The traditional rule-of-thumb for asset allocation is to keep 6 mos living expenses in liquid cash, then allocate the rest of your investable assets (your age)% bonds, (1-your age)% stocks. For example, if you are 32, then 32% bonds, 68% stocks.

But Terry clued me in to another strategy you also might want to consider-- I'll call it "rolling availability" strategy but we didn't invent this and I don't know what it's officially called. Keep as much in cash & high-quality bonds as you need to cover expected expenses over the course of a market cycle, the rest in stocks and other investments.

If you are more than a market cycle away from retirement, then I don't think you need to consider this model. Unless you have plans to use your investments to fund some other goal (most likely downpayment on a home or college tuition) within one market cycle, then you can use this method for the portion of your assets you'll need to fund that expense.

For example, we've decided to use 10 years as the length of a market cycle (meaning we think stocks can stay depressed for up to 10 years, so we wouldn't want to be forced to sell during that time). We want to keep a minimum of 15% investable assets in cash and cash equivalents (CDs and T-bills). This is high based on expected expenses (it will cover about 3 years for us since we're living simply now), but it's a hedge in case both the stock and bond markets go to @#$ in a handbasket, which we think is within the realm of possibility. Plus we're essentially already in retirement, so we must be very conservative so we're not forced to get real jobs in the future. So with three years of expenses covered by our cash reserves we'll need the equivalent of 7 years of expenses to be covered by bonds in order to get through a market cycle.

The rest of the investable assets will be in what I call "other" investments. Not only stocks (equities), but also commodities, real estate, junk bonds, local businesses, jewelry, art, et al.

Overall, this strategy should have the same general effect as the age-based allocation, in that the percentage of bonds within your overall portfolio should increase over time, but the percentage is determined less by an arbitrary age-based formula than by your actual needs and existing portfolio balance. For us, we're actually only about 2% off our age-based guideline, but this may change when we re-allocate in the future. If the market does well, the percentage in bonds will decrease, if it does poorly, the percentage will increase, but no matter what, we shouldn't be forced to sell investments during a downturn (assuming the downturn doesn't last longer than 10 years). The percentage will also change if our expected expenses change.

It's more complicated than the simple age-based formula, but if you're investing for either an early retirement or something like college tuition for your kids, it may provide you with a more meaningful risk-management strategy than the age-based guideline which assumes you won't start spending down your assets until you're in your 60s.