bought the line from the Bank but not sure I get your conspiracy theory view about surprise inflation. They don’t need votes …. curious to hear your theory'

Glad to hear I'm not the only one to think BoE might be up to some tricks.

[Glenn]

Isn't the Phillips curve just one of the most over-used and poorest economic theories out there? I've never seen hard evidence to justify it ever being mentioned any more.