The easiest way to outperform most investors and grow your wealth is based on a simple concept that all investors should understand, both as a tool for investing and as an example of logical thinking about the markets. Consider a mutual fund that buys every stock trading on a major US stock exchange, investing in each company a percentage equal to that company’s percentage of the total value of all the US stocks. Thus the fund behaves like the entire market, with the same daily percentage price changes and dividend payouts. This means if the oil giant Exxon has a market value, computed as share price times number of shares outstanding, of $400 billion and the total market value of all stocks is $10 trillion, then the index fund puts 4 percent of its net worth in Exxon, and so on for all the other stocks. A mutual fund like this that replicates the composition and investment results of a specified pool of securities is called an index fund, and investors who buy such funds are known as indexers.
Call any investment that mimics the whole market of listed US securities “passive” and notice that since each of these passive investments acts just like the market, so does a pool of all of them. If these passive investors together own, say, 15 percent of every stock, then “everybody else” owns 85 percent and, taken as a group, their investments also are like one giant index fund. But “everybody else” means all the active investors, each of whom has his own recipe for how much to own of each stock and none of whom has indexed. As Nobel Prize winner Bill Sharpe says, it follows from the laws of arithmetic that the combined holdings of all the active investors also replicates the index. Although this idea is well known, and I’m not sure where it first appeared, I first heard it from him and he has given the clearest exposition I’ve seen. I’ll call it Sharpe’s Principle.
I met Bill Sharpe in 1968 or 1969, when we were both young professors at UCI. Highly regarded, he had already completed the work for which he was awarded the Nobel Prize in 1990. Unfortunately he was in UCI’s School of Social Sciences and I didn’t really get to know him before he was recruited by Stanford just two years after his arrival. Had he still been at UCI after Princeton Newport Partners was well under way, might we have collaborated? He contributed a key simplification for understanding options, the binary model, and I might have been able to convince him that markets have significant inefficiencies—in other words, opportunities for abnormal risk-adjusted returns. Discussing this in 1975 when I invited him to lecture at UCI, Bill argued that my rewards from PNP didn’t demonstrate market inefficiency, because you could argue that I and my associates were simply getting paid according to our worth. Had we turned our talents to other areas of economic endeavor we could expect the same.
Before costs, each passive investor gets the same return as the index. This is also true for the active investors as a group but not for each one individually. Holding a larger percentage than the index in some stocks and less in others, they may do better or worse than the index in various periods. Although the results (before costs) for the entire group of active investors matches the return on the index, their individual returns are statistically distributed around it with most fairly close and some quite different.
They have more risk without the expectation of more return. Reducing risk through diversification is one reason to buy an index, but an even more important one is reducing the costs that investors bear. Index funds trade infrequently, with stocks turning over just a few percent a year as the “keepers” of the index occasionally add and remove stocks, or because cash flows into or out of the fund. On the other hand, active investors as a group have been trading more than 100 percent of their portfolio per year. This imposes a substantial cost on them from commissions and by their impact on market prices.
To illustrate the losses from market impact, suppose XYZ stock has a “true” price of $50 a share. Assuming for simplicity that it trades in 10-cent increments, between trades there will be buyers bidding for various amounts at $49.90, $49.80, $49.70, and so forth. Similarly, sellers will be asking $50.10, $50.20, et cetera. Someone who places an order to buy at whatever price is available in the market, called a market order and one of the most common types, will pay $50.10, a little above the true price. This 10-cent difference between the price paid and the “true” price is called market impact. Market impact increases with order size since, to continue our example, a large market order may clean out not only the offering at $50.10 but also stock offered for sale at higher prices, resulting in an average purchase price above $50.10 and a market impact greater than 10 cents per share.
When Steve Mizusawa and I operated Ridgeline Partners, we reduced these costs by dividing large orders into smaller ones of $20,000 to $100,000, and waiting a few minutes between transactions to allow the market price to recover. We know the “true” price is somewhere at or between the highest bid price (the Bid) and the lowest asking price (the Asked), but not exactly where. On average, it is about halfway between the two. To see that market impact is a real cost, suppose in our example that just after buying stock at $50.10 the buyer wants to sell it at market. He gets $49.90, for an immediate loss of 20 cents or about 0.4 percent.
Investors who don’t index pay on average an extra 1 percent a year in trading costs and another 1 percent to what Warren Buffett calls “helpers”—the money managers, salespeople, advisers, and fiduciaries that permeate all areas of investing. As a result of these costs, active investors as a group trail the index by 2 percent or so, whereas the passive investor who selects a no-load (no sales fees), low-expense-ratio (low overhead and low management fee) index fund can pay less than 0.25 percent in fees and trading costs. From the gambling perspective, the return to an active investor is that of a passive investor plus the extra gain or loss from paying (on average) 2 percent a year to toss a fair coin in some (imaginary) casino. Taxable active investors do even worse, because a high portfolio turnover means short-term capital gains, which currently are taxed at a higher rate than gains from securities, the sales of which have been deferred for a year. For instance, if $1,000 is invested at 8 percent and gains are taxed when realized, table 1 compares the result of paying the tax every year versus paying only at the end of a certain number of years. I used 35 percent for short-term capital gain taxes and 20 percent for long-term capital gains. Actual rates will vary with the investor’s tax bracket and changes in the law.
Influential private equity and hedge fund managers have persuaded their friends in Congress to grant them the benefits of deferring taxes on their overseas income for many years and, even better, then paying the tax not at the rates for ordinary income that are paid by wage earners, but rather at much lower long-term capital gains rates. The difference between the first and last columns of table 6 indicates the magnitude of the benefits.
Value of Investment |
|||
Investment Ends at Year |
Pay 35% Tax Every Year |
Pay 20% Tax Every Year |
Pay 20% Tax at End |
|
0 |
1,000 |
1,000 |
1,000 |
|
1 |
1,052 |
1,064 |
1,064 |
|
10 |
1,660 |
1,860 |
1,927 |
|
20 |
2,756 |
3,458 |
3,929 |
|
30 |
4,576 |
6,431 |
8,250 |
If the index beats the pool of active investors by 2 percent each year, does that mean it also beats most equity mutual funds? Widely publicized, year-end annual reports show the S&P Index of 500 stocks beating a majority of mutual funds most years but not always. Why? For one thing, we’re comparing apples and oranges: The S&P 500 Index isn’t the whole market—if our universe is the total stock market index then it’s an active investor, although one with low costs—since it doesn’t include most small companies, so the assets of a mutual fund that are not part of the S&P 500 are not subject to Sharpe’s Principle as applied to that index. The S&P 500 stocks are selected by the Standard & Poor’s Corporation, with occasional deletions and additions. Although these five hundred large companies account for roughly 75 percent of the market value of all publicly traded stocks, it omits some very large companies, notably, before 2010, Berkshire Hathaway, one of the ten largest US companies by market value. In fact, the compound annual return on small companies for the eighty-two years from 1926 through 2007 was 12.45 percent, compared with 10.36 percent for large companies. Yet the extra boost for mutual funds from having had some of their assets in smaller stocks still hasn’t offset their extra costs.
Another aspect of the apples-and-oranges comparison is the impact of cash balances. Since fund investors continually add or withdraw money, funds are partly invested in fluctuating cash balances. When the market rises strongly, the interest on this cash doesn’t keep up and the fund return lags the return on the equity portion of its holdings. Conversely, when the market is down sharply, the losses on the fund’s equity position are reduced to the extent it is in cash and by the interest it gets on that cash. The impact of this cash drag is generally small.
Also, non-index mutual funds are only part of the total pool of active investors. Conceivably, their managers could be relatively skilled, in which case the mutual fund group would outperform the rest of the active investors. In this case, though the active investors lagged overall, the mutual fund group might excel compared with the others. However, academic studies of the historical returns of mutual funds show little evidence of such managerial skill on the part of mutual funds. Third, it is not the number of active investors that must lag the index, according to Sharpe’s Principle. Instead, it is the return on the total pool of actively managed assets invested in the index that must underperform.
Morningstar, which tracks mutual fund performance, does periodic studies comparing fund performance with indexes. The 2009 results are typical. After adjusting for risk, size, and investment category, only 37 percent beat their benchmark over the previous three years, with similar results for five and ten years.
The benefits from indexing are shown in table 7. Here I have used historical returns on large stocks, like those in the S&P 500, with my assumed costs. More details are given in appendix B. After costs and inflation, tax-exempt passive investors gained 6.7 percent annually compared with 4.7 percent for the actives, or one-third more. After taxes it is 2.0 percent for the actives and 4.8 percent for the indexers, more than double.
Index |
Passives |
Actives |
|
|
Before costs |
10.1% |
10.1% |
10.1% |
|
After costs |
— |
9.7% |
7.7% |
|
After inflation |
7.1% |
6.7% |
4.7% |
|
Tax-exempt after inflation |
— |
6.7% |
4.7% |
|
After taxes |
— |
4.8% |
2.0% |
If you index, select a fund with annual expenses less than 0.2 percent. Reject funds that add management fees, sales loads, or other charges. The one charge you can ignore is a penalty for selling before a short holding period, such as thirty days, which funds introduced to prevent costly large-scale rapid in-and-out trading by certain investors.
Each year, typically at the end of October, US equity mutual funds assign the year-to-date taxable gains or losses to their current investors. If you were to make an investment shortly before this in a year when the fund had a lot of gain, you could experience the inequity of paying taxes on an amount far larger than your real economic gain. On the other hand, in a year when the fund allocated large losses, a purchase shortly before the time to receive the losses could let investors reduce their tax bill without having had a corresponding economic loss.
Tax-exempt investors such as IRAs, 401(k)s, employee benefit plans, and foundations ought to consider swapping their active investments in equities into a broad no-load index fund, unless they have strong reasons to believe their current investments give them a significant edge. In my experience, superior stock-picking ability is rare, which means almost everyone should make the switch.
Taxable investors need to review their holdings on a case-by-case basis. For instance, in 2015, with a cost basis of about $1,000 a share, a market price of $225,000 a share, and a combined federal and state tax rate of, say, 30 percent, I would net about $157,800 per share after a sale of my Berkshire Hathaway Class A stock. An index fund purchased with this smaller amount would have to do about 43 percent better than Berkshire in the future for me to catch up. This seems extremely unlikely.
Like me with Berkshire, investors who don’t trade, and use no advisers, will avoid the usual expenses paid by active investors. In fact, their costs may be even less than those of indexers. If such a buy-and-hold investor were, for instance, to choose stocks at random, purchasing an amount of each proportional to its market capitalization, we could show, by reasoning like that used to prove Sharpe’s Principle, that the expected return is the same as for the index from which the stocks were chosen minus the presumably small costs of acquiring the stocks.
The main disadvantage to buy-and-hold versus indexing is the added risk. In gambling terms, the return to buy-and-hold is like that from buying the index then adding random gains or losses by repeatedly flipping a coin. However, with a holding of twenty or so stocks spread out over different industries, this extra risk tends to be small. The threat to a buy-and-hold program is the investor himself. Following his stocks and listening to stories and advice about them can lead to trading actively, producing on average the inferior results about which I’ve warned. Buying an index avoids this trap.
For another way to look at index investing, suppose the same percentage of each US stock were put into a low-cost index fund and all the rest went into a giant pool actively managed by the world’s best managers. Then a clerk managing the index fund with a computer to do the bookkeeping would beat the team of the best managers on earth, by the amount of their extra commissions and fees. In contests promoted by journalists, random portfolios of stocks selected using chance devices such as darts, dice, or (figurative) chimpanzees hold their own against the experts.