Skip to: search, navigation, or content.


Indiana University Bloomington

Center for Econometric Model Research

United States Forecast Summary

August 2015


There has been a lot of data to digest recently. In terms of trying to clarify what lies ahead, the overall result for us has been mainly indigestion. Some data points look positive, but an equal number suggest caution. In our model the latter ultimately carry the day, resulting in a slightly weakened near term forecast.


The release of advance data on economic activity during the second quarter also included revisions to the data back through 2012. Two aspects of the revision warrant special note. First, the recent recovery loses most of what little luster it had appeared to have. Growth for 2013, which appeared to have been the strongest four-quarter stretch of the recovery, was revised from 3.1% to 2.5%. Growth for the entire revision period was lowered to just 2.0% from 2.2%. Second, the pattern of first quarter weakness during the recovery is reduced, but not eliminated.

The new data for the second quarter are entirely consistent with this more sluggish scenario. They are also pretty much in line with our forecast of three months ago. Total output growth came in at 2.3% compared with our May forecast of 1.8%. A flat inventory change number instead of the reduction we had assumed explains all of the difference. Final sales (that is, leaving out the inventory component) grew by 2.4%, which exactly matches our expectation.

Overall, the NIPA data suggest that the present is fully consistent with the pattern of the past six years – slow growth, with a few positive items (e.g. solid spending by consumers and the housing sector) and a few that have been disappointing (e.g. business investment and international trade).

Recent Monthly Data

Recent data have been a nearly equal mix of positives and negatives. Household income growth has been relatively weak over the past six months, and consumer sentiment was off significantly in July. But spending has held up reasonably well, and auto sales were very strong in both June and July. On the producer side, industrial production has been weak since December, and the ISM manufacturing index, although in positive territory, is down from its level early this year. On the other hand, the July ISM index for non-manufacturing was very strong.

Friday’s release of labor market data for July was a nonevent. Both payroll employment and the unemployment rate came in exactly in line with the WSJ board of economists’ forecasts. The former was an increase of 215 thousand. That is just 7 thousand higher than the average for the past four years. There were increases in all major sectors except mining, which includes crude oil production. Manufacturing added 15 thousand employees, all of them and then some in nondurables production. In the household survey the unemployment rate was unchanged at 5.3%, the same as June. Calculated to two decimal places the rate was down, but not quite enough to round to 5.2%. Labor force growth remained weak, with the participation rate again at 62.6%, its lowest level in nearly four decades.

Baseline Forecast

We are a little more pessimistic about the near-term compared to our outlook three months ago. For the next six quarters (through 2016) we now expect growth to average 2.5%, down from 2.7% in our May forecast. For the current quarter our new estimate is 2.2%, also 0.2% below May.

At a disaggregate level our forecast for the next six quarters is generally very similar to the patterns over the past six. The domestic private sector (consumption, business investment and housing) is all nearly identical. Note, however, that for business investment this represents a large improvement from the first half of this year. The forecast contains a lessened drag from trade (based partly on our expectation that the appreciation of the dollar will decelerate). We also expect slow growth in government spending. This is similar to the outcome in the second quarter, but a reversal from the pattern during the previous five years.

In the labor market we now see employment growth averaging 222 thousand per month over the next year, a little above our May forecast. Our model may be ambitious in this regard. If it is not, the rising employment would be a mixed blessing. Combined with our modest forecast for output growth it implies very slow productivity growth. The model’s productivity variable (which measures output per hour in the non-agriculture private sector) shows an increase for the next year of only 0.5%. This would just match the disappointing productivity growth for the six years since the end of the recession.


The performance of the U.S. economy has nearly ground the optimism of our model into dust. The model, of course, mostly extrapolates past trends in the economic data into the future. But the specification of many equations means that recent history is given more weight than earlier patterns. Those earlier patterns (the “old normal” if you will) combined labor force growth a little above 1% with productivity increase somewhat above 2% to produce growth that averaged over 3%. The past six years (the period since the Great Recession / the “new normal”) has not come close to these values. Both labor force growth and productivity have been well below 1%. Output has managed 2% growth only because employment growth has greatly exceeded the labor force due to declining unemployment. Our forecast for the next year is a continuation of this situation with a little more labor force growth.

We still think the long-run potential is above these values, with labor force growth a little below 1% (although falling), and an achievable productivity increase of perhaps 1.5%. That implies potential output growth at about 2.5%. To reach this potential will require removal of some of the regulatory sand that has been shoveled into the gears of the economy during the recovery. But even if this is accomplished (a big if) growth will average almost a full percent below the post-WWII norm.