Risk-On/Risk-Off And Blackjack: Why RORO Is Always Doomed To Fail

Includes: DIA, SPY
by: Chris Richey, CFA

The second quarter of 2012 was beset by Greek, Spanish, and Chinese worries: the Greeks ended up having to hold two elections in order to pick a government which has no hope of solving their debt crisis, the Spanish had to go hat in hand to the EU, ECU, and ECB to gather $125 billion to shore up Spanish banks, and China continued to slow economically, thus affecting commodity prices and general investor sentiment.

Meanwhile, in the U.S., Ben Bernanke and the Federal Reserve decided to kick the can a little further down the road on "Operation Twist", the stimulus scheme they started last summer wherein the Fed sells their short-term notes and buys long-term bonds to drive down long-term interest rates. The program that was supposed to end in June has now been extended to the end of 2012 with another $267 billion of "twisting" to be undertaken in the interim. Lower long-term interest rates are good for borrowers (such as would-be mortgage refinancers) and bad for savers (such as anyone with positive net worth or a pension).

Based on the charts below, it seems apparent that all the "easing" and "twisting" are producing diminishing effects as we roll into each subsequent proposed solution. Once the cheap money, in whatever form, stops flowing, markets slump. While we are sure Bernanke has the best intentions, we worry that the patient is developing a resistance to the medicine while remaining on life support. We have to wonder how long can Bernanke keep "goosing" markets to get them to leap out of their hospital bed.

(Click to enlarge)

The other issue we take with cheap money is that it punishes savers with low interest rates. We understand that the theory is that low rates help those that need to refinance their homes or service their floating rate loans, and that it forces many whom would otherwise hold their money in cash and near cash form to push that value into riskier assets that presumably promote economic growth. Nevertheless, the fact is that rates held this low for this long has a real, negative impact on wealth generation given the heavy weighting in bonds carried by most individuals and institutions.

That is just using today's officially reported Consumer Price Index (CPI) of 2%: your CD earns 1.5% and so you lose a real .5% on the deal. Imagine that we were still using the definitions 1980 and 1990 inflation definitions, which were much stricter and more inclusive of real purchases made by consumers. Per the charts below from ShadowStats.com, we can see that under the 1980 definition (modified in 1984 under President Reagan), we'd be reporting an inflation rate of 9% and under the 1990 definition of CPI, current inflation would be 5%. The primary differences between the 1980, 1990, and 2012 definitions of inflation are mainly the exclusion of food and energy costs from today's calculation, while 1980's definition bore the full load of those factors, and 1990 had a lesser weighting in food and energy.

(Click to enlarge)

(Click to enlarge)

What Can Intelligent Investors Learn From Business-like Gamblers?

Benjamin Graham said that "Investment is most intelligent when it is most business-like." We take "business-like" to mean that the investment practitioner operates in a consistent, systematic, and patient manner, basing their decisions on data and analysis. Underlying this approach is the belief that these efforts will increase the likelihood that the investments chosen will produce positive returns that are superior to those based on whims, hunches, and occasional blind luck. Note that Graham does not assert that a business-like approach is a guarantee of success, just that when we invest in a "business-like" fashion, we are using our intelligence and, thus, presumably increasing our odds of success. It would be fair to say, then, that un-businesslike investment decisions are more like gambling. Professional investors are not paid to gamble: the client can make arbitrary and whimsical investment decisions for themselves and avoid the management fee.

Conversely, we have observed some "business-like" forms of gambling that share many of the features of business-like investing in as much that a thoughtful system is employed in a disciplined and patient manner. The best examples of this gambling/investing are "Blackjack teams" such as the MIT Blackjack Team, a formal corporation actually known as Strategic Investments formed by MIT and Harvard students in the 1990's, as well as the "Holy Rollers", a more informal group of church-going Christians who formed a Blackjack team in the 2000's. Both groups netted millions in profits working as coordinated Blackjack card counting enterprises.

Of course, some of what the MIT Blackjack Team and the Holy Rollers did is, in fact, illegal, not the card counting part, but the teamwork, conspiracy, and signaling parts certainly were, as you cannot work in league with others to "beat the house." Likewise, it is illegal to use mechanical or electrical machines to follow the count of the cards. Card counting itself is legal in all 50 States. In fact, under New Jersey law, you cannot be barred from a casino simply because the casino suspects or even knows you are counting cards (Uston v. Resorts International Hotel Inc., 445 A.2d 370 (N.J. 1982)). Card counters are also known as "advantage players", i.e. they try to move that 51% or greater House advantage over to their side of the ledger, which sounds like the intelligent approach to take.

Most people gamble at casinos as a form of entertainment, realizing the odds of their leaving with more money than they walked in with are rather small. For the most part, they are there for the excitement of sometimes winning (and mostly losing, which is why casinos are so plentiful, nicely decorated, and happy to hand out free drinks). These "excitement" gamblers, if they have any approach at all, employ a couple strategies, half-heartedly or incorrectly, in a sporadic fashion. One of these approaches is the "Hot Table" approach: they look for happy Blackjack tables where the players appear to be winning more than losing and are thus said to be "on a hot streak".

Behavioral Finance aficionados, such as us, will immediately recognize this sort of faith in "hot streaks" as the phenomenon known as "Gambler's Fallacy" wherein the observer predicts wins or losses simply on the basis of the prior outcome. However, over the past few years, we have seen this type of speculation adopt the new moniker "Risk On/Risk Off" in the world of investments.

Since 2008, Risk On/Risk Off, or "RORO", has dominated global stocks markets known as investors of all sizes and nationalities have undertaken a manic cycle of buying and selling so-called "risky" assets in the course of hours or minutes. In simplest terms, RORO the practice of being all in or all out of those assets classes defined as "risky", primarily stocks, but also including commodities and certain currencies.

HSBC Bank has done an excellent study on RORO, creating the correlation "heatmap" graphs below comparing 2005 and 2012. Separating investment choices into four basic categories of stocks, commodities, bonds, and currencies, the HSBC team then color coded the graph to show correlations in red as they approached 1.0 (the highest positive correlation, i.e. the two assets rise and fall exactly in sync) and correlations in blue as they approached -1.0 (the lowest correlation, i.e. the two assets move exactly in opposite directions).

(Click to enlarge)

What we can tell from this graphic is that the largest western stock markets, western currencies, and western sovereign debt are each much more positively correlated with their brethren than they were six years ago: red dominates the northwest and southeast quadrants. From the deeps blues in the 2012 graphic, we can tell that bonds and stocks are now very negatively correlated: if one is being bought, the other is being sold, regardless of the quality of either.

There are two downsides to RORO, however. The first is that the RORO phenomenon requires it's practitioner to be superhuman in their ability to foresee upcoming, short-run swings in market sentiment. This is akin to being able to predict the next card turned in a game of Blackjack. The second downside to RORO is the transaction costs associated with jumping in and out of the market: the bid-ask spread, brokerage commissions, settlement charges, and taxes.

There is ample evidence that predicting market swings is and has been a complete failure as a strategy for almost all investors. Between 2000 and 2010, the average mutual fund investor earned about 1.7% annualized versus a 3.2% annualized return for the average fund ("Bad Timing Eats Away at Investor Returns", Russel Kinnel, Morningstar FundInvestor, February 15, 2010.). Further proof is found in the annual study by Dalbar, a market research firm, into investor behavior and its consequences: in 2011, the average equity investor lost 5.7% while the S&P 500 was up 2.1%, meaning 780 basis points of underperformance (2012 Quantitative Analysis of Investor Behavior, pp 9, 10, Advisor Addition, Dalbar, Inc., April 2012). This underperformance is prominent and consistent over longer periods: in the same Dalbar study, over the past 20 years, the average equity investor has underperformed the S&P 500 by an annual 432 basis points. The combination of being out of the market when they should have been in, as well as the aforementioned transaction costs, meant that investors are swimming against the tide in their efforts to beat the market by using the "hot table" strategy.

In contrast to the "hot table" RORO theory, the disciplined, patient approach to winning at Blackjack, card counting, is designed to help the player win by tracking the shifting of the odds between the House and the player as cards are dealt. The keys in card counting are two-fold: first, like the casino itself, play for the long run, which means staying at the table, through thick and thin, even if it means betting the minimum for long stretches, and counting the cards, and, two, betting significantly larger amounts when the count favors the players, not the casino.

The first key is crucial because over the short run, the odds can play out in seemingly impossible short terms streaks and runs. Card counting is not foolproof, nor does it guarantee winning, even if applied with maniacal accuracy.

The second key is the willingness to bet larger amounts when the count turns in the player's favor. We think of concentrated portfolios as the equivalent of "betting large" when the count clearly favors the player over the House. By concentrating our portfolios on what we consider the best hands, we believe we skew the odds back in the favor of our clients and ourselves.

However, unlike casino Blackjack tables, we do not have to make the minimum bet or play every hand in order to continue counting the cards. In our instance, the most important count we track are the multiples we are asked to pay for the lifeblood of an enterprise: it's free cash flow. The lower the multiple of sustainable free cash flow, the more we are interested because we believe those metrics best represent the odds that an enterprise will prosper in the coming years.

We believe that, ultimately, a business-like, "card counting" approach to investment will outperform a RORO, "hot table" approach because the laws of real world economics, which have been around since people began trading and sharing goods, are very hard to repeal: quality outlasts junk, discipline beats whim, and price matters. The last few years since the Lehman Brothers collapse with stock, bond, commodity, and currency pricing rising and falling in lockstep with often little or no substantial reason have certainly caused some to question the utility of intelligence in the world of investments, but not us. These times, too, shall pass. Seventeenth century jurist and politician Francis Bacon perhaps sums it up best, "Truth is the daughter of time, not of authority."

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.