Pity the Fed

October 29 was a “data rich” day with the Bureau of Economic Analysis releasing data on personal income and outlays for September as well as its first release for the National Income and Product Accounts (NIPA) for the third quarter. The importance of the personal income and outlays data is that it includes the personal consumption expenditure (PCE) price index. On a month-over-month basis, this measure of inflation rose from an unacceptably high 3.95% (August) to 6.05% (September). To be sure, there is considerable variation in this measure and the year-over-year measure rose more modestly, from 3.44% (August) to 3.65% (September). The Gomme-Rupert measure of trend PCE inflation also increased from 3.84% to 3.91%.

The Fed’s preferred measure of inflation is based on core PCE (excluding food and energy). The monthly rate rose from 3.19% to 3.86%; the annual rate from 3.03% to 3.12%; and the Gomme-Rupert trend from 2.37% to 3.11%.

Meanwhile, output growth rose from 2.22% (second quarter) to 3.37% (third quarter). While consumption growth was unchanged at 3.77%, investment growth rose from 4.65% to 19.42%. Within investment, nonresidential equipment grew 20.88% in the third quarter, up from 13.44% in the second. Nonresidential structures grew 5.39%; this category has actually exhibited negative growth over the past year or so. Investment in residential structures was -3.34% (previously 2.81%) and has also experienced negative growth over the preceding year. Export growth was -5.74%, down from 5.04% in the second quarter; and imports grew 12.58% (down from 19.67%). Some readers may recall that on August 22, the U.S. imposed 50% tariffs on a variety of goods from Canada, and that on September 8 Canada levied countertariffs of 15, 25 and 50% on a number of U.S. goods. It seems unlikely that the third quarter exports and imports numbers were significantly impacted by these tariff increases since they were unexpected and came late in the third quarter.

So, would the Fed have made a different decision had they met after having these data? Overall, the economy is killing it, with GDP growth well over the long run average of about 2%. Indeed, there have been six times in the past couple of years that GDP growth has been over 3%, once over 4%. Inflation continues to run well above the Fed’s 2% target. Well, it took just a day to accumulate more info:

Inflation remains elevated. The accumulation of more information will help the Committee determine the policy stance appropriate to ensure a timely return of inflation to the Committee’s 2 percent goal. The Committee will deliver price stability. (FOMC Statement)

September CPI Spikes

By Paul Gomme and Peter Rupert

The BLS announced that the CPI jumped 8.07% on an annualized basis in September. The year over year increase was 3.72% and our trend measure shot up to 4.62%.

As mentioned in earlier inflation posts, the trimmed mean is another measure that people have been looking to; it that increased to 3.53% from 2.67% in August. The core CPI (ex food and energy) increased 4.09% and our trend measure increased 3.11%.

Although the Fed has a preference for the PCE inflation measure, it seems likely, given the correlation between the CPI and PCE, that we will be seeing an increase in those numbers as well. The timing of the PCE release is unfortunate for the FOMC: the PCE data will be released on October 29, one day after the conclusion of the FOMC’s October 27-28 meeting. Further complicating matters for the FOMC: mid-term elections less than a week later. Those wishing to hold the Fed funds rate unchanged can, in addition, suggest that the effects of the Fed’s earlier tightening have not yet been realized. On the other hand, the committee may look foolish if one day after their rate decision the September PCE inflation rate is reported to have risen.

Just as a bit of an aside, why does inflation measurement seem so confusing? In any month there are possibly large price swings across the broad categories measured by the BLS or BEA. This month, for example, see what happened to energy commodities (motor fuel and fuel oil), food and medical care commodities:

While fuel and motor oil skyrocketed, food prices have been pretty tame and medical care commodities (such as prescription drugs) have been declining. The difficulty, then, is how best to understand what is happening to average prices across all the commodities? This is where the controversy comes in. Throw out food and energy? Look at an underlying trend measure? Or, in any month remove the highest and lowest 8%, i.e., trim off about 16% of price changes. The jury is still out.

September Jobs Report

By Paul Gomme and Peter Rupert

The BLS announced that the labor market was pretty flat in September. Employment rose a lean 29,000 while unemployment rate barely changed, from 4.14% to 4.18%. The flatness spanned almost all major sectors, no big gains or losses anywhere. The diffusion index (percent of industries with employment increasing plus one-half of the industries with unchanged employment, where 50 percent indicates an equal balance between industries with increasing and decreasing employment) sat at 49.0, so a fairly equal balance of increasing and decreasing industries.

The employment gains recorded by the Establishment Survey, 29,000, are considerably lower than the average over the past year. Indeed, by historic standards 29,000 new jobs seems anemic. However, our discussion of the July Employment Report pointed to reasons why the U.S. economy simply doesn’t need to create as many jobs as it once did. Central to that discussion was the notion of “breakeven'” job gains: the number of new jobs that need to be created to leave the unemployment rate unchanged. Two factors have lowered breakeven job gains: a falling labor force participation rate, largely due to an aging population, and slowing population growth. The fact that the unemployment rate rose very slightly, from 4.14% in August to 4.18% in September, suggests that 29,000 is close to breakeven: about what’s needed to hold unemployment steady.

The household survey painted a slightly different picture, with the labor force rising 485,000 and the number employed up by 406,000. The chart below explains who gets counted in the two surveys.

The bottom line: with breakeven this low, headline payroll numbers that once would have signaled a slowdown now look roughly consistent with a stable labor market.