August 2007

Growth Picks Up in Second Quarter 2007

The growth of real gross domestic product (GDP) accelerated in the second quarter to a 3.4 percent annualized rate, up from 0.6 percent in the first quarter. Every category but consumption participated in the improvement. Nonresidential fixed investment and exports made stronger positive contributions, while inventories, government purchases and imports swung from negative to positive contributions. Residential investment made a smaller negative than it had in the first quarter.

Chart 1 breaks down the contributions to recent GDP growth by component. The green bars represent average contributions to growth over the period 2002–06, the yellow bars represent 2007:Q1 and the blue bars 2007:Q2.

The most notable change from Q1 is the slowing in consumption growth. Real personal consumption expenditures (PCE) grew an annualized 1.3 percent in Q2, contributing only 0.9 percentage point to overall growth. This compares with 3.7 percent growth and a 2.6 percentage point contribution in Q1, and 3 percent growth and a 2.1 percentage point contribution, on average, over 2002–06. Higher prices for food and energy and the cumulated effects of higher interest rates no doubt all played a role in the slowing. Higher food and energy prices, in particular, eroded households' real incomes in the second quarter. After increasing at a 5.9 percent annualized rate in Q1, real disposable personal income fell 0.8 percent in Q2, as nominal income growth of 3.5 percent was more than offset by inflation of 4.3 percent in the price index for PCE.

A Continued Drag from Housing
As noted above, Q2 also saw a smaller drag from residential construction on overall growth. Real residential investment subtracted 0.5 percentage point from growth in Q2, compared with 0.9 percentage points in Q1. Does this mean that the downturn in residential investment has finally reached bottom? Unfortunately, the answer is probably no, as monthly data on the housing sector continue to show deterioration. In last month's National Update, Evan Koenig discussed the relationship between movements in housing permits and subsequent changes in real residential investment. At the time, the relationship predicted (correctly) a lessening of the drag from residential investment in Q2. Based on the most recent permits data, though, that relationship predicts a Q3 decline in residential investment comparable with what we observed in Q1. In other words, the drag from residential investment in Q3 will be closer to Q1's –0.9 percentage points than to Q2's –0.5 percentage point. Chart 2 plots the actual and fitted values from Koenig's forecasting relationship.

Employment Growth Moderates
Employment growth slowed somewhat in July, with an increase in nonfarm payrolls of only 92,000, compared with 126,000 net new jobs in June and an average of 144,000 net new jobs per month over the first six months. Employment growth may yet rebound, but at least one signal points to a further slowing—the recent behavior of temporary employment. Changes in temporary employment tend to lead overall payroll employment, and lately temporary employment has been declining. Chart 3 plots six-month changes in nonfarm payrolls together with six-month changes in temporary employment, the latter shifted forward by six months. The units for both series are thousand of jobs per month, and the scales in the chart have been chosen based on a regression of changes in payroll employment on changes in temporary employment—in other words, the end-point of the temporary series is the regression's prediction for nonfarm payroll gains over the next six months.

Inflation Gradually Decelerating
Core rates of consumer price inflation have been drifting gradually downward over the past few months, and, given sharp declines in the price of gasoline in June, even headline price indexes posted only moderate gains. Rather than focus on one particular measure, however, Chart 4 combines information from seven different published measures of consumer price inflation: headline rates (CPI and PCE), ex. food and energy rates (CPI and PCE), trimmed mean rates (CPI and PCE) and median (CPI). The shaded area is the range of 12-month inflation rates in these measures, and the blue line is the median of the seven 12-month inflation rates at each date. The red line is the 12-month inflation rate in a particular weighted average of the seven indexes, with weights based on the relative volatilities and correlations among the seven inflation measures.[1]

Looking collectively at a set of consumer price gauges is useful to the extent that the relative performance of the individual measures has varied over different sample periods and over different performance criteria.[2].

The range, median and average shown in Chart 4 all seem to agree on two points: the rate of consumer price inflation picked up in 2005 through late 2006, and it has since been drifting down, albeit gradually and irregularly.

—Jim Dolmas

About the Author Dolmas is a senior economist in the Research Department at the Federal Reserve Bank of Dallas. Notes Imagine constructing a portfolio from the monthly inflation rates in the seven indexes, with portfolio weights that sum to one. The weights used in constructing the series in Chart 4 are the weights that would minimize the variance of that portfolio. Note that some indexes enter into the linear combination with a negative weight—hence the weighted average can, occasionally, fall outside the shaded range, as it does at the beginning of the sample. This is a point made in a recent paper—presented at the Dallas Fed's May Conference on Price Measurement for Monetary Policy—by Robert Rich and Charles Steindel of the New York Fed ("A Review of Core Inflation and an Evaluation of Its Measures," Federal Reserve Bank of New York Staff Reports no. 236, 2005).