Hacker News Reader

Comment by Animats | Hacker News Reader

With index funds so big, who determines prices? An index fund tied to the S&P 500 just buys stocks in the proportion that they're in the S&P 500. The price of the stock plays no role in that decision. At some point, this has to create problems, but so far it hasn't. It does mean the active traders, who are basically moving the same money around all day, have an outsized influence on prices.

Index funds are so successful because managed stock funds, as a class, underperform the market indices. So do hedge funds, which are a net lose for their investors. People are finally aware that Wall Street's stock pickers mostly aren't very good.

View on HN

> With index funds so big, who determines prices? ... At some point, this has to create problems, but so far it hasn't.

Any active trader remaining in the market. Fortunately, active traders still make up a large part of the market. And the bigger indices grow, the larger the opportunities for active traders to profit. It's not a real problem — it's self-correcting.

> And the bigger indices grow, the larger the opportunities for active traders to profit.

I'm having trouble understanding why this is the case. Care to clarify?

Index funds are buyers that don't think. If you have information that a stock has the wrong valuation you can sell to an index fund.

Index investments are still investments, they help push up the prices of the stocks they buy too. That creates opportunities.

You also have the situation that index funds must buy or must sell a stock if it enters or leaves an index. This creates an opportunity for active traders.

Most indexes are fairly predictable, so the changes are usually frontrunnered by index funds and traders.

Index funds buy or sell blindly, at whatever the prevailing best price is. Naturally, if there were only index funds in the market, the price would vary randomly.

That being said, consider what would happen if stock prices did start to vary randomly - if you had actual research suggesting the price was too high or too low, you could trade accordingly. This would net you a profit, and also help push the price in the opposite direction, towards whatever a reasonable price is.

The larger the deviation from the "correct" price, the larger the potential profits are to be had. So if the problem ever starts to be significant (i.e. a few cents of deviation caused by index funds), this means a very large potential profit for any active funds or traders out there. And so we would expect the system to reach an equilibrium - where there are just enough active funds and traders to snatch up the profits that arise from tiny price errors and distortions caused by index funds. In effect, the index funds are paying those remaining active funds and traders a tiny "management fee" (in the form of exploitable trading behavior) to figure out the appropriate price of stocks for them!

stock prices Do vary randomly.

I'm assuming that is a reference to the random walk theory. That theory suggests that "the past movement or trend of a stock price or market cannot be used to predict its future movement". It's random in the sense of unpredictable, not in the sense of actauly being determined by chance.

The implication of the random walk theory is that current information is already uilt into the price of the stock. That is only true insofar as there are active traders acting on that information. The parent is merely explaining why those traders will exist, and why you only need a small part of the market to actively trade.

Random walk depends on all participants acting rationally based on the same information released at the same time. However, index funds are not rational, they always buy in the same ratios regardless of news.

I'd suspect all those HFT firms will have robot agents (trading based on some ratios/logic) as simulations for their testing environment.

Resuming, as I understand it:

Good stock pickers > Index Funds (blind money) > Bad stock pickers

By investing on index funds you are betting in the average of the average (a fund is already an average of undervaluated and overvaluated stocks).

However, by investing on managed funds your probability of having profits over index funds is low: even if stock pickers that beat the market (by definition) amount to 50%, extra fees make most of them still less profitable to investors than passive funds.

[edited]

> Why not invest on a good stock picker?

You can only tell them after the fact

Sure, but I still prefer to bet on Buffett than on a managed fund that is consistently beated by the market.

[edited]

Buffett would take the other side of that bet.

http://longbets.org/362/

Pretty close to finishing. Is Buffett still winning?

He made a recap last year, was winning by quite a bit: https://www.yahoo.com/news/buffett-most-mportant-investment-...

Thank you :) . Makes the people investing in Vanguard happier then.

But the "correct" price is the one the investors are willing to buy at, right?

Imagine that 99.98% of all investors are index funds, and there is a company BigCo that is at some point is at the top of the market. So pretty much every index fund invests in it.

Then suppose BigCo makes some move that would traditionally be a mistake. Like it has some scandal, the sales drop off, etc. Say it even has a bad quarter.

How do I, an active investor (among the remaining 0.02%), make profit from the arbitrage? For the stock price of the company to fall, there'd have to be no buyers at the given price. But the index funds will keep investing in it. Why would it drop off even a little?

Even if it doesn't drop off, you can make money by investing into businesses who give you better long-term growth and (related) dividends than the one that everyone else is buying through index funds which you know isn't doing well.

This would require a long-term view similar to how Berkshire Hathaway is acquiring businesses. If active investors' short-term price speculation were reduced in favor of long-term bets, that might be a good thing for businesses and the economy overall.

And of course, if this situation actually becomes commonplace, chances are the market for index funds is going to self-correct since a well-performing stock needs either solid dividends or above-average growth. A market with 100% indexing can not give you above-average growth, and dividends depend on actual business performance.

> And the bigger indices grow, the larger the opportunities for active traders to profit. It's not a real problem — it's self-correcting.

I appreciate that finally someone puts forward a rational argument as to why index fonds will keep working. Books and online resources tend to not take a critical look at the system at all or they offer an answer along the lines of "Trust me!"

Having said that, only hindsight is 20/20! It feels like we are rushing into the next great financial experiment. In a complex world there is simply no telling what's actually going to happen. I assume (any sources?) that currently more money than ever is flowing into index fonds. Furthermore, chances are this is just the beginning. For instance, the buy-and-hold hype is just arriving in Europe. Blogs and online communites on this topic are currently mushrooming here! Yesterday I even saw an ad on Germany's biggest TV station right before the evening news. This is very unusual to say the least as the common people of Germany by and large have an incredible amount of distrust in anything but savings books! This current gold rush mood is just scary to me as usually, when something hits mainstream media, the magic is gone!

Anyhow, the question I am asking myself is: What will the market do now that it has access to more cash than ever? Are the markets even productive enough to put those sums of money to good use? Or is this bonanza just FU-money that encourages more reckless behavior?

I can only speculate as to what is going to happen but my hunch is that in the long term the gap between returns from index fonds and savings books is going to become a great deal smaller as more people are willing to shoulder risk for companies and thus risk premiums go down. There may still be active traders who try to find opportunities but I suspect structurally the percentage of passive investors will expand quickly and won't go back below today's percentage.

You're talking as if there's a swarm of fresh money flowing into the market, whereas it's more a case of people shifting away from traditional actively managed mutual funds etc. into indexing.

Money is cheap at the moment because growth is low, and that in turn means risk premia are lower and so on, but I don't think that's related to the rise of index funds. Then again I never understood why active management was so popular in the first place.

> You're talking as if there's a swarm of fresh money flowing into the market, whereas it's more a case of people shifting away from traditional actively managed mutual funds etc. into indexing.

Yeah, the question seems to be about fresh money. You could be right that we are mostly witnessing a shift from actively managed funds to indexing. As far as my home country (Germany) is concerned indexing seems to become more attractive to people who never invested, though. Now that I think about it I am not sure whether or not this kind of money would be "fresh money" as these people stored their money in banks who were probably investing it.

> Money is cheap at the moment because growth is low, and that in turn means resk premia are lower and so on, but I don't think that's related to the rise of index funds. Then again I never understood why active management was so popular in the first place.

As I understand it, risk premia is not related to growth. It is simply the costs to transfer risk to someone else. Active management was probably high in the past as banks had little incentive to sell passive investment plans: If people don't constantly buy and sell they don't cause transaction costs and thus income for the bank. Active and passive management are both neither inherently wrong or right. Until now active investment has been irrational but it could theoretically change if the share of money passively invested is high enough, say (made up number incoming) 80%.

> As I understand it, risk premia is not related to growth. It is simply the costs to transfer risk to someone else.

My point was that investors targeting a particular rate of return are having to take on more risk (because growth is low), which in turn lowers the risk premium through ordinary supply and demand.

"Money is cheap at the moment because growth is low, and that in turn means risk premia are lower and so on"

Maybe. Money is cheap if you are a bank or a government backed borrower (like a conforming mortgage loan in the US).

If you have collateral, like the car you're borrowing against, money is kind of cheap ... also if you have a perfect credit history.

But I am not so sure that money is cheap right now out in the real world. If you are a new business with no track record or a consumer with poor credit history I think money might be quite expensive for you ...

> But I am not so sure that money is cheap right now out in the real world. If you are a new business with no track record or a consumer with poor credit history I think money might be quite expensive for you ...

Really? My understanding was that (non-mortgage) subprime lending was higher than ever, business loans were cheaper than ever...

Doesn't that suggest people on average have had a misplaced fear of stock (indices) in the past, and that capital is now more efficiently allocated? Sounds like it would make the world better off.

> Doesn't that suggest people on average have had a misplaced fear of stock (indices) in the past.

Yes, I think this bit is pretty much agreed upon today. Hence the run for indices.

> and that capital is now more efficiently allocated?

That sounds likely to me.

> Sounds like it would make the world better off.

As tempting as it is to argue one way or another ... I think that's an impossible statement to make.

But like I initially said, it's all just speculation anyhow. Tbh these kind of discussions are inherently whacky. Chances are everything you've quoted from me is flawed on so many levels if one takes a closer look. Discussing finance is mostly a fool's errand.

Does anyone know what fractions of American equity currently belong to active and index investors? That seems like an important question, but the article doesn't address it.

http://www.zerohedge.com/news/2017-04-09/hedge-fund-cio-expe...

http://www.zerohedge.com/news/2017-04-09/horseman-global-unv...

We'll see how the next large correction plays out. That the "problem" may be "self-correcting" is not very reassuring (the dot-com bubble and the subprime fantasy also ended with some nice self-correction).

I've been meaning to read this essay which I believe argues that with index funds so big, prices are determined by the remaining investors who think they know better than the market -- which in theory should mean that the market gets even better information than it has now and determines prices even more accurately: http://www.philosophicaleconomics.com/2016/05/passive/

I'm not 100% sure I understand the argument or have presented it correctly. It's reassuring.

The other takeaway here is that if you have a theory that ETFs are going to become increasingly popular, you could "test" that theory by investing directly in companies with run ETFs.

For example you could buy some NYSE:STT or NYST:BLK and that might help you invest in "people pay a premium for the liquidity and other benefits of ETFs". Of course, you'd want to believe that that theory will outperform the S&P 500 :)

I had assumed - perhaps incorrectly - that as more and more money is invested in index funds and similar passive vehicles, it becomes increasing easy to beat the market as an active trader.

My thinking goes like this - as less money in invested actively, the market becomes less efficient at pricing. As the market becomes less efficient at pricing, it becomes easier to make money as an active trader. As it becomes easier to make money as an active trader, more money is invested actively, and the market becomes more efficient at pricing.

I may be using "active trading" incorrectly -- I just mean any kind of non-passive trading, both long and short term. But presumably, any kind of active trading theoretically increases the pricing efficiency of the market?

HFT is very efficient. Individuals will simply not beat HFT unless they actually have insider information.

I believe the comment was talking about any kind of active trading (i.e. picking individual stocks to buy and sell), not only HFT, which uses millisecond fluctuations in security price to make a profit.

HFT doesn't need price fluctuations (though they are used if necessary). It is mostly about capturing spread and rebates.

Your intuition seems correct but with a big caveat. If you divide the market in two, active to one side and passive to the other, both of them will have the exact same returns as a group. That's because all the passive investor is doing is replicating the market average and not changing it so the mix of stocks the total passive investors hold is the same as the mix of stocks the total active investors hold. That means as more money gets invested in index funds pricing becomes worse and thus there are more active funds that make a lot of money and more active funds that lose a lot of money and on average they still make the same returns before fees than the passive funds.

It only stops being like this if pricing gets so poor that active funds can start to trade in ways that the passive funds can't replicate, either because they're faster or more frequent or have some other trading advantage that makes it very hard for the passive fund to replicate it effectively.

"My thinking goes like this - as less money in invested actively, the market becomes less efficient at pricing."

Wouldn't it be that the market becomes as efficient at pricing as the accuracy of index inclusion ?

That is, we have replaced the widely varying performance and heuristics of active stock pickers with the more formulaic S&P 500 pickers ?

Presumably that's a lot more efficient, but I suspect there is room for errors and games in index inclusion ...

What do you mean by "pricing"? If there were only passive investors in the market, all the stocks would move together. Apple could announce that the next iPhone will be delayed until 2020 or Google could discover how to get energy from water and there would be no reaction in their stock price.

Reply on news.ycombinator.com