2024-03-01: Time
Time allows us to deal with price volatility and for compounding to work its magic. Time is what defines investing as such and what distinguishes it from other activities in the market. In fact, investing is not about making money (we don’t control returns) but about budgeting (we can, partially, control how we spend, and hence how much time we have to put capital to work). The more time we have to keep our money unspent the more price volatility we can afford to take. Long-term time series of returns indicate this means we should have more equity exposure the longer we can keep our money invested or the longer we don’t have to spend it. Stocks for the long run. Well known stuff.
Or is it? How long a run do we need to be certain that stocks are the thing to buy? Turns out that it depends. A recent academic paper puts this age-accepted posture in doubt by, well, lengthening the time period over which the analysis is carried out (the data set in the study starts from the end of the eighteenth century instead of the end of the nineteenth century; one hundred years of additional data). It would appear that if we do the analysis over this larger data set the conclusion is not ‘stable’: over time, asset prices have experienced different return regimes.
The fact that the time span over which financial analyses are carried out changes is well known, whether because of new discoveries (more data becomes available) or because of intentional manipulation (good or bad). It applies not only to long-term or structural asset studies like the one in the link but also to the calculation and use of performance statistics. Independently of how far back in time you go, the truth is we simply don’t have enough years of data to make accurate definitive statements. Charles Ellis, a well-respected American consultant and writer of many academic papers, implicitly arrived years ago at the same conclusion by observing that to be certain (at the statistical standard 95% confidence interval level) that a manager is skillful or simply lucky you need 83 years’ worth of quarterly returns. In other words, not even Charlie Munger or Warren Buffett realistically make the cut.
Read the article. It’s very thought provoking.
******************************
If you are like me, weary and suspicious of news concentration, then you must have noticed the predominance of artificial intelligence in this week’s releases: Musk (who else?) suing ChatGPT/Microsoft, Apple (without Musk’s intervention) pulling out of the self-driving car project, Microsoft (again) buying Mistral, Alphabet botching diversity with Google Gemini, the US Justice Department hiring a chief AI officer for the first time and Nvidia simply carrying along after its great earnings the previous week.
Aside form this, markets continued to behave well in a declining volatility environment.
[Cover Source: Edward F. McQuarrie, ‘Stocks for the Long Run? Sometimes Yes,
Sometimes No,’ Financial Analyst Journal Vol 80, No. 1, 2023]
