Last month I wrote about a soon-to-be-published paper by Wade Pfau, PhD ("A 'Safe' Retirement Spending Rate"). Since that piece ran, I have received a substantial number of messages and phone calls from financial planners that found Dr. Pfau's approach fascinating. You will be able to read Pfau's paper, "Safe Savings Rates: A New Approach To Retirement Planning Over The Lifecycle," in the May issue of the Journal of Financial Planning.  The comments I received came in basically two categories.  First were criticisms or discussions of the assumptions used by Prof. Pfau. The second group was comments and questions about what to do with this information from a practical standpoint.

I plan to dedicate the next few columns for Financial Advisor to some of the issues raised by readers. Today I'll focus on one of the assumptions readers found fault with in the Pfau paper, namely that it illustrated a steady spending pattern beginning with a specified amount that was increased in lockstep with inflation throughout retirement.

This spending pattern assumption is hardly unique to Pfau. Most studies of sustainable withdrawal rates or portfolio survivability use this assumption or assume a fixed and steady rate of inflation such as 3% or 4%. Pfau states his assumption clearly and I believe it a reasonable way to incorporate inflation into his examination of the issue. I do not think it wise to dismiss any study based upon these types of assumptions.

In my first column for this retirement publication of Financial Advisor, "A Little More To It" (December 2010), I discussed how the real-world often conflicts with theory. Assumptions like these are necessary to frame an issue and we simply can't simulate every possibility. In the end, the studies can be quite useful as long as the practitioner understands the studies' limitations.

In the real world, I have yet to see a client begin with a specified spending amount and either only increase that amount each year by a fixed rate to compensate for cost of living increases or increase the spending amount in lockstep with the actual inflation rate. Actual spending patterns vary greatly, however, according to countless conversations with colleagues around the country, they fall into a few basic categories.

The least common is an endowment approach. Like an endowment fund at an educational institution, a set percentage of the portfolio's balance on specified dates, typically annually, is distributed for spending. The principal advantage to this approach is that it is mathematically impossible to deplete the portfolio. Even if the asset pool dropped to $1, a nickel is paid out and $.95 would remain.

There are negatives. The higher the percentage payout, the less likely the payouts would keep up with inflation over the long term.  However, the primary issue that prevents people from using such an approach is the unpredictability of the amount available to spend. This approach can yield large "pay cuts" and raises even if a smoothing technique such as averaging the balances of three years to determine the payout is employed. Further, using a conservative portfolio virtually guarantees a diminishing cash flow stream due to the portfolio's likely struggle to earn more than the payout rate.

Probably the most common approach to retirement spending is the "as needed" method, the default method for most Americans. They add up their bills, add on a few discretionary expenses, and if their pension and Social Security payments don't cover these costs, they pull the difference from their assets. Most of the time, in two successive years, if there are no "off budget items" (a substantial "if" for some), the withdrawals will be nearly identical with no real thought about inflation. An increase occurs typically only when clients begin to feel the effects of rising costs.

If one models withdrawals that are increased for inflation only in every third year rather than in each year, the sustainability of those withdrawals will be enhanced. This is true of any spending pattern model that does not increase withdrawals for inflation in a steady manner.

Which leads me to an approach to retiree spending that seems to be gaining more widespread use in the financial planning community. I give credit for this to Jon Guyton and his work surrounding "decision rules." Guyton recognized that throughout a person's career they face both prosperous and challenging times in which their income varied accordingly. Doing similar in retirement might not be that hard.

In addition to some portfolio issues, Guyton specifically created rules for withdrawals.   Among them a 6% cap on increases and, in tough years for the portfolio, no change in cash flow would be instituted.  Theoretically, clients would be willing to accept constraints on changes if it meant a higher initial withdrawal rate.  The results showed a safe initial withdrawal rate closer to six percent vs. the more often cited 4% rate.

For the most part, I find people strongly attracted to a higher initial withdrawal rate for a couple of reasons. First is the tendency for many Americans to want it all and want it now. Second, juxtaposed with the fear of outliving one's assets, is a concern about never getting a chance to enjoy one's assets. Almost everyone approaching retirement tell me of things they want to do and places they want to see before their health makes such things impossible. Most people seem to believe that they will spend more earlier on in retirement than they will in their later years.

A few years ago, the U.S. Department of Labor, Bureau of Labor Statistics (BLS), released data indicating that the amount spent by persons over age 75 was dramatically lower than the amounts spent by younger retirees. This spawned a debate about why that was the case. One theory is that it simply reflects a generally higher level of health and vitality among younger retirees.  The more active, the more funds are needed.

Some believe that it is a matter of pent-up demand brewing within new retirees.  After working and not having a lot of free time, the new retiree is more likely to spend money to knock things off their bucket list, travel being possibly the most common manifestation of this phenomenon.

Another theory is that the lower spending is a generational issue.  Older retirees often have been impacted by vivid memories of the teachings about money they received from their parents who were adults in during the Great Depression.  Therefore, perhaps statistics like those from the BLS simply reflect a continuation of prior lower spending levels and not a decrease from younger years.