Goldman: "Forecasters Need To Cut GDP Estimates A Lot Further"

Tyler Durden's picture

From Jan Hatzius and the Goldman economic team. Check to you David Bianco: time for another S&P estimate raise?

  • Over the last few months, the US economic indicators have shown a broad-based slowdown.  Such a slowdown around the middle of 2010 has long seemed likely given the dependence of growth over the prior year on the boost from the inventory cycle and fiscal policy.  Our forecast is that real GDP will grow at a 1½% (annualized) rate in the second half of 2010 and in early 2011, and the risks to it are tilted to the downside.
  • But the forecasting community has only partially caught up with the deterioration in the numbers.  Last week’s FOMC statement suggests that Fed officials still expect the economy to grow at a slightly above-trend rate over the next year or so.  Likewise, most private forecasters predict that GDP will grow at roughly a trend rate in the second half of 2010 and a somewhat above-trend rate in 2011.  If our view is correct, substantial further downward revisions are coming.  In turn, these are likely to trigger downgrades to consensus earnings forecasts toward our strategists' more cautious views, as well as a return to large-scale asset purchases or other forms of "unconventional" monetary policy by the Fed.
  • Will the economy in fact return to a technical recession, marked by declines in real GDP?  We think the odds are still against such an outcome, but the risk is a substantial 25%-30%.  In our view, this exceeds the likelihood of the trend/above-trend growth scenario envisaged in the consensus forecast.

Over the last few months, the US economy has shown signs of slowdown across a broad range of indicators:
 
1. GDP and its components.  We estimate that the Commerce Department will announce on Friday that real GDP grew just 1.1% (annualized) in the second quarter.  We do not yet have much “tracking” information for the third quarter, but we believe that the sparse data released so far are consistent with our forecast of a 1½% growth.
 
2. Business surveys.  At present, the ISM surveys for the manufacturing and nonmanufacturing sector—at 55.5 and 54.3, respectively—are still consistent with trend or above-trend growth.  However, other indicators such as the NFIB small-business survey and regional surveys such as the Philly Fed index look a lot softer.  We expect the ISM indexes to fall sharply in coming months, to around 50 in the manufacturing sector and to the low 50s in the nonmanufacturing sector.
 
3. Labor market indicators.  Both the establishment survey and the household survey of employment have shown a sharp slowdown in underlying job growth over the past few months.  Moreover, initial jobless claims have crept up in recent week and hit 500,000 in the latest week, with only some of this increase attributable to temporary factors.  Although we do not have a forecast for the August employment report yet, the early indications are that private sector job growth may have stalled (or possibly gone into reverse) in the last month.
 
A slowdown around the middle of the year has long seemed likely given the dependence of GDP growth since mid-2009 on the boost from the inventory cycle and fiscal policy.  Over the last four quarters, the swing from inventory liquidation to accumulation has contributed 1.9 percentage points to real GDP growth, and overall fiscal policy—federal, state, and local—has contributed a little over 1 percentage point to real final demand growth.  These two numbers are additive, which implies that almost all of the 3.2% growth in real GDP over the past year was due to temporary factors, and that final demand excluding the impact of fiscal policy grew by less than ½% over the past year.  (Final demand excluding the impact of federal fiscal policy but including the impact of state and local policy has declined slightly.)
 
Given these relatively easy-to-measure factors—and given that it is difficult to tell a compelling story for why underlying final demand growth should accelerate sharply from here—we find forecasts that do not look for GDP growth well below trend quite implausible.  If the inventory and fiscal effect in combination are zero, which is a relatively generous assumption in our view, underlying final demand growth would need to accelerate by more than 1 percentage point to reach even our 1½% growth pace for real GDP growth.
 
Nevertheless, most official and private forecasters still expect trend or above-trend growth.  Last week’s FOMC statement noted that “…the Committee anticipates a gradual return to higher levels of resource utilization…,” which we interpret as growth of around 3% coupled with a modest decline in the unemployment rate to 9% or a bit below over the next year.  And according to the August 2010 survey by Blue Chip Economic Indicators, Inc., the average private-sector forecaster still expects real GDP growth of 2.4% in the third quarter, 2.7% in the fourth quarter, and 3% in 2011 (on a Q4/Q4 basis), as well as a drop in the unemployment rate from 9.5% now to 8.8% at the end of 2010.  We believe that the GDP forecasts will need to fall by at least 1 percentage point and the unemployment rate forecast will need to rise by at least 1 percentage point.  As we discussed recently, such GDP revisions are likely to trigger downward revisions to the consensus forecast for corporate earnings toward our strategists' more cautious views (see Andrew Tilton, "Reconciling 'Micro' Strength with 'Macro' Weakness," US Economics Analyst, 10/32, August 13, 2010).  They are also likely to persuade the Federal Reserve to resume large-scale asset purchases or engage in other forms of "unconventional" monetary policy.
 
Will the economy in fact return to a technical recession, marked by declines in real GDP?  As explained recently, we think the odds are still against such an outcome (see Ed McKelvey, “Private-Sector Insurance against a Double Dip,” US Daily, August 12, 2010).  This is mainly because several components of economic activity that usually help drag the economy down during recessions have already suffered large hits and are unlikely to fall much further, if at all.
 
Despite this, we continue to believe that the risk of a renewed technical recession—defined as a return to quarter-on-quarter declines in real GDP—is an uncomfortably high 25%-30%.  In our view, this exceeds the likelihood of the trend/above-trend growth scenario envisaged in the consensus forecast.