During my usual morning reading process I came across a posting on PragCap by Cullen Roche regarding a chart of U.S. recession probabilities.  The chart can be found at the St. Louis Federal Reserve website and is derived from a study by J. Piger and M. Chauvet, from the University of Oregon, which was published in the Journal of Business and Economic Statistics in 2008.  (For other economic geeks the full paper is attached)

Cullen points out that "What’s interesting about this index is the current reading.  At 20%, the index is at a level that has ALWAYS been followed by a recession. As you can see below, the index has never approached 20% without a subsequent recession. All 6 recessions since 1967 have coincided with 20%+ readings in the US Recession Probabilities index."  Currently, that index, as shown in the chart below, is approaching that 20% level as of August which is the latest reported data.

The recession probabilities, as stated in the research article, are obtained from a "dynamic-factor markov-switching model applied to four monthly coincident variables: non-farm payroll employment, the index of industrial production, real personal income excluding transfer payments, and real manufacturing and trade sales."  Without getting to far into financial econometrics a Markov switching model involves multiple equations that can characterize the time series behaviors between different data sets.  By permitting the switching between equations the model is able to capture more complex dynamic patterns.  The question that we want to answer is whether the indicator is currently correct in its prediction of a U.S. recession or "is this time different."

We have written recently about the ECRI's recession call, which has been extremely early, and is probably the most widely debated recession call to date.  However, there have been other calls as well from notables such as Gary Shilling and John Hussman.  I also have been laying out the foundation in recent reports about the pending recession which will most likely "officially" set in in 2013 (see here, here and here)  I say "officially" set in because while many of the composite and economic indicators that we follow are already signaling a recessionary economy - it is the National Bureau of Economic Research (NBER) that officially dates recessions in the U.S.     

From the NBER Website:  "The NBER does not define a recession in terms of two consecutive quarters of decline in real GDP.  Rather, a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales."

Furthermore, the NBER awaits the final revisions to those data points before determining the peaks and troughs of economic activity.  Therefore, it is usually several months after the fact before the NBER officially announces the start or end of a recession. 

Why is this important?  As we have discussed in the past understanding when a recession has begun is hugely important to investors.  The table below, which uses monthly S&P 500 data, shows the price declines during recessionary periods going back to 1873.   The average drawdown to investor's portfolios is a little more than 30%.  During recent recessions the damage has been far worse.  

However, the declines in the market, and subsequent damage to investors portfolios, happened WELL before the "official announcement" by the NBER.  As an example the majority of the decline in the stock market in 2008 had already occurred by the time the NBER announced in December of 2008 that the recession had begun a year earlier.  (As a side note I stated in the December 8, 2007 newsletter that the recession had already started) 

Understanding this is important.  While the media continues to look at data points from one month to the next and proclaim that no recession is in the offing - the reality is that by the time the data conclusively points to a recession it has generally been far too late to be of use.  The chart below shows the probabilities of recession as discussed by J. Piger and M. Chauvet.

As stated above, each time this indicator has signaled the probability of a recession at 20%, or higher, the economy has either been in, or was about to be in, a recession.  However, while the indicator is currently at a level that is indicative of a recession, along with a host of other economic indicators confirming the same (LEI Coincident to Lagging Ratio, ADS, GDI, Final Sales, STA Economic Composite, etc.), the NBER will wait to date the recession until the data revisions are in.  

Historically, as shown in the table below, the amount of time between the recession probability indicator hitting 20% and the NBER confirming the start of the recession has been on average about 8 months.  However, since the turn of the century that lag has moved to roughly 11.5 months.  

Therefore, if the recession probability indicator is once again signaling the onset of a recession, and we use the average lag time of 8 months for the NBER to confirm, then the official dating of a recession starting in August or September of 2012 should occur in April of 2013.  However, due to the impact of QE programs, suppressed interest rates, and other artificial interventions, it is likely that the lag could be more equivalent to recent history pushing the official dating into July or August of 2013.

Regardless of when the NBER officially announces the start date of the next recession - the damage will have already been done to investors.  The current decline in earnings and revenue, the drop in exports and the weakness in incomes will likely impact stock prices long before the mainstream media recognizes that a recession has set in.  As stated above there are many confirming indications that a recession is already underway in the U.S. economy - the only question is how long it will be before the stock market recognizes it and begins to realign prices with revised valuations.

A Comparison of the Real-Time Performance of Business Cycle Dating Methods