US Views: Above the Trend at Last

1. Despite the 1% drop in real GDP in the first quarter, we believe that the US economy is now growing at an above-trend pace. The best way to see this is via our current activity indicator (CAI), which grew at an annualized rate of 3.4% in May, similar to the average of the prior two months. Although an estimated ½ percentage point of this sequential growth is due to a bounceback from the weather distortions of the first quarter, even the year-on-year CAI now stands at 2.7%, the fastest pace of the expansion so far and above our estimate of potential growth of 2%-2½%. In our view, the CAI is a far more reliable indicator of economic activity than real GDP because it is more timely, more broadly based, less noisy, and less subject to revision.

2. One key reason why we expect a further pickup in the underlying growth pace to 3%+ is an improvement in the housing sector. The impact of the 100bp increase in 30-year mortgage rates last summer should now be mostly behind us; in fact, the 40bp drop in rates since the start of the year could provide a modest boost in coming quarters. More fundamentally, we are optimistic on household formation. From 2006 to 2012, the share of 18-34 year olds who live with their parents increased by nearly 5 percentage points, equivalent to an extra 3½ million individuals. Although the reversal of this surge—which seems to have started last year—is unlikely to be rapid given the gradual pace of labor market improvement and financial factors such as the high levels of student debt, ultimately we do expect much of it to reverse. This probably means that household growth will exceed population growth in percentage terms, and supports our forecast that household formation will average 1¼ million per year over the next half-decade.

3. The only meaningful surprise in the May employment report was that the unemployment rate did not reverse any of the 0.4-point drop to 6.3% in April, only 0.9 percentage points above the FOMC’s current estimate of the structural rate. But other measures of labor underutilization remain significantly higher than the unemployment rate by itself would suggest. We estimate that the U6 underemployment rate of 12.2%—which adds marginally attached workers and involuntary part-timers—would normally be consistent with an official unemployment rate of 7%-7¼%. The shortfall becomes even larger if we include the entire gap between actual and structural labor force participation, i.e., not only the “marginally attached” but also those who would be in the labor force under normal labor demand conditions but have not searched for work in the past year. In a recent study, Andrew Levin of the IMF estimates that the total employment gap—defined as the sum of the unemployment gap, the participation gap, and the part-time employment gap, converted into full-time equivalent jobs, and expressed as a share of the labor force—remains above 3 percentage points, larger than the peak following the 1990-1991 recession. Levin was a close adviser to then-Vice Chair Yellen until 2013, and her recent speeches suggest that her own thinking may be similar.

4. The participation and part-time gaps imply that the FOMC is still much further away from reaching its employment goal than suggested by the unemployment gap alone. This is important for monetary policy, both in its own right and because there is at least some evidence that wage growth—the linchpin between the labor market and inflation—depends not only on unemployment but also on broader measures of labor market slack. Our own analysis found some evidence for this using national data. Moreover, a new Federal Reserve study that uses more granular data finds that wages at the state level grow more slowly if there are more individuals who are out of the labor force but want a job, and also if there are more involuntary part-timers. This is likely to reinforce Chair Yellen's focus on broad measures of labor market slack.

5. Despite still-subdued wage gains, price inflation has firmed in recent months. The core PCE index now stands at 1.4% year-on-year, up from 1.1% at the start of the year, partly because of a positive base effect but partly also because of higher sequential inflation in March/April. Some predict a rapid further acceleration in coming months because of increases in rents and pass-through from higher commodity prices. But neither our top-down models nor our own bottom-up analysis support this view. The current pace of rent inflation looks quite consistent with fundamental factors such as vacancy rates, and with bottom-up measures such as the rent growth reported by the apartment REITs. We also expect commodity prices, and therefore core goods prices, to remain roughly flat in the next year. If so, inflation should move back only slowly toward the Fed's 2% target, even if other factors such as healthcare inflation pick up a bit from their very low current levels.

6. Our forecast for the first hike in the funds rate remains early 2016, about six months later than the Fed’s own projections and the market consensus. One risk to this forecast is the increasing likelihood that balance sheet reduction will begin only after the first rate hike. If the Fed has two instruments and decides to keep one at a more expansionary setting, the other needs to be tightened earlier to achieve a given stance of monetary policy. However, this is a relatively small issue. At present, the Fed reinvests about $10-$15bn in mortgage securities per month. If the FOMC decides to keep the reinvestment going for 6-12 months longer than originally anticipated, this means that the balance sheet at the end of that period will be perhaps $120bn larger than originally anticipated. Standard models of the effectiveness of QE suggest that this should reduce Treasury bond yields by only 2-3 basis points relative to the baseline, a minor effect.

7. A more important issue is the continued easing in financial conditions. Our basic view has long been that the Fed should aim to keep financial conditions on a path that is likely to be consistent with its mandate for employment and inflation. The details of how to translate this idea into actual policy can be tricky, but the basic principle is simple. Assuming we start from a point at which financial conditions look broadly appropriate, the instruments of monetary policy—the funds rate, forward guidance and/or QE—should be tightened if financial conditions ease significantly further while employment rates and inflation rise. At present, financial conditions do look roughly appropriate based on our FCI rule, and this seems to be the FOMC's own assessment as well. Moreover, the forecasts from our strategists—higher bond yields, softer equity prices, and a stronger dollar—suggest that conditions should tighten modestly in coming quarters. If that happens, and if the economy evolves as we expect, our current forecast for monetary policy liftoff still feels about right. But if financial conditions continue to ease while the economy makes further progress toward the Fed's goals, a somewhat earlier liftoff for the funds rate may end up looking more appropriate.

Jan Hatzius

Goldman, Sachs & Co.

Legal and Certification Disclosures
I, Jan Hatzius, hereby certify that all of the views expressed in this report accurately reflect my personal views, which have not been influenced by considerations of the firm's business or client relationships.
For Reg AC certification, see above. For other important disclosures, go to