Deep Risk

3 Flares Twitter 1 Facebook 2 Filament.io 3 Flares ×

I’m reading a fascinating book lately that has me thinking a lot about risk. (There will be a more in-depth discussion of said book at a later date.)

Risk is such a fascinating and tricky concept.

One problem with risk, is that by its very definition it is unpredictable.

What we usually try to do is to look at past destructive events, in order to estimate future risks.  But such an approach ignores the fact that the destructive events in question were entirely unpredictable when they happened.

(If we could have predicted the damage that the surprising events would go on to cause, then the events wouldn’t have been nearly so destructive in the first place.)

Furthermore when most surprising and destructive events happen, the magnitude of their destruction could not have been foreseen beforehand by back testing (Since the largest such events had, by definition never before been recorded.)

Put another way even the biggest economic calamities like the Great Depression, are only impressive because they were not surpassed by prior or subsequent economic crises.

So if we look back over 100 years of the American stock market and determine that in the worst recorded crisis an investor fully invested in the American stock market lost 75% of his portfolio value, this only tells us that it is possible to lose at least 75% of your net worth by investing in the American stock market. (It remains perfectly possible that you could lose 100% of your worth (or worse if using leverage) in some future event that will make the Great Depression look like a minor correction.

So it seems safe to assume that the greatest risk out there is one that we can’t see coming.

Furthermore risk is quite situation dependent. One person’s worst fear is another person’s greatest hope.

Even talking about investing in general, there can be multiple different ways of defining risk.

Volatility risk.

This is the risk represents the uncomfortable up-and-down movements of stock prices or commodity prices or bond prices.

Each time a price starts tumbling downwards and we see thousands (or millions) of dollars of our net worth disappear in a matter of days or weeks, we can’t help but imagine the negative trend continuing on forever (and it could.)

This is an uncomfortable thing. It is quite a different experience from, say, having your money sitting in a safe FDIC insured bank account (or even of simply using your money to buy something that you like.)

And beyond the psychological toll of volatility, volatility is destructive in and of itself. Consider the “low volatility anomaly.” In an efficient market model where investors are paid simply for taking risks, one would expect the portfolios of highly volatile stocks to outperform those of low volatility stocks. And yet historically portfolios of low volatility stocks have outperformed their higher volatility brethren on a risk adjusted basis.

(So even though volatility feels bad, at least it also destroys wealth!)

Underperformance risk

This is the flipside of volatility risk. If you decide that you cannot sleep at night with your money invested in volatile assets, and you put all of your savings in FDIC insured bank accounts, then the chief risk is that your money will slowly lose its buying power year after year as it does not keep up with inflation.

Although you will sleep well at night, your money will almost always shrink instead of grow, and it will become almost impossible for you to reach financial independence (unless you make, and save, an awful lot of money.)

Behavioral risk

This is probably the biggest risk of all for you, and me, and everyone else who invests. Our genetic makeup makes it likely that we will react in exactly the wrong way at exactly the wrong time.

Getting aggressive in order to claw back losses, and being overly conservative when it comes to collecting gains are exactly the wrong strategies when it comes to acquiring wealth. Unfortunately these are also the approaches that will feel most comfortable to us in most instances.

Probability wise, behavioral risk is a risk worth keeping in our consciousness at all times.

blargh.0

Hunter Pence’s diving-catch-tongue-position is an example of a  risky behavior.

In any case, thinking about deep risks has made me try to imagine the worst risks for each investing approach that I have personally considered adopting.

Buy-and-hold indexing.

The biggest risk here is that the underlying assumptions are false.

For instance what if the global economy and it’s underlying companies stop growing?

What if US t-bills stop being a risk-free asset?

What if there is a long downward trajectory of the market such that periodic rebalancing weakens an already weak portfolio over the course of years? (Something like the last two decades in Japan.)

(These risks will damage almost any portfolio, unless there is a minimum of equity risk incorporated into it (ie a permanent portfolio or a risk parity portfolio?))

Value investing.

What if the value premium has been arbitraged away?

What if the value premium doesn’t actually exist for retail investors?

What if there will be a unique economic stressor which disproportionally effects vulnerable companies (value companies) for a long stretch of time in the future?

Dual momentum investing.

What if there is a flash crash that concentrates the loss of the momentum portfolio’s value within one month (because of an increased equity exposure?) And what if the momentum strategy forces the investor to switch out of equities prior to a rapid recovery?

What if the momentum anomaly is arbitraged away?

What if there is a long period of whipsawing such that asset classes ping-pong between over performance and underperformance on a monthly basis, forever leaving the momentum investor in the wrong asset class at the wrong time?

What if short-term treasury/T-bills stop being a risk-free asset class?…

… And these are just worst-case scenario’s that I came up with off the top of my head. By definition there are much worse scenarios and I can’t even conceive of because I am trapped by my own empirical experience.

(Please add to the post in the comments section by helping me to imagine some new worst case scenarios, as well as some different investment strategies’ Achilles’ heels which you have considered. )

But the point here is that the future is unknowable, and even the most probablistic and best laid plans are subject to the whims of randomness.

So what are we left with***?

Not to beat that old drum, but it probably makes sense to just focus on the basics that we as individuals can actually control.

  1. We should save more of our own money.
  2. We should minimize our own investment costs.
  3. We should choose an investment strategy that best matches our own personality and behavioral tendencies (it’s tough to go wrong with buy-and-hold indexing here). Such an approach will be the one that we will be more likely able to stick with through thick and thin.
  4. We should play the probabilities as we best understand them.  (Long periods of backtesting are preferable to short ones.)

*** (Subtle cue for me to dispense the my own particular (and repetitive) brand of wisdom.  You can try to change the topic, but my wisdom is very robust and unchanging.  Inputs can vary but the output seldom does.)

 

3 Flares Twitter 1 Facebook 2 Filament.io 3 Flares ×

4 Responses to “Deep Risk”

  1. Robert October 28, 2014 at 7:46 pm #

    Some good discussion starters there! Risk is indeed an important and complicated subject. To be perverse, I’ll challenge the statement, “the greatest risk out there is one that we can’t see coming”. I’d like to suggest that the greatest risk is the one you can’t avoid and can’t recover from, whether you saw it coming or not. In 1900, Galveston was mostly destroyed by a hurricane. Over 6000 lives were lost. http://en.wikipedia.org/wiki/1900_Galveston_hurricane

    There were some indications that the storm was coming, but most did not understand or heed the signs until too late to evacuate the island.

    Death is a risk one cannot avoid and cannot recover from. Some Galvestonians experienced it ahead of schedule, but all of us face this “risk” and cannot avoid it. Barring supernatural events, no recovery is expected either.

    Most stock market declines have been temporary. Provided one stayed invested, the effect was neither fatal nor final. Exceptions include undiversified holdings (i.e., 100% Enron) and investing in speculative bubbles where there wasn’t intrinsic value to support the valuation (after the crash recovery never occurred). Example of this would be the tulip mania.

    Another aspect of risk is timing. We can be pretty certain there will be another large market decline. We just don’t know when. Thus, we can see it coming, but we can’t see how far away the event is. Options are priced on that basis; valuation is proportional to the square root of time. If you are buying an option, the amount you risk for a bet lasting a given period is proportional to the square root of that time period. Thus, the per month cost of this “insurance policy” is lower if you buy it for a longer time period. By the same token, if you buy equities and hold them for a long time, your risk is lower than if you hold them for a short time, if you express risk as potential price change (decline) per period of time. Interestingly, even some bubbles may be recovered from if investments are sufficiently diversified. Thus, the .com bubble burst resulted in many stocks going to $0, yet if one had invested in a diversified portfolio of .com stocks, including eventual winners like Amazon and Apple, one would have eventually recovered even from this period of speculative excess.

    • Miles Dividend M.D. October 29, 2014 at 10:12 pm #

      Robert,

      Welcome back!

      Knowing little about the Galveston hurricane other than Wikipedia entry that you linked,I would wager that the Galveston hurricane was a black swan event. I am sure in retrospect that there were indications that it could’ve been predicted. But given our recent history of Katrina with satellite imaging and the rest, I suspect it was entirely unpredictable in 1900. As such I would file the Galveston hurricane into my file of events that were incredibly risky and unpredictable.

      It is always possible for a stock market to go to zero.it hasn’t happened yet in America, but it did happen in China and Russia. And the more that income equality continues to widen,The more possible government seizure of assets becomes. Right now it’s probably a 1% proposition.

      Eventual death is not a risk, it is a certainty.

      The book in question, which I will write more about later is “anti-fragility” by Nassim Taleb.

      I imagine you would enjoy it as most of the messages are quite consistent with a libertarian agenda. Taleb is a very original thinker, and his construct where risk is unpredictable but fragility (and robustness, and anti-fragility) are predictable is a revelation.

      AZ

      • Robert October 30, 2014 at 7:07 am #

        I’m not sure the distinction between risk and certainty is useful, at least if we understand risk to mean more than annual volatility (and that is what I thought one of your points was). Even black swans are certain to occur, even if infrequently. They are, after all, just the tails of the distribution. Mandelbrot showed the application of a power law distribution (vs. normal distribution) to events like this, and in that sense, black swans are just the tails of a power law distribution.

        The Galveston storm is deeply embedded in the culture around these parts. 114 years later, it still motivates (most) people to evacuate when a storm is coming. Forecasting storm tracks is still not precise, but forecasting ability is vastly better than in 1900. If someone stays now, it is generally because they want to take the risk. But whether Galveston will be hit with another major storm is not a question of if, but when. It is a certainty. Galveston built a seawall and raised the elevation of the island, so the storm intensity required to wipe out the city is higher than it was in 1900. Ike recently flooded parts of the island, but not with damaging surf.

        Another certainty is that the earth will get hit by a large meteorite. We just don’t know when. I don’t worry about it. But, if you do the math, the risk of dying from such an event is higher than many things we do worry about, such as dying from a venomous snake bite or a shark attack (actual risk is debated, but these numbers are from http://www.killerasteroids.org/interactives/risk/index.php). Rare (black swan) events with high impact (pun intended) can have higher average death rates (deaths/year) than events that happen with high frequency but low impact. When we estimate risk, we tend to worry more about high frequency events than low frequency events without adequately accounting for impact. We also tend to fear the more sensational events: an ISIS beheading generates a far greater emotional (and thus military) response than a street shooting in Chicago.

        If one defines market risk as volatility, then it is also a certainty. Even black swans are also a certainty, just with a low frequency. (We may not know which black swan will appear, but we can predict that “a” black swan will appear, on a certain frequency). So, I’m not so sure that a distinction can be drawn between certainty and risk. The key is how you deal with it. In the case of markets, one can greatly reduce impact by long hold times. That won’t protect against markets going to 0, but it will protect against volatility. The impact of markets going to 0 can be reduced by diversification across markets and asset classes.

        There is another point about risk, however. The risk of dying in a plane crash is about 1 in 4.7 million. http://www.planecrashinfo.com/cause.htm. However, these are averages over the population. if you never fly, the risk is 0! One can avoid market risk by avoiding the market. (There are other risks, however, like inflation or spending all your savings before you die). We take risks because we are compensated for doing so. Whether risking a plane crash to enjoy traveling around the world, or risking a market crash to enjoy a higher standard of living, we take risks for the return they offer. To the extent we can avoid a risk, though, why not? If you don’t need or enjoy living on the beach, don’t buy a house in Galveston. If you don’t need the money to live on, you can just buy a diversified bond portfolio and sit tight.

        • Miles Dividend M.D. October 30, 2014 at 12:05 pm #

          Robert,

          Great comment.

          I may be missing something, but my impression is that we are saying the exact same thing in different ways.

          “The risk that we can’t see coming” is simply another way of saying black swan events.

          We can say with certainty that highimpact, low frequency, Black swan events (Such as meteorite hits, Solar flares frying our infrastructure,or gaping holes in the space time continuum) will happen,but we can neither predict when they will happen, nor specifically what they will be.

          Your example of someone not flying in order to eliminate the small risk of plane crashes is an example of someone who is “robust” to the risk of plane crashes. A person who is anti-fragile to the risk of plane crashes is someone who actually gets stronger when his plane crashes!

          I find Nassim taleb’s observation that Blackswan events are impossible to predict (though they appear predictable in retrospect)to be entirely convincing. What he argues is that robustness, Fragility, and anti-fragility is possible to predict a priori.

          As an example no one knows when the stock market will crash next, (though we do know that at some point it will.) But someone invested entirely in physical gold and cash is robust to a simple stock market crash (even a 100% loss of value), and someone shorting the stock market is be anti-fragile to a stock market crash.

Leave a Reply


Visit Us On TwitterVisit Us On FacebookVisit Us On Google Plus