VR Logo

Fund Managers Vs Algorithms

Index funds don't involve any research or active investment management what they invest in is automatically decided by the index. But the word automatically isn't really suitable. The index is chosen by people. So for quite some time to come, fund managers will easily beat the index.

One morning last week, as I sat in my car waiting for a green light at a crossing in an area full of government offices, I idly wondered whether any of the hundreds of babus crossing the road knew that they'd just become investors in a hot real estate stock. Or rather, they would, if the new pension system were to be implemented thoroughly. This would happen because equity investments under the new pension system would be done through a fund that would be an index fund and that morning I'd just read about the Unitech stock being added to the National Stock Exchange's (NSE) premier Nifty index. The news of Unitech being added to the Nifty caught the eye of many people because it wasn't too far back that the Unitech stock was considered a fairly adventurous one, so to speak. On the other hand the Nifty ('the stock of the nation', according to the NSE's advertising) is generally expected to be the abode of non-adventurous stocks, if I may put it that way.

Now I'm sure that adding Unitech to Nifty is just fine by the stated norms for the Nifty's composition. As the NSE website explains (in an FAQ that is shockingly well-written by the general standard of such webpages), the stocks in the Nifty are chosen algorithmically as the fifty most liquid stocks by a certain way of measuring liquidity. I guess if Unitech turns up in that list then that's that. The BSE Sensex, on the other hand, appears to be more of a committee-driven creature. The BSE does have some liquidity and other norms but it's clear that index's composition is essentially subjective. However, in my mind, this does bring up the question of index fund investing and treating indices like investment portfolios.

Index funds don't involve any research or active investment management-what they invest in is automatically decided by the index. Except that as we've seen above, the word automatically isn't really suitable. The index is chosen by people. Perhaps by a committee, or by a committee reading the results of some algorithm designed and chosen by people. Perhaps that algorithm produces results that are correct in a statistically satisfactory proportion of cases. Perhaps any process, whether committee-driven or algorithm-driven, produces a large number of correct cases and a small number of mistakes. And perhaps, as in physics or cricket's Duckworth-Lewis method, we should be aware that using the results of a system can itself modifies that system. After all, Unitech's liquidity seems to have taken a major jump ever since its liquidity-induced inclusion in the Nifty was announced.

All of which leads me to conclude that the frequently-predicted death of active investment management is a long way off. The NSE's indices FAQ says that “Movements of the index should represent the returns obtained by 'typical' portfolios in the country”. I guess the quotes around 'typical' creates some wriggle-room but if one considers mutual funds as typical portfolios then across all periods, a good majority of funds beat the indices handily. Given what I've observed above, the reason is not difficult to figure out. The competition is not between active fund management and passive fund management. Instead, the competition is between active management by a fund manager and active management by an index committee or an algorithm filtered through an index committee. I predict that for quite some time to come, most fund managers will win.