What would happen if everybody invested only in index funds?
The reason indexing works is that the stock market is largely “efficient.” That is, all the publicly available information (and maybe even, illegally, a little of the inside information) has been factored into the price of any given stock. Some people will be wrong for very much more sophisticated reasons than others (an oil analyst may have stayed up all night crunching numbers based on tanker bookings, yet misinterpreted how this data would affect Exxon’s stock), and some complete nincompoops may find that they’re right (“X is my lucky letter — I love stocks with X’s in them!”), but it is very hard to tell whether a stock is likely to outperform or underperform the market.
It may not be impossible to tell the bargains form the bozos, especially with smaller, overlooked stocks, and especially when a market seems to have gone to extremes of panic or euphoria. And it is certainly fun to try. But it is very hard. (How well would you have done taking advantage of the panic of 1917 to buy Russian stocks? Or taking advantage of Microsoft’s wild 20-fold increase to short it and watch it climb 20-fold again?)
Once you have hundreds of smart people trying really, really hard to decide whether a stock should be bought or sold — and thereby setting its price through the convergence of supply and demand — will you really get a much more accurate or sensible price if you persuade thousands of others to chime in with their opinions? No.
If literally everyone just blindly bought the S&P 500 index, the market would of course become completely inefficient and astonishing opportunities would emerge. Which is why we’ll never get that far. There will always be a balance between people trying to beat the averages, some succeeding, most failing, and those taking the boring, prudent, low-cost course.