Can You Really Game Index Funds?
bloombergview.com
bloombergview.com
As always, Levine is fucking fantastic.
Remember, the main rationalization for Bernie Madoff's unbelievably consistent returns was that he was front-running, and fund after fund after fund after fund lined up to give him money. Besides, front-running isn't really where the big money is anyway. Insider trading is way, way more profitable from an individual standpoint.
Just because Vanguard claims to be smart about it, doesn't mean necessarily they actually are incentivized to be smart about it or actually are in practice. People can still judge them by how close they track the index, but that is reported separately from fees, which are all a lot of current and future retirees look at after having fees fees fees drilled into their heads. And the indexes themselves take a hit, so you need to adjust for that with a much more complicated measure.
They can effectively launder bad (or even good) tracking of the index into lower reported fees, by letting the order handlers profit on the inanity (and kickback via order-flow payments), allowing the fund managers to give themselves higher compensation without commiserate alarming fees.
To what extent, if any, do they actually do this? Do they report their income from paid order-flow in the fund prospectuses? Do they break it out by the managed funds, vs retail flow from their clients? Do they get major concessions to their retail trading costs in tacit exchange for being dumb with their etfs?
It is basically order flow that often blindly takes liquidity. Traders pay to get it, execute it prop at the best price (though they also can dump straight to the market without taking it prop if it isn't behaving dumb enough). Then they resell the position on the market at a more deliberate pace, providing liquidity (which lets them capture the spread, and lets them earn kickbacks from exchanges that pay for liquidity). Or they match the positions against future dumb flow coming in on the other side.
The ETFs can be smart about it and try and make the trades over time instead of all at once, and Vanguard likely does. This then feeds back in and lowers the amount that the actual index takes a hit. The equilibrium in reality is that both the indexes and the shareholders in the ETFs take a bit of hit.
I would think that would be a HUGE scandal if it were true and ever came out.
But price fixing is also illegal--nevertheless, two gas stations across the street at a profitable intersection can engage in it solely through price signal tit-for-tat[1]. This effectively masks intentionality.
Much more fantastical and speculative: machine learning algorithms at both firms could now, or in the future, arrive at this cooperative strategy, even with the only communication being through price signals. Without any human ever even knowingly giving the explicit go-ahead.
[1] https://en.wikipedia.org/wiki/Prisoner's_dilemma#The_iterate...
http://vrs.vista-one-solutions.com/reports/1-6/vang/
Short answer is that Vanguard Brokerage Services did not receive compensation for equities order flow in 1QTR 2015, but did receive compensation for options order flow.
I have no idea what the laws are surrounding their index funds or how those index funds are allowed to/actually do interact with Vanguard Brokerage Services.
---
As for proxy voters choosing the management, proxy vote research usually doesn't make sense. You see strategies amongst large hedgefunds like this: keep their valuable proxy vote research private, make a vote based on the most predictable-to-them-but-not-to-you outcome based on the research (positive or negative for the company, doesn't matter unless it will be picked up by others immediately, even then they can do things like exiting their position through an obscure hedge), and then hold or sell their shares based on the overall vote outcome and its implications in the research.
Researching and making proxy votes out of naive benign interest of the company is often just doing altruistic work for a greater collective, something markets frown on and usually punish.
Another interesting point is that the Volcker Rule has more or less caused a massive shift of this type of strategy away from US investment banks and into hedge funds. I don't have real data on this - just my observations.
http://kiddynamitesworld.com/where-bloomberg-discovers-that-...
But saying "index funds free-ride on the work done by active investors" and then following with "no one thinks that active managers should be able to charge for their services, is a world that will spend too little time and effort on allocating capital to the right businesses" is FUD.
The value of the market represents the sum total opinion of everyone in it (plus noise), not just the managers of mutual funds losing business to index funds. Frankly, it sounds like the griping of someone telling fund managers that they deserve their fees, but the supposed loss from using index funds described in the original article (~.2%) is still dwarfed by the increased fees of actively managed funds.
Most index fund expenses are around .1-.2%, while most active funds come in at a whole 1-2%. To justify the cost of an actively managed fund, a manager has to not just beat the market, but trounce it. Very, very few can do so for any length of time, and they know it, which is why articles trying to convince people of the virtue of active fund management are constantly written. Unless your manager is as good as Buffett, buy an index fund.
The math is simple, but there's many fund managers out there trying to convince you otherwise.
However, we are in no danger of running out of active traders, so there's no need for anyone to run out and sell their indexed investments to save the free market ...
The little coda about active management makes more sense if you read him religiously, because this is a schtick of his. Passive management helps most investors. But the market as an entity benefits from active management, because active management makes prices more accurate. This despite the fact that for the most part, contributing to the accuracy of prices comes at the expense of the actively-managed funds.
So without active management, the passive funds would perform more poorly, because their prices wouldn't benefit from the corrections of people trading into them to profit from mispricing.
There's also a kind of second-order version of market efficiency that says that active fund managers that can actually beat the market will increase their fees until their post-fee returns are the same as everyone else. So even if active _fund managers_ get compensated for making prices more efficient, there's no reason to believe that _fund investors_ will be.
This is evidence for the "second order version of market efficiency" your mention: it was uncanny, and put me into index funds for life (that, and the fact that there was no way of predicting from year-to-year which funds would beat the market the following year.)
There is a good reason that index funds are very difficult to beat consistently, and it's not because they are copying all the hard work everyone else does for fees. It's because you don't get paid for specific risk.
But like tptacek said, you probably need to read Levine regularly to unpack his argument there; it was extremely self-referential.
The market needs active traders for stock prices to accurately reflect investor opinion, but actively managed funds are not the only source of active trading.
Another good article where he discusses the Goldman software developer who 'stole' company code:
http://www.bloombergview.com/articles/2015-07-07/goldman-cod...
Also by this logic total market funds should outperform other indexes by a healthy amount over time. Also maybe not what we are seeing. Granted total market funds invest in something that is different from what other indexes track.
To an extent the amount the ETFs are effective at this lowers the amount the index underperforms, because they (the etfs) are themselves the driver of the liquidity crisis the index is getting subject to. So you would expect some sort of equilibrium, and the claim is therefore to be taken that this equilibrium settles down at of 20+ basis points.
https://www.google.com/finance?q=NYSEARCA%3ASPY%2C+NYSEARCA%...
I'll admit it's looking for a needle in haystack though. In all my looking I've only found these three. Really just VHT and POGRX, since the success of BRK (Warren Buffett's company) is common knowledge.
Hint: The people constructing the index are doing as much of an active choice of stocks as the active traders are. Nothing stops a trader from selecting stocks following the same strategies as how an index is selected.
Still makes me wonder if that means I am going to get an extra 20 basis points for free out of VTSAX.
Good luck finding a total market index fund elsewhere.
Yes, there are funds that try to match the entire market. Vanguard runs a couple "total market" indices (VT and VTI, for example), and there are more.
But that's not to say that they actually invest in the entire market. They don't; they use sampling to try to match the performance of the thing they index. (Vanguard, at least, gets very very close to this...)
Disclosure: I am invested in in VTI.