Decumulation Ch. 3: The Bucket Structure — Building the Buffer

阅读中文版

How many years of cash, where the buffer actually comes from, how buckets refill, and an honest account of what the structure does and does not do.

🔊 Listen to Article (Chinese Audio)

Retirement Decumulation Mechanics — Chapter 3: The Bucket Structure — Building the Buffer

"The buffer's function is not higher returns. It is the ability to follow your rules precisely when following them matters most."

The Problem This Chapter Solves

Chapter 2 showed that sequence risk does its damage by forcing equity sales during declines. So the most direct defense is ensuring that declines never force you to sell equities.

That is the entire logic of the bucket strategy. It is not an investment technique but a way of organizing assets by when you will need them.

The Three Layers

Bucket Horizon Holdings Purpose
One 1–2 years of spending Cash, money market, T-bills Immediately available, no price risk
Two 3–10 years Short/intermediate bonds, Treasury ladder Resupply when bucket one runs low
Three Beyond 10 years Broad equity index Growth and inflation protection

Bucket three is the point: its whole purpose is that you will not need to touch it for a decade. With the first two in place, a three-year bear market forces no equity sales at all — you live from bucket one and let bucket three recover.

How Many Years of Buffer

Split annual spending into two categories first:

Category Examples Property
Essential Housing, food, health insurance, utilities Must be paid regardless of markets
Discretionary Travel, dining, gifts, replacing a car Can be deferred or reduced

The buffer only needs to cover essentials minus fixed income:

Buffer = (annual essential spending − Social Security and other fixed income) × years covered

Example: $50,000 essential spending, $30,000 Social Security — a $20,000 gap. Covering three years requires $60,000, not the $150,000 many people assume (3 years × total spending).

This distinction typically halves the buffer requirement, and oversized cash has a real cost: it loses to inflation over time. This is the most practically valuable calculation in the chapter.

Where the Buffer Comes From

A common error is building the buffer on the day you retire — which may mean liquidating a large equity position at a market high or low.

Build it gradually over the 3–5 years before retirement:

  1. From five years out, direct new savings into buckets one and two rather than more equities.
  2. Use rising markets to trim equities into the buffer, rather than selling reactively later.
  3. By retirement day, the first two buckets should already be funded.

Note how step 2 echoes Chapter 2: topping up the buffer during rallies inverts sequence risk in your favor — shares sold at a high consume a smaller share of the future compounding base.

How Buckets Refill

Situation Action
Equities above target (say 65% vs. a 60% target) Sell the excess to refill buckets one and two
Equities within the target band Refill from dividends and interest; sell no principal
Equities below target (bear market) Do not refill. Keep drawing from bucket one and let equities recover
Bucket one nearly empty and the bear market persists Refill bucket one from bucket two; still do not touch equities

The third row is the core of the structure. Not refilling during a bear market is the entire reason the buffer exists — refilling from equities in a downturn converts the bucket strategy into an elaborate procedure for selling at lows.

Honestly: What It Does and Does Not Do

There is a legitimate academic criticism of bucket strategies, and this book should relay it rather than dodge it.

The criticism: mathematically, a bucket portfolio and a single portfolio with the same overall allocation and proportional withdrawals produce nearly identical long-run results. Labeling assets "bucket one, two, three" does not change your overall stock/bond ratio, and the overall allocation is what drives returns. In that sense buckets are mental accounting — the bias our Misbehaving series examines.

The criticism is mathematically correct. It misses one thing: buckets solve a behavioral problem, not a mathematical one.

What the structure actually provides: when markets fall 40%, you know the next three years of living expenses already sit in an account that did not fall. That knowledge is what stops you selling equities. And not selling equities is precisely the variable Chapter 2 proved to matter.

So the honest conclusion is: the bucket strategy's value is not that it optimizes the math but that it makes you more likely to execute the behavior the math requires. If you are genuinely certain you would not panic at −40%, proportional withdrawal with annual rebalancing works just as well and is simpler. Most people are not certain.

Procedure

  1. Split spending into essential and discretionary; compute the buffer only against the gap.
  2. Cover 2–3 years of that gap. Beyond five years, cash drag becomes significant.
  3. Begin building 3–5 years before retirement, using up years — not all at once on retirement day.
  4. Write down the refill rules, especially "do not refill during a bear market," into your annual checklist.
  5. Do not treat buckets as three separate portfolios. Your overall stock/bond ratio must still match the plan; buckets are only its organization.
  6. Recheck annually that the buffer still covers 2–3 years of the gap as spending and benefits change.

Relevance to a Retirement Portfolio

Our Defensive Investor's Operating Manual Chapter 4 argues that a retiree's "unbeatable position" is a structure rather than a stop-loss; this chapter is that structure's concrete form. And our Art of War for Trading Chapter 6 identifies the individual's greatest edge as the freedom not to act — the buffer is the mechanism that makes that freedom real. A retiree without one must sell into a bear market; a retiree with three years of buffer can choose not to.

That is one principle stated three ways across three books: the strongest position is being able to choose not to act while the market cannot compel you.