I almost always agree with Paul Krugman’s views on inflation. We’ve both been skeptical of claims that US monetary and fiscal policy will produce high inflation. Krugman points to the economic slack in the economy, I focus on TIPS spreads. In a recent post, Krugman made the following observation:
. . . if we think of wages as the ultimate core price, I don’t see any mechanism in today’s America whereby rising commodity prices translate into higher wage contracts.
But what does the historical record say? It depends on which era you’re looking at. . . .
The two big commodity price shocks of the 70s did, in fact, feed quickly into core inflation. Since then, however, nada.
Why the difference? The obvious point is that back in the 70s many labor contracts included cost of living adjustments (COLAs). This in turn partly reflected stronger worker bargaining positions and also real doubts about whether monetary policy would contain inflation. Today, none of that: COLAs are rare, and commodity-price fluctuations don’t feed into wages at all.
My only observation here is that we don’t even need to bring COLAs into the picture. Wages respond to trend NGDP growth, and those growth rates were extremely high during the 1970s, even during recession years. For instance, NGDP grew at an 11% rate between 1971:4 and 1979:4, and then grew another 9.6% in 1980, despite a mild recession. Since monetary policy determines the trend rate of NGDP growth, this analysis is consistent with Krugman’s observation that in the 1970s there were “real doubts about whether monetary policy would contain inflation.” But the phrase “contain inflation” is a tad misleading—suggesting that inflation was like a wild animal on the loose and needed to be reined in by the Fed. The Fed caused the Great Inflation.
As you know, I often say “never reason from a price change.” I think we’d be better off talking about the impact of NGDP on wages, which is fairly stable, rather than the impact of prices on wages, which is highly dependent on whether prices are rising because of more demand, or because of supply-shocks. As we saw in mid-2008, high headline inflation doesn’t translate into big pay increases if not associated with fast NGDP growth. The Economist made a similar observation for Britain:
But a jump in inflation caused by higher commodity prices and a rise in VAT—in an economy with spare capacity—is quite different from one caused by excess demand and a pay-price spiral. It intensifies the squeeze on households from other tax rises and curbs consumer spending. Although the central bank is facing calls to tighten monetary policy soon, that would be warranted only if there were signs of inflation getting embedded into expectations and feeding through to higher wages.
Unsurprisingly, households are now expecting inflation to be higher over the coming year. But other official figures published this week showed no sign of a pay-price spiral. Average earnings are rising by just 2.1%, a very muted rate by historical standards. It is difficult to envisage wages taking off when the public sector is shedding jobs and facing a two-year pay freeze and there are 2.5m people unemployed, close to 8% of the labour force. Indeed the youth-unemployment rate has reached 20.3%, the highest since comparable records began in 1992.
The economy clearly retains quite a bit of spare capacity—the main reason why the Bank of England has insisted that the flare-up in inflation will be temporary. The bank has lost credibility as the inflation overshoot has persisted and its forecasts have proved incorrect.
Many Keynesian focus on . We saw in the 1970s that wages can grow rapidly during recessions, if the trend rate of NGDP growth is high. Of course more sophistacated Keynesian models can handle this by adding expected inflation. And models of slack do seems to explain wages better during an economic recovery, when NGDP growth may be fast, but wage gains may remain muted.
Keynesians often focus on economic slack as a determinant of wage gains. That’s a bit too simple (as we saw in the 1970s); the more sophisticated Keynesian models now account for expected inflation. And in fairness, the “slack” model of wages does seem to outperform the NGDP growth rate model during recoveries, when wage gains are often quite moderate, despite fast NGDP growth.
But there’s a good reason why wage gains often trail NGDP growth during a recovery—wage cuts trail declines in NGDP during most contractions. Thus wages are still adjusting to the previous drop in AD during the early stages of a recovery. The Keynesian model is good at explaining cyclical variations in wages, but not so good at explaining long run changes in trend wage growth, and trend NGDP growth. For that you need the quantity of money.
Britain is facing an interesting dilemma. The BOE clearly understands that NGDP, not prices, are the best indicator of demand. They understand that it would be a mistake to react to the 3.7% headline inflation in December (yoy), by tightening monetary policy. But they are constrained by the fact that almost everyone (wrongly) thinks it’s the central bank’s job to control inflation.
In contrast, fiscal policy is almost always evaluated in terms of real GDP growth. This dichotomy makes no sense. There is no macro model of any school of thought that says monetary policy controls inflation by shifting demand, and fiscal policy controls RGDP by shifting demand. Yet when RGDP growth in Britain came in at a disappointing minus 0.5% in Q4, almost all the reports pointed not to the BOE, but to the fiscal austerity of the new British government. Even Brad DeLong, who devoted much of his talk at the recent AEA meetings to emphasizing the importance of NGDP, quotes an article giving a RGDP number for Britain. RGDP numbers tell us nothing about whether demand stimulus is succeeding. In fairness, I believe the UK NGDP numbers come out with a lag, but the problem is much deeper than that. The media overwhelmingly sees fiscal policy as a RGDP issue and monetary policy as an inflation issue. Until both are seen as NGDP issues we will not be able to come up with coherent policy regimes, which assign fiscal and monetary policy their appropriate roles.
There are two ways British policy might fail. First, the BOE might fail to hit its implicit NGDP target. Why would that occur? Not because Britain is in a liquidity trap. They are actually expected to tighten policy this summer. And Britain can always depreciate its currency if it wants to. It could even cut rates another quarter point. No, with 3.7% inflation it’s madness to even talk about liquidity traps. If they fail it will be because fear of politically embarrassing headline inflation numbers caused BOE officials to take their eye off the ball (NGDP.) However if BOE policy does fail to boost NGDP, I predict fiscal austerity will be blamed.
Alternatively, policy might fail because the Gordon Brown government did significant damage to the supply-side of the UK economy, by increasing the size of the state. A supply-side failure would show up as stagflation, i.e. appropriate NGDP growth but high inflation and slow RGDP growth. Although we saw a bit of stagflation in the 4th quarter, it’s too soon to draw any conclusions. Inflation will probably slow over the next few years.
I predict that if policy does fail for supply-side reasons, the failure will be widely attributed to demand-side factors, especially fiscal austerity. That’s because everyone focuses on RGDP, and almost no one pays any attention to NGDP.