Although the purpose of this paper is not to discuss performance, please make sure you read the disclaimers at the bottom.
I have managed or co-managed Tocqueville’s largest account since the end of 1974: first with my partner Christian Humann at Tucker Anthony, then alone and, since 1992, with the help of my partners at Tocqueville Asset Management.
The reason for choosing this account as reference is that, besides being our largest, at over $1 billion, it is the only account for which we have audited statements going back nearly four decades. Therefore it should be useful for drawing some lessons about investing for the long run. This is the third version of a paper originally entitled The Rear Long View, now updated and edited to include the experience of the recent crisis and recovery.
A look back at the 39-year record calls for some observations:
1. When measuring investment performance, the starting date is all important.
I always remembered this account being worth around $12 million when we took over its management but the audited record showed it multiplying by less than 38 times from the end of 1976 to the end of 2006, when I wrote the first version of this study. Something did not jibe: 38 times $12 million is “only” $444 million, not $1 billion. Then, while doing some file cleaning, my secretary uncovered an old, hand-made (un-audited) performance analysis, which included the years 1975 and 1976.
The account indeed was worth less than $12 million at the end of 1974. But in its first two years it had gained a touch over 78%, as the world’s stock markets rebounded spectacularly from the depths of the 1973-1974 bear markets. When the measuring period was lengthened to cover the thirty-two years 1974-2006, instead of the thirty years since 1976, the portfolio was shown to have multiplied by a bit more than 67 times which, after adjusting for capital received and disbursed over the life of the account, was generally consistent with both my recollection and the auditors’ work.
We have now updated the table to incorporate the years 2007 through 2013, which include the crisis years of 2007 and 2008, and the recovery years of 2009-2013. As can be seen in the appendix, the account’s performance index did make a new high (on a year-end basis) in 2010, less than three years after the onset of the crisis, and its annual compound rate of growth over 39 years averaged 12.5 percent per annum, implying that, without additions or withdrawals of capital, the original capital would have multiplied almost 98 times.
So, the rate of appreciation of our portfolio increased greatly by adding two years of strong gains (1975-1976) at the beginning of the measuring period. When investment managers present their record of performance, it is important to check the period they chose to put forward and the kind of market environment that prevailed while this performance was being achieved. This is especially true for short periods, like five years for example, which are often used for presentations but in my opinion are somewhat meaningless out of context.
2. A dollar isn’t what it used to be.
In the last 39 years, the purchasing power of a dollar was eroded by inflation, even at a slowing rate. Back then, one dollar used to purchase roughly 4.7 times what it does today. Our client’s true wealth, as opposed to his “paper worth”, therefore increased by 21 times – not 98 times – over 39 years. On the one hand, stocks seem to have been a good inflation hedge over that period; on the other, inflation makes performances look better that they have been in terms of “purchasing power”.
3. The miracle of compounding.
Albert Einstein once marveled that “the most powerful force in the universe is compound interest”. Over 39 years, our largest account grew at an average compound rate of 12.5% (after charging management fees). This sounds pretty good but, in fact, it is only 0.5% more per year than the 12.0% achieved by the S&P 500 index when the index’s dividends are assumed to be automatically re-invested.
One might wonder if a mere additional 0.5% per year was worth all our efforts. Judge for yourselves: Over thirty-nine years, without any inflows or outflows, $12 million growing at 12.0% per annum will become $997 million, whereas if it grows at 12.5% per annum, it will become $1,186 million – a $189 million difference on an initial $12 million investment!
Moreover, although the last three years saw our portfolio appreciate 24.2%, this represented a period of underperformance against the major U.S. indices, as the S&P 500, for example, gained 56.8% over the same three years. This being said, this kind of behavior has not been particularly rare in the past. Just as adding a couple of good years at the beginning of a period can markedly improve the total performance over the period, adding a couple of bad years at the end can weigh down on the overall return.
[Note that there exist many different stock market indexes with varying characteristics of company size, geographic location, etc. The S&P 500 is U.S. based and one of the most commonly used. In addition, in recent crisis-affected years, it also has been one of the best performing.]
4. Performance is uneven.
The stock market is, by nature, a cyclical beast and there will always be some great stock market years, followed by mediocre or even negative ones.
Surges like the one in 1975-1976 have not been that exceptional. Our portfolio experienced other outsized gains throughout the period: for example, 70% in 1979-80; 65% in 1982-83; 105% in 2003-06 and 98% in 2009-2013.
But, to average to 12.5% per year, there must have been some more subdued years – and there were. One satisfaction we take from our long-term record, however, is that down years, overall, have been few and rather mild: -5.3% in 1977; -1.4% in 1984; -1.0% in 1990; -4.1% in 2002. The only true exception was 2008, when the market declined 37.0% as measured by the S&P 500 and our portfolio lost “only” 33.1% but was still a disappointment when compared to the historical pattern.
It is also fairly frequent for the value approach to bounce quite strongly in the early recovery following a severe decline but then to underperform on the upside once momentum begins to rule the market – for a while. For example, in the 1982-1983 bounce that followed the 1980-1981 bear market, our portfolio gained 65.5% vs. 48.7% for the S&P 500. In 1984-1987, however, it gained only 34.5% while the S&P 500 rose 73.9%. Similarly, the portfolio gained 59% in 2009-2010 vs. 45.5% for the S&P 500; but it only gained 24.2% vs. 56.8% for the S&P 500 in 2011-2013. These, by the way are measured by comparing calendar year-end figures, but…
5. The calendar makes it look better.
Customarily-used annual performance figures can be misleading. A well-known mutual fund manager, whose sixteen-year streak of beating the S&P 500 Index every year was brutally interrupted in 2006, had once commented that at least some of his extraordinary record was due to the vagaries of the calendar. In many interim periods, he pointed out, his fund had performed poorly or actually declined. The same could probably be said of most managers with superior long-term records.
6. Volatility is not risk.
Risk comes either from shaky investment fundamentals, unforeseen outside factors (“black swans”) or gross discrepancies between stock prices and companies’ values. It implies the possibility that you may lose all or part of your investment permanently. From that perspective, short-term or medium-term volatility (unless extreme) is mostly “noise” to be largely ignored. As Warren Buffet is fond of saying, “time is the investor’s friend and the speculator’s foe”.
7. To time or not to time the market: that is not the question.
I don’t know of any investment manager who has made a lot of money over a significant period of time (several major cycles) by timing the market. In fact, various studies stress the opportunity loss of being un-invested during market advances that, generally, are near-impossible to forecast – at least in their timing. One such study, from Townley Capital Management and the University of Michigan, covers thirty years. It shows that $100 invested in the stock market in 1963 would have earned $2,333 by 1993. But missing the best 90 days of that period (an average of 3 days a year) would have cut the earnings from $2,333 to just $110.
Some of my partners feel a responsibility to be more or less fully invested at all times. I don’t feel as strongly about this. But all of us always give precedence to stock picking over economic or market considerations. We follow principally contrarian-value criteria to identify and select fundamentally-researched stocks globally: If we find compelling new ideas, we buy them regardless of our market views; if we don’t, we do not buy and cash tends to build up as other stocks in our portfolios reach full value and are sold.
Differences among us really better qualify as nuances about the definition of “compelling idea”, especially when we view stocks as generally overvalued. But interestingly, in the particular case of the account that is the object of this nostalgia trip, and for which we use several in-house managers, my observation is that, though our individual performances have diverged at times, they show a surprising convergence over the longer term.
8. Catch a rising star; Avoid the falling knife.
Receiving fresh cash to manage in a rising market constitutes a conundrum. Even though we cannot “time” the market to try and get in at the precise bottom, there are some guidelines we can, and do, follow.
At any particular point in time, our existing portfolios typically are made up of 1) cheap stocks at or near their buying point; 2) stocks that have moved up, but are neither dirt-cheap nor excessively overpriced; and 3) stocks that are approaching our selling target price. For a new account, stocks in the first category clearly can be bought immediately and those in the third category should not. For stocks in the second category, however, the decision is more difficult. Often, we will buy some partial positions but, sometimes, we will just wait for a better opportunity.
In such situations, some view of where we stand in the market cycle can be helpful. A glance at the table below, derived from a study by Crestmont Research, makes it fairly clear that it is generally better to buy when prevailing price/earnings ratios are low than when they are high.
|87 Twenty-Year Periods|
This observation, though not perfect, is consistent with our contrarian-value approach. We cannot “time” the market, but we can identify periods when the ratio of perceived risk to potential return augurs poorly for future stock market gains. In such periods (usually ones of high price/earnings ratios), we believe that “constructive inaction” is the best response: we keep in mind that, just as missing the best market days can hurt performance, avoiding the worst days should help protect it.
So, we keep actively looking for new, compelling ideas and, until we find them, we buy nothing. On the other hand, if we do find them, we forget about the market’s position and buy the ideas.
9. If you are so smart…
As value-contrarian investors, we do worst when the market is carried away from its fundamentals by a strong upward momentum. In the technology and Internet bubble of the late 1990s, for example, our selected account badly trailed the S&P 500, advancing 101.3% from 1994 to 1999, while the index surged 251.3%. Fortunately, before our clients lost patience with our “discipline”, the bubble burst and the market suffered three consecutive years of decline while we kept plodding along. As a result, the account finished the 1994-2003 cycle up 97.1% against the 82.1% chalked up by the S&P 500 – with a lot less volatility.
After this episode, over lunch, another client recalled: “I knew all along this was crazy, but I must say that, quite a few times I wished François were a bit crazier. Now I know why I stayed with Tocqueville.”
Upon reading the first draft of this review, one of my younger partners claimed to be surprised: as a rule, I seldom mention performance and in fact advocate that we not publicize it too much, because “performance chasers” seldom turn out to be very desirable (or successful) clients. I answered him that doing so once every thirty-two years is acceptable, and that when I do it next, he probably will be retired.
In fact, this article was not about “relative” performance and the futile competition in which our industry tends to engage. It really was about putting wealth protection and growth into historical perspective, and drawing from that exercise some pearls of wisdom. So, here they are:
There are no miracles or magical black boxes in investing for the long term.
Patience, discipline and common sense will always prevail over restlessness and “genius”.
For the rest, we can trust compound interest.
This article reflects the views of the author as of the date or dates cited and may change at any time. The information should not be construed as investment advice. No representation is made concerning the accuracy of cited data, nor is there any guarantee that any projection, forecast or opinion will be realized.
The returns discussed in this article and the Appendix are based upon the annual returns for each of the last thirty-six years for fully-discretionary accounts managed by Tocqueville Asset Management and François Sicart, founder and Chairman, for its largest client. The client account predates the formation of Tocqueville on January 1, 1990, and was managed by Mr. Sicart initially as an executive of Tucker Anthony, R.L. Day, Inc. beginning in 1974 through the formation of Tocqueville. Other accounts were managed by Mr. Sicart during the same period, and may have had different investment objectives and achieved different results.
A new account with similar investment objectives and style may not achieve similar results. Performance data quoted represents past performance and does not guarantee future results. The returns were calculated using a time-weighted monthly rate of return formula and are presented net of advisory fees, commissions and trading expenses and, assumes reinvestment of capital gains and dividends. The accounts are valued monthly and transactions are recorded on a trade date basis. Dividend income is recorded on a cash basis. Cumulative rates of return for multi-year periods are calculated by linking the annual rates with such periods. The annualized rate of return is equivalent to the annual rate of return which, if earned in each year of the indicated multi-year period, would produce the actual cumulative rate of return over the time period.
The client account includes investment in foreign securities which involve greater volatility and political, economic and currency risks and differences in accounting methods. The S&P 500 Index is a market-value weighted index consisting of 500 stocks chosen for market size, liquidity and industry group representation. The S&P 500 Index returns include reinvestment of dividends. The volatility and other risk characteristics of the S&P 500 Index may be greater or less than those of the client account. You cannot invest directly in an index.
References to stocks, securities or investments in this writing should not be considered recommendations to buy or sell. Past performance is not a guide to future performance. Securities that are referenced may be held in portfolios managed by Tocqueville or by principals, employees and associates of Tocqueville, and such references should not be deemed as an understanding of any future position, buying or selling, that may be taken by Tocqueville. We will periodically reprint charts or quote extensively from articles published by other sources. When we do, we will provide appropriate source information. The quotes and material that we reproduce are selected because, in our view, they provide an interesting, provocative or enlightening perspective on current events. Their reproduction in no way implies that we endorse any part of the material or investment recommendations published on those sites.
Client Account 1974=100