Two weeks ago, I wrote a post on research that showed that cash flows into index funds distort the relative momentum between stocks that are included in the index and those that are not as well as between larger and smaller stocks in that index.
In reaction to that post, I received two kinds of responses. First, a large number of readers wrote me that my claim that nobody had ever heard of that argument before was wrong and that there are plenty of people who had been making this argument before. So, I stand corrected in my ignorance. I just hadn’t caught up on that discussion or that argument.
The second type of reaction was by my friend Larry Siegel, who sent me an article he wrote in 2017 for AJO Partners that elegantly debunked much of that argument. Here is the link to his article which you absolutely need to read. I recommend you subscribe to his E-mail list which presents his new articles as he writes them. They are roughly once a month, not as frequent as some newsletter writers but very intelligent and highly recommendable.
Larry shows nicely that as long as there are no primary market transactions (i.e. no IPOs or capital raises), and as long as the two classes of funds use the same benchmark, index funds, and active fund managers fish in the same pond and thus no matter how much index funds grow relative to actively managed funds, there will be no distortion between large and small stocks in the same index. (There will be distortions right at the boundary between stocks in and not in the index, such as the S&P inclusion effect wherein stocks added to the S&P rise in price because there are so many more dollars indexed to that benchmark than to any other.) He also makes the argument that if there were index trackers that tracked all stocks from large to small with the same amount of money relative to the stocks’ market caps, there would be little distortion between stocks that are included in an index and those that just miss out. Under current conditions, where the S&P 500 is the dominant benchmark for index funds, there is distortion but not so much to be meaningful in a macro sense.
I take pride in the fact that I am willing to change my mind if the facts change or the logic of the argument is overwhelming. I even have an entire chapter in my book on Seven Mistakes Every Investor Makes dedicated to stubbornness. So, I have no problem admitting that Larry’s argument convinced me to change my mind about the possible distortions of index fund flows between large and small stocks in the same index.
But I still think there are some micro-effects of index fund flows that matter on the margins. But just like Larry, I think they mater probably only temporarily due to another piece of research that I didn’t remember when I was walking my dog today (An old post of mine for CFA Institute on how I generate ideas has recently gained traction again, so here is another example of how I generate ideas: walking the dog).
First, the original research I quoted last week makes the important distinction to say that they measure an impact on the relative valuation between stocks included in an index and those that miss out due to new money flowing into the investment market. They are not arguing about relative flows from active to passive funds, but the money that wasn’t invested in the stock market at all before. This is in a sense similar to raising money in the primary market through an IPO or an equity raise. It does distort relative valuations because the recipients of these new funds see their market cap increase while other members of the index do not. The difference to an IPO or an equity raise is that in an IPO it is one company that sees its market value increase relative to all other companies while fresh money that is invested in stock markets for the first time amplifies the relative valuation differences at a specific moment in time. So, there probably is a magnifying glass effect, but it likely is too small to matter in practice when it comes to the large and small stocks within an index.
However, where I think this magnifying glass effect is somewhat more important is in the inclusion of stocks in an index. Large-cap stocks are tracked by far more index funds and active funds than small and mid-cap stocks. In fact, for clients of Liberum, I have shown last year that small and mid-cap stocks that have a higher share of ownership by index trackers tend to perform better because there is more money chasing stocks that have reduced liquidity. But while walking the dog, I remembered another study published in the Financial Analysts Journal that looked at the impact of country reclassifications in MSCI and FTSE Country indices between developed, emerging, and frontier market indices. It showed that countries that are being reclassified from an index that is used by less money as benchmark (again, it doesn’t matter if it is a passive fund or an active fund using the index as a benchmark) to an index that is used by a larger pool of money experience a short-term boost to their market capitalisation in anticipation of the reclassification. However, the study also showed that prices revert within a year.
In sum then, I stand corrected insofar as there probably is no relevant impact of index fund flows to the relative valuation between larger and smaller stocks in an index as long as there are no massive new flows of money into the equity market that have never been invested there before (i.e. most of the time), and as long as we are not looking at the very smallest stocks in the index which are affected by the decision to include them. There is pretty clearly a small effect of including a stock into an index that is tracked by bigger pools of money (i.e. typically by being promoted to the dominant large-cap indices), but this effect probably is short-lived and disappears pretty quickly once a stock has been included in an index.
But if you disagree with me, please feel free to reach out and argue. I am happy to change my mind if presented with compelling empirical evidence or conclusive arguments. But note that I don’t waste my time on ideologically-driven arguments or arguments that are not supported by evidence. In fact, that may just put you into my cross hairs like these guys.
Really interesting article - as per usual Joachim, but it did lead me to think if spreads are different between index and non-index constituents. The round-trip cost might be cheaper to trade stocks that are in the index and volumes could have an impact from a liquidity and portfolio management perspective. This could then have knock-on impacts from the investment risk departments with regard to tracking error, tracking tolerances and the like. Also, I wonder if any work has been done relating to the derivatives side of things and the counterparty risks associated.