Hacker News Reader

Comment by notadoc | Hacker News Reader

Well you could always avoid the tech heavy indices, but then you're sort of actively managing your positions rather than letting it be purely passive.

Theoretically, is not another growing peril of indexing that it removes incentives for companies to behave well or outperform, since if they're part of an index their shares will be bought automatically by retirement plans and investors anyway, irregardless of performance or competency?

And yet another theoretical peril question would be, if everyone is indexing, it surely must lose it's efficacy because it is no longer efficient, will indexing then not underperform? I suspect if or when that happens, active management will regain interest.

Indexing by the masses is a fairly new trend, it will be interesting to see how the markets handle the behavior long term.

View on HN

> it surely must lose it's efficacy because it is no longer efficient, will indexing then not underperform

Well, it's just a question of competitive advantages. In a world where so many people are trying to outperform the market, it's really hard to outperform the market and index funds essentially piggy-back to the aggregated wisdom of the best investors. In a world when the majority of investment and trading (the price is set by the marginal investor, not by the average investor) is index funds, outperforming the market will be easier. In reality, you get a sensible balance.

Index funds don't have to do the majority of trading as long as they cannibalize the trading that would have occurred. Say that the majority of investable dollars is held in index funds that rebalance once/year, only trading then. The effect on liquidity & prices is that fewer trades are made and relevant information on the stock's future prospects takes longer to get reflected in the price. This opens up profitable investment opportunities for active fund managers to move in and buy up underpriced assets before the passive investments get in there, which then corrects the pricing imbalance (but not before the active fund managers have made a good profit).

I suspect we will reach an equilibrium where most people are indexing and a few very good active managers are in play, while most of the less competent managers have been booted out. There might be a few swings towards either side on the way to that steady state.

The problem is that the pool of "few very good active managers" is changing all the time.

And if you happen to find an outperforming active manager, it's like buying high -- managers, like the market, always revert to the mean.

I totally get what you are saying, I am just saying that people will still fall for active managers like they do today, although in far fewer numbers and active managers will not completely go away. Also, when almost all people are indexing there will be a few arbitrage opportunities for active managers.

This is not quite correct. There are other weighting methods (specifically, equal weighting) which would prevent this situation, whilst still remaining passive.

You would have to actively rebalance to keep equal weighting, so it is not as passive as typical market cap weighting.

You still have to rebalance market cap weighted funds, just much less so. Market cap weighted is indeed more efficient, but that's only assuming most stocks don't do things like pay dividends.

I'd argue an equal weighted index is no longer passive. You're actively making a bet that smaller companies will outperform bigger companies. Someone else will be taking the other side of that bet. Me with my cap weighted index will get the average of the two and be actually passive.

But many investors are going to take the side of big companies, not because they are choosing the optimal approach, but because they have to, because they have too much capital to invest in small companies. So it's not logical to assume their returns must be equal on average.

I didn't say their returns would be equal, I said my returns will always be equal to the average of the market exactly because I'm a passive investor. The sum of all bets is what I'll be making. The standard text on that:

https://web.stanford.edu/~wfsharpe/art/active/active.htm

As for there not being enough of the small companies available to do the equal weighted S&P500, the smallest companies in the index have 3.5B$ in market cap. So there's almost 2T$ of that mix available and there would be much more if that amount of capital suddenly decided to implement that strategy as the market caps of the bottom of the S&P500 index would certainly rise significantly.

"Theoretically, is not another growing peril of indexing that it removes incentives for companies to behave well or outperform, since if they're part of an index their shares will be bought automatically by retirement plans and investors anyway, irregardless of performance or competency?"

You don't have to look far to find this ... a great example is Restoration Hardware, which very recently borrowed money to buy back roughly half of the outstanding shares:

"The move has caused the company to consume basically all of its available cash balances and debt to increase by $500 million. In consequence, stockholder's equity has swung from $920 million at the end of the company's last fiscal year into slightly negative territory."[1]

This squeezed the shorts as well as (to the uninformed) bolstered their EPS dramatically as there are half as many shares now outstanding.

Nobody but an industry insider or other market-specific trading desk would have any interest in a disaster of a stock like this, and yet RH is in the Russell 3000 and is therefore auto-purchased by savers and retirees everywhere.

"if everyone is indexing, it surely must lose it's efficacy because it is no longer efficient, will indexing then not underperform?"

This was explained, I think, very succinctly in the Dave Collum "year in review"[2] wherein he explains:

"In his must-read book The Wisdom of Crowds, James Surowiecki posits that a large sample size of non-experts, when asked to wager a guess about something—the number of jelly beans in a jar, for example—will generate a distribution centered on the correct answer. Compared with experts, a crowd of clueless people offers more wisdom. I submit that this collective wisdom extends to democracies and markets alike. A critical requirement, however, is that the voting must be uncorrelated. Each player must vote or guess independently."

And that is precisely the issue we are worried about facing with index funds - we are buying indexes so as to access the (real, efficacious) wisdom of the crowd. However, if nobody actually makes a best guess - if there are no participants in the "guess voting" the wisdom of the crowd is lost.

[1] https://seekingalpha.com/article/4104757-shorts-restoration-...

[2] http://www.zerohedge.com/news/2017-12-23/dave-collums-2017-y...

The crowd's wisdom extends to the indexer. Another issue with these index-linked funds is that they create a single point of failure, i.e. the few people at Russell that curate the index. It would be very very easy for someone to, um, externally compensate an individual or two at Russell to include a dog in the index for the purpose of a massive stock pop. Similar to the bond-rating agency fiasco(s) leading up to the 2008 crisis.

However, once an index is compromised like this, its overall returns will begin to suffer and people will move out of, say, Russell-linked funds. If it gets bad enough across the board, active-managed funds will begin to consistently outperform index funds and people will move their money there ... which presumably will rebalance the indexes. As far as I can see it, over the long term the system should self-correct.

It would also be pretty easy to spot something like that happening. If you see a stock in the Russell 3000 that is substantially less valuable than the distribution's tail would lead you to expect, then somebody is probably fucking with things. I would be astonished if Vanguard and other fund managers do not have automated monitoring to detect cases like this.

I think you hit something right on the head. The index itself is defined by a formula that we generally agreed upon. We only need to have more choices of formulas...Russell is capitalization based - maybe that is too simple? Can we make a Hacker News 5000?

Reply on news.ycombinator.com