Bucket strategies and cash reserves: two- and three-bucket setups, refilling rules and the evidence versus plain rebalancing

What you will learn

• How a two-bucket and a three-bucket structure are sized in years of spending.
• Which refilling rules exist and why the refilling rule, not the buckets, decides the outcome.
• What a bucket plan is in asset-allocation terms and how it behaves in a crash compared with rebalancing.
• What the research (Estrada, 2019) found when buckets were tested against static allocations.

Explanation

A bucket strategy divides the portfolio by time horizon instead of by asset class. The idea is to never be forced to sell equities in a downturn (lesson 3), because near-term spending sits in cash. Rebalancing with bands was covered in the courses ETFs, Funds & Long-Term Investing and Portfolio Construction; here we compare the two approaches in the withdrawal phase.

Two-bucket setup. Bucket 1 holds one to three years of planned withdrawals in cash or a money-market fund. Bucket 2 holds the rest in the long-term mix. You spend from bucket 1 and refill it from bucket 2.

Three-bucket setup. Bucket 1: one to two years in cash. Bucket 2: five to eight years in short- and medium-term bonds or bond ETFs. Bucket 3: everything else in global equities (and possibly property). Spending flows 3 → 2 → 1 over time.

Refilling rules are where plans differ, and they matter more than the number of buckets:

• Calendar refill: once a year, sell from bucket 3 (or 2) to top bucket 1 back to its target, regardless of markets. This is mathematically almost identical to a rebalanced portfolio with a fixed cash share.
• Conditional refill: refill from equities only after a positive year; after a negative year, refill from bonds or let cash run down. Cash can be depleted after two or three bad years, so the rule must say what happens then.
• Opportunistic refill: sell equities whenever they exceed their target weight by a band, and refill cash with the proceeds.

Translate every bucket plan into an effective asset allocation. Two years of a 4 percent withdrawal is 8 percent cash; six years of bonds is 24 percent; the rest, 68 percent, is equities. A bucket plan is therefore an 8/24/68 portfolio with a particular rebalancing rule. Seen this way, two things become visible. First, the cash share is a drag on expected return in normal years. Second, conditional refilling moves money only in one direction, from risky to safe, and never buys equities after a fall, which is what rebalancing does.

Evidence. Javier Estrada (2019, "The Bucket Approach for Retirement: A Suboptimal Behavioral Trick?") tested a two-bucket plan with two years of withdrawals in bills against eleven static stock/bill allocations on US data from 1900 to 2014, using three different refilling rules. Every static allocation with at least 40 percent stocks had an equal or lower failure rate than all three bucket versions; the 60/40 static mix did not fail in any 30-year period, while the bucket rules failed in roughly 3.5 to 4.7 percent of periods. His explanation: buckets avoid selling low, but they also forgo buying low, which static rebalancing captures. Other researchers reach similar conclusions. The honest summary is that buckets are mainly a behavioural tool. If a visible cash reserve keeps you from panic-selling the equity bucket in a crash, it can be worth a small expected cost; it is not a source of extra return.

Rules of thumb: keep bucket 1 at one to two years, not more; prefer calendar or band-based refilling so that equities are also bought after falls; count the cash as part of your allocation when you compute your equity share; hold bucket 1 at a bank or in a money-market fund covered by investor protection rules, not in the brokerage cash account alone.

Worked example

Hypothetical example: €900,000, spending €36,000 per year (4 percent). Three buckets: bucket 1 = 2 × €36,000 = €72,000 cash (8 percent); bucket 2 = 6 × €36,000 = €216,000 bonds (24 percent); bucket 3 = €612,000 equities (68 percent). Year one: equities −30 percent, bonds and cash +2 percent. Spending is taken from cash at the start of the year. Fees and taxes excluded.

Bucket plan, conditional refill (no equity sales after a loss): cash (72,000 − 36,000) × 1.02 = €36,720; bonds 216,000 × 1.02 = €220,320; equities 612,000 × 0.70 = €428,400. Total €685,440. Refill cash from bonds by €36,000: cash €72,720, bonds €184,320, equities €428,400. Equity share now 428,400 / 685,440 = 62.5 percent. Nothing was bought at low prices.

Static 68/32 mix with pro-rata withdrawal and rebalancing: equities (612,000 − 24,480) × 0.70 = €411,264; defensive part (288,000 − 11,520) × 1.02 = €282,010. Total €693,274. Rebalancing back to 68 percent means buying 0.68 × 693,274 − 411,264 = €60,162 of equities at the lower price.

After year one the totals differ by only €7,834, but the positions differ a lot: the rebalanced investor holds 68 percent equities bought partly at −30 percent; the bucket investor holds 62.5 percent and has spent his buying power. If equities recover 20 percent in year two, the rebalanced portfolio gains about 0.68 × 20 = 13.6 percent on equities against 12.5 percent for the bucket plan, before new withdrawals. Over several cycles this gap is what Estrada measured. The bucket investor's advantage was psychological: he never sold a single equity unit during the fall.

Common mistake

Practice activity

On paper: Sofia has €1,200,000 and spends €42,000 a year. She wants bucket 1 = 18 months of spending, bucket 2 = 5 years in bonds, bucket 3 = equities. (a) Size the three buckets in euros and percent. (b) In a year in which equities fall 20 percent and the other buckets are flat, she spends €42,000 from cash and refills cash from bonds. What is the equity share afterwards? (c) How much would a rebalance back to her original equity share require her to buy?

Key takeaways

• A bucket plan is an asset allocation plus a refilling rule; always translate it into percentages.
• Conditional refilling only moves money from risky to safe and never buys equities cheaply; that is why static rebalancing historically beat it.
• Buckets are a behavioural tool against panic-selling, worth a small cost only if they change your behaviour.
• Keep cash to one or two years and use calendar or band-based refilling.

Next step

Lesson 6 moves from strategy to execution in Germany: Auszahlplan products versus selling ETF units yourself, the order of sales and the FIFO effect on taxable gains.