3 ms·

One of the ideas of cap weighting is that it has minimal impact. If company A is worth 10x what company B is worth then I can buy 10x of company A and 1x of co

by vostok 9y ago

One of the ideas of cap weighting is that it has minimal impact. If company A is worth 10x what company B is worth then I can buy 10x of company A and 1x of company B to spread my impact as evenly as possible.

cbanek 9y ago
But if everyone does that, that inherently pushes up the price of company A a lot faster than company B. Which reinforces the feedback loop of the bigger they are the more you buy. This is all done without anything needing to happen to the actual value of the company. Only active traders and institutions would sell individual stocks to bring that value down. Selling the index keeps the market weights even. The minimal impact of cap weighting is actually that you don't have to rebalance your investments to keep them tracking the same. By them going up and down, they keep the same percentage by market cap weight. What you seem to be saying is good is diversification, which it is, but you have to make sure you are well diversified.
- vostok 9y ago
  > But if everyone does that, that inherently pushes up the price of company A a lot faster than company B. Can you explain how this works? Suppose Company A is worth $100 billion and Company B is worth $10 billion. Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. You're buying 0.0000001% of each company so you'd expect that you'd push up the price of both companies at the same rate. > The minimal impact of cap weighting is actually that you don't have to rebalance your investments to keep them tracking the same. By them going up and down, they keep the same percentage by market cap weight. Yes this is another benefit of cap weighting.
  - cbanek 9y ago
    > Suppose Company A is worth $100 billion and Company B is worth $10 billion. Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. You're buying 0.0000001% of each company so you'd expect that you'd push up the price of both companies at the same rate. Sorry, you're totally right. The prices should increase in the same weighting when buying just like they go down in the same when selling. The difference is that you're weighted by market caps, so you're increasing the P/E of the entire market (in a super small way). That is, you're not investing more in companies that are valued more "attractively" (lower P/E multiple) than those that might be "over valued" (high P/E), which is what the market typically does to determine prices.
    - vostok 9y ago
      Yes I agree that cap weighting doesn't over buy the low P/E stocks. I'm not sure I agree that low P/E means that a stock is undervalued. For example a company that is guaranteed to make $10 every year should probably have a higher P/E than a company that is expected to make $10 with $1 variance. If we move to a more abstract notion of undervalued then I still agree with you that cap-weighting isn't guaranteed to overweight the undervalued stocks.
  - cs702 9y ago
    > Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. Not necessarily. The price could change in VERY different ways depending on the marginal supply and demand for each stock (i.e., how many more/fewer shares of the stock are other investors willing to buy/sell at different prices when the index fund goes out to buy that 0.0000001% of float). A tiny 'buyer-seller imbalance' in the stock market cause a big price adjustment.

by vostok 9y ago

cbanek 9y ago

But if everyone does that, that inherently pushes up the price of company A a lot faster than company B. Which reinforces the feedback loop of the bigger they are the more you buy. This is all done without anything needing to happen to the actual value of the company. Only active traders and institutions would sell individual stocks to bring that value down. Selling the index keeps the market weights even. The minimal impact of cap weighting is actually that you don't have to rebalance your investments to keep them tracking the same. By them going up and down, they keep the same percentage by market cap weight. What you seem to be saying is good is diversification, which it is, but you have to make sure you are well diversified.

vostok 9y ago
> But if everyone does that, that inherently pushes up the price of company A a lot faster than company B. Can you explain how this works? Suppose Company A is worth $100 billion and Company B is worth $10 billion. Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. You're buying 0.0000001% of each company so you'd expect that you'd push up the price of both companies at the same rate. > The minimal impact of cap weighting is actually that you don't have to rebalance your investments to keep them tracking the same. By them going up and down, they keep the same percentage by market cap weight. Yes this is another benefit of cap weighting.
- cbanek 9y ago
  > Suppose Company A is worth $100 billion and Company B is worth $10 billion. Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. You're buying 0.0000001% of each company so you'd expect that you'd push up the price of both companies at the same rate. Sorry, you're totally right. The prices should increase in the same weighting when buying just like they go down in the same when selling. The difference is that you're weighted by market caps, so you're increasing the P/E of the entire market (in a super small way). That is, you're not investing more in companies that are valued more "attractively" (lower P/E multiple) than those that might be "over valued" (high P/E), which is what the market typically does to determine prices.
  - vostok 9y ago
    Yes I agree that cap weighting doesn't over buy the low P/E stocks. I'm not sure I agree that low P/E means that a stock is undervalued. For example a company that is guaranteed to make $10 every year should probably have a higher P/E than a company that is expected to make $10 with $1 variance. If we move to a more abstract notion of undervalued then I still agree with you that cap-weighting isn't guaranteed to overweight the undervalued stocks.
- cs702 9y ago
  > Buying $100 of Company A and buying $10 of Company B shouldn't push up the price of Company A faster than Company B. Not necessarily. The price could change in VERY different ways depending on the marginal supply and demand for each stock (i.e., how many more/fewer shares of the stock are other investors willing to buy/sell at different prices when the index fund goes out to buy that 0.0000001% of float). A tiny 'buyer-seller imbalance' in the stock market cause a big price adjustment.