How well do our models fit actual behavior in retirement?

I’ve seen a lot of new research and thinking recently that, taken together, threatens to upend everything we think we know about retirement distribution planning. I think this may be important as more clients march into retirement under the guidance of their financial planner.

At the AICPA Personal Financial Planning conference in January, Jim Shambo, president of Lifetime Planning Concepts of Colorado Springs, Colo. – an out-of-the-box thinker if ever there was one – asked a deceptively simple question. Our Monte Carlo engines assume that clients will spend a certain percentage of
their total retirement portfolio in the first year (4% is the safe harbor amount), and then require that same dollar amount, going forward, for the rest of their days, adjusted for inflation. Analyses by Jon Guyton, a principal of Cornerstone Wealth Advisors in Minneapolis, and others show that we can adjust that safe harbor upward a bit if clients are willing to freeze their spending in down markets and limit their raises.

But what Shambo wondered is this: Is the constant-dollar spending goal an accurate or reasonable assumption?

UNEQUAL COSTS

To find out, he looked at inflation calculations by the Bureau of Labor Statistics and found something interesting: Inflation tends to strike retirees harder than preretirees. Most notably, health care costs are rising faster than the inflation rate.

Beyond that, the CPI calculation factors out cost increases that are attributable to improvements in the goods and services you purchase. A car may cost 4% more
this year than last, but if there are new fancy electronics in the standard model, the government may decide that inflation only counts for a half-percent of the increase. Of course, if you buy the car, you still have to pay the full higher cost. Add it all up, and people aged 65 to 74 appear to be experiencing an inflation rate that is a remarkable 1.11 percentage points a year higher than CPI, and this grows to 2.09 percentage points (a year!) when retirees get past age 75.

MAKING A MODEL

A sophisticated retirement spending model, created by California Lutheran University Professor Somnath Basu, combines Shambo’s revised inflation estimates with these revised expenditure figures. Each line item gets its own estimate of higher or lower consumption, and that’s multiplied by an individualized inflation factor.

Health care spending will tend to go up as clients age, and medical costs rise faster than CPI. Food expenditures tend to go down as people age and
food inflation is typically lower than the average, and so forth.

I asked the audience of my Inside Information newsletter what they thought of all this, what their experience has been with retired clients, and how (or whether) this more sophisticated retirement spending model would change their advice to clients. The most common response is that generalizations have to be pushed aside when you’re dealing with an individual client.

When David Jacobs, who practices in Honolulu, estimates a client’s retirement spending, he takes into account such driving factors as the size of the house and maintenance expenses, a habit of providing significant financial assistance to children, a dependent parent, and/or costly activities on the bucket list. This latter line item gets particular attention from Tom Murphy of Temaa Financial, who practices in Dallas. After studying the behavior of his clients closely, he has started planning for a spending blip in the first two years of retirement. That, he says,
is when clients take full advantage of their newfound freedom.

Elyse Foster, of Harbor Financial Group in Boulder, Colo., builds a larger portfolio buffer for what she calls “spendy” clients. Meanwhile, both Neal Van Zutphen of Delta Ventures Financial Counsel in Phoenix and Cindi Conger of Conger Wealth Management in Little Rock, Ark., are pioneering a detailed budget planning service for their clients.

Each expenditure is assigned its own line item and there are projections both for spending increases or decreases and individualized inflation rates. Clients get in the terrific habit of tracking their expenses against the projections before they enter retirement, and are far less likely to get off-track when their paycheck runs out.

At the other end of the spectrum, Pat Raskob, who practices at Raskob Kambourian Financial Advisors in Tucson, Ariz., finds that some retirees who were champion savers in preretirement are unaccustomed to the frivolity of leisure activities. She has to coax
them along, brainstorming fun things for them to do, helping them purchase trips and plan activities with their children as well as their grandchildren.

This suggests new services for planners to offer their clients. Raskob has been called a memory creator because she helps her clients plan activities and family events that they can enjoy recalling over and over again. They need the advisor to help them fully enjoy retirement without the guilt or fear associated with spending.

Others, who have trouble understanding why they can’t safely spend a bigger chunk of their multimillion dollar retirement portfolio, will need advisors like Van Zutphen and Conger to help them track their budgets. Tools like Mint.com are making these services much easier to provide.

A DIFFERENT APPROACH

Some advisors willing to brainstorm see an easier way to manage the running-out-of-money risks. Vince Schiavi of Schiavi + Dattani in Wilmington, Del., pays particular
attention to a client’s core expenses – those needed for survival – and then trusts clients to adjust their discretionary spending as market circumstances dictate. Kevin Kroskey of True Wealth Design, who practices in Akron, Ohio, uses Mint to identify those core expenses, then uses his own system to estimate how those life-sustaining expenditures will rise over time. He also looks for ways to reduce unpleasant surprises. Getting all debts paid off before retirement, and having both good long-term care and Medigap policies in place are all ways to narrow the range of possibilities.

We still don’t know if data from the Bureau of Labor Statistics’ Consumer Expenditure Survey data – which covers the spectrum from the impoverished to the ultrawealthy – is a good measure of how financial planning clients will behave as they age in retirement. But there is clear anecdotal evidence that it might be reasonably accurate.

According to Conger’s voluminous spreadsheets, clients spend more on travel and
hobbies in early retirement, and less on clothing and food over time. She reports that, somewhere around age 75, people start spending less on groceries.

THE ANNUITY ARGUMENT

Why does any of this matter? Because some of this analysis rebuts the key argument against recommending immediate annuities to clients in retirement.

Most annuities don’t adjust their payments for inflation. But if these core life-sustaining expenditures tend to go down as clients age, rather than rise at or (per Shambo) above the inflation rate, then this is not a problem – and might, in fact, be the simplest way to give clients peace of mind that they won’t be eating dog food in their later years.

Also, if the profession is overestimating those core expenditures in retirement habitually, particularly after age 75, then the standard pre-retirement (Monte Carlo) calculations may be overestimating how much a client needs to accumulate by as much as 25% – a substantial amount.
Once we see a flood of new thinking about an aspect of planning, it’s only a matter of time before we see a major shift in the way the service is provided.