Hello everyone,
I've been reading lately a lot about portfolios, "time diversification" and asset allocation. According to books like "Stocks for the long run", the efficient frontier for periods of 20 years or so are very different than the "normal" graphs that I normally see. In fact, in 20 years, 100% stocks have an expected return of almost 7% while the SD is +-3%. In the minimum risk portfolio the return is a little above 5% for a SD of +-2.5%. You sacrifice almost 2% of return to decrease only 0.5% of the SD! Does that make any sense?
I mean, after reading all those texts I can't find a reason NOT to invest 100% in the stock market IF:
- You want to go for the very long-term (15-40 years)
- You can sleep at night watching 50-70% drawdowns
I've been reading some articles (like this one) criticizing the book, but the arguments in my opinion are not valid, because:
1- I know that dealing with the very long term, although the volatility of the expected returns decreases, the range of where you may end increases, like if you invest 1k you can end between the big range of 7k-112k in 40 years. I don't know how this makes the idea less interesting, because even 7k is still a 5% return, but you have a far better probability to end with a substancially bigger amount.
2- The probabilities. They argue that there is a 10% chance of stocks returning less than bonds in 20 years, and that the shortfall of only 2% means 50% less in 20 years, etc. etc. But what about the whooping 90%, where you normally end with something like 5x than if you had invested 100% bonds? Where do you want to stay, in the 90% or in the 10%?
I mean, if in a 20 year time horizon stocks have a BIGGER expected return and LOWER volatility than bonds, why invest in bonds at all if you want to stay in the market for that long? Why invest 10 or 20% in bonds to decrease the volatility somewhat at a huge expected return cost (remember a small difference in 20 years becomes a huge one)?
I know that everything can be different in the future, in fact the 20y return of the S&P 500 is now lower than bonds (unlucky 10% I mentioned above), but I can't see a better strategy than having a 95% chance of having the best returns (more $$$ means lower % of withdrawal, which is safer).
BTW, I'm a little confused of how to apply this "theory" with the withdrawals after retirement, because you neither stay fully invested (because you use 3% of your initial portfolio, for example), neither you sell everything at day 1.
I've been reading lately a lot about portfolios, "time diversification" and asset allocation. According to books like "Stocks for the long run", the efficient frontier for periods of 20 years or so are very different than the "normal" graphs that I normally see. In fact, in 20 years, 100% stocks have an expected return of almost 7% while the SD is +-3%. In the minimum risk portfolio the return is a little above 5% for a SD of +-2.5%. You sacrifice almost 2% of return to decrease only 0.5% of the SD! Does that make any sense?
I mean, after reading all those texts I can't find a reason NOT to invest 100% in the stock market IF:
- You want to go for the very long-term (15-40 years)
- You can sleep at night watching 50-70% drawdowns
I've been reading some articles (like this one) criticizing the book, but the arguments in my opinion are not valid, because:
1- I know that dealing with the very long term, although the volatility of the expected returns decreases, the range of where you may end increases, like if you invest 1k you can end between the big range of 7k-112k in 40 years. I don't know how this makes the idea less interesting, because even 7k is still a 5% return, but you have a far better probability to end with a substancially bigger amount.
2- The probabilities. They argue that there is a 10% chance of stocks returning less than bonds in 20 years, and that the shortfall of only 2% means 50% less in 20 years, etc. etc. But what about the whooping 90%, where you normally end with something like 5x than if you had invested 100% bonds? Where do you want to stay, in the 90% or in the 10%?
I mean, if in a 20 year time horizon stocks have a BIGGER expected return and LOWER volatility than bonds, why invest in bonds at all if you want to stay in the market for that long? Why invest 10 or 20% in bonds to decrease the volatility somewhat at a huge expected return cost (remember a small difference in 20 years becomes a huge one)?
I know that everything can be different in the future, in fact the 20y return of the S&P 500 is now lower than bonds (unlucky 10% I mentioned above), but I can't see a better strategy than having a 95% chance of having the best returns (more $$$ means lower % of withdrawal, which is safer).
BTW, I'm a little confused of how to apply this "theory" with the withdrawals after retirement, because you neither stay fully invested (because you use 3% of your initial portfolio, for example), neither you sell everything at day 1.