What's the greatest investment feat ever? Consider a neglected study from The Journal of Investing entitled, "The Greatest Return Stories Ever Told," which reports that a fund called BGI-TAA posted the best risk-adjusted returns from the period from 1977-2000. Yes, this dark-horse BGI-TAA bested Warren Buffett's Berkshire Hathaway, the Ford Foundation, Harvard University's endowment, and myriads of other celebrated contenders over this 23-year period, achieving the highest Sharpe Ratio of them all. (We'll leave the shortcomings of the Sharpe Ratio for another article.)
BGI-TAA stands for Barclays Global Investors-Tactical Asset Allocation. The fund was launched in 1977 by Wells Fargo (BGI came later), the brain child of William Fouse, the greatest investor you've (probably) never heard of. So who is he? And how did he do it? And why are Fouse and BGI-TAA virtually unknown today? And what can we learn from his example?
Who was Fouse?
William Fouse was the father of the world's first index fund (in 1971) as well as the father of tactical asset allocation. Peter Bernstein's Capital Ideas devotes an entire chapter to him. You may be surprised to learn that the first index fund (an equal-weight fund of NYSE issues for the Samsonite pension fund) was not originally conceived as a static investment vehicle at all, the way index funds are often understood today. Fouse believed the market is relatively efficient and that gains are more likely to come from asset allocation than from security selection. Therefore, low-cost exposure to US stocks as an asset class was needed to better implement tactical asset allocation. Imagine what Fouse could have done with today's toolkit of low-cost asset class exposure using ETFs like SPY, JKL, JKF, RSP, ACWI, GLD, VIG, SHY, ITE, TLT, etc.
How did he do it?
Fouse's insight could not have been more basic: buy more stocks when they're cheap and fewer stocks when they're expensive. Compared to how Tactical Asset Allocation is often understood today, Fouse's TAA was relatively passive and incremental, shifting allocation infrequently, in measured steps, and only when merited by major shifts in the opportunity set presented by Mr. Market.
It's not really all that difficult to tell when stocks are expensive and when they're cheap-the hard part is having the discipline to act on it, against our natural tendencies toward inaction (as they rise) and panic (when they drop!). For example, in late September 1987, risk-free US Treasury bonds were priced to deliver higher returns than large US stocks over the intermediate to long term, based on the earnings yield of the S&P500 and Treasury yields. Not only that, but this negative risk premium was approx -2% (after allowing for earnings growth), the lowest risk premium since 1870! For anyone awake at the time, it was a no-brainer that stocks would have to fall to restore a normal risk premium. They did, of course, just a few weeks later, on Black Monday, October 19, 1987, when the market shed 22.6% in a single day.
Going into Black Monday BGI-TAA had just 10% allocated to stocks. However, after the crash, BGI-TAA quickly raised its stock allocation to 30%, scooping up stocks that had suddenly become very cheap. For the month of October 1987, the S&P500 experienced a 22% drawdown, 60/40 portfolios were down 12%, and BGI-TAA gained 6%.
Following this method of valuation-driven dynamic allocation, BGI-TAA captured 90% of the upside of the S&P500 and just 50% of the downside (comparing performance during the S&P's 57 best quarters and 27 negative quarters from 1977-1992).
Why is Fouse an Unknown?
It's hard to say for sure, but it is clear that Fouse's genius runs at cross-currents with the finance industry as a whole. On the institutional side, pension consultants formulate tightly-constrained "strategic asset allocations" that permit relatively little opportunistic shifting. Once a static policy allocation is set, the consultants focus on selecting managers and making sure those managers maintain "style purity," which precludes dynamic strategies. It's easier for them to simply outsource portfolio management and they can always blame the managers if things go wrong.
On the individual side, brokers and personal finance gurus generally preach similar dogmas, calling for model portfolios with infinite tweaks on conventional static asset mixes (30/70, 60/40, 80/20, etc.), all the while oblivious to the opportunities for profit and loss presented by Mr. Market.
To the mainstream finance world, Fouse is a heretic and his record is a stick in their eye. They are quick to dismiss any dynamic strategy as "market timing," which has become a kind of four-letter word in their circles.
What can we learn from Fouse?
To the open-minded, Fouse shows that equity-like returns can be achieved with substantially lower risk than equities. He also showed us the way to do this:
1) use prevailing market valuations and bond yields to assess the opportunities and risks to investors at any given time,
2) re-evaluate this changing opportunity set on a regular basis,
3) outline a plan for how to scale into and out of positions in risky and defensive assets given various valuation levels and bond yields, and
4) implement this plan in a disciplined way.
Dynamic allocation raises many legitimate questions and counterarguments. We'll tackle some of these in subsequent articles.
Dynamic allocation is "simple but not easy," in the famous words of Warren Buffett. You can do it yourself or you can work with a like-minded investment adviser. But be forewarned: most advisers and brokers will dismiss dynamic allocation out-of-hand. Ask them if they've ever heard of William Fouse.
Siegel, Lawrence, Knoer, Kenneth, and Clifford, Scott, "The Greatest Return Stories Ever Told," The Journal of Investing, Summer 2001, p.2,7-8.
Deringer, Janice and Lawrence Tint, "Development and Implementation of a Tactical Asset Allocation Model at Wells Fargo Nikko Investment Advisors," in Lederman, Jess and Robert Klein, eds., Global Asset Allocation - Techniques for Optimizing Portfolio Management, New York: John Wiley & Sons, 1994, p.237.
Arnott, Robert and Peter Bernstein, "What Risk Premium is 'Normal'?" Financial Analysts Journal, March-April 2002, p.79,84.