Back

Blogs

Speaker Highlights from the Grant’s Conference, Plaza Hotel in New York, April 9, 2019 – Part 2

Benjamin Lavine, CFA, CAIA

4/15/2019

The following are excerpts of selected presentations taken from the Grant’s Conference held at the Plaza Hotel in New York on April 9, 2019.  This is Part 2 where we highlight Edward McQuarrie’s presentation, “Asset Allocation Reconsidered.”  Permission has been granted by Grant’s Interest Rate Observer for 3D Asset Management to publish these excerpts.  3D interjects its own commentary and views throughout the excerpts.

Afternoon Presentation: Edward McQuarrie, Professor Emeritus, Santa Clara University

Edward McQuarrie gave the last presentation at the conference.  He argued that the ‘Siegel Constant’ of 6.6% real return for stocks used as the basis for the traditional 60/40 portfolio needs to be reconsidered in light of additional research that he and his team uncovered on the historical performance of U.S. stocks and bonds.  McQuarrie did disclose that his research is preliminary and a work-in-progress.  You can find McQuarrie’s research on ssrn.com such as “Stock Market Charts You Never Saw (October 2017)”.    

Titled “It’s Not What You Don’t Know that Hurts You – It’s What You Know for Sure that Just Ain’t So.  Revisiting Stocks for the Long Run,” McQuarrie’s divided his critique of Professor Jeremy Siegel’s seminal work, “Stocks for the Long Run,” into two parts:

• Referencing Siegel’s existing research into stock and bond performance, McQuarrie argued that the historical outperformance of equities over fixed income is episodic at best and that Siegel fails to present a systematic case for equity outperformance (i.e. do we have an enduring “equity premium” or just a one-time bond deficit?).
• Using an expanded dataset going back to 1793, McQuarrie calculated the historical real return to stocks and bonds as follows:
  • All periods (220 years ending 2012): 6.11% stocks and 4.26% bonds.  The 6.11% real return for stocks is lower than the 6.6% return calculated by Siegel.  Rather than representing the expected average return, McQuarrie characterized the Siegel Constant of 6.6% real return as the “peak return.”
  • Through January 1940, the equity premium shrinks substantially with stocks returning a real 6.01% return versus a real 5.41% for bonds.
  • Through January 1897, the equity premium vanishes with stocks returning a real 6.02% return versus a real 6.18% return for bonds.

If confirmed and validated, McQuarrie’s research has significant implications for buy-and-hold investing in order to capture historical equity risk premiums over bonds.  Rather than representing a systematic phenomenon, equity outperformance over bonds is relegated largely to a post-World War II market regime. 

Even using Siegel’s existing research, McQuarrie observed that stocks underperformed bonds for the 60 years leading up to the Civil War.  After the Civil War, stocks matched bonds.  Stocks also matched bonds after the late 1960s.  Much of the post-World War II outperformance of stocks over bonds can be isolated to 1949 – 1968. 

Looking at the entire US historical record, McQuarrie observed that there are only three isolated periods where stocks significantly outperformed bonds: 1) 1862 – 1872; 2) 1897 – 1925, and 3) 1949 – 1968 or ~60 years of history.  In contrast, over ~150 years of history, stocks only managed to match bonds: 1) 1802 – 1861; 2) 1873 – 1896; 3) 1926 – 1948; 4) 1969 – 2012.

McQuarrie then recalculated stock and bond returns over the period from 1802 through 2012 and found a significant reduction in the historical equity premium.

For bonds, McQuarrie found the following:

• An extra 0.60% in annualized real return from 1/1802 through 1/1871 by aggregating multiple bond indices, not incorporating the minimum yield from Homer (see “A History of Interest Rates” from Sidney Homer and Richard Sylla) and using corporate only bond from 1857 on.
• An extra 1.05% in annualized real return from 1/1871 through 1/1926 by relying on data from Snowden (1990) who recast Macaulay’s railroad bonds dataset into holding period returns.
• An extra 0.42% in annualized real return from 1926 through 2012 by swapping Ibbotson SBBI Corporate for SBBI Long Government.

For stocks, McQuarrie found the following:

• A reduction of 0.64% in annualized real return from 1/1802 through 1/1871 by observing dividends, including Philadelphia banks, and including shares of the 1^st and 2^nd Bank of the United States.  By contrast, Siegel’s work excluded major stock markets like the Philadelphia market as well as shares of the Bank of United States.  McQuarrie credited Richard Sylla for compiling the expanded stock market dataset. 
• A reduction of 0.20% in annualized real return from 1871 through 1925 by assuming annual re-investment of dividends rather than monthly.  McQuarrie cites Goetzman et al (2001) in arguing that “total return must be calculated on an annual basis (McQuarrie 2018).” 
• A reduction of 0.19% in annualized real return from 12/1925 through 12/2012.  Again, McQuarrie applies annual reinvestment rather than monthly reinvestment of dividends.

McQuarrie’s recalculation of bond returns is the primary contributor to the reduction in Siegel’s equity risk premium calculation.  McQuarrie maintained that Siegel’s bond compilation only represents 1/100 amount of bonds traded at the time.  This is primarily due to Siegel’s insistence on only using risk-free bonds rather than including corporate bonds for calculating the historical bond return.  However, Siegel’s desire to only reference risk-free bonds resulted in the inclusion of tiny, obscure issues from small towns across Massachusetts and Maine (1865 – 1914) as well as omitting most Federal bonds that traded before 1835, such as Hamilton Refunding US Bonds (see Garber 1991). 

McQuarrie’s dataset expands bond coverage from Siegel’s coverage through 1926.  McQuarrie acknowledged his insistence in including corporate bonds for calculating the historical bond return, in contrast to Siegel, but McQuarrie wanted to replicate the average bond investor experience whose bond portfolio would likely include bonds not part of Siegel’s bond dataset. As a result, McQuarrie’s calculation of historical bond returns is 3.5x greater than Siegel’s calculated returns. 

In addition, McQuarrie’s stock dataset has much greater coverage prior to 1871 with a complete observation of the dividend record – dividends accounted for ~100% of the total return experienced by early stock investors.  McQuarrie’s research tapped a wide variety of early sources such as legislative records, early compilations, and miscellaneous sources (i.e. corporate biographies).  It wasn’t until the 20^th century that stock appreciation became a greater component of total return versus dividend payouts.  Without the early dividend history, estimates of total return “are just guess work.”

McQuarrie concluded that the “Siegel gambit” has failed.  The 19^th century of U.S. stock and bond performance does not resemble the performance experience of the 20^th century – will the 21^st century resemble the latter or revert to the former?  There is little certainty that stocks will beat bonds over extended periods of time, covering multi-decade intervals.  And it is unlikely that stocks will return the Siegel estimated real return of 6.6% over any lengthy planning interval.

Our two cents: two major critiques come to mind concerning McQuarrie’s analysis:

• The nature of capital market structures and level of transparency are clearly different between the 19^th versus 20^th and 21^st centuries.  There are differences in regulations and investor protections not to mention technology in pricing, information processing and communication, trading, etc. that has produced more refined risk pricing for equity and bond issuance.  A corporate financial officer has more tools at his/her disposal to better determine the company’s weighted average cost of capital (WACC).  As a result, the risks to equity and fixed income investors are more clearly spelled out that should result in a better discernment of how much risk should be priced into each instrument type.
• Yes, capital markets have a long history of boom/bust cycles going all the way back to the Dutch tulip mania in the 1630s– history may not repeat itself, but it does rhyme (Mark Twain).  And one can certainly point towards multiple lapses in rationality and episodes of misbehavior by investors caught up in excess fear/greed.  But underpinning capital market pricing is a rational view of risk-taking based on the Capital Asset Pricing Model (CAPM) framework introduced by Jack Treynor, William Sharpe, John Lintner, and Jan Mossin throughout the 1960s.  CAPM has been criticized by the likes of Fama/French, not for its theoretical underpinning of investor behavior towards risk, but that the CAPM model is too simplistic and leaves out other risk premiums such as small cap and value.  Unfortunately, McQuarrie’s research fails to produce a theoretical framework for explaining his observations, so one must assume that McQuarrie views capital market pricing more from a behavioral perspective (i.e. investors overestimate/underestimate capital market risk due to behavioral tendencies) rather than a rational perspective (i.e. investors expect to be compensated with capital market premiums such as stocks over bonds for the extra risk taken).  Perhaps, when McQuarrie does publish his final work, he will provide a literature review of academic work that would explain why capital markets truly follow a random walk regardless of the risks they present to investors.

Disclosure: The above is the opinion of the author and should not be relied upon as investment advice or a forecast of the future. It is not a recommendation, offer or solicitation to buy or sell any securities or implement any investment strategy. It is for informational purposes only. The above statistics, data, anecdotes and opinions of others are assumed to be true and accurate however 3D Asset Management does not warrant the accuracy of any of these. There is also no assurance that any of the above are all inclusive or complete.

Past performance is no guarantee of future results. None of the services offered by 3D Asset Management are insured by the FDIC and the reader is reminded that all investments contain risk. The opinions offered above are as April 15, 2019 and are subject to change as influencing factors change.

More detail regarding 3D Asset Management, its products, services, personnel, fees and investment methodologies are available in the firm’s Form ADV Part 2 which is available upon request by calling (860) 291-1998, option 2 or emailing sales@3dadvisor.com or visiting 3D’s website at www.3dadvisor.com.