[dgcupps]

Why does anyone buy 10-year bond? It's because they value stability in returns over absolute returns. While it is true that over the long run the stock market will outperform current bonds, it isn't always true that it will in the short-term.

If your a pension fund, a retiree, or anyone else who will need to spend money in the short-term, it often makes sense to forgo some long-term returns for short-term stability.

You're correct that past returns are no guarantee of future performance; however, there is 100 years of history (not to mention economic theory) that supports the assumption that equities will outperform current interest rates over the long run.

[jbay808]

I guess I didn't think of 10 years as the short term. Let's say the 30 year bond then. The 30 year bond rate is something like 2.5%... Who would buy that if it was vanishingly unlikely that equities would return less than that after 30 years?

I guess what I'm trying to get at is, is there a way I can estimate the probability that equities will return less than 2.5% over the next 30 years, given the information that the 30-year bond rate is 2.5% and the expected return on stocks is, say, 10%, and the market is efficient and some people buy bonds anyway?

[dgcupps]

One thing to keep in mind is that people who are buying 30 year bonds often aren't planning on holding them to maturity; rather, they are looking for an asset that is relatively stable in value while also generating a small return.

Can you estimate the probability of your scenario? Sure, you could use the historical volatility of equity returns to calculate the chance of that happening. However, unless you add in other factors that you think could realistically hurt future equity returns, I think you'll find that probability to be very small.