Home > Investment Strategies > Buy & Hold is Dead

Buy & Hold (hereafter referred to as “B&H”) is an asset allocation strategy.  A clear majority of well-diversified, turnkey asset management solutions follow B&H.  Often, such turnkey solutions will claim to utilize some type of an active asset allocation solution (i.e., not B&H).  But upon closer examination, it is found that in the vast majority of such cases, their claims are false.  Their adjustments to the asset selection or asset weightings are far too small to make any practical difference.  So despite their claim to be actively asset allocating, in practice they remain B&H.


B&H is closely related (if not identical) to Strategic Asset Allocation (hereafter referred to as “SAA”).  Unfortunately, no universally accepted definition exists for SAA.  In general, SAA is considered to have little or no active component.  If it is implemented with no active component, then it becomes what is commonly known as the investor’s Policy Asset Allocation.  If the SAA implementation retains some degree of active asset selection or weightings, these consist of slight deviations away from the investor’s Policy Asset Allocation.  Generally, SAA consists of what could be called “intelligent” diversification and is based in large measure on three academic developments:


  1.  Modern Portfolio Theory, MPT  (Harry Markowitz 1952)
  2.  Capital Asset Pricing Model, CAPM  (Jack Treynor 1961, William Sharpe 1964, John Lintner 1965, Jan Mossin 1966)
  3.  Multi-factor pricing models, e.g., Fama-French Three-Factor Model  (Eugene Fama 1993, Kenneth French 1993)


In practice, B&H and SAA are essentially equivalent.  The underlying logic on which they are founded is as follows:


  1. There exist well-defined and meaningfully differentiated asset categories
  2.  These have inherent, fundamental investment properties
  3.  These properties are stable and dependable over time, e.g., stocks earn an average 10% per year subject to a 20% standard deviation
  4. Prices follow a random walk
  5. Investment markets efficiently price and reflect all currently available information
  6. In the aggregate, investors are logical and sensible, correctly processing the available information


If these underlying assumptions were actually true, then it would be impossible for investment markets to go through such long episodic periods during which they either perform or don’t.The long-term history of asset class returns demonstrates that the assumptions underlying B&H are false.  Consider the following performance results (all returns are after inflation has been taken out and have been annualized) [1]:


US stocks, S&P 500


15.61% per year for 18.1 years  (7/31/1982 to 8/31/2000)

– 0.03% per year for 19.2 years  (5/31/1963 to 7/31/1982)


US treasuries, 10-year maturity


6.96% per year for 30.8 years  (9/30/1981 to 7/31/2012)

– 0.01% per year for 84.6 years  (2/28/1897 to 9/30/1981)


US corporate bonds, long-term maturity, AAA-rated


7.86% per year for 30.8 years  (9/30/1981 to 7/31/2012)

– 0.01% per year for 49.6 years  (2/28/1932 to 9/30/1981)


Commodities, Reuters Jeffries CRB


4.88% per year for 14.6 years  (11/30/1993 to 6/30/2008)

– 0.04% per year for 14.3 years  (8/31/1979 to 11/30/1993)


Obviously, given these return patterns, an investor using B&H would have to wait 40 years for the ultra long-term averages to be realized.  Very few investors can afford to wait 40 years.  So if B&H ignores reality, then what is its basis and where did it come from?  In truth, there were four contributors to the crime, the first of which was academia.


Academia develops abstract, over-simplified models that are intended to help us advance our thinking of how the real world works.  Abstraction and simplification are important components when attempting to understandisolated relationships.  The problem comes when one applies such over-simplification to the real world.  The two main bodies of academic work that led to B&H were:


  1.  Efficient Market Hypothesis  (Eugene Fama early 1960s)  –  Investment markets correctly reflect all currently available information
  2. Random Walk Hypothesis  (Louis Bachelier 1900, Paul Cootner 1964, [2])  –  Investment markets don’t have trends


If these two hypotheses were actually true, then the market is always right, and it would be a waste of time to do anything other than simple B&H.


The second contributor to the crime was the investment industry itself.  Unfortunately, our industry has in large measure introduced investment products that are likely to underperform or, at best, match markets over the very long-term [3], [4].  In other words, most investment industry solutions fail to side step or otherwise mitigate market downdrafts.  The prevalence of such non-performing products only served to reinforce the notion that the market is always right and that it is a waste of time to do anything other than B&H.


The third contributor was the institutional consulting business.  As Charlie Ellis observes in his FAJ article “Murder on the Orient Express: The Mystery of Underperformance,”[4] institutional consultants suffer from and are defined by their potent “agency interests.”As Dr. Ellis points out, it is not in the consultant’s interest to focus on those types of decisions and structures that lead to long-term superior performance results for their pension, endowment, or foundation clients.  Instead, they focus on obtaining another fee-paying project (manager search, asset allocation study, liquidity study, risk audit, asset/liability modeling, etc.) and engage in valueless hyper-diversification.


The final contributor to the crime of B&H was the investment market itself.  Here in the United States, we experienced a history-making 18-year bull market during which virtually every single asset category appreciated at the same time  –  and by a lot.  This near uniform positive return pattern, spanning almost all asset classes, cemented the notion that B&H was the way to go and clearly worked.  Academia contributed the theories, models, fancy names, and Nobel Prizes.  Industry delivered products that added no value.  Institutional consultants were conflicted and suffered from their “agency interests.”  Mr. Market, across the board, went straight up.  With this level of support, how could anyone suggest that B&H was an emperor unaware of his own nakedness?


The problem is that no asset category has sufficiently stable return properties for B&H to have the slightest degree of applicability.  Just as a single example, consider the long-term history of US stocks (S&P 500 total returns after inflation has been taken out, annualized).  Quality data exist for the last 138 years (12/31/1874 to 10/1/2012).  During the last 138 years, US stocks experienced just six periods during which they generated attract investment returns (net of inflation) [1]:


11.05% per year for 13 years  (6/30/1877 to 6/30/1890)


10.04% per year for 13 years  (8/31/1893 to 8/31/1906)


12.15% per year for 13 years  (8/31/1916 to 8/31/1929)


13.34% per year for 13 years  (5/31/1932 to 5/31/1945)


16.50% per year for 13 years  (11/30/1948 to 11/30/1961)


15.04% per year for 13 years  (7/31/1984 to 7/31/1997)


Unfortunately during the remaining 60 years, the US stock market returned an average annual loss of -2.10% per year!  Obviously, any asset category that loses the investor money over a 60-year period can hardly be of any use in a B&H strategy (even if it was broken into separate non-contiguous time segments).  Similar results are found for every other major asset category.  Here are the results for US corporate bonds, long-term maturity, AAA-rated (again, all net of inflation) [1]:


8.62% per year for 13 years  (12/31/1874 to 12/31/1887)


4.89% per year for 13 years  (12/31/1887 to 12/31/1900)


10.49% per year for 13 years  (6/30/1920 to 6/30/1933)


2.54% per year for 13 years  (12/31/1959 to 12/31/1972)


10.65% per year for 13 years  (6/30/1982 to 6/30/1995)


5.78% per year for 13 years  (7/31/1999 to 7/31/2012)


Disastrously, during the remaining 60 years, US corporate bonds returned as average annual loss of -1.08% per year!  In contrast with the flawed assumptions underlying the B&H strategy, all asset categories go through long periods during which they are working or they are not.  One should think of these as long-duration episodic eras.  These eras are not random.  Instead, powerful economic, capital market, and political drivers define them.  It is these “primary drivers” that determine and define an episodic era  –  and therefore which asset categories will be favored and which will be disadvantaged.  Recall the episodic era for US housing.  After adjustment for inflation, US home prices returned 4.72% per year for the 13 years from 1/31/1993 through 1/31/2006 [5].  Since that time, they have fallen off a cliff.  Remarkably different episodic drivers define these two eras, pre-2006 and post-2006:


Pre-2006 Episodic Drivers

  • Leveraging
  • Deregulation
  • Rapid growth of global economic trade
  • Falling marginal tax rates
  • Huge benefits realized from the 1980s restricting and reinvention of American business
  • Greatest fall in mortgage rates in US history
  • Technological advancement
  • Loss of manufacturing jobs


Post-2006 Episodic Drivers

  • Deleveraging
  • Regulation and massive regulatory uncertainty
  • Global trade stagnation and risk of global trade wars
  • Rising marginal tax rates
  • Demographics resulting in reduced housing demand
  • Government growth reaches its limits
  • Return of manufacturing jobs


These primary drivers are remarkably different and immediately obvious.  Moreover, their prospective impact on asset category returns is relatively straightforward.  The solution is not Market Timing, nor is it Tactical Asset Allocation.  Instead, the solution is to identify the primary drivers defining the forward-looking episodic era and then assembling an asset mix in strong alignment with those drivers.  This is the highly successful approach used over many decades by institutional global macro hedge funds.  But that will be a story for next month’s article.


Leave a Reply