Here are my thoughts on a 100% stock allocation: A gain of 1000 dollars followed by a loss of 500 dollars is better than a gain of 300 dollars without a following loss. I feel that many people criticizing the 100% stock portfolios always look at the downturn from the peak and compare it to a more conservative AA. But that is only valid if you invested a lump sum on the peak. If you, like me, continuously invest over a long period of time, most of your money will have been in on the runup as well, and even after the pullback they will be better off than with a conservative allocation. Simply put, most of the volatility is on the upside. This is why I'm not very worried about "selling stocks when they're down", because even if stocks are down from their peak, they are likely to still be up when compared to the bonds!
Here's a quote that summarizes my though well and also provide some data, from an excellent paper by Javier Estrada. You can find the entire paper here: http://blog.iese.edu/jestrada/files/201 ... path-2.pdf
Full disclosure: I'm 100% stocks (pretty much no emergency fund) when it comes to the portfolio I can control, but I have some money locked up at my employer's and I also have a pension waining for me at 65 (or whenever the politicians decide I will be allowed to officially retire). As I'm not a US citizen, I can expect my pension to be somewhere close to my salary if I were to work until 65, though in practice I expect to retire much earler than that without the help of that pension. I would, however, probably keep all my money in stocks (internationally diversified to avoid single-country risk, of course) even if I had control of all of this money.Consider again the US first and ask why (in fact, whether ) the 100×30 strategy is riskier than the other strategies shown in the exhibit. Relative to the best dynamic and static strategies considered here, the strategy that fully invests in stocks has the lowest failure rate (tied with the best static strategy, which in this case is 90% invested in stocks) at 3.7%; provides the same or better downside protection when tail risks strike (measured by P1, P5, and P10); and provides much higher upside potential (measured by the mean, median, P90 , P95, and P99). Therefore, as discussed before, the higher standard deviation of this strategy only indicates uncertainty about how much better off (not worse off) a retiree will be after 30 years.
Results for the world market and the average country in the sample are similar. Relative to the best dynamic and static strategies considered here, the strategy that fully invests in stocks has the lowest failure rate (6.2% for the world market and 26.4 % for the average country), provides the same or better downside protection when tail risks strike, and provides much higher upside potential. The only exception to this statement is in terms of downside protection as measured by P1 for the average country in the sample; in this case, the best static strategy provides a slightly higher terminal wealth ($12 versus $3 for the all‐equity strategy). Importantly, note that this panel shows average figures for the best static strategy for each country, and which one is the best changes across countries. In other words, no single static strategy can be said to provide better downside protection when 1% tail risks strike than the all‐equity strategy.
In short, although a strategy that fully invests a retirement portfolio in stocks can be perceived as riskier than most alternatives, is that really the case? Is a strategy that has the lowest probability of failure, provides the same or better downside protection, and higher upside potential really riskier than other strategies simply because a retiree is more uncertain about (how much higher will be) his bequest? If not, then having a retirement portfolio fully invested in stocks is a strategy that should be seriously considered by retirees.