The stock market. Source of unknown riches - but not necessarily for investors. So-called "professional" investors offer to manage your money. However, their fees are based on the level of assets managed, not performance. Hence their goal is to maximize assets, not performance, and prey for markets to behave. You will never hear a bad word about stocks from a professional money manager. the by-laws of many mutual funds do not allow the manager to have cash levels above 5% of assets. He has to be invested at least 95% at all times. On one hand, it is probably right to force money managers to concentrate on stock picking, not market timing. On the other hand, this puts the onus of market timing onto the inidiviual investors. Lighthouse's Alex Gloy's excellent presentation below proves finance doesn't have to be complex (people make it complex).
Via Lighthouse Investment Management's Alex Gloy,
The money management industry would like to have their clients' assets indefinitely, through bull- and bear markets. Ride the wave during good times. And simply state that "nobody could have foreseen this", "we don't have a crystal ball" or "it's too late to sell now" in case of a crash.
There must be a better way to invest.
This publication tries to assess the following questions:
1. What kind of return can be reasonably expected from stock market investments? Is that rate
2. What kind of simple tools exist to tell if the stock market is cheap or expensive?
3. Are stock market returns mean-reverting?
4. Are we going to continue to see similar cyclical fluctuations in the future, or are we in the midst of a structural break?
I will try to keep things as simple as possible. Finance doesn't have to be complex (people make it complex). A picture says more than a thousand words - I hope the following charts help.
Performance: How to Visualize It
How do we look at performance?
Above you see the S&P 500 Index since 18711. By looking at the black line (nominal, non-logarithmic scale) you would think there was no point in investing before 1981. That's why you should look at longterm data on a logarithmic scale. The green surface is the real (inflation-adjusted) S&P 500. Should we look at nominal or real returns? What good is a 10% rise in the stock market if inflation runs at 20%? Conventional wisdom has it that inflation is good for stocks. It that true? Compare the chart on the next page:
Performance: Nominal or Real
Look what the inflationary period of the 1980's did to stocks: not much in nominal terms (black line), but devastating in real terms (green surface). From 1973 to 1982, the nominal S&P remained stable (117 versus 118 points). However, in real terms, the index fell from 640 to 286 (-55%). Yes, you would have lost purchasing power, too, if you kept your money in cash. But that is a different question.
For performance measurement, real returns count.
Today, the S&P 500 is around 1,800 compared to 82 (real) in 1871, yielding a real return of 2.2%3. But the S&P 500 is a price index (as opposed to total return), so we must account for dividends (and reinvestment of those). Including dividends, the total real return is around 6.5%. Impressively, this shows how important dividends are (2/3 of total return) in the long run.
We don't live 142 years, so the average total real return from 1871 to 2013 is not so useful for the individual investor. But you can slice those 142 years into periods of 10, 20 and 30 years. Take the returns over those periods and plot their frequency (see above).
You will notice that among all 10-year periods (blue) you had a few with negative returns. When investing over 20-year periods, you would have suffered only one (ending in 1921) with close to zero return. The longer your investment horizon, the closer the returns are clustered around the average, or expected, value. You can see it visually as the distribution of returns gets "slimmer" (green surface) and contains less "outliers".
The more data points we add, the closer the annual returns lie around the same mean (average). This serves as indication that stock market returns are mean-reverting.
Conclusion: It makes little sense to invest in stocks with a time horizon of less than 10 years.
1. In 20 years, many different people will have been at the helm of the job as money manager
2. Career risk: most money managers get terminated after a few quarters of unsatisfactory
performance (hence nobody dares to stick his neck out)
3. End-user risk: very few investors would be willing to accept multiple years of disappointing performance (changing strategy mid-term and hence messing up performance)
And here lies the conundrum: almost nobody is investing according to what theory prescribes.
It doesn't help that you can check on the value of your investments every minute via your smart phone.
Do you check every day what your house is worth? No, because, luckily, that is impossible.
It would probably be beneficial for most investors if their investments traded only once a year. The constant availability of pricing information, coupled with swings from one minute to the next add to psychological pressure, leading to mistakes.
The previous chapter assumes you don't try to time the market (you just invest whenever funds are available and lock them up for at least 20 years). But the stock market rarely trades at fair value. It is either over- or undervalued. What if you could actually determine those valuations? And what do you base valuation on?
In the long run, stocks are driven by earnings:
The problem: company profits are very cyclical. Meaning: in every recession they decline by large amounts, only to recover strongly afterwards.
From 2006 to 2008, for example, real earnings for the S&P 500 declined from $94.70 to $28.50 (-70%).
The price-earnings multiple, or P/E-ratio, rose from 15.7 to 52.7 despite a drop in share prices. Stocks seem expensive when they are not and vice versa.
So Professor Robert Shiller (Yale) came up with a simple solution to smooth out cyclicality: take the average earnings from the last 10 years. Boom and bust should even out.
The "cyclically-adjusted price-earnings"-ratio (CAPE, or Shiller-P/E) was born.
It actually does a much better job in pointing out when the stock market is "cheap" or "expensive". It also shows the extent of the stock market bubble in 1999/2000.
The average 10-year CAPE-ratio since 1871 is 17 (low: 7 in 1933, high: 42.5 in 2000). Today, we are at 24.6.
This puts us pretty far towards the expensive side.
What you do know is the starting CAPE-ratio, and assume a regression to mean (17). With today's CAPE (25), we are facing strong headwinds for returns over the next 10 years. Sliding down the above regression line, the expected annual real return for -8 CAPE points is only around 1%. This does definitely not compensate for the risk associated with stocks. As a result, you should lighten up on stocks.
Gloy goes on to discuss the link between GDP and Profits, War, Inflation, and its effect on all markets.