In a recent column, "Which Inflation Target?" I wrote, "It wasn't 'fear of inflation' that spooked the markets on June 5, but fear of the Fed." Any doubts about that were put to rest on Wednesday, when Federal Reserve Board Chairman Ben Bernanke presented a much different testimony to the Senate Banking Committee -- predicting that "core inflation should decline from its recent level over the medium term."
"Clearly we don't want to tighten too much" he said, noting that "lags between policy actions and their effects" mean "we must take account of the possible future effects of previous policy actions -- that is, of policy effects still 'in the pipeline.' "
Stocks soared just after the prepared remarks were released, with the Dow-Jones index showing the fastest daily gain in three years. The interest rate on 10-year bonds dropped dramatically, below the federal funds rate. Chairman Bernanke was asked about comments in the past, which attributed such low bond yields to a "global savings glut." But U.S. bond yields certainly did not fall so sharply on Wednesday because global saving increased in a single day. Foreigners do not have to invest their savings in the U.S. Like Americans, they would buy short-term bills rather than long-term bonds if they were seriously worried about U.S. inflation. They aren't. We aren't, either.
Both stocks and bonds do poorly when inflation speeds up, as Chairman Bernanke rightly remarked. It follows that before he spoke, the financial markets were far more worried about the Fed overdoing the interest rate hikes than they were about the Fed not having done enough. Were they wrong?
"As measured by the price index for personal consumption expenditures excluding food and energy [called the core PCE deflator]," Mr. Bernanke noted, "inflation is projected to be 2-1/4 percent to 2-1/2 percent this year and then to edge lower, to 2 percent to 2-1/4 percent next year." Why should interest rates edge higher if core inflation is about to edge lower? Note that this was in marked contrast to the chairman's market-crashing June 5 talk, when he worried that "core inflation as measured by the consumer price index excluding food and energy prices was 3.2 percent over the past three months."
Nobody doubts that a chain-weighted price index is more accurate than a fixed-weight index like the consumer price index, which may be why Mr. Bernanke's testimony discarded the familiar CPI and only alluded to the chain-weighted PCE deflator. Yet there is now a chain-weighted CPI, too, and it merits more attention than it gets.
The three-month "trend" of the chain-weighted core CPI showed an inflation rate of 1.4 percent in March, 1 percent in May and 0.4 percent in June. That chain-weighted CPI is not seasonally adjusted because is too new (it began at the end of 1999) to reveal reliable seasonal patterns. To suggest that core inflation is higher "on a seasonally adjusted basis" -- even though it is down sharply in reality -- is to say some prices did not fall as much as they usually do this time of the year. That is not what most people mean by inflation.
If seasonal adjustments are the main reason for the difference between the appearance of higher inflation in the conventional CPI and that of lower inflation in the chain-weighted CPI, then the premature June 5 alarm about faster inflation (aside from energy) may have been nothing more than a statistical illusion.
Seasonal adjustments are irrelevant when comparing a month or quarter with the same period a year ago. May was the same season last year as it is this year. The year-to-year increase in the index Bernanke referred to was 2.1 percent this May compared with 2.0 percent a year ago and 2.0 percent a year before that. As increases go, that's scarcely worth mentioning. And the "market-based" version of the PCE deflator was up only 1.8 percent for the year ending in May.
Continued... |