Almost nothing written about investing is actionable. It’s why I’ve read less and less on investing and tried to instead do more and more work myself—to reach my own conclusions, conclusions which can be applied over long investing horizons. The action most appropriate is almost always: do nothing.
One thing that is written about a lot (way too often) is the level of buyback activity in the market (I’ll use the S&P 500 to represent the market). While certain aspects of buybacks are indeed very actionable at the stock level (i.e. using buyback yield as an investment “factor”, read here and here), the market’s overall buyback yield is not something on which you should base any investing decisions.
Quickly, here is the history of buybacks, first in raw dollars terms (a useless but frequently quoted statistic—imagine someone saying “beware: earnings in raw dollars at all time high”) and then in yield terms (more useful/interesting). I show both gross buybacks and net buybacks (subtracting share issuance) in both charts.
Let’s focus on net buyback yield to see if there is any relationship (since 1987 when the best data on buybacks and issuance becomes available on the statement of cash flows) between current buyback yield and forward returns.
You don’t have to get too fancy with the statistics here to see there is no real relationship over the past 30 years. With R-squared’s of 3% and 9% for forward 1- and 3-year returns, not much of the variance of stock returns are explained by starting buyback yield.
The periods on the lower right are all the periods surrounding the global financial crisis: when buyback yield was high, and subsequent returns were very low. Net buyback yield back then was 4.0%. Today we are at about 2.5%. Note: this 2.5% is still inflated, because my way of calculating net buybacks is simple cash used for buybacks minus cash raised through issuance. This does not take into account share issuance which results in no cash inflow (like in a cash + stock or all stock acquisition). Because large-cap companies tend to acquire smaller ones, this brings the S&P 500’s weighted average change in share count down close to zero (0.6% or so).
Other measures like “Shiller’s CAPE” of my friend Jesse Livermore’s “Percentage allocation to equities” have much stronger statistical relationships with forward returns—and even those may not be actionable (CAPE has been “elevated” for decades). For more on buybacks at the market level, this is very sensible.
I am not saying buybacks aren’t interesting and important. I am just saying that they aren’t a good tool for getting excited or worrying about the market and its near-term potential for returns. Stay tuned for a much longer story that focused on buybacks at the stock level—which is far more interesting!