Some thoughts on the extent to which index providers’ decisions affect not only the flows of index funds but also the everyday investment dections of active investment managers, sometimes forcing them to buy stocks they are negative on. #Tesla’s recent rally is an example: 1/15

Active fund managers have official or unofficial benchmarks by which their bosses & investors evaluate their performance. These are typically stock indices but can also be peer groups of other funds (in turn likely benchmarked to an index). 2/15

Their personal targets will stipulate that they should aim to to beat their benchmark by say 2% or 3% p.a. To reach these targets they may take a given amount of risk. Each active fund manager will have a limit on the active risk she is able to take when pursuing her targets.3/15

Active risk results from diverging from the benchmark index, from being overweight or underweight a stock versus the weight it has in the benchmark index. These limits must not be explicit but can be institutional norms within the asset management firm. 4/15

At some firms these active limits can be as small as 0,4% or 0,6% per stock, at others it can be as big as 1% - 3% and very few firms will have none. The ultimate risk limit is of course implicit career risk. Tesla’s weight in the S&P 500 is ~1,9% & ~4,5% in the Nasdaq 100. 5/15

As a firm’s share price rises, so does its weight in the index (most are market cap weighted). Tesla has an index weight of 2% in an index. A fund manager bearish on Tesla may start out with only a 1% position in Tesla stock and thus be actively underweight Tesla by 1%. 6/15

Tesla joined the S&P500 with a weight of ca. 1.6% weight. A fund manager bearish of Tesla may decide to hold only a 0.6% absolute holding in Tesla, and thus be actively underweight Tesla by 1%. As a result of the recent share price rally Tesla’s weight has increased to 1.9%. 7/15

An increase of 0.3 percentage points (or 19%, 1.9/1.6). For the fund manager underweight Tesla, her position now has increased in value from 0.6% to 0.71% in absolute terms, meaning the active underweight has increased from 1% to 1.19% (1.9-0.71). 8/15

To comply with her risk appetite (or that of her firm), she will have to buy 0.19% additional Tesla stock into her fund in order to increase the position to an absolute weight of 1.3%, and to return to an active underweight of only 1%. 9/15

Tesla’s addition to the S&P thus forced a large group of _active_ managers benchmarked to the S&P who were _bearish_ on its stock to buy at least some of it, as being underweight 1,9% in such a volatile stock will be too large a single stock risk for many managers. 10/15

S&P estimates there is >$11.2trl indexed or benchmarked to the index, of which $4.6trl is in the form of index funds. So, while ETF inflows may contribute to Tesla’s rally, it may be active funds, and managers bearish of Tesla, that are causing its stock to go up further. 11/15

If $4.6trl is indexed, this leaves $6.6trl that is benchmarked. If each of these fund managers has had to buy 0.3% additional Tesla shares that is $19.8bn of buying equating to $20bn in quasi forced buying of Tesla from active funds in the past 3 weeks alone. 12/15

Add in funds benchmarked to the Nasdaq, the MSCI USA and the MSCI World and this figure is likely 2-3x that size. Tesla is a highly liquid stock, so $40bn-60bn equates to just 10-15% of the volume that traded over the past 3 weeks, but it is significant nevertheless. 13/15

Conclusion: (a)index provider decisions have consequences on the entire capital market, not just on index funds. (b)active managers’ benchmarks force them to buy stocks that go up in order to maintain their initial active weights and thus contribute to share price momentum. 14/15

For more on the topic see the following paper by @johannes_petry and @fichtner_jan https://www.tandfonline.com/doi/full/10.1080/09692290.2019.1699147

As well as this by @adri_robertson
https://digitalcommons.law.yale.edu/yjreg/vol36/iss2/6/

And this by @scotthirst
https://heinonline.org/HOL/LandingPage?handle=hein.journals/bulr99&div=36&id=&page=

15/15