Thomas Piketty, Apple fan-boy and sample biacist

Thomas Piketty, in Capital in the Twenty-First Century, is apparently an Apple fan-boy:

Note, too, that Steve Jobs, who even more than Bill Gates is the epitome of the admired and talented entrepreneur who fully deserves his fortune, was worth only about $8 billion in 2011, at the height of his glory and the peak of Apple’s stock price. That is just one-sixth as wealthy as Microsoft’s founder (even though many observers judge Gates to have been less innovative than Jobs)…

[As long as Piketty is deciding who deserves to be rich, why did he not set aside any $billions for the shareholders of Xerox who financed the development of the modern personal computer at PARC? Or for the researchers who programmed the Alto?]

This is part of a chapter where Piketty says that he has determined that extremely rich people earn a better return on investment than average schmoes (in the case of Steve Jobs, of course, he might be right due to Jobs’s ability to grant himself backdated stock options (nytimes)). Piketty did this by looking at the Forbes 400 and similar journalist-produced lists of rich people around the world.

it is perfectly possible that wealthier people obtain higher average returns than less wealthy people. There are several reasons why this might be the case. The most obvious one is that a person with 10 million euros rather than 100,000, or 1 billion euros rather than 10 million, has greater means to employ wealth management consultants and financial advisors. If such intermediaries make it possible to identify better investments, on average, there may be “economies of scale” in portfolio management that give rise to higher average returns on larger portfolios. A second reason is that it is easier for an investor to take risks, and to be patient, if she has substantial reserves than if she owns next to nothing. For both of these reasons—and all signs are that the first is more important in practice than the second—it is quite plausible to think that if the average return on capital is 4 percent, wealthier people might get as much as 6 or 7 percent, whereas less wealthy individuals might have to make do with as little as 2 or 3 percent.

This is a surprising hypothesis since generally the larger the fund the closer the results are to indices. Also, the main investment vehicle that is available to rich people but not to the rabble is the hedge fund. Yet hedge funds, on average, have underperformed the S&P 500 in recent years (see this Bloomberg article, which notes that “Hedge funds last beat U.S. stocks in 2008”) and, except for the high fees, may not differ from the S&P 500 in the long run (see “The 20-Year Performance Of Hedge Funds And The S&P 500 Are Almost Identical”).

Piketty cites one or two examples of tracking individuals, e.g.,

Take a particularly clear example at the very top of the global wealth hierarchy. Between 1990 and 2010, the fortune of Bill Gates—the founder of Microsoft, the world leader in operating systems, and the very incarnation of entrepreneurial wealth and number one in the Forbes rankings for more than ten years—increased from $4 billion to $50 billion.

This 13.5 percent annual growth (compared to 8.5 percent for the S&P 500 over the same period) is presented to support the proposition that rich people get high returns, with no discussion of whether or not it might have more to do with the worldwide growth of the PC, the explosion of the consumer Internet, and the substantial monopoly that Microsoft enjoyed until its corporate suicide with Windows 8.

Piketty does the rest of his analysis basically by looking at the enormously rich as a group in 1987 and then looking at the enormously rich in 2010. Thanks to vibrant economic growth worldwide, today’s super rich are indeed ridiculously richer than the super rich of 1987. From this Piketty concludes that super rich people get great returns on investment. A potential problem with his analysis is that he has made no attempt to track the extent to which these are the same rich people/families in 2010 as in 1987. Thus he has succumbed to sample bias (classic example of sample bias: interviewing people in baggage claim and asking “what percent full was your flight?” will result in an overestimate of load factor because there are more people on full planes than on half-full planes).

The sample bias in the case of the Forbes 400 is stated right at the top of the article: “these are the richest bastards in the world.” Anyone who had a big pile of cash in 1987 and invested it in underperforming assets will, by definition, not be on the Forbes 400 list in 2010. And someone who made big leveraged bets that went well (see John Paulson) is a likely candidate for the list.

For completeness, here is the rest of Piketty’s analysis:

Furthermore, today’s global growth rate includes a large demographic component, and wealthy people from emerging economies are rapidly joining the ranks of the wealthiest people in the world. This gives the impression that the ranks of the wealthiest are changing rapidly, while leading many people in the wealthy countries to feel an oppressive and growing sense that they are falling behind. The resulting anxiety sometimes outweighs all other concerns.

[Note the French academic focus on the anxiety being felt by rich people in rich countries. Did he actually interview rich people in Europe and North America to come to the conclusion that they worry 24/7 over whether a family in India has a bigger private jet or nicer house?]

Yet in the long run, if and when the poor countries have caught up with the rich ones and global growth slows, the inequality of returns on capital should be of far greater concern. In the long run, unequal wealth within nations is surely more worrisome than unequal wealth between nations.

[Why the focus on national borders? Because I do not have $70 billion in my Bank of America account, I feel like a pauper compared to Carlos Slim. Would I feel even worse if I moved to Guanajuato and had less than $70 billion in a Banamex account?]

The oldest and most systematic ranking of large fortunes is the global list of billionaires that Forbes has published since 1987. According to Forbes, the planet was home to just over 140 billionaires in 1987 but counts more than 1,400 today (2013), an increase by a factor of 10 (see Figure 12.1). In view of inflation and global economic growth since 1987, however, these spectacular numbers, repeated every year by media around the world, are difficult to interpret. If we look at the numbers in relation to the global population and total private wealth, we obtain the following results, which make somewhat more sense. The planet boasted barely 5 billionaires per 100 million adults in 1987 and 30 in 2013. Billionaires owned just 0.4 percent of global private wealth in 1987 but more than 1.5 percent in 2013, which is above the previous record attained in 2008, on the eve of the global financial crisis and the bankruptcy of Lehman Brothers (see Figure 12.2). This is an obscure way of presenting the data, however: there is nothing really surprising about the fact that a group containing 6 times as many people as a proportion of the population should own 4 times as great a proportion of the world’s wealth.

The only way to make sense of these wealth rankings is to examine the evolution of the amount of wealth owned by a fixed percentage of the world’s population, say the richest twenty-millionth of the adult population of the planet: roughly 150 people out of 3 billion adults in the late 1980s and 225 people out of 4.5 billion in the early 2010s. We then find that the average wealth of this group has increased from just over $1.5 billion in 1987 to nearly $15 billion in 2013, for an average growth rate of 6.4 percent above inflation.2 If we now consider the one-hundred-millionth wealthiest part of the world’s population, or about 30 people out of 3 billion in the late 1980s and 45 out of 4.5 billion in the early 2010s, we find that their average wealth increased from just over $3 billion to almost $35 billion, for an even higher growth rate of 6.8 percent above inflation. For the sake of comparison, average global wealth per capita increased by 2.1 percent a year, and average global income by 1.4 percent a year, …

To sum up: since the 1980s, global wealth has increased on average a little faster than income (this is the upward trend in the capital/income ratio examined in Part Two), and the largest fortunes grew much more rapidly than average wealth. This is the new fact that the Forbes rankings help us bring to light, assuming that they are reliable. Note that the precise conclusions depend quite heavily on the years chosen for consideration. For example, if we look at the period 1990–2010 instead of 1987–2013, the real rate of growth of the largest fortunes drops to 4 percent a year instead of 6 or 7.4 This is because 1990 marked a peak in global stock and real estate prices, while 2010 was a fairly low point for both (see Figure 12.2). Nevertheless, no matter what years we choose, the structural rate of growth of the largest fortunes seems always to be greater than the average growth of the average fortune (roughly at least twice as great). If we look at the evolution of the shares of the various millionths of large fortunes in global wealth, we find increases by more than a factor of 3 in less than thirty years (see Figure 12.3). To be sure, the amounts remain relatively small when expressed as a proportion of global wealth, but the rate of divergence is nevertheless spectacular. If such an evolution were to continue indefinitely, the share of these extremely tiny groups could reach quite substantial levels by the end of the twenty-first century.

For example, if the top thousandth enjoy a 6 percent rate of return on their wealth, while average global wealth grows at only 2 percent a year, then after thirty years the top thousandth’s share of global capital will have more than tripled. The top thousandth would then own 60 percent of global wealth, which is hard to imagine in the framework of existing political institutions unless there is a particularly effective system of repression or an extremely powerful apparatus of persuasion, or perhaps both. Even if the top thousandth’s capital returned only 4 percent a year, their share would still practically double in thirty years to nearly 40 percent. Once again, the force for divergence at the top of the wealth hierarchy would win out over the global forces of catch-up and convergence, so that the shares of the top decile and centile would increase significantly, with a large upward redistribution from the middle and upper-middle classes to the very rich. Such an impoverishment of the middle class would very likely trigger a violent political reaction.

As we will see, only a progressive tax on capital can effectively impede such a dynamic.

Piketty’s analysis has, I think, the same logical and mathematical merit as the following process:

  1. look at People magazine’s World’s Most Beautiful List for 2013 and compute that the average age is 35
  2. look back to find that Michelle Pfeiffer was top of the list in 1990 at age 32
  3. conclude that therefore Michelle Pfeiffer and other 1990 winners are aging at only 3/23rds or 13% of the rate as the rest of us (comparatively ugly) people

[This process has the advantage that the researcher gets paid a comfortable university salary to spend time looking at pictures of Michelle Pfeiffer. Combining Piketty’s love of French history and the challenge of thoroughly investigating the slowed-down aging of People’s Most Beautiful one would need to start by viewing Dangerous Liaisons]

[Finally, for those who are curious, “sample biacist” in the headline is my own coinage.]

8 thoughts on “Thomas Piketty, Apple fan-boy and sample biacist

  1. I’m curious what you thought of Piketty’s discussion of the rates of return achieved by university endowments (later in the same chapter), as summarized in Table 12.2. That seemed like pretty strong evidence of scale effects.

  2. Russil: What do I think of Piketty’s Table 12.2? shows that the total real (inflation-adjusted) return on the S&P 500 during the 1980-2010 period was 7.7 percent per year. The universities as a group got 8.2 percent, according to Piketty. That’s not an enormous boost. Schools with small endowments did even worse? I guess that shows that schools that are bad at raising money are also bad at managing money.

    The table provides an example of sample bias. He singles out the universities that are currently the richest, e.g., Harvard, Yale, and Princeton and points out that they got good returns. But if those schools had invested their money very unwisely and were no longer rich/prominent, he wouldn’t have put them at the top of the table.

    Also, remember that these big schools use leverage, either directly or by investing in hedge funds and private equity funds that are leveraged. So their returns will exceed the returns on the underlying investments any time that the underlying stuff is going up. “Buy the S&P 500 with a bit of leverage/margin” is not the most common strategy but it is one available to investors with less than $1 billion and, as a lot of observers are fond of pointing out, it gets one very close to the performance of the best regarded private equity partnerships (using leverage to the same extent as the PE partnership) without paying the 20% fees.

  3. Russil: Note further that Piketty’s Table 12.2 shows the selected-after-the-fact schools getting a return of 10.2 percent per year. So if you were a rich person with unique access to a hedge fund that had the same return on investment you would get… about 6 percent per year (10.2 percent return every year, minus 2 percent fee, minus 20 percent of the profits).

  4. Thanks for your take, Philip. If “buy the S&P 500 with a bit of leverage” is sufficient to get Harvard levels of risk-adjusted return — 10% annual real return over 30 years — why is it so uncommon? Why hasn’t everyone been doing it? (Could you point me to a reference?)

    The nice thing about the university data is that it avoids the Michelle Pfeiffer problem — we can track the returns achieved by the individual endowments.

    Is there still a survivorship bias problem? We can check by going back and looking at the top endowments 20 or 30 years ago. Top 5 in 1990: Harvard, University of Texas, Yale, Princeton, Stanford. Top 5 in 2013: Harvard, Yale, Texas, Stanford, Princeton.

    In other words, this doesn’t look like the typical mutual fund, where a period of higher returns is followed by reversion to the mean, and funds which are in the top quartile one year are unlikely to be there again the next year.

    The 2010 Yale endowment report argues that where there’s few opportunities to beat the average in efficiently priced markets like large-cap US stocks, there’s many more opportunities in the “infrequently traded, illiquid world of private assets.”

    This is actually a pretty deep question, with literally billions of dollars at stake: does active management work, or should large investors just stick with a passive approach (buying index funds, which is the usual recommendation for small investors)? A recent news story notes that over the last eight years, the Canada Pension Plan’s sovereign wealth fund (currently at $220 billion) has achieved $3 billion in additional gains beyond the corresponding passive benchmark.

  5. Russil: Thanks for the references. Yale thought back in 2010 that they could do better than public equities? shows that they failed to demonstrate that for the period 2009 through 2013. In fact, they underperformed a 60/40 mix of stocks and bonds by about 3% per year.

    My understanding is that the richest schools of 1990 were rich because of asset allocation, i.e., they weighted stocks more heavily than other schools starting in the 1950s (bonds were the traditional choice).

    Why doesn’t everyone leverage the S&P? Everyone who is in the private equity world does! They just don’t always realize it. Mitt Romney at Bain Capital, for example. When running for President here in the U.S., he touted his great performance from 1984 through 1999 (in borrowing money and picking companies to buy with that borrowed money) but he very likely could have achieved a similar return by leveraging the S&P 500 with the same factor.

    [Similar to how someone who picks 50 stocks, watches them go up, and thinks of him/herself as a genius.]

  6. Thanks, Philip. I found a couple references to leveraged investing for individual investors. Unsurprisingly, Canadian commentators tended to be negative: MoneySense on “Lifecycle Investing” (which argues that young people should be 200% invested in the stock market, by Ayres and Nalebuff). Rob Carrick.

    Carrick does mention an inspiring story about a 36-year-old financial adviser who took out a $250,000 investment loan in December 2008:

    A little over a month ago, with the stock market sloshing around its low point for the past year, Andrew Guilfoyle did something gutsy and brilliant.

    Or maybe crazy and destructive. Only time will tell how smart Mr. Guilfoyle, a married 36-year-old financial adviser with three young boys, was to borrow $250,000 and invest it in stocks.

    … Mr. Guilfoyle believes it’s impossible to time the exact right moment to get into the market. His thinking is that the market is low, and it’s bound to rise at some point.

    “Please keep in mind that I have a very long horizon,” he said. “I think the minimum time frame to look at something like this is five years.”

    What Mr. Guilfoyle is doing is called leveraging – borrowing to get more exposure to the market than you could with your own cash. It’s a risky strategy, so much so that Mr. Guilfoyle has never actually had one of his clients use it.

    Some advisers have promoted leverage to clients, in some cases to generate fees and commissions. But Mr. Guilfoyle worries about the emotional state of someone who loses a lot of money in a portfolio bought with borrowed money. “How could you ever push leverage on somebody? It’s a dynamite stick waiting to blow up.”

    For our own retirement savings, my philosophy is: “More money has been lost reaching for yield than at the point of a gun.” A simple 50/50 split between equities and fixed-income isn’t going to maximize our returns, but we don’t need to take any higher risks.

    Thanks for reading Piketty and giving us your thoughts! I think you’re pretty close to the end — footnotes take up the last 75 pages of the book.

    If reading Piketty has failed to sate your appetite for mainstream economics books dealing with Big Issues, I’d recommend Atif Mian and Amir Sufi’s House of Debt. It hasn’t gotten nearly as much attention as Piketty, but it’s perhaps even more important. If Piketty’s book deals with inequality and its causes, Mian and Sufi tackle the question of why the current slump has gone on so long (compared to the bursting of the 2000 bubble, for example). They conclude that the key problem is not lack of bank lending, it’s the overhang of household debt due to the housing crash. The Obama administration (particularly Geithner) appears to have thought that their work was done once they’d saved the banks, but they were very wrong. Summary from the Economist.

Comments are closed.