It’s a common misconception that markets and market trends cannot be predicted. The usual response is “markets are efficient, random, and everything is discounted” and that there therefore the past is of no use for predicting the future.
But as I discuss in my rationalism and predicting series, this is not always true:
Markets exhibit a high degree of randomness and efficiency in the short-term, but in the long-term obvious trends emerge, and these trends are based on underling systems that can be understood and used to predict with a high degree of accuracy.
The reasons are: prices and markets obey underlying systems, and capital tends to flow to the most optimal allocations, with some diversification due to liquidity constraints and to minimize risk, of all possible choices. If given a choice between 10 investments, capital will flow to the best three or so, not the worse three. If one understands the underlying economic system or fundamentals, then knowing what will happen becomes as easy as solving a simple physics problem, such as modeling the trajectory of a ball throw strait in the air. Eventually is becomes so obvious and easy that one cannot help but to be right most if not all of the time.
One such strategy based on a system is the HBD investing thesis, which posits that smart, competent counties outperform less intelligent ones. This can explain why the Nasdaq (if the US tech sector, which is mostly concentrated in the Silicon Valley can be isolated as single sovereign entity, much like a country) has outperformed all markets since 2009 on a real basis, but also why the S&P 500 has outperformed all foreign markets too.
To understand the predictive power of systems for an example unrelated to the stock market, within 20 minutes of the Titanic hitting the iceberg, the lead engineer determined that the Titanic was doomed because five of the bulkhead compartments had been breached, which was one too many to remain afloat. Furthermore, he ascertained that the Titanic only had 2 hours left. Many passengers however were oblivious for most of those 2 hours until near the end, when the tilting of the ship had become obvious, not that there was anything that could be done as there there were not enough lifeboats. Imagine if the engineer was a speculator and was taking bets on how long the Titanic would stay afloat. By being ‘short’, because he knew the ship was doomed, he would have made a big profit by taking the opposite bet of the prevailing narrative that it was unsinkable.
Likewise, if someone knows that a certain company will succeed hugely, substantial profits can be made without insider trading, but by understanding the underlying system.
A common objection is, the efficient market hypothesis means that investments with ‘good’ fundamentals will be overvalued (high PE ratios) due to the expectation of higher prices, reducing future returns. Conversely, markets and investments in which the fundamentals are poor, will be undervalued (in terms of low PE ratios) due to the expectation of lower prices. Future expectations are baked into the present price, making excess returns unlikely.
However, there are obvious counterexamples that fly in the face of supposed efficiency.
First off, historically, the S&P 500 has posted annual returns of around 11% (including dividends), which represents ‘real’ returns of around 6-8%. 43% of returns are from dividends, which is indicative economic activity and underlying fundamentals, than just random price movement. In a random system with 4% annual drift, the market would have only gained 4%/year; with no drift, it would be flat; yet it has gained 11%/year. This is also known as the equity risk premium, “Equity risk premium refers to the excess return that investing in the stock market provides over a risk-free rate. This excess return compensates investors for taking on the relatively higher risk of equity investing.”
But the point is, if markets were truly efficient and random, there would be no underlying drift in excess of interest rates, and yet the empirical evidence has shown that the returns of the S&P 500 exceed this by 6-8%/year.
But what about stocks with ‘good’ fundamentals being overvalued and subsequently lagging in performance, due to the market discounting future expectations? Although there is some evidence overvalued stocks under-perform due to this discounting mechanism, the market can only look so far ahead. This is a really important detail. Even in spite of the best attempts of the market to discount the future, things may be vastly better/worse than expected by even the most optimistic or pessimistic of forecasters, making stocks that seem overpriced even more expensive, and investments that are undervalued even more so.
Consider Google stock. In 2004, after its IPO, it had a PE ratio of over 100 due to the expectation of higher prices due to strong growth, yet in spite of being overvalued, over the past 15 years Google stock has posted excess real returns over over 1,500%, and presently has a PE ratio of only 22. Even the most optimistic forecasts in 2004 underestimated Google’s actual growth. The same is true for Facebook, which after its IPO in 2012 had a PE ratio of 100, but now the stock is at $170, a for a ‘real’ return of 300%, and the current PE ratio is only 22, versus 100 after going public in spite of the price being considerably higher. If the underlying system shows that an overvalued investment will exceed all expectations, because the system says so and it’s predestined to, then it’s a good investment even if overvalued now.
But what about the possibility that in spite of excess returns for some investors and investments, in the aggregate, abnormally high returns are not possible. For every investor who does really well, most will not, and it’s impossible to disentangle the role of luck versus skill.
Market efficiency does not rule out the possibility that some investors will earn above-normal returns. Over any period of time, some investors will beat the market, but the number of investors who do so, will be no greater than expected by chance and not without accepting increased risk.
This ultimately boils down to a philosophical problem of how much is knowable or not, and what is deterministic or not. My systems approach assumes that things that seem random upon cursory glance, such as certain types of markets, are knowable, but given that the efficient market hypothesis debate has been raging for over 50 years, this will not be resolved anytime soon.