Faced with a complete lack of other attractive income-generating investments, many retirees have turned to stocks as their last refuge to produce the cash flow they need to cover their living expenses. With many dividend stocks yielding more than bank CDs, Treasury bonds, and even high-grade corporate debt, those who need to live off income from their portfolios have boosted the stock portion of their asset allocations to above-normal levels.

If you've boosted your stock allocations because of low interest rates on bonds and other fixed-income investments, you're not alone. But if a recent study is accurate, you may want to rethink your strategy.

An inconvenient truth?
Prevailing wisdom has long suggested that a substantial equity allocation makes sense even for retired investors. Given the need to see assets grow over time, the argument goes, the appreciation potential of stocks gives many retirees their best chance to produce enough in income and capital gains for their portfolios to outlast them. In fact, some asset allocation recommendations advise owning more in stocks than in bonds and cash.

But a new study from the Putnam Institute casts that conclusion into question. According to the study, most retirement portfolios should have only 5% to 25% allocated to stocks, with a 10% stock allocation being the most common figure.

What explains the disparity? The study makes many assumptions, some of which you often don't see in retirement planning analysis:

  • The study takes mortality into account, and as a result, it assumes that you'll be able to spend significantly more of your remaining nest egg than most similar studies. In particular, the study assumes 5% to 8% withdrawal rates for investors who are age 65, 9% to 12% withdrawal rates for 75-year-olds, and rates of 17% to 23% for 85-year-old investors. That's well above the 4% rule many advisors use.
  • Return assumptions are inflation plus 6 percentage points for stocks, 3 percentage points more than inflation for bonds, and inflation plus 1 percentage point for cash. Those returns are largely based on historical averages.
  • Rather than looking at actual portfolio values at specific points in time, the analysis condenses a continuum of portfolio values throughout retirement into a single discounted figure. The discount rate is determined by the specific mix of investments used.

The net result of these assumptions is that riskier portfolios have higher discount rates applied to them, thereby reducing the net present value of their cash flows. Given that the biggest risk retired stock investors face is a bear market shortly after retirement, it may be appropriate to use different discount rates for riskier portfolios.

Why you shouldn't change a thing
Putnam's analysis is quite elegant and does a good job of simplifying a complicated topic. But in my view, several of the assumptions it makes aren't realistic.

First, the analysis necessarily assumes that people won't change their spending behavior in response to changes in the financial markets. Yet many people take steps to reduce their spending when they take a financial hit.

Second, average returns simply aren't available in today's market. Cash investments today have returns that lag inflation by 2 to 3 percentage points. Bond ETFs iShares Barclays Aggregate Bond (NYSE: AGG) and Vanguard Total Bond Market (NYSE: BND) have Securities and Exchange Commission yields of just 2.5% to 2.7%, which also lag inflation over the past 12 months. iShares Barclays TIPS Bond (NYSE: TIP) has a real yield of less than 0.3% over inflation.

Of course, stocks are risky, too. Over the past 10 years, stocks have returned around 3% annually, just treading water against inflation. But you can find many blue-chip stocks that offer earnings yields of 10% or more, including ConocoPhillips (NYSE: COP), AstraZeneca (NYSE: AZN), and Hewlett-Packard (NYSE: HPQ). And with the current earnings yield on the Dow Industrials approaching 7%, even a broader investment like the Dow-tracking ETF Diamonds Trust (NYSE: DIA) can give you reasonable value right now.

In other words, by shifting to bonds, you'd be buying them at potentially their highest levels -- levels at which historical returns are mathematically impossible. That's not a recipe for success.

Stay the course
Perhaps the best lesson from the Putnam study is that studies are only as good as the assumptions that they make. Although it's helpful to have perspective on the factors that affect your retirement finances, you need a plan that can adapt to your own situation and preferences. Only then will you maximize your chances of having a prosperous retirement.

I still think that dividend stocks offer a strong compromise between risk and the need for growth and income for retiree portfolios. For some good prospects, the Fool's free special report with 13 great dividend stocks is a good place to start.