Leaks in the Bucket Method

After reading Wade Pfau's new article on bucketing, I concluded that bucketing has three flaws as a retirement income generator. It calls for equities in the last bucket, not longevity annuities; timing risk makes it labor-intensive; and it's not for everybody.

As a tool for retirement income planning, time segmentation offers something for almost everyone. Combining elements of behavioral finance with asset-liability matching and a dash of equity premium, time segmentation—aka bucketing—is broad enough and flexible enough to satisfy the needs of many different types of advisors and clients. It’s very popular.

If you’re not familiar with bucketing, it’s the practice of dividing an individual’s or couple’s retirement years into as many as six discreet time segments and designating a specific bucket of assets—an annuity, a bond ladder, an equity portfolio—as the current or future source of income in that segment. It breaks a big problem into manageable tranches.

Time segmentation has critics (mostly in academia). They’ve belittled it as a form of mental accounting that offers at best a thin sense of financial security, questioned its assumption that stocks pay off “in the long run,” and warned that transitions between buckets will be fraught with timing risk and sequence risk. The jury is still out.

The judge, however, is in. Wade Pfau, Ph.D., the eminent economist, professor and retirement specialist at The American College, examines the practice of bucketing in a new three-part series in the online publication AdvisorPerspectives.com. In the first article, which appeared this week, he puts bucketing into historical perspective and reviews its strengths and limitations. In future articles, he intends to dig deeper.

Hazards of bucketing

The concept of bucketing has always appealed to me. When I write articles, I use a five-bucket outlining method. One of the first people I met in the retirement business was David Macchia,whose Income for Life Model began as a bucketing solution but has since moved passed bucketing to embrace a hybrid solution that includes lifetime guaranted income. I entered this field via the annuity world, and annuities are easy to plug into a time-segmentation model. So RIJ is a bucket-friendly zone.

Pfau’s article raised my awareness of the hidden wrinkles of bucketing. He offers fresh analysis of the conflicts or tensions that bucketing might pose for advisors. For instance, he points out that bucketing can frustrate advisors who want to maintain a specific asset allocation and risk exposure. He acknowledges the sequence risk that can occur during transitions between buckets. After reading his article, I came to a few new conclusions about bucketing: 

Reserves shouldn’t be risky, so using stocks for the final bucket may not be wise. A deferred income annuity strikes me as a more appropriate income provider in the last bucket than stocks. That runs counter to bucketing orthodoxy, but the more I think about it, the more sense it makes.

In standard bucketing, you put equities into the final bucket, the one that you tap last, or whose appreciation you occasionally take off the table and put into an earlier bucket. Equities certainly belong in a retirement portfolio, to protect against inflation or provide a bequest whose basis steps up. They satisfy a universal yen for upside; retirees don’t like to feel that they’re making a “dead-stick landing” to eternity.

But if a retiree wants to reduce longevity risk with the final bucket, equities don’t fill the bill. They carry market risk, which could potentially make longevity risk worse instead of making it go away. Pfau observes that people who have not saved enough may want to hold equities as a long-term catch-up strategy. That’s risky. Banks don’t hold equities as their reserves; why should retirees hold equities as their reserves?

For retirees who have under-saved, I think it makes more sense to rely on the mortality credits of a DIA (or, for tax-deferred savings, a qualified longevity annuity contract) rather on than the uncertain equity premium when assigning assets to the final bucket.  

Retirees want simplicity, so advisors shouldn’t make their finances complicated.

Pfau’s article leaves the impression that the bucket method, if practiced with appropriate zeal, is likely to be labor-intensive for the advisor and perhaps fee-intensive for the client.

For instance, the more buckets there are in the plan, the more transitions there are between buckets. A transition might involve the exhaustion of one bucket and the liquidation of the next. Or it might involve the periodic refilling of a near-term bucket via sales of bonds or equities. Either way, transitions imply timing risk. To ensure that assets don’t need to be sold at unfavorable times or at unfavorable prices, advisors will have to monitor the portfolio closely.

A floor-and-upside income generation method seems more manageable. As flooring material, income-producing annuities would be easier than bond ladders. Manufacturing and distribution costs raise the price of income annuities by an estimated 15% above actuarial value, but they offer simplicity and peace of mind.  

Bucketing isn’t necessarily for everyone, and advisors should assess their clients’ financial situations before choosing an income generation method. Academic comparisons between bucketing and total return investing or a floor-and-upside approach imply that the method is the priority, and that the best method is the one that produces the highest return, income rate, or final balance, according to Monte Carlo simulations. I would argue that that’s not necessarily client-centric.

The most logical first step in retirement planning might be to determine the client’s funding level. Canadian advisor Jim Otar assigns clients to a “green,” “orange” or “red” zone. Clients in the green zone have lots of money relative to their income needs; they can choose almost any income style they want. A total return method would be simpler than bucketing. Tax-efficient distributions may be their highest priority. 

Clients in the red zone by definition have too little money relative to their income needs. Dangerously vulnerable to sequence risk, market risk or longevity risk, Otar wrote, red zoners typically need to transfer one or more of those risks to an insurance company through the purchase of annuities.

Clients in the orange zone pose the most interesting challenges for retirement income specialists. Orange zoners usually have a limited risk budget, which they can spend on a variety of combinations of risky and risk-free assets. Bucketing might suit them well. Bear in mind that clients can move to a better-funded zone simply by reducing their expenses. Bottom line: Clients’ funding levels help determine their income generation methods. 

Stay tuned

Pfau frames bucketing as a compromise between the probabilistic (risky) total return approach and the deterministic (low-risk) floor-and-upside approach. He promises in the second part of his series (published today) to examine a hypothetical time-segmentation strategy that combines a bond ladder (using techniques described by Brent Burns and Stephen Huxley in their 2004 book, Asset Dedication) with equity investments. “Part 2 will formalize three different rules for how to implement time segmentation in practice,” Pfau wrote. “Part 3 will then compare time segmentation approaches to total-return investing strategies to determine whether it is a superior investing strategy.” I look forward to reading both.

© 2017 RIJ Publishing LLC. All rights reserved.