Credit ratings help facilitate an efficient capital market. They provide transparent third party information that is not only forward-looking but standardized for consistency’ as per the S&P Global Ratings on their website. Ratings range from AAA (highest rating – obligor has very strong solvency) and D (obligor default) obligations).
The AAA rating is usually reserved for a few of the world’s most solvent governments and best-managed companies. Before the global financial crises began (circa 2007), thousands of mortgage-backed securities were given an AAA rating by Standard & Poor’s (they were financial instruments that allowed investors to bet that someone else would default on their debts). The AAA paper indicated that there is only a 0.12 percent chance (or 1 chance in 850) that it may fail to repay over the next five years. As the crises hit, internal S&P figures indicated that about 28 percent of CDOs (secured debt obligations) rated AAA had defaulted (some independent estimates are higher). This means that the actual default rates for the secured collateral have expired
two hundred times than expected by Standard & Poor’s.
Let’s take another example. In December 2007, economists at
The Wall Street Journal The forecasting committee projected only a 38 percent chance of a recession over the next year. We know what happened, but this was especially notable because the data would later reveal that the economy was already in a recession at the time of the forecast!
In 1971 it was claimed that we would be able to predict earthquakes within a decade, a problem we are not close to solving in fifty years. And the list goes on.
Then the “prediction problem” gets worse because we’re so convinced of our predictability that we’re raising our bets to the limit.
In 2007, the total volume of home sales in the United States was only about $1.7 trillion, but the total volume of transactions in mortgage-backed securities was about $80 trillion. Every time someone borrowed a $1 mortgage, Wall Street was making a $50 side bet.
Or take nuclear power plants. The World Nuclear Association has calculated the life-cycle cost of various energy sources and concluded that nuclear energy is cost-competitive with other forms of electricity generation. Then the seismologists looked at the previous data and built the Fukushima nuclear reactor to handle an 8.6-magnitude earthquake, because “anything bigger was impossible.” In March 2011, Japan experienced a terrible 9.1-magnitude earthquake and the resulting tsunami destroyed three active reactors at the plant. Loss of standby power led to overheating and breakdown. According to the International Nuclear and Radiological Events Scale (INES), it was a major scale 7 accident (the highest possible accident) and the only accident other than Chernobyl to date. After the accident, the authorities closed 54 nuclear power plants in the country. The Fukushima site is still radioactive, and about 30,000 people had to be evacuated. The cleanup job would take 40 years or more and cost tens of billions of dollars.
In the context of markets, this situation (the forecasting problem is exacerbated by leverage) is exacerbated by its unique features. Last week, I wrote that it is the marginal/most motivated buyers and sellers who decide prices in the markets; The opinion of the majority does not matter.
Add to this the fact that technological developments have exacerbated this problem. Now, fast algorithms and computers rush to execute the trade if the proposed rule is triggered.
Now let’s get it all together. We expect the stock to rise, talk to several people who agree with our expectations, and look at research by analysts who all think the stock is a buy. Then we “augment” ourselves to get the most out of it. Predictions are based on things that happened in the past, but things that never happened before happen all the time now. When this becomes clear, the majority of the contributors who were positive continue to be positive, but the person with the most influence (the marginal/most motivated seller) is forced to sell. Algos also chime in because they are required to act if certain rules are triggered. The stock falls below its fair value and then the cycle repeats in the opposite direction. This creates cycles.
How does one face this? First, being aware of the existence of cycles is critical. For a company that grows its core value (eg EPS or FCF growth) by 15 percent each year for example, the assumption that the stock price will rise steadily, at the same rate each year, is risky. As the two graphs below show, we can follow shorter cycles (as in the case of
), or much longer cycles (as in the case). Enter at the wrong time, and we’ve performed poorly for years, if not decades.
Second, applying the plausibility filter helps. I might think that iOS offers better privacy than Android, and therefore chose to use iPhone. But I’m a potential consumer only until my iPhone is offered at a marginal premium on Android. If an iPhone costs five times the price of a competitor, they lose me as a consumer. What we pay for what we buy matters. As consumers, we know this; As investors (when we buy a company), we sometimes tend to forget about it.
Finally, as humans, we are certified to recognize the patterns we see. We may not have a lot of natural defenses – we’re not very fast, or all that powerful. We don’t have claws, fangs, or bulletproof shields. We cannot spit out poison or camouflage ourselves. And we can’t fly. However, we are at the top of the natural food chain, and a baby can recognize the basic facial pattern within months of birth. It is not the learning of the individual, it is learned through evolution. Our investment experience will be much better if we allow evolution to take over and realize how these cycles have always existed and incorporate that into our journey rather than fight them.
(The writer is
Co-founder of Buoyant Capital)