In the last two posts we looked at an incredibly accurate set of models for predicting future 10 year annualized returns in the market. These models utilize the equity allocation of investors to predict future long-term returns. When investors allocate a high amount to equity, future long term returns will be diminished, while if they allocate a low amount to equity, future 10 year returns in the market are very likely to be high. Given this, I thought it would be instructive to reconstruct what the equity allocation of investors looked like in the past. We only have real data going back to 1945, however we have long term total return data for the market going back to about 1871 as provided on Robert Shiller’s website. Our models can calculate long term returns from equity allocation of investors, but they can also do the reverse and calculate equity allocation of investors from market returns. We can use this to estimate equity allocation of investors before 1945 to see what the trends in the data look like.
Figure 1 shows the equity allocation of investors over the entire period 1871 – 2024. This data was calculated using the more advanced and accurate model for equity allocation of investors that I developed in the last post. However, using the more simple and traditional model for equity allocation of investors gives qualitatively the same result. The blue line in Figure 1 is estimated data for equity allocation of investors from 1871 – 1945, while the red line is actual data from the Federal Reserve Bank of St. Louis found here.
We notice some quite interesting things from this data. The first is that the equity allocation of investors is usually between 0.4 and 0.48 the majority of the time. Investors seem to prefer to have this allocation to equity across all possible investments. This allocation is usually only briefly breached to the downside during significant events. When these events occur, the equity allocation of investors typically drops to 0.3 – 0.35 with occasional dips even lower (down to 0.25-0.3 or so). These events seem to shift investor mindset to a lower value whether due to panic, liquidity needs or some other event. They are usually relatively brief dips of at most a couple of years. The only times there were more sustained dips was when inflation was running high for a time (far above the historical average of 2.1%). However, after this inflation was brought under control the equity allocation of investors would, after a delay of a few years, return to its normal range of 0.4 – 0.48.
The long term average equity allocation of investors over the entire time period is around 0.37. The all time high in equity allocation of investors was at the 1929 market peak at an estimated 0.553. The dot-com bubble came in second at an actually measured value of 0.515. The all time low equity allocation of investors was around 0.219 in 1982 when inflation was running very hot and treasury yields were sky-high.
More recently, equity allocation of investors has been trending above the historical range for “normal times” of 0.4-0.48. It is currently just over 0.50 (last measured in Q1 of 2024), and likely has moved higher with the run up in the market during Q2 and Q3 of 2024 (equity allocation measurements for these quarters are not yet released as of this writing). Why might this be? Well, since the 2008 housing crash, interest rates have been held quite low. Treasury bonds have not had much yield to the point that I have read news stories asking if the old 80/20 or 60/40 allocations to stocks and bonds is dead and whether it makes sense to go 100% stocks or at least 90/10 stocks / bonds. Even though interest rates have started to return to normal, the previous lower rates plus this shift in attitude could have something to do with it.
Will the market return to historic norms for equity allocation of investors? If I had to guess, I would say most likely.There are of course conditions under which the equity allocation of investors can be permanently higher. If investors are willing to accept lower returns indefinitely then they may choose to allocate a greater amount to equities. This permanent attitude shift and/or long-term lower or higher interest rates mentioned previously would be the most likely causes of such a permanent change in condition.
Alternatively, a more optimistic explanation would be if the earning power of the stock market is permanently increased and remains so relative to other asset classes, that could also justify a permanently higher allocation to equity. To take an extreme example to illustrate the point, if we supposed that equities earned massive returns and all other asset classes earned almost nothing, then a near 100% allocation to equities would be perfectly justified.
However, the trick here is that these phenomena that increase productivity must differentially affect publicly traded stocks as an asset class relative to all other investment classes (bonds, real estate, private businesses, cash, commodities, collectors items, etc.) While possible, it is difficult to envision such a lopsided scenario. For example, one could imagine that machine learning or the rise of the internet could lead to increases in productivity for publicly traded companies. However, these same factors should improve the prospects of private businesses or the ability of companies to pay debts and expand their debt through bonds. Even real estate assets would be improved by the improved productivity of the businesses and people that use this real estate. So, unless a particular effect is substantially improving the publicly traded stock market relative to other asset classes, this wouldn’t necessarily move the needle to permanently higher equity allocations.
I find it much more likely that the classical allocation to equities will prevail in the future, and we will return to that eventually. The open question is whether that will be a slow process (like flat markets for a decade), a quick one (with a large crash), or in a more roundabout way (with ups and downs).
More recently, interest rates have started to climb up which changes the situation with bond yields. However, it remains to be seen if this is a more permanent situation with rates stabilizing in a historically normal range, or if we will return to near zero interest rates in the future with that policy being the new normal.
In any case, we are at historically one of the highest investor allocations to equity, which has only been observed a few times before (in 1929 pre-Great Depression, in 2000 during the dot-com bubble, and in 2021 during the post covid phase of which we may still be a part of now in 2024). Even though this data cannot be used to time the market next week or even next year, I think it will be useful to use for retirement planning. We live in interesting times, and it might be wise to increase your savings buffer a bit more if you are thinking about retirement within the next 10 years before you finalize that decision.
Leave a Reply