In his 2013 shareholder letter, Mr Buffett said he would instruct trustees to invest 90 per cent of his wife’s wealth in standard and poor 500 index funds and 10 per cent in short-term us treasuries.
An academic background study shows that this non-traditional allocation not only produces higher returns than traditional 40/60 and 30/70 portfolios, but also has a lower failure rate.
The essence of buffett’s portfolio is to divide the future into short periods where money is needed and long periods where stock yields are high.
In the lingo of bond managers, the effect is a “barbell” portfolio that focuses on very short and very long maturities and excludes the intermediate-berkshire term structure.
Barbells, it turns out, may be the best way to match maturities to the needs of pension funds, institutions, insurance companies and individuals, including some retirees.
The following passage from buffett’s 2013 annual shareholder letter caused quite a stir:
The Suggestions I make here are basically the same as some of the instructions listed in my will. The estate provides that the cash will be paid to the trustee for my wife’s use. (I have to use cash for a personal bequest because my Berkshire shares will be fully allocated to certain charities for the decade after my estate closes.) My advice to trustees cannot be simple: 10% cash in short-term government bonds and 90% cash in standard and poor 500-stock index funds. (I recommend Vanguard’s.) I believe the long-term results of this trust policy will be better than most investors – whether pension funds, institutions or individuals – hiring high-fee managers.
To his surprise, he chose the s&p 500 index over finding a way to keep his Berkshire model stock (brk.a) at 90% (brk.b). One could argue that, in order to be logically consistent, he should also mandate that all Berkshire’s future free cash flow should also be invested in the s&p 500. While this may not be Berkshire’s biggest possible cash future – it would be very bad to pay large taxable dividends – it would be a mistake. Investing in Berkshire is similar in principle to leaving an investment trust to one’s wife, but choosing specific investment vehicles is another matter. Mr Buffett has made clear that Berkshire has low risk and high returns, including buying back his own shares.
My guess is that the main reasons for choosing the s&p 500 for my wife’s legacy are simplicity and diversity. Buffett doesn’t seem to like halimakowitz or risk differentiators, but he is considering the basic statistical advantages of diversification when choosing a standard & poor’s 500 index fund over Berkshire. What buffett’s heirs have in common with most retired investors is the lack of professional complexity in investing. That’s the question buffett posed for Vanguard standard & poor’s 500 index fund or VFAIX. That’s most of us.
Mechanical rebalancing using the s&p 500 would go a long way towards eliminating attributions in human processes. When choosing effective portfolio managers, there is no possibility of stock picking or misjudgment. This is actually a prerequisite for the automatic rebalancing of index investments and bonds as individuals approach retirement. Buffett accepted the first premise, but rejected the latter. It’s still surprising that buffett considers himself an index investor in his own heirs.
One of the subtleties of buffett’s choice was that he chose the s&p 500 over the overall stock market index (MUTF). Some contrarian studies suggest that small-cap stocks, which account for about 25 per cent of the overall market index, have outperformed large-cap stocks, the largest 500. Why not include them? The reasons may be twofold. First, big capitalization reflects past success, which has gained some durability. Second, organizations based on the principle of maximum weighting ensure that they include innovative companies that become an important part of the index when they succeed and are recognized for their success. One can argue about the validity of both points, but it’s likely that buffett’s premise is S&
That’s the dividend. The dividend yield on the s&p 500 has been about 2 per cent in recent years, providing a good start to an annual exit rate of 4 per cent. With more than 2% of money-market funds now in use, about 20% of the cash allocation needs to be withdrawn before it can be rebalanced to 90/10. Sure, buffett points out that it’s easy and tax-efficient to get the same amount of revenue by selling off some of Berkshire’s mold inventory – more on that later – but a strategy that includes paying regular dividends is one way to reduce the amount required. The rest of the portfolio, thereby minimizing the need for action.
The amazing thing about buffett’s 90/10 allocation strategy is that it actually seems to work. Javier estrada, a professor at the prestigious IESE school of business at the university of navarro in Spain, has published a paper in Madrid and Barcelona (” buffett’s asset allocation advice: build on the distortions of October 26, 2015 “) that assumes an exit rate of 4 per cent over a 30-year series from 1900 to 2014. Failure is defined as running out of money in 30 years.
Surprisingly, buffett’s failure rate in a 90/10 portfolio is just 2.3%. More surprisingly, the failure rate of a 90/10 portfolio is much lower than that of a 40/60 or 30/70 portfolio. These are almost universally recommended retirement or individual retirement equity allocations. For Vanguard’s target retirement fund (65 years in 2015), the allocation is 40% equity / 60% bonds.
Reducing the equity allocation to 80% – the level of the vanguard’s most aggressive life-cycle fund – seems to offer more “divergence” protection than the “buffett” model. However, it did not increase the statistical failure rate, and the average rate of return decreased by 20%. In fact, only 70/30 (1.2%), 60/40 (0.0%) and 50/50 (1.2%) portfolios have very low failure rates, with mean and median yields falling off the cliff. Some simple rules for constructing withdrawals – estrada’s “twist” – significantly improved the Numbers. Buffett’s instructions may include these or similar adjustments.
Vanguard does not choose here. It happens to be where I keep most of my assets. Only vanguard is the most recognized standard of traditional wisdom, so it is the ideal basis for comparison.
These two main questions presented his wife with striking signs of buffett’s legacy – his vitally important 90% stock allocation and his choice of the s&p 500 as a vehicle. One can speculate on his choice of the s&p 500, but Mr Estrada’s paper suggests that a 90 per cent equity allocation may not be as crazy as it sounds.
As far as I know, however, no one is looking at the more interesting side of his 90/10 portfolio – a very unorthodox term structure. In bonds, it’s called a barbell.
Buffett’s distributional problems go beyond the overweight of stocks and the underweight of bonds. The shocking truth is that there are no bonds at all. No. The only other component of the portfolio is treasuries. It’s basically cash. Bond managers, whose portfolios have life-or-death structures, call them barbells.
Gym barbells are the perfect image. There are weights at both ends and only one empty bar in the middle. This is exactly what buffett’s 90/10 portfolio looks like. Cash and stocks create the ultimate barbell. Imagine a person carrying a small amount of cash – if they suddenly decide to eat out – and depositing the rest in stock. This could be buffett’s ideal choice. Of course, his wife may need to spend more money, and his other great love, Berkshire, may need more cash for things that come up. Otherwise, all stocks. It’s no coincidence that buffett’s legacy to his wife, and the way he invests with Berkshire, has taken the form of a barbell. The barbell is the basic structure of Berkshire Taurus itself. It’s also a complete departure from buffett’s normal insurance model. Various types of insurance companies have models of possible payment ranges, which may be what they have written in recent years. Life insurance is relatively easy, but property, casualty and reinsurance are harder because all these liabilities are hard to calculate – and new ones are harder
The traditional solution to this problem is to buy bonds of the same maturity to eliminate these possible obligations – even if they are in the distant future. That’s why insurance companies like higher interest rates. Insurance ceos like to sleep at night, and they’re convinced that their projected liabilities in 2030 are completely offset by safe assets, so I hit T, T’ cross that line. In the insurance industry, a bond portfolio is generally considered a conservative option.
Buffett disagrees. The role of the traditional approach is to pull the future into the present in a way that eliminates opportunities for time use.
Traditional insurance company methods use specific assets (mainly bonds) with the same maturity date to offset future liabilities. This is achieved by investing in “floating” future liabilities and increasing underwriting profits. Is that why insurance companies like Banks do better at higher rates
Long-term fixed income investing is almost unthinkable for buffett. This is especially true at the far end of the currently available yield curve. Articles have been written about buffett’s deployment of insurance buoys, but little attention has been paid to his strategy of allowing for almost 100% equity investments. It has a simple elegance. What buffett has done is keep a large enough position in treasuries to take care of future liabilities. He has said repeatedly recently that he will not allow available cash to fall below $20bn – a figure that will rise steadily as Mr Buffett’s insurance arm does more business. The number is not as arbitrary as his casual manner. Since he typically refers to an unacceptable figure of $20 billion in response to questions about deploying cash for acquisitions, note its main implications.
$20 billion is the short end of Berkshire’s barbell. Given that Berkshire generates more than $2 billion in cash each month, this is how much cash Berkshire needs to handle frequent or unexpected cash that may come up for the foreseeable future. It’s the daily items Berkshire needs to pay for in its wallet, the possibility of a bad hurricane season, the new normal of severe fires in California, and worse things like terrorism.
Buffett does not hold bonds because they are an extremely inefficient way to offset future risk. Demand that has been dormant for years will take care of itself as it rolls over its short-term cash position. Why spoil long-term returns by dragging particular distant future moments into the present? There is no need to focus on 2025 or 2035. Some hedge funds have tried to mimic this approach, throwing insurers together and seeking returns that go beyond bonds. We’ll have plenty of time to see if they can pull it off like buffett