Perhaps overly generalizing, most everyone's portfolio is a poor approximation of the overall market with extra fees taking away even more -- hence the index funds tracking the overall market (or just the traditional top companies like S&P 500) will generally do better. When the market goes tits up, pretty much everyone will take a hit at the same time, index-fund holders and people with more specific portfolios alike. The people that won't be hit hard will be those with more "safe" investments (that don't outpace inflation...) like bonds, those who saw it coming and invested in negative-indexes, or those who get lucky with e.g. penny stocks / startups.

> What is the risk when a large percentage of the middle class and others are all in low fee index funds?

One risk is that index funds aren't really unmanaged, they are managed by people who set and apply rules as to what is included in the index. The greater proportion of total investment that is passively (i.e., without deep research into the practices behind the index) invested in such funds (and, particularly, the proportion that is passively invested in any one such fund), the greater the risk of the rules and practices of setting the index being manipulated.

The risk worries me too. Look at it this way: the whole purpose of a market is to allocate capital to productive ends. If everyone invests in index funds, market capitalization (and thus capital allocation) becomes a function of inertia and ceases to have anything to do with actual performance (because nobody is interested in looking at actual performance). As a result, I'd expect productivity growth to slow.

Guess what? Productivity growth has been slowing.

I doubt that the world economic slow growth projections are due to the 36% of equity and bond assets which are indexed instead of actively managed.

If that were true, then the 64% of actively managed funds would have a clear opportunity to capitalize on this fact for higher returns.

Index funds expose you to essentially the entire market; it's the sum of thousands of tiny pieces of businesses all across America. We're not putting all our eggs in one basket, rather, we're all distributing them across many.

If you can pick an investment that will weather an economic downturn, and still provide a decent return — let me know, because I want to get in on it ;)

But when everyone invests in everything equally, we all end up investing in nothing!

The previous post on the same blog debunks this. http://www.philosophicaleconomics.com/2016/05/passive/

http://www.ironbridgellc.net/PDF/hidden_risks_of_passive_inv... is an interesting countervailing read

What you're missing is that the stock market isn't a zero sum game. More money comes out of it than goes into it, because it represents companies that generally do useful things.

Except the biggest index by far is only 500 of the companies, not even remotely "the entire market".

You are thinking about this wrong. The S&P 500 is a capitalization weighted index meaning the larger the market cap the larger the weight. If you constructed a total market index using the same capitalization based weighting scheme, you'd find the S&P 500 would account for about 80% of the index value. In that light it is actually pretty close to "the entire market."

Vanguard total stock market fund (VTSMX) has over 3000 stocks, right?

Oh, there are definitely indexes/funds with broader coverage but the S&P500 funds continue to get a disproportionate share.

Are you sure about that? I mean, have you actually verified that with any particular fund family? I did, for the Vanguard family.

Vanguard S&P 500 index fund has $227.5B net assets (across all share classes)[1]. Vanguard Extended Market Index fund (which is the set difference of Vanguard Total Stock Market Index Fund and the S&P 500 fund) has $42.8B net assets (across all share classes)[2].

That is 18.8% of funds in Extended Market. The true ratio is 19% of funds in Extended Market[3]. Not so different.

Now you can argue that funds like the SPYder are the ones which receive a disproportionate share of attention. I would counter that (1) SPY is a smaller fund than Vanguard's (2) How much of that is investors vs. traders who use it for the liquidity.

In conclusion, I'm unconvinced that S&P500 index funds receive a disproportionate share of attention in investment, apart from investment professionals discussions. Even Buffet doesn't appear to care, and may simply use the S&P 500 because it is more well-known:

> Okay, so gold is not a screaming buy to Buffett. What should a typical upper-middle-class person in the U.S. buy to prepare for retirement?

> "Equities," Buffett answers without a moment's hesitation.

> "The VTI?" I ask.

> "That's good enough. Maybe a selection of high-dividend-paying stocks that are likely to raise their dividends. Maybe the top 100 dividend payers of the S&P 500."

>Then, after a second's thought, he adds, "Well, maybe not that, but equities."[4]

[1]: https://personal.vanguard.com/us/funds/snapshot?FundId=0040&...

[2]: https://personal.vanguard.com/us/funds/snapshot?FundId=0098&...

[3]: https://www.bogleheads.org/wiki/Approximating_total_stock_ma...

[4]: http://archive.fortune.com/2010/10/18/pf/investing/buffett_b...

I think you are already on the right track. In the end it's always hard to predict how something will turn bad. But you can already tell that a monoculture of simplicity will turn bad at some point. That's already enough to make a judgement call.

There's a number of risks associated with ETFs that isn't associated (or less) with individual stocks.

As a rule of thumb you might assume that if the index goes down, the index fund will add 1-5% going down (large index funds only), and if the index goes up, the index fund will "lose" 0.1-0.5% of that value going up. ETFs are similar. Why ? Lots of reasons:

1) gating (during periods of heavy activity, the adding or removing shares process can block. Additionally, due to new regulation the management of the ETF can impose this without recourse for you. So if the ETF drops 20% in an hour, but you notice this 5% into the drop and want out, chances are good you won't be able to)

2) slippage (due to time delays, contracts, fees, the fact that fractional share ownership doesn't exist, ...) $100 added to an ETF does not result in $100 increase in ETF value. This affects smaller ETFs more than larger ones, but it means that when the S&P 500 goes up 10%, SPY goes up 9.9% or so, when it goes down this makes the drop worse (so, say 10% -> 10.1%). There have been cases where slippage has been large, especially in smaller ETFs).

Note that the reverse is also true, unless you're a billionnair (or at least dozen millionaire). Maintaining "your own", say, SPY equivalent account by owning individual shares will likely cost you more in fees than it would cost you to own the equivalent SPY shares. Additionally if you have less money than the ETF, you'll have to have a rougher approximation.

4) slippage due to add/removes from the S&P 500 index. Since the ETF will have to make suboptimal (very large) trades when the index changes, this will cause the value of the ETF to "slip" compared to the index itself. Putting it separately since it results in bigger amounts.

So the SPY index tracks something between the S&P500 and this list : http://siblisresearch.com/sp-500-additions-removals/

This has, in the past, affected the difference between an S&P 500 index and an ETF more during recessions and that will continue.

5) Global synchronization. Many people use an ETF like SPY as an alternative for a 50% bonds 50% stocks portfolio. Needless to say, this is less diversified, and carries more risk.

6) Generally speaking, an ETF is of course vulnerable to the same risks as the individual stocks, to a lesser extent. However, bad things will happen to the ETF when bad things happen to any large holding. The thing is that the odds go up much faster than I bet people would expect.

Let's say the odds of something really bad (getting economic sanctions, default, bankruptcy, ...) happening to a top-10 S&P 500 company are 1% (happens once every century). We all (should) know the formula that determines risk across SPY as a result of that, but do you know what value it gives for this scenario ?

It's just shy of 10%.

So under these assumptions you can expect a 6-7% drop in the SPY etf every decade that won't be reflected in the index (index can, "for free", just remove failed stocks and replace them, the ETF can't). I hope you agree those are reasonable assumptions.