Risk Defined, Generally

The Merriam-Webster dictionary defines risk as "the possibility of loss or injury," an intuitive definition to most people.  The average individual typically associates risk with some manner of negative outcome.  But the negative association is in many ways counterproductive because every activity, every decision, carries with it some degree of risk.  As human beings, our success as a species is due naturally in part to our aversion to risk.  But our instincts for survival have also resulted in us being quite poor at objectively evaluating risk and assessing the probabilities of events.  Our emotions often work against us, and many of the mathematical and statistical tools we have developed for appraising probabilities and risk have made us worse at the task, not better - at least where investing is often concerned.

Risk vs. Uncertainty

However subtle the distinction, it's impossible to comprehend risk without understanding how it differs from uncertainty.  The difference might best be characterized as follows:

Uncertainty is what meteorologists grapple with in trying to forecast the weather. 

Risk is what a gambler confronts in placing bets on a roulette wheel. 

Meteorologists have yet to construct a model that consistently and accurately forecasts the weather because there are simply so many possible outcomes.  In addition to the almost limitless list of outcomes, the weather has so many inputs and influences that scientists have yet to produce a model capable of translating the deluge of data into reliable forecasts.

By comparison, roulette has a defined set of possible outcomes - an American roulette wheel contains 38 numbers, and the ball ultimately has to land on one of them.  

Predicting the weather is a matter of uncertainty because of the superabundance of inputs and outcomes.  Were there not nearly so many factors influencing the weather, forecasting might be more a function of risk than uncertainty.  In that sense, one can see how risk takes on the unpredictability of uncertainty when the totality of inputs and outcomes becomes so numerous and complex that predicting a single outcome with any reliability becomes functionally impossible.  Predictive reliability as to a specific outcome means that the subject outcome has a high probability of occurring.  This is the interactive relationship between certainty, risk and probability. 

Investment Risk Defined, Incorrectly and Correctly

Most professional investment managers define risk as volatility in price, and they measure that volatility utilizing standard deviation (if you are unfamiliar with standard deviation, variance, and other statistical measures, please see Primer on the Misapplication of Statistical Analysis to Investing, which discusses these issues in much more detailed and technical terms). 

As Harry Markowitz, one of the intellectual founders of Modern Portfolio Theory ("MPT"), set out in his 1959 text Portfolio Selection: Efficient Diversification of Investment, "I use standard deviation as a measure of risk." 

Standard deviation, as well as variance (a related statistic), indicate risk by measuring the volatility in security prices, or how widely dispersed those prices are.  The more a price bounces away from its average, the more it's deviating or varying from the average, and hence the more volatile.  It's this volatility that's taken under MPT to be a reliable indicator of the amount of risk in an investment.

Thus, under Modern Portfolio Theory, risk is ultimately defined by volatility in security price:

Standard Deviation of Security Price  =  Volatility of Security Price

 

Volatility of Security Price  =  Risk of Security

This is both misleading and incorrect.

Some of the "smartest minds in investing" would of course vehemently disagree with the preceding statement.  And I quote:

"Risk is a function of volatility.  These things are quantifiable."

The above assertion was uttered by Peter Rosenthal, press spokesman for Long Term Capital Management.  If you're not familiar, LTCM was an investment partnership that was thought to comprise the keenest intellects in the investment world, including two Nobel Prize-winning economists.  It imploded in 1998, and were it not for the intervention of fourteen banks that committed a total of $3.65 bil. in capital, it might have taken a great part of the US financial markets with it.¹

By way to assuming another vantage point on the topic, consider the following example (it's an extremely simple one):

There's a company, "X," whose shares are trading at $50.  Early one morning, a rumor circulates that this company is going bankrupt.  There's no factual basis for this, but the shares drop to $25 on the "news."  Let's say the company was actually a fairly attractive buy at $50.  Because it is now trading at $25, a 50% discount to the prior day and with no actual change in the financial aspects or economics of its underlying business, according to MPT this company has actually become riskier because it has become cheaper. 

But if the same assets can be purchased for half of what they would have cost the day before, that investment isn't more risky - it's less risky because the lower price affords an even greater margin of safety.

Price volatility has nothing to do with risk.  There is a much more intelligent, more logical definition:

Investment Risk = The Possibility of Financial Harm

"The possibility of financial harm" doesn't mean that the price of the security drops, but rather that the prospective returns on the investment drop.  (For more information on calculating future returns, consult the Valuation section of this website; more broadly, value is addressed in the Philosophy section.)

Imagine now that the total market capitalization for company X is $100 mil. if the shares are trading at $50.  Assume that X generates $10 million in net income, which under the circumstances is attractive.  At $100 mil. market cap, that $10 million X generates translates into a 10% net margin. 

When the price of the entire company, which in this case is the market capitalization (X has no debt, so one can purchase all claims on the business by simply purchasing all of the outstanding shares), drops to $50 mil. because the price of the shares has dropped by 50% from $50 to $25, that means that X is now generating 20% net margins ($10 mil. divided by $50 mil.).

The question is:

If X is still generating that $10 mil. in net income year in and year out, how has the risk increased because the stock price has dropped?  And doesn't the level of risk decline if one purchases more shares of the same business at lower prices?

Risk is about the likelihood of the intrinsic value of the business declining,

not the price behavior of the stock.

If I owned 100% of Company X, would I care if one day someone walked in the door and offered me $50 mil. for a business that was worth $100 mil. to me?  I'd ignore him - and this should still be the case if one owns 1% of a company, or even less.

Forget about the price; focus on the business.

In 1963, American Express plummeted from $65 per share to $35 in a matter of days.  At that point running his first partnership, Warren Buffet proceeded to put 40% of the partnership's total assets into Amex.  The shares tripled in price in 2 years, and he made a fortune.  He didn't care about the volatility, which had gone through the roof, and he didn't care about his concentration, because he knew great investment opportunities don't come along that often.  The key is to bet big when they do, and when he looked closely and saw an attractive business at AmEx that was fundamentally intact, that's exactly what he did.

Portfolio Diversification and Concentration

Wide diversification is, generally speaking, little more than a crutch against ignorance.  Why in the world, after weeks of searching for a company that's worth owning for the long haul, would one want to diminish the impact of owning that truly great company by surrounding it with a bunch of mediocre ones?  Or by only owning a fractional position in it?

Owning a concentrated portfolio with fewer companies is defensive, not more risky.

Position size is indeed critical - the data confirm this.  In You Can Be a Stock Market Genius, Joel Greenblatt offers the following statistics:

• Owning two stocks eliminates 46% of non-market risk of just owning one stock

• Four stocks eliminates 72% of the risk

• Eight stocks eliminates 81% of the risk

• 16 stocks eliminates 93% of the risk

• 32 stocks eliminates 96% of the risk

• 500 stocks eliminates 99% of the risk

These statistics highlight that diversification beyond perhaps 10 stocks does little to reduce risk.  Beyond approximately a dozen stocks, diversification only serves to diminish returns. 

For the investor who limits herself to only the most outstanding opportunities, the above is ultimately irrelevant - there are only a handful of truly worthwhile investments that can one discover in the space of a year anyway.  Even the best investors in the world only come up with a handful of good ideas.  But they always bet big when they do.

Bet big when the odds are with you, and not at all when they're not.

The shrewdest investors have embraced the fact that they can only have in-depth, detailed knowledge of a very small number of situations.   

Knowledge is what insulates us from risk;

knowledge decreases with diversification, but risk decreases as knowledge increases.

The reality of the "business" of investing (vs. investing solely in order to produce the highest returns possible) is that very few money managers get fired by owning a stock that is well known and considered "solid" by the financial press.  If a mutual fund manager loses money because he bought IBM stock, well, IBM is a great company - everybody knows that, right?  IBM will come back.  This is what they said about the "nifty fifty" stocks like Polaroid.  Taken a look at Polaroid lately?

Conversely, if the same manager buys Widgets Inc. and the stock drops in price, even if he knows the company inside out and it's a great company, he's dismissed as foolish.  A year later, when Widgets Inc. fulfills its potential and trades up 140%, he won't be there because he'll have been fired for buying something that nobody had heard of that went down in price.  This is why 90% of mutual funds not only fail to match the performance of the S&P, but dramatically underperform the index.  Better to be a live lemming than a dead oracle.  This is the conclusion most fund managers come to, consciously or otherwise.  They're in business to be in business.  Call it The Institutional Imperative.  Or simply survival.  But regardless of what you call it, it's diametrically opposed to investors' interests.

Caveats and Criticisms

The chief criticism leveled at the above approach is that it can only work if the investor not only has superb judgment, but also an extremely detailed knowledge of their investment, neither of which is apparently achievable.  Thus, the investor is ultimately advised to give up before they start; you can't know, so don't try. 

Given how little research most analysts and portfolio managers actually do on their investments, the above viewpoint is ironic - in essence they're arguing that because they don't do the work, others are incapable of doing it.  But individuals should give their money to the former anyway because they're the professionals and they know what they're doing?  Huh?

Defining when one has an authentically thorough grasp of a company's fundamentals is an admittedly slippery practice.  Because every company or asset is different, what constitutes a complete or adequate analysis can be a moving target.  As for what it means to achieve that level of knowledge, in addition to my general comments in the Philosophy page, in the Investment Analysis page I offer some specific criteria that I personally find very practical in maintaining an objective viewpoint on the rigour of my own process.

¹ When Genius Failed, Roger Lowenstein, p.64.