Inflation bounced back in March, as downward base effects are fading. This is, however, not the sign of inflationary pressures building up as core inflation remained unchanged at 1.6%. The whole developed world is suffering from large overcapacities: disinflation at home is supplemented and fuelled by disinflation abroad. Inflation is not the problem: the lack of it is.
In March, consumer prices were up by 0.2% m/m, leading the year-on-year rate of inflation to slightly accelerate, at 1.5%, after a cyclical low of 1.1% in February. This development should not be overestimated: inflation pressures remain non-existent in the US. Indeed, the acceleration was mainly due to base effects, as the CPI had dropped 0.3% m/m in March 2013 after a booming 0.6% the month before.
On top of that, the March 2014 slight acceleration mainly came from energy prices, which year-on-year rate of growth accelerated from -2.5% in February to 0.6% in March. Indeed, the core index, which excludes food and energy components from the consumer price index, kept rather the same trend as over the previous twelve months, up 0.2% m/m and 1.6% y/y. A narrower measure of underlying inflation1, which we borrow from the Bank of Canada, is trending downwards even further: on a monthly basis, it was down 0.2% in both February and March, leaving the year-on-year rate of growth at 0.5%, the lowest reading since December 2010.
As for early signals of inflation pressures, there are none. At early stage of processing, the producer price indices still are on a soft trend. The core index for intermediate materials was up 0.3% y/y in March, after +0.4% in January and February. That rate has been constantly below 1% since September last year. As for crude materials, the core PPI, the pressures are downwards (-3.6% y/y in March), even if this trend is somewhat softening: the index was down by an average 5.2% in 2012 and 2013.
Import prices are as soft. Excluding food and fuel, they are declining on a year-on-year basis since March last year. With inflation running low all around the world, there is no surprise that imports are no source of inflation in the US. While downward pressures have been at play for a long while, the phenomenon was more about cheap labour-intensive goods from the developing world. But lately, manufactured goods from industrialised countries have been a powerful disinflationary force: down 1% y/y in March, this was the twelfth month in a row of decline.
For inflation to accelerate, domestic demand would have to strengthen markedly. As it remains weak, firms still have to compete for consumers, and this comes through better service and quality but also through lower prices. As for now, this did not result in lower profit margins, as costs are well under control, especially labour costs, thanks to productivity gains and slow growth in wages. As long as available workers will remain that numerous, there is no reason for wages to accelerate. This large level of under-employment is illustrated by our SLACK Index2, which is currently way below the trough reached following the 2001 recession.
Recently, some have been arguing that the overall rate of unemployment was irrelevant as to the determination of wages. The idea is that short-term unemployment is the key determinant, as employers look down candidates who have been unemployed for too long, probably fearing a loss of skills. Answering a question about the relevance of the research linking wages with short-term unemployment at the latest post-FOMC press brief, Janet Yellen answered: “I wouldn’t endorse, and I certainly don’t think our Committee would endorse, the judgment of the research that you cited”.
Those models are derived from a sample period covering the 1970s as well as the 1980s. We live in very different time, and it makes sense to think that this long-term relationship does not hold currently. A way to support that hypothesis is that, usually, the long-term and the short-term unemployment rate move together: in short, even if, as an explicative variable, the former provides with more robust statistical results, the underlying story may still be about the overall rate of unemployment. To that matter, some research tends to conclude that the length of unemployment episodes is very dependent on the timing of transition to unemployment: in short, if one loses his job during a recession, there is a greater probability that he will remain unemployed for a longer period of time.
Our own results3 show that the relationship between unemployment and wage inflation probably shifted with the 2008-2009 recession, mainly because of downward rigidities, but also probably because the unemployment rate may be overstating the health of the labour market. As Eric S. Rosengren, the Boston Fed President, said this week, the slack in the labour market will take a very long time to get absorbed: “anywhere between April 2016 and January 2020”. And as even Governor Daniel K. Tarullo – the one de facto in charge of financial stability issues at the Fed – the Fed will wait for evidence that the level of underemployment has been diminishing before tightening policy: “We should remain attentive to evidence that labor markets have actually tightened to the point that there is demonstrable inflationary pressure. […]. We should not rush to act preemptively in anticipation of such pressures based on arguments about the potential increase in structural unemployment in recent years”. Crystal-clear, isn’t it?
BY Alexandra ESTIOT