This is an interesting area of study, but I don't see how it concludes that bonds should do as well as stocks if the premium is truly arbitraged away. That is, if the premium is really gone, then it seems one should still expect stocks to do better than bonds because they make up a faster part of the economy’s growth curve relative to bonds.
Yes, they are riskier investments. But that is why you buy and hold them for the whole century (or at least as long as you can so the day to day risk smooths out a bit).
I don't understand what you mean about the economy's growth curve. But if what you think is correct then that implies you have discovered a trading rule! You should go short bonds and long stocks, and make yourself a fortune. ;-)
In the real world, capital markets tend to become more efficient over time as inefficiencies are arbitraged away. So it's unrealistic to expect that any particular asset class will outperform others on a risk-adjusted basis over a century.
What I mean is quite simple. Suppose the economy is growing at x%. That x% is made up of components. Some things are growing at smaller than x%, some larger. But when you sum them all together in a weighted fashion, you get x%.
Now bonds, on average, grow somewhere near x%, while stocks, on average, grow more than x%. And other things, like most savings accounts grow at less than x%. The reason stocks should grow faster is that they represent money that is being invested in new innovation (often technology), which have generally higher growth prospects than where the bond money is being invested (e.g. infrastructure).
If you look at the economy's growth curve, you can consider it made up of several different curves corresponding to different asset classes, which when summed together in a weighted fashion, yields the overall curve. The stock part makes up more of the growth component than bonds and that is what I meant by makes up a faster part of the growth curve.
This doesn't constitute a trading rule by any means. I'm not talking about any premiums or arbitrage opportunities. In fact, my comment said if the premium is really gone.
All I'm saying is I don't see how the area of study you referenced concludes that bonds as an asset class will grow at the same rate as stocks. And I still don't see it. Please enlighten me.
Note that I am not suggesting stocks outperform bonds on a risk adjusted basis. That's why I ended the comment saying they were riskier, and that, as a result you should buy and hold them.
The overall point here that I suppose I am making is that just because asset classes grow at the same rate when adjusted for risk doesn't mean you should be indifferent to which you invest in. Buying low, selling high aside, if you can smooth out the risk via buy and hold and other strategies, it is wise to invest some in the assets with the higher growth curves.
If stocks really grew faster than bonds (outside of risk), wouldn't it make sense for investors to take some of the money they have in bonds and put it in stocks, thereby increasing their total return? That'll drive the price of stocks up and the price of bonds down, driving total returns for stocks down and returns for bonds up (since you can get the same bond more cheaply). The process should only stop when investors are indifferent, on a risk-adjusted basis, to stocks & bonds.
That's what the grandparent means by "inefficiencies are arbitraged away".
As for equities funding more high-growth areas of the economy - you need to make a distinction between primary and secondary markets for capital. The primary market is where actual, producing firms offer part of their capital structure (either debt or equity) up for sale. This is often associated with entrepreneurs, investment bankers, venture capitalists, and private equity. The secondary market is where financial firms trade securities of existing companies - hedge funds, mutual funds, and retail investors.
In the secondary market, investors always have the option of bidding up the prices of securities that they really, really want. As a result, it doesn't matter how well the economy does, it only matters how well the economy does relative to how people expected it would do. If people expect the economy to shrink 10% but it really shrinks 5%, stocks (in the secondary market) will jump. If they expect it to grow 20% but it really grows 15%, stocks will crash, even though absolute growth rates are significantly higher. That's why only unexpected earnings have an effect on a company's stock price.
The primary market is much less liquid and hence less subject to arbitrage opportunities. If a company is growing very well, it will be able to charge a higher price for its securities, and the proceeds will go directly to its balance sheet.
So, the fact that high-growth industries tend to be financed with equity rather than debt means a lot for entrepreneurs and venture capitals, but very little for investors in the public markets. For that matter, a number of studies have found that "growth" stocks do significantly worse than "value" stocks, because investors consistently overweight their growth prospects.
I said in my comment specifically Note that I am not suggesting stocks outperform bonds on a risk adjusted basis. It seems your whole comment assumes the contrary as it opens with If stocks really grew faster than bonds (outside of risk). This assumption is simply not true.
However, given your comment, would you suggest an investor invest in stocks at all in a buy and hold strategy (as opposed to merely speculation on the price being low at the moment)?
I think what you're trying to say is that is in the long run, equities (relative to debt securities) will contain an increasing share of the total wealth in the whole economy. That may or may not be correct, but it's irrelevant to Warren Buffett's original statement or the equity risk premium. What we're discussing here are the returns an individual investor should expect to receive from stocks in general.
Buying and holding individual stocks is a bad strategy for a long-term investor. Eventually a lot of those stocks will go to zero. In most cases you'll be better off with a passively-managed index fund which periodically rebalances based on market capitalizations.
No, that's not what I'm saying at all, but let's just move on. (We can take it off-board if you want though.)
Also, I never suggested buying and holding individual stocks. I agree that one should only buy and hold that type of index funds.
As for the relevancy of this discussion, I was just responding to your comments. The original post was just about a dumb argument made by Warren Buffett, and not about equity premiums or any other underlying issue at all. In fact the post closed with This is not a discussion about whether we are actually in for similar, higher, or lower returns. Let's leave that for another day. This is just to say that this argument for lower returns is ridiculous!
That being said, with your belief in the equity premium theory, do you believe there is any scenario where investing in stocks long term (say buy and hold for the century) makes sense?
Sure, it makes sense to put a share of your assets into low-cost equity index funds which base allocation weights on market capitalization. Note that the DJIA (mentioned in the original article) isn't a suitable index for this because its 30 components are picked more or less arbitrarily by the Dow Jones staff.
But this is what I was getting at. You say to buy stocks, in a good index mind you, but you also say they are no better as an asset class when you adjust for risk. So why buy them?
Or do you agree with my point from above?:
The overall point here that I suppose I am making is that just because asset classes grow at the same rate when adjusted for risk doesn't mean you should be indifferent to which you invest in. Buying low, selling high aside, if you can smooth out the risk via buy and hold and other strategies, it is wise to invest some in the assets with the higher growth curves.
I partially agree with you. While I don't think equity index funds will have higher risk-adjusted returns than any other asset class, they are still useful for diversification. By building a portfolio of uncorrelated or negatively-correlated investments, investors can somewhat reduce the total risk while retaining most of the gains. Personally I try to maximize the expected Sharpe ratio of my portfolio as a whole. This isn't exactly a buy-and-hold strategy; it requires rebalancing the portfolio whenever market moves cause the current asset allocation weights to drift too far from the target.
I could tell you, but the answer wouldn't be applicable to your or anyone else. Almost all of my long-term investments are in a certain retirement account that only provides a limited set of mutual funds as investment options. If I could invest in anything then I'm sure I would have different allocation targets, but since I can't then I haven't bothered to figure it out.
What I have done is write a small program which can take a set of allocations for portfolio components and then calculate an expected Sharpe ratio for the whole portfolio by using a Monte Carlo simulation run over hundreds of years using historical returns data. It then uses simulated annealing to do a constrained optimization over the solution space. This is only feasible because there are only about 20 available portfolio components. There are also no capital-gains taxes, transaction costs, shorting or borrowing, all of which keeps things reasonably simple.
Of course historical returns are a poor predictor of future performance. But historical correlation coefficients are a reasonably good predictor of future correlations, which is mainly what I am after with trying to find the best diversification.
Actually at the moment I have the majority in a money-market fund just because I'm trying to time the market and play a hunch. But that's usually a poor strategy and I wouldn't recommend it to anyone else.
Yes, they are riskier investments. But that is why you buy and hold them for the whole century (or at least as long as you can so the day to day risk smooths out a bit).