In Depth Analysis: CalculatedRisk Newsletter on Real Estate (Ad Free) Read it here.

Saturday, July 03, 2010

Recession Dating and a "Double Dip"

by Calculated Risk on 7/03/2010 06:18:00 PM

My forecast is for U.S. economic growth to slow in the 2nd half (a sluggish and choppy recovery), but not slide into recession. However a recession is a possibility, and the following describes how NBER differentiates between a "double dip" and a new recession.

The National Bureau of Economic Research (NBER) Business Cycle Dating Committee is the recognized group for dating recessions in the U.S. It is always difficult to tell when a recession has ended, especially with a sluggish recovery. If the economy slides back into recession - a possibility right now - the NBER has to decide if it is a continuation of the previous recession, or if the new period of economic decline is a new separate recession.

This is just a technical question: for those impacted by the recession it makes no difference if it is called a "double dip" or a new recession.

Yesterday an AP story quoted Robert Hall, the current Chairman of NBER on a "double dip": So what exactly is a 'double-dip' recession?

"The idea -- hypothetical because it has yet to happen -- is that activity might rise for a period, but not far enough to complete a cycle, then fall again, and finally rise above its original level, only then completing the cycle."
The closest we've seen to a "double dip" was in the early 1980s - and the NBER dated those as two separate recessions. We can use the NBER memos from that period to look for clues. From July 8, 1981 announcing the end of the 1980 recession: Business Cycle Trough Last July
The period following July 1980 will appear in the NBER chronology as an expansion. An important factor influencing that decision is that most major indicators, including real GNP, are already close to or above their previous highs.
And from the January 1982 announcing the beginning of the 1981-1982 recession: Current Recession Began in July
The committee also reviewed its earlier decision that a peak of economic activity occurred in January 1980 and a trough in July 1980 and reaffirmed that decision. Although not all economic indicators had regained their 1979-80 peaks by the summer of 1981, the committee agreed that the resurgence of economic activity in the previous year clearly constituted a business cycle recovery.
And from The NBER's Recession Dating Procedure
In choosing the dates of business-cycle turning points, the committee follows standard procedures to assure continuity in the chronology. Because a recession influences the economy broadly and is not confined to one sector, the committee emphasizes economy-wide measures of economic activity. The committee views real GDP as the single best measure of aggregate economic activity. ...

The committee places particular emphasis on two monthly measures of activity across the entire economy: (1) personal income less transfer payments, in real terms and (2) employment. In addition, the committee refers to two indicators with coverage primarily of manufacturing and goods: (3) industrial production and (4) the volume of sales of the manufacturing and wholesale-retail sectors adjusted for price changes.
GDP is the key measure, and the NBER actually uses two measures of GDP: 1) real GDP, and 2) real Gross Domestic Income (GDI). For a discussion on GDI, see from Fed economist Jeremy Nalewaik, “Income and Product Side Estimates of US Output Growth,” Brookings Papers on Economic Activity.

Below is a look at four of the measures mentioned: real GDP (and real GDI), industrial production, employment and real personal income excluding transfer payments.

Note: The following graphs are all constructed as a percent of the peak in each indicator. This shows when the indicator has bottomed - and when the indicator has returned to the level of the previous peak. If the indicator is at a new peak, the value is 100%. On all graphs the recent recession is marked as ending in July 2009 or Q3 2009 - this is preliminary and NOT an NBER determination. GDP is quarterly, the other data is monthly.

Recession Measure GDP Click on graph for larger image in new window.

The first graph is for GDP through Q1 2010.

This is the key measure and real GDP is only 1.3% below the pre-recession peak - and real GDI 2.0% below the previous peak. GDP probably increased in Q2 too (probably at close to a 3% annualized rate), and at the end of Q2 both of these measures will be even closer to the previous peak, but not there yet.

If you look at the period between the two early '80s recessions, both real GDP and real GDI returned to pre-recession levels before declining again.

Recession Measure Industrial Production The second graph is for monthly industrial production based on data from the Federal Reserve through May.

Industrial production is still 8.1% below the pre-recession peak -and now it appears that growth is slowing in the manufacturing sector (although still expanding). Even if growth continues, it will take some time before industrial production is back to pre-recession levels.

Between the early '80s recessions, industrial production didn't quite return to pre-recession levels - but it was only about 0.5% below the previous peak.

Recession Measure EmploymentThe third graph is for employment through June.

Between the two recession in the early '80s, employment returned to the pre-recession peak.

This time employment is barely off the bottom.

Recession Measure IncomeAnd the last graph is for real personal income excluding transfer payments through May. This bottomed in Sept 2009, but has only increased slightly since then and is still 6% below the pre-recession peak.

Once again - looking back - this measure returned to the pre-recession peak between the 1980 and 1981/1982 recessions.

Based on these graphs and the NBER memos, it would seem pretty easy to date two recessions in the early '80s. However, if another recession starts this year, it will almost certainly be dated as a continuation of the "great recession" that started in 2007. If so, I'll need more blue ink to shade all my graphs ...