The Bureau of Labor Statistics reported that inflation as measured by the seasonally adjusted consumer price index for all urban consumers rose only 0.1% in February (a 1.2% annual rate), down from 0.7% (an 8.4% annual rate) in January. Those who view the monthly CPI as the most important inflation indicator breathed a sigh of relief, perceiving the economy to have lurched from hyperinflation back to price stability within the space of 30 days.
| Month | 12-month change | 1-month change (annual rate) |
|---|---|---|
| Jan-04 | 1.9% | 2.2% |
| Feb-04 | 2.0% | 2.3% |
| Mar-04 | 2.1% | 2.6% |
| Apr-04 | 2.2% | 2.6% |
| May-04 | 2.3% | 2.2% |
| Jun-04 | 2.4% | 2.6% |
| Jul-04 | 2.4% | 2.2% |
| Aug-04 | 2.4% | 2.6% |
| Sep-04 | 2.3% | 1.6% |
| Oct-04 | 2.2% | 2.2% |
| Nov-04 | 2.2% | 1.6% |
| Dec-04 | 2.3% | 2.2% |
| Jan-05 | 2.3% | 2.6% |
| Feb-05 | 2.4% | 3.1% |
| Mar-05 | 2.3% | 2.2% |
| Apr-05 | 2.3% | 2.6% |
| May-05 | 2.3% | 2.0% |
| Jun-05 | 2.3% | 2.0% |
| Jul-05 | 2.3% | 2.6% |
| Aug-05 | 2.3% | 2.0% |
| Sep-05 | 2.4% | 2.8% |
| Oct-05 | 2.4% | 2.0% |
| Nov-05 | 2.4% | 2.6% |
| Dec-05 | 2.5% | 2.8% |
| Jan-06 | 2.5% | 2.5% |
| Feb-06 | 2.5% | 3.6% |
On the other hand, those like Macroblog’s Dave Altig who take a longer view found some reasons to be concerned about the seemingly quite favorable February CPI data. Dave noted that the median CPI, which summarizes the price change that half the value-weighted items in the CPI exceeded, rose by 0.3% in February (a 3.6% annual rate), up from its more modest 0.2% increase of previous months. Dave was likewise concerned that 40% (on a value-weighted basis) of the items in the CPI increased in February at a 3-4% annual rate, suggesting that a higher rate of inflation may be becoming incorporated more broadly throughout the economy.
I personally prefer to look at the median change over a 12-month rather than a 1-month interval. This measure had been pretty steady around 2.3% for most of 2005, and has been slowly but steadily creeping up to 2.5% over the last six months. We’ve also seen a corresponding 20-basis-point increase in the market’s expectation of the average 10-year inflation rate over the last few months.
Is 20 basis points that big a deal? If expected inflation is, say, 2.3%, that sounds to me like “around two” and something I would think the Fed could live with. But when it starts to get over 2.5%, it’s more like “around three”, and something I don’t expect the Fed to tolerate.
2.3 is “around 2” and 2.5 is “around 3” and this is totally different? I can’t buy that.
More troublesome is that there is this trend from 2.3 to 2.4 to 2.5 over the past six months, pretty steady. The real worry is that it’s going to continue, 2.6 to 2.7 and so on. Looking at your chart, 2.6 is due next month. If that happens on schedule it will confirm the trend and indicate that more needs to be done. If we stay at 2.5 or drop, then it’s more of a “wait and see” situation.
Isn’t it surprising that world labor deflation and commodity inflation more or less balance out with respect to the dollar?
Jim –
So what is the policy lesson here for Congress? I appreciate your concern that tax cuts may not be as magical as some conservatives have been advocating, and the fiscal spending binge of recent years may be just as responsible for overstimulating Aggregate Demand. But I think it’s a fair characterization of your median legislator that inflation is not their concern since that’s the Fed’s deal. Is there a pithy comeback?
If the Fed were not concerned they would not have raised rates as much as they have, nor would they have said that they will continue to raise rates.
I think the belief that you can accurately measure inflation changes of 0.1% is wrong. There is simply too much guesswork involved in the calculations, such as adjusting for differences in quality. The adjustments become even more difficult when so much of the economy consists of services. Nevertheless, inflation caused by debasement of the currency is real. I think the best you can do is a rough estimate, say distinguishing between 2% and 4%.
Rich, I agree that inflation itself cannot be measured to the nearest 0.1%, but the spread between TIPS and nominal yields can. Although it’s true that factors other than changes in expected inflation might produce a change in that spread, I would think that inflation expectations is the most natural interpretation, particularly in conjunction with the other evidence.
Kane, I believe that inflation is primarily the Fed’s responsibility. Although fiscal deficits can make the Fed’s job harder, it’s the Fed rather than Congress that’s most likely to be paying attention to what I see as the recent trends.
I wonder how intolerant of inflation the Fed will be, since Ben Bernanke is sometimes referred to as “helicopter Ben”, although the speech that refers to was actually about deflation.
Also, I should think our Congress critters will continue to spend money like it’s free, which to them it is. And a wide swath of our middle class probably figures high inflation would pay off their overextended mortgages for free, as it did for others in the 1970s. Those attitudes won’t help.
James, out of interest, why do you prefer to focus on the median CPI as opposed to exclusion-based measures of underlying inflation (such as the chain-linked core CPI or the core PCE deflator)?
Kieran, food and fuel are often outliers, but not always, and I don’t see a general argument why leaving them out is (a) always sensible, or (b) always solves the outlier problem. Some of the issues are discussed here.
Professor-
The implied inflation based on 10 Year TIPS versus nominal treasuries is down this month to 2.5% as of 3/21. It was 2.6% on 3/1. As far as I can gather, the CPI does not track the TIPS measure precisely, so you really can’t use the TIPS measure to verify that your CPI is “correct”. Furthermore, I believe that the implied inflation reflects market buy/sell balance rather than a forecast of inflation. A very useful indicator, but just one of a number of useful indicators.
Good points, Rich. On the other hand, the implied inflation as of Jan 3 was 2.34%, and I was talking about this broader trend up over the last several months compared with the broader trend over these same months in the median CPI. You’re right that the specific market reaction to the Feb. CPI release seems to have been more optimistic than mine. Another indicator of that point is that Altig’s funds rate probabilities suggest that the market’s response to the new CPI data was a perceived decrease in the probabiity of a 5.0% fed funds target for May.
Professor-
Thanks for the tip on macroblog. It looks like a good one to add to my economics bookmarks.