Bucketing doesn't work

CIO Global Blog

21 Feb 2019

Here's a short list of things that don't work if you do them incorrectly:

  1. Modern portfolio theory
  2. Tactical asset allocation
  3. Strategic asset allocation
  4. Active management
  5. Passive management
  6. Fundamental analysis
  7. Technical analysis
  8. Economic forecasting
  9. Bucketing
  10. Liability driven investing
  11. Smart beta
  12. Rebalancing

I could go on, and frankly there are a few things on that list that I don't think ever work, but that's not the point of this blog post. The reason I bring this up is a recent paper by Javier Estrada titled "The Bucket Approach for Retirement: A Suboptimal Behavioral Trick?"He posted the paper on SSRN in December, but Barron's just picked it up so I feel like I have to respond to it at this point. If nothing else, my inbox could use a break.

Here's how Estrada describes bucketing:

"The three bucket strategies consist of two buckets, one with funds parked in bills (Bucket 1) and the other with funds invested in stocks (Bucket 2). All three strategies set aside two years worth of inflation-adjusted withdrawals and park them in bills, investing the rest in stocks. Also, all three strategies determine whether to make the annual withdrawal from Bucket 1 or Bucket 2, and whether to refill Bucket 1 when it has less than two years of real withdrawals, depending on the performance of the stock market; the specific rules are outlined in Exhibit 1."

As much as I don't want to believe it, there probably are some investors out there using such a framework. But I don't need 18 pages of math to know that the strategies he describes will end poorly. I see a lot of other bucketing strategies that are more along the lines of "bucket 1 is 5 years of cash, bucket 2 is for years 6-10 and is mainly bonds, bucket 3 is years 11-20 and is mainly equity, etc." Details are always slightly different, and those strategies won't destroy capital quite as fast as Estrada's strategy, but they are certainly suboptimal too.

More importantly, Estrada correctly points out that not many academics have looked at bucketing. Frankly we haven't either. Our Liquidity. Longevity. Legacy. framework isn’t bucketing for the sake of bucketing. We focused on creating a goals-based investment framework, and determined that liability- driven investing, which has deep academic roots, is the most effective way to do it. A behavioral finance overlay meant that we ended up with a 3- part segmentation strategy, but there's 30 years of academic work and institutional success behind the 3Ls that we covered in our white papers.

All that being said, confusion around this Estrada's paper and how it relates to the 3Ls has helped me realize a few things:

  1. We can go too far in our desire to simplify. I appreciate that investors gravitate to the intuitiveness of the 3L segmentation, but that intuitiveness is a gateway for us to be able to add a technical layer at the portfolio management level that improves their outcomes.
  2. Details matter. We need to refocus on some of those details to clearly differentiate the 3L strategy from some of the less-effective ideas out there, and be willing to discuss the problems with other strategies in order to compare and contrast some of those details. The Estrada paper is a great start.
  3. The term "bucketing," which we've generally shied away from, needs to be stricken from our vocabulary.

Liquidity. Longevity. Legacy. disclaimers: Timeframes may vary. Strategies are subject to individual client goals, objectives and suitability. This approach is not a promise or guarantee that wealth, or any financial results, can or will be achieved.

Author: Michael Crook, Head Americas Investment Strategy, UBS Financial Services Inc. (UBS FS)