Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
Many high-profile analysts have predicted recessions for 2010, 2011 and 2012 – incorrectly, at least thus far. A far more reliable source for recession forecasting is the unemployment rate, which can provide signals for the beginnings and ends of recessions. The unemployment rate model (article link), updated with the November figure (7.7%), does not signal a recession now, nor does it support Economic Cycle Research Institute's recent claim that a recession started in July 2012, nor their September 2011 recession call, nor any of their many "follow-up" recession calls.
The model relies on four indicators to signal recessions:
- A short 12-period and a long 60-period exponential moving average (EMA) of the unemployment rate (UER).
- The 8-month smoothed annualized growth rate of the UER (UERg).
- The 19-week rate of change of the UER.
The criteria for the model to signal the start of recessions are given in the original article and repeated in appendix A.
Figures 1, 2 and 3 shows the various indicators for the time periods 1968-77, 1987-94 and 2000-12, respectively. The pattern of the indicator graphs preceding recessions is always similar and reflect at least one criterion of the three which signal recession starts.
Figure 1 encompasses the 1973-75 recession to which ECRI referred in their Tell-Tale Chart article when they discussed the current improving employment figures which do not support their position that a recession started in July 2012, claiming that in the severe 73-75 recession job growth stayed positive eight months into the recession as well. However, the historic unemployment rate figures say otherwise and contradict this claim.
One can see that the short EMA of the UER (the blue graph) formed a trough in December 1973 at the start of the recession and that the UERg (the green graph) had a positive slope from May 1973 onwards to the end of the recession. None of this indicates employment growth and does not support ECRI's statement that employment was growing initially during this recession. According to the Bureau of Labor Statistics, the unemployment rate increased in the first eight months of the recession from 4.9% in December 1973 to 5.5% in July 1974.
Figure 2 is included here to show Moore's recession call which was reported by "The Wall Street Journal" on March 9, 1990. It would appear that Moore, who founded The Economic Cycle Research Institute in 1996, was far more guarded when forecasting the 1990 recession than the current ECRI spokesman is now. (Note the contents of the text box in figure 2 and compare this with the certainty of ECRI's September 2011 message (appendix B), which implied a 100% chance of recession at the time of their call. Also see appendix C in this regard.
Moore then put the odds of a recession occurring in the first half of 1990 at two-to-one, which translates into a probability of 67%. It is interesting that he used an employment index for his forecast, which appears to be a similar approach as this model uses.
Note, that at the date of his recession call the indicator graph pattern closely resembled the typical pattern prior to recessions. One can see that the unemployment rate had formed a trough (the red and blue graphs), and the short EMA of the UER was in the process of rising and crossing the long EMA to the upside. Additionally, UERg was greater than zero then and was rising (the green graph). (The model signaled recession later than Moore because after his call the UER and UERg were flat for a few months, only beginning to rise from the middle of 1990 onwards.)
Referring to figure 3 below, and looking at the end portion of the chart, one can see that none of the conditions for a recession start are currently present.
- The UER is not forming a trough and its short EMA is well below its long EMA - the blue and red graphs, respectively.
- UERg is currently at a low level, approximately minus 10%, and declining – the green graph.
- Also the 19-week rate of change of the UER is now at about minus 5%, far below the critical level of plus 8% - the black graph.
For a recession to occur, the short EMA of the UER would have to form a trough and then cross its long EMA to the upside. Alternatively, the UERg graph would have to turn upwards and rise above zero, or the 19-week rate of change of the UER would have to be above 8%. There is no indication that any of this will happen anytime soon; currently the trajectories of the unemployment rate's short- and long EMA are still downwards - none having a positive slope, UERg is far below zero, and the 19-week rate of change of the UER is also way below the critical level.
Based on the historic patterns of the unemployment rate indicators prior to recessions one can reasonably conclude that the U.S. economy is currently not in a recession, and that ECRI's recent claim that a recession began in July 2012 is wrong.
The model signals the start of a recession when any one of the following three conditions occurs:
- The short exponential moving average (EMA) of the unemployment rate (UER) rises and crosses the long EMA to the upside, and the difference between the two EMAs is at least 0.07.
- The unemployment rate growth rate (UERg) rises above zero, while the long EMA of the unemployment rate has a positive slope, and the difference between the long EMA at that time and the long EMA 10 weeks before is greater than 0.025.
- The 19-week rate of change of the UER is greater than 8.0%, while simultaneously the long EMA of the UER has a positive slope and the difference between the long EMA at the time and the long EMA 10 weeks earlier is greater than 0.015.
Appendix B: ECRI's September 2011 recession call
When examining the indicator pattern prevailing at the end of September 2011 (figure 3), the date of ECRI's initial recession call, one can see that the indicator graphs, when projected forward in time along their trajectories, could have then been aligned so as to possibly indicate recession at a later date.
At that time the short EMA of the UER was forming a trough and was poised to rise and cross the long EMA to the upside, and UERg was increasing and not declining. If this trend had continued for a while then the indicators could have signaled recession a few months later. However, the trend did not continue after the date of the call and the indicators never aligned themselves in a pattern consistent with a recession start signal. It is evident that ECRI's September 2011 recession call was not supported by the model then, nor anytime later.
Their call was wrong, since no recession occurred within a reasonable time afterwards. Moreover, the S&P 500 is up about 25% since the September 2011 recession call.
Here is the recession call:
On Wednesday, September 21, 2011 the Economic Cycle Research Institute issued its U.S. cyclical outlook, summarized with "Economy on Recession Track - The jury is in, and the verdict is recession."
| Today, we must sound the alarm bells loud and clear. ECRI's leading indices of U.S. economic activity have turned down in a textbook sequence - first the U.S. Long Leading Index, then the Weekly Leading Index, and finally the U.S. Short Leading Index. Their growth rates are also in cyclical downswings, as are the growth rates of every one of ECRI's sector-specific leading indexes. Under the circumstances, there is no indication that a reacceleration in economic growth is near at hand.
In the process of scrutinizing the evidence, we examined every one of these leading indexes to check whether they are in pronounced, pervasive and persistent (three P's) downturns consistent with a 'hard landing,' namely, a recession, rather than a non-recessionary slowdown. After examining the three P's for all of these leading indexes, we found that the overwhelming majority of their trajectories are currently in recessionary configurations. In practice, such a finding is sufficient to justify a recession call.
A useful way to summarize the evidence we see pointing to recession is to examine the spread of weakness among the components of ECRI's U.S. leading indexes of economic activity… In that context, the recessionary decline is a summary measure of numerous reliable leading indicators, coupled with an ominous drop in a broad measure of current economic activity representing facts, not forecasts, constitutes a compelling recession signal.
And here is the video and transcripts from it:
On Friday September 30, 2011, Lakshman Achuthan of the Economic Cycle Research Institute reviewed the weight of ECRI's research, observing "Now it's a done deal. We are going into a recession." Lakshman Achuthan on CNBC
| This is a done deal. We are going into a recession. We've been very objective about getting to this point, but last week we announced to our clients that we're slipping into a recession. This is the first time I'm saying it publicly. A broad range - this is not based on any one indicator - this is based on dozens of indicators for the United States - there is a contagion among those forward looking indicators that we only see at the onset of a business cycle recession.. These leading indicators, which are objective.. they have a certain pattern that they present in front of a recession, and that is in, that is in right now.
A recession is a process, and I think a lot of people don't understand that; they're looking for two negative quarters of GDP. But it is a process where sales disappoint, so production falls, employment falls, income falls, and then sales fall. That vicious circle has started. You're looking at the forward drivers of that, which are different indicators - there's not one - everything's imperfect. The Weekly Leading Index .. that is saying unequivocally, this is recession. Long Leading Index, which has a longer lead, is saying recession. Service sector indicators, non-financial services where 5 out of 8 Americans work, plunging. Manufacturing, going into contraction. Exports, collapsing. This is a deadly combination, we are not going to escape this, and it is a new recession.
Extract from Nate Silver's Message for Financial Advisors:Don't believe predictors who are overconfident
|…. In a society accustomed to overconfident forecasters who mistake the confidence they express in a forecast for its veracity, the expressions of uncertainty are not seen as a winning strategy. Furthermore, even if you know that the forecast is dodgy, it might be rational for you to go after the big score. There is often more upside in confidently making a bold prediction that turns out to have been right after the fact, than there is downside in having been wrong.|
Georg Vrba is a professional engineer who has been a consulting engineer for many years. In his opinion, mathematical models provide better guidance to market direction than financial "experts." He has developed financial models for the stock market, the bond market and the yield curve, all published in Advisor Perspectives. The models are updated weekly. If you are interested to receive theses updates at no cost send email request to firstname.lastname@example.org.
Past performance is no guarantee of future returns or that the models referred to in this article will be profitable at all. The opinions stated here could be wrong due to false signals generated by the models, or that the models were incorrectly structured and/or interpreted.