March headline inflation (all goods) in the US came in higher than expected at 0.5% while core inflation (all goods minus food and energy) came in lower than expected at 0.1%. Rising gasoline and food prices accounted for almost three quarters of the jump in headline inflation. These latest inflation figures have added fuel to the debate what measure of inflation central banks should aim to control.
The ECB and the Bank of England traditionally target the headline consumer price index (CPI), which measures the change in the prices urban consumers pay for a basket of consumer goods and services including food and energy. A spike in headline inflation in the eurozone triggered the ECB recently to increase the benchmark policy rate to 1.25%.
In the past the US Federal Reserve has targeted headline CPI as well. But in the early 2000s, with Alan Greenspan still at the helm, the Fed quietly changed its preferred measure of inflation twice.
In January 2000, at the onset of the housing boom, the Fed replaced headline CPI with the personal consumption expenditure (PCE) index because the latter is less dominated by the cost of housing than the former, as Ben Bernanke explained in Atlanta in January 2010 (Bernanke 2010). He failed to mention, however, that PCE inflation generally is about 0.5% below the corresponding CPI measure. Since the Fed left the (implicit) target level of inflation unchanged, it effectively loosened monetary policy considerably (Mees 2011).
Subsequently, in July 2004, the Fed replaced headline PCE with core PCE, thus excluding food and energy prices. According to Bernanke, the Fed at the time expected any changes in the price of food and energy to be temporary in nature (never mind that Alan Greenspan actually supported the war in Iraq because of the country’s oil supplies).
Relying on core inflation instead of headline inflation, as the Fed does, is justified in case core inflation is a better predictor of headline inflation than headline inflation itself. Core goods and services tend to be subject to nominal price rigidities, while non-core goods, like agricultural commodities, oil, natural gas etc., have their prices set in auction markets.
For much of the 20th century, core inflation has been both less volatile and more persistent than the inflation rate of non-core goods. However, the integration of China and India in the global market added more than 2.3 billion consumers and producers to the global economy. They entered as suppliers of core goods and services and as demanders of non-core commodities. The result has been a major, persistent, and continuing increase in the relative price of non-core goods to core goods (Buiter 2008).
In Atlanta Bernanke asserted that both the Fed and private forecasters correctly assumed that the increases in energy prices in 2007 and 2008 would subside. While it is true that the financial crisis and the collapse of global trade in the second half of 2008 led to a sharp decline in energy prices and a corresponding drop in headline inflation, energy prices were back at their 2008 level by 2010, despite an anaemic economic recovery in both the US and Europe. A barrel of oil now costs three times what it cost in July 2004, when the Fed replaced headline PCE with core PCE.
At the first post-Federal Open Market Committee press briefing that was held on April 27, 2011, chairman Bernanke nevertheless insisted that the Committee expects the increases in the prices of energy and other commodities to be transitory.
A closer look at US labour market data suggests, however, that there is little room for complacency about runaway headline inflation. Hourly wages in the US track headline inflation much more closely than core inflation (see Figure 1). This is also the case when the unemployment rate is higher than the natural rate of unemployment. From 2000 through 2010 hourly wages of production and non-supervisory employees increased on average 0.7% more than headline CPI inflation (3.2 versus 2.5%) and 1.3% more than core PCE inflation (3.2 versus 1.9%).
Figure 1. Hourly wages, headline CPI and core PCE
Source: Federal Reserve
That hourly wages increase on average more than headline inflation is mainly due to downward wage rigidity. When commodity prices fall, like they did in mid-2008, hourly wages remain stubbornly high. Unit labour costs, on the other hand, are not downward rigid at all. Unable to lower the costs per hour, US employers cut the number of hours instead. During the recessions that started in 2001 and 2008, unit labour costs in the US dropped precipitously, reflecting massive lay-offs that resulted in high and prolonged unemployment (see Figure 2).
Figure 2. Headline CPI and unit labour costs
Source: Federal Reserve
Even if the current spike in headline inflation proves to be transitory, past experience suggests that it may well lead to a permanent increase in real hourly wages. Unless monetary policymakers in the US favour feverish boom-and-bust cycles with prolonged periods of high unemployment, they had better start paying close attention to headline inflation, like their counterparts at the ECB do.
Bernanke, Ben S. (2010), “Monetary Policy and the Housing Bubble”.
Bernanke, Ben S. (2011), FOMC Press Briefing April 27.
Buiter, Willem H. (2008), “Central Banks and Financial Crises”.
Mees, Heleen (2011), “US Monetary Policy and the Interest Conundrum”.