Does ECRI's Weekly Leading Index Remain a Usable Indicator?

April 2nd, 2013

by Anton Vrba and Georg Vrba

Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.


In our continued endeavor to understand Economic Cycle Research Institute's Weekly Leading Index (WLI), and what it actually represents, we now discovered that the index is excessively driven by one of its components, the ratio of "ten year treasury bond yield" to "BAA corporate bond yield." A further component of the WLI is M2, a downgraded and obsolete leading indicator of the economy. These two components, which reflect US monetary policy, seemingly are causing the excessive movements of the WLI. We consequently believe that the WLI has become unable to provide leading information for the U.S. economy.

The known 1999 components of the WLI, published for a short time on the ECRI website are:

  1. Initial claims for unemployment insurance,
  2. Money supply (M2 plus),
  3. Industrial materials price index,
  4. Mortgage loan applications,
  5. 10 year treasury bond yield / BAA corporate bond yield
  6. BAA corporate bond yield; and
  7. Stock price index.

Component No. 5 puzzled us. The spread between the yields of short-term and long-term treasuries is widely used as an indicator, however the ratio of long treasury and corporate bond yields is unusual. How does one derive an indicator from two long term interest rates? In order to investigate this, we used the FRED weekly data series 10-Year Treasury Constant Maturity Rate (WGS10YR) and Moody's Seasoned BAA Corporate Bond Yield (WBAA) and produced an indicator X1 = WGS10YR divided by WBAA. Then we calculated X1g, the smoothed 6-month annualized growth rate of X1, using the same formula that ECRI uses to determine WLIg, the growth of the WLI. The graphic below compares X1g to WLIg.

The near exact match of the two graphs from 2008 onwards is astounding. As interest rates were reduced, any small movements in the treasury bond and/or BAA corporate bond yields gave disproportionate high inputs into the WLI, thereby explaining the wild swings of the WLI after the great recession.

The visual match is quantified mathematically by calculating a Pearson moment correlation of WLIg with X1g, shown in the table below. The high correlation coefficients 0.91 and 0.93 indicate the great influence of the "ten year treasury bond yield" to "BAA corporate bond yield" ratio on the WLI, which has lately swamped the other component data of the WLI.

We have now demonstrated that ECRI uses two highly questionable components (number 2 and 5 in the list) to calculate the WLI. This has severely impaired the WLI, at least from 2008 onwards. WLIg gave two incorrect recession signals since 2010 alone, equal to the number of false signals in all the years before, as can be seen in more detail on the chart below. We consequently believe that the WLI has now become unreliable to provide leading information for the U.S. economy.

Be warned, tasseography may be more reliable in its predictive prowess than what the WLI has been lately. This index has provided wrong signals for the US economy for several years now. A consequence of this may have been the ill conceived September 2011 message from ECRI that the U.S. economy was going into recession at the end of 2011. ECRI is on record as having stated that their recession call was based to a significant extent on the WLI. Their insistence to this date that they are not wrong could have significantly contributed to lost investment opportunities and false investment decisions for those who heeded that message.



Anton Vrba & Georg Vrba

iM imarketsignals.com

Anton Vrba is an electrical engineer. He pursued a career in R&D, manufacturing and construction project management. His interests are mathematics and physics. He is a lateral thinker and has ideas that challenge the established explanations to the workings of the universe, which he is in the process of publishing on his website.

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, yield curve, gold, silver and recession prediction, all published in Advisor Perspectives. The models are updated weekly at http://imarketsignals.com/.

Website by the Boston Web Company