Tuesday, July 12, 2011

Book summary - 'A Random Walk down the Wall Street' by Burton Malkiel

My first proper reading on investments (after living for 3 years without money :-)). Main points are described below, which can serve as a summary of the book too:

Why do we need to put effort and mind in investing?
Beating inflation is the most common explanation given to investing. In recent past, real returns on Bank FDs have been negative. Although the theoretical difference between inflation and FD return is 3-4% (it goes up to 6-7% after taxes), I have a feeling that the actual difference is a lot higher for common household (a separate detailed analysis is required to prove it). So we are losing purchasing power if we invest in FDs. Question is: Is that a good enough reason to lose your sleep by investing? You can react to this question in two ways: 1) Yes it is, and I will go aggressively to beat inflation by investing in high risk-high return assets; 2) More risk-averse reaction could be: Not really, but can I come up with safe and intelligent (or rather unknown as of now) ways to reduce the difference.
Another less talked about reason to invest (due to our age group) is to save for old age. Other reasons could be investing for a particular purpose like kids' education, buying house, etc.
I don't think fun and adventure should be reasons to invest. We have places called Casinos for that.

Investment theory #1: The Firm foundation theory
Each stock/asset has an intrinsic value which can be computed based on present conditions and future prospects. For a stock, it can be computed based on future dividend payouts. There are four determinants of value of a stock:
  1. The expected growth rate of dividends: Apparently Einstein described compounding as the greatest mathematical invention of all time. A rational investor should be willing to pay a higher price for a share the larger the growth rate of dividends and earnings and the longer the growth is expected to last.
  2. Expected dividend payout: Other things being equal, a company paying higher proportion of its earnings in cash dividends should have a higher price. Stock dividends or splits are not beneficial for stock owners.
  3. Degree of risk: Lower the risk, higher the price of a share (and hence, perhaps lower the returns). Blue chip stocks sell at a premium due to this.
  4. Level of interest rates: Lower interest rates -> poor returns on debt/FDs -> more money for equity -> higher the price.
There are two important caveats here: 1) Expectation about future cannot be proven in the present. 2) Precise figures cannot be calculated from undetermined data (Excel tricks played by analysts in DCF computation can move intrinsic value anywhere. Just change growth rate or terminal growth rate, or number of years, or discounting factor)

It is the P/E of a stock, not the absolute price of the stock, which determines its value for comparison. PEG multiple (Growth/ P/E ) is a more accurate multiple: For a stock with high growth and high P/E, market has already accounted its growth in its price. Higher the PEG ratio, more attractive is the stock.

It has been seen that markets do follow logic, and the 4 determinants are followed most of the times (thus a stock with higher expected growth rate will have higher P/E). But what is a good enough P/E? We should take long term historical averages to understand how much premium market has given in the past for a particular growth rate/dividend payout/risk/interest rates.

Investment theory #2: The Castle-in-the-air theory
Famous economist Keynes proposed this theory in 1936 - people invest based on crowd behavior and human psychology, rather than sound fundamental intrinsic value. No one can predict future earnings or dividend payouts. So look out for latest 'hot' thing in the market!

Stock prices are in a sense anchored to certain 'fundamentals', but the anchor is easily pulled up and then dropped in another place. Both the above theories work in parallel to result in 'controlled madness' or a 'balancing act'.

How Pros play with above two theories
Technical (Theory #2) and fundamental (Theory #1) analysis. Chartists believe that market is 10% logical, 90% psychological; fundamentalist believe the opposite.

Technical analysis
Author has severely criticized technical analysis. It is trading, not investing to begin with. Trends might be built up due to two reasons: crowd instinct/behavior and unequal access to information. There are logical shortcomings such as any information is already accounted for in price, by the time a trend is noticed.

Fundamental analysis
For its modeling, fundamental analysis can only provide a range or an anchor point. There are three potential flaws: 1) information and analysis may be incorrect, 2) analysts' estimate of 'value' may be faulty, 3) market may not correct its 'mistake' and stock price may not converge to its value estimate.

Using both analysis together
Rule1: Buy only companies that are expected to have above average earnings growth for five or more years. Both earnings and multiple might increase - resulting in double benefit.
Rule2: Never pay more for a stock than its firm foundation of value: Growth stocks selling at multiples in line with or not very much above the market multiple often represent good value. Keep away from growth stocks with high multiples.
Rule 3: Look for stocks whose stories of anticipated growth are of the kind on which investors can build castles in the air.

Does technical analysis work?
Academicians don't like chartists because: After paying transaction costs, the method does not do better than a buy-and-hold strategy. Results reveal that past movements in stock prices cannot be used reliably to foretell future movements. The stock market has little memory. Correlation between present and past price movements is slightly positive. So there exists some momentum in prices, but too small to be economically significant. Flip of a coin can produce the same chart as a stock market. Reading charting patters produces no significantly better results than placebo strategy of buy and hold. Frequent trading in technical analysis also leads to early realization of capital gains tax.
If past prices contain little or no information for the prediction of future prices, there is no point in following any technical trading rule for timing the sale and purchases of stocks.

Does fundamental analysis work?
Past earnings growth cannot predict future growth. Let us consider a naive forecasting model that every company would enjoy a growth in earnings equal to long-run rate of growth of the national income. A study showed that this model would make smaller errors than those used by professional analysts. Analysts did poorly in 5-year growth estimates, and worse in one-year estimates across industries. This difficulty in predicting future can be explained by 1) Influence of random events 2) creation of dubious reported earnings due to creative accounting, 3) incompetence of analysts, and 4) Loss of best analysts to sales desk or portfolio management

Mutual Funds also do not show any consistency in performance. Some funds might outperform the market during a certain period, but not consistently. Good past performance of a fund does not imply good performance in future (in fact, reverse may be true due to law of averages which seemingly works for funds).

Timing the market: Holding on your stocks as long term investments works better than market timing because your gains from being in stocks during bull markets far outweigh the losses in bear markets. A market timer would have to make correct decisions 70% of the time to outperform a buy-and-hold investor.

Thus, random walk theory suggests that Fundamental analysis cannot produce investment recommendations that will enable an investor consistently to outperform a buy-and-hold strategy. Both Peter Lynch and Warren Buffett admitted that most investors would be better off in an index fund rather than investing in an actively managed equity mutual fund.

Investment Theory #3: Modern portfolio theory
The only method of beating the market is to assume more risk. However a portfolio can be created to reduce risk and still generate higher returns (upto a limit) through diversification. Two asset classes with covariance <1 can be combined to reduce risk. If covariance = 1.0, no risk reduction, +0.5 - moderate reduction, -0.5 -> most risk can be eliminated, -1.0 -> all risk is eliminated. However risk can't be reduced indefinitely with increasing returns, and after a point, risk increases with returns.

There is no premium for bearing risks that can be diversified away. So to increase returns, you need to increase risks which cannot be diversified away. Risks can be divided into two main types: Systematic (market risk) and asystematic (diversifiable risk). Systematic risk cannot be eliminated by diversification. Unsystematic risk is the stock specific risk which can be diversified away. So market won't pay for this risk. It will only pay for systematic risk (Beta of a portfolio). Thus higher the beta of a portfolio, higher the returns. Note that portfolio might still have some non-diversified unsystematic risk, but there will be no additional premium for it.

In practice, CAPM has failed as it was shown that there is no relationship between beta and returns!

Thus it seems that both technical and fundamental analysis do not outperform markets. The only way to get higher returns is to assume higher risks. But there is no measure of higher risk, i.e. I do not know if I have taken greater risk.

Predictable patterns in the behavior of Stock Prices
  1. Stocks do follow momentum
  2. Eventually stock prices do change direction and hence stockholder returns tend to reverse themselves: Poorly performing stocks over the last three years may give higher returns. Reversals may be due to economic factors.
  3. Stocks are subject to seasonal moodiness such as diwali, end of the week, beginning of year: Buy your stocks on monday afternoon at the close, not on Friday afternoon, or monday morning, when they tend to be selling at slightly higher prices.
Patterns with Fundamental analysis
  1. Smaller is often better: Smaller companies give higher returns, but it may be due to the fact that only those giving higher returns survive over a longer period.
  2. Stocks with lower P/E outperform those with higher P/E: But beware of accounting and real laggards. Results are not consistent over time.
  3. Stocks having low multiples of their book values tend to give higher returns: Results are not consistent over time. Survivorship bias, not easy to determine book value especially due to real estate,
  4. Higher initial dividends (D/P ratio) and lower P/E multiples have meant higher returns:
Investment guide
  1. Cover yourself with protection: Invest over long term. So you should have noninvestment resources such as medical and life insurance to draw on any emergency. Go for term insurance. In addition, you should keep at least one year's expenses in safe and liquid investments.
  2. Know your investment objectives: Determine your risk appetite. In increasing risk profile, bank accounts, money market deposit accounts, money market funds, high quality corporate bonds, diversified portfolio of blue chips, real estate, gold, common stocks of smaller companies.
  3. Tax should be considered while computing returns and avoid taxes wherever possible
Guide to investing
A person should look at his attitude towards risk and capacity to take risk. The most important investment decision is balancing of various asset categories at different stages of risk. More than 90% of an investor's total return is determined by asset categories that are selected and their proportional representation. Four Asset allocation principles are:
  1. History shows that risk and return are related.
  2. The risk of investing in common stocks and bonds depends on the length of time the investments are held. The longer an investor's holding period, the lower the risk: Follow buy-and-hold strategy over long periods of time.
  3. Dollar cost averaging can be useful: This is similar to SIP. You need to buy more stocks at lower prices to gain.
  4. Risks you can afford to take depend on your total financial situation
Typical portfolio for mid-twenties
5% cash with maturity 1 to 1.5 years, 20% bonds, 65% stocks, 10% real estate

Three strategies
  1. Invest in index funds: no capital gains, low transaction costs: choose the right index to invest in - for popular indices, stock prices increase just on listing.
  2. Hiring a professional mutual fund: Choosing funds is tricky. Choosing 'best' funds with high performance has shown to under-perform market, even after ignoring load fees and expenses. So recent performance is not a good indicator of future performance. There are three factors in selecting mutual funds: 1) Risk level - understand beta of fund's portfolio 2) amount of fund's unrealized gains - do not buy funds with high unrealized capital gains due to capital gains tax and 3) fund's expense ratio - should be low
  3. Pick up stocks as per the following rules:
  • Confine to companies which can sustain above-average earnings for at least five years
  • Don't pay more than intrinsic value of stock (you can estimate a range of the value)
  • Buy stocks with the kind of stories of anticipated growth on which investors can build castles in the air.
  • Trade as little as possible to avoid brokerage and capital gains tax

No comments: