I've written about sequence-of-returns risk and the Retirement Red Zone and the 1966 retiree who did everything right and still ran out of money. The fix for all three of those problems is the same. It is structural. It is the three-bucket income floor โ the way of organizing retirement money so that you are never forced to sell shares in a down market to pay your bills. Today's article is the practical walkthrough. Let me show you how to actually build it. Three buckets. Plain English. Nothing exotic. Just discipline about which bucket pays the bills and which bucket sits and waits, right?
Start with the spending number, not the portfolio number
Before any bucket math, you need to know what you're solving for. Most retirees can produce a savings number off the top of their head โ "we have $1.2 million" โ but cannot produce a monthly spending number with the same confidence. Yet the spending number is the one that drives the bucket sizes.
Sit at the kitchen table for an hour with last year's bank statements and credit-card bills and produce two numbers:
- Essential monthly spending โ housing, food, utilities, insurance, healthcare, transportation, taxes. The stuff that doesn't pause if the market drops 25%.
- Discretionary monthly spending โ travel, hobbies, gifts, dining out, the things you choose. The stuff you could throttle back temporarily if you needed to.
For most pre-retirees in southeastern Massachusetts, essentials run somewhere between $4,000 and $7,000 a month. Discretionary spending varies wildly. Add the two together and you have your total spending target. Subtract Social Security, any pension income, any rental or part-time work income โ what's left is the gap your portfolio has to fill. That's the number that drives bucket sizing.
Bucket 1 โ Cash
Bucket 1 holds one to two years of total spending. For a couple spending $80,000 a year, that's $80,000 to $160,000.
What goes in it:
- High-yield savings account at an FDIC-insured bank or credit union. Spring 2026 rates: 4% to 4.5%.
- Money market mutual funds at brokerage accounts. Spring 2026 yields: 4% to 4.5%.
- Short-duration Treasury bills if you want explicit federal backing โ 4-week, 13-week, 26-week T-bills yielding similar rates with maturity dates aligned to your near-term spending.
What doesn't go in it: long-duration bonds, stocks, real estate, anything with a "term" longer than about a year.
The job of Bucket 1 is to be the checking account that pays the bills no matter what the market does. You spend from it monthly. You replenish it from Bucket 2 (or Bucket 3 in good market years) on a schedule, not in a panic. The peace-of-mind value of having 12 to 24 months of bills locked in cash is, in my experience, the single biggest improvement most retirees can make to their actual nervous-system response when markets get ugly.
Bucket 2 โ Income floor
Bucket 2 holds five to ten years of essential spending. For a couple with $5,000/month essentials, that's $300,000 to $600,000.
What goes in it:
- Bond ladders โ laddered Treasuries, laddered investment-grade corporates, laddered municipal bonds (especially in-state MA munis for the state-tax exemption).
- MYGAs โ multi-year guaranteed annuities, locking in 5%+ rates for 5 to 7 years.
- Sometimes a SPIA โ single premium immediate annuity, if you want lifetime income covering essential expenses.
- CDs โ laddered 3- to 5-year CDs from FDIC-insured banks.
- Conservative bond funds as a simpler alternative to individual bond ladders, with the trade-off that bond funds don't have a guaranteed maturity value.
What doesn't go in it: stocks, growth-oriented funds, anything that can drop 30% in a bear market.
The job of Bucket 2 is to be the tank that refills Bucket 1 in years when Bucket 3 is down. If the market drops 25% in year three of retirement, you do not sell from Bucket 3. You let Bucket 1 run down a bit (it has 12-24 months of buffer), and you use maturing rungs from Bucket 2 to refill Bucket 1. By the time you've drawn meaningfully from Bucket 2, it's usually been three to five years โ long enough that the market has typically recovered enough that you can refill Bucket 2 from Bucket 3 at decent prices.
The five-to-ten year sizing is calibrated to historical bear-market recovery times. The S&P 500 has historically recovered to its prior peak within four to seven years even after the worst bear markets, accounting for dividends. A ten-year Bucket 2 covers virtually any historical bear-market recovery. A five-year Bucket 2 covers the typical case but can feel tight in a 1966-style stagflation decade.
Bucket 3 โ Growth
Bucket 3 holds everything else โ usually the largest bucket, and the one with the longest time horizon.
What goes in it:
- U.S. equity index funds or ETFs โ broad market exposure, low cost, tax-efficient.
- International equity funds โ for global diversification.
- Some allocation to longer-duration bonds if you want to dampen volatility within Bucket 3.
- Real estate funds (REITs) if you want diversified real-estate exposure beyond your home.
The asset allocation within Bucket 3 looks like a typical "growth" portfolio โ 60/40 stock/bond, 70/30, 80/20 depending on age, risk tolerance, and how much of the total portfolio is already in Buckets 1 and 2 in conservative form. A retiree whose Bucket 1 + 2 already represents 30% of their total savings can afford to be more aggressive in Bucket 3 than a retiree whose Bucket 1 + 2 is only 15%.
Bucket 3's job is to grow over decades and replenish Bucket 2 in good market years. It is not the bucket that pays this month's bills. It is not the bucket you watch on a daily basis. It is the long-horizon engine.
The refill rules
Here's where the structure earns its keep. The refill rules are what turn three piles of money into a working system:
Every month: spend from Bucket 1.
In a good market year (broad equities up): refill Bucket 1 from Bucket 3 at end of year. Refill Bucket 2 if Bucket 2 is below target.
In a flat or modestly down year: refill Bucket 1 from Bucket 2. Don't touch Bucket 3.
In a deep bear market year (broad equities down 15%+): refill Bucket 1 from Bucket 2 if needed; spend reserves down. Don't touch Bucket 3 unless Buckets 1 and 2 are exhausted.
When markets recover: refill Bucket 2 from Bucket 3 first, then Bucket 1.
The discipline of these rules is what makes the structure work. The temptation in any down market is to "do something" โ sell stocks to lock in losses, panic-rebalance, abandon the plan. The bucket structure gives you permission to do nothing. Bucket 1 covers this month's bills. Bucket 2 covers the next several years. Bucket 3 sits and waits for the recovery. You don't have to make any decisions during the downturn โ you just spend from Bucket 1, occasionally refilling from Bucket 2, and let Bucket 3 do its long-horizon job.
An example: $1.2M total, $80K spending
Let me put a real number on this. Couple with $1.2M total in retirement assets, spending $80,000/year ($55,000 essentials, $25,000 discretionary), receiving $40,000 in Social Security:
- Annual gap to fill from portfolio: $40,000 ($80K spending - $40K SS)
- Bucket 1 (cash, 1.5 years of spending): $120,000 in high-yield savings + T-bills
- Bucket 2 (income floor, 7 years of essentials): $385,000 in bond ladder + MYGAs
- Bucket 3 (growth): $695,000 in equities + diversifiers
Total: $1.2M, divided 10% / 32% / 58%. The "growth" allocation is more conservative than a typical 60/40 portfolio because the Bucket 1 and Bucket 2 dollars are doing the conservative work. Bucket 3 can stay aggressive because it's protected by the income floor underneath it.
Common mistakes
- Building Bucket 1 too small. Six months of expenses is not enough for retirement. Twelve to twenty-four months is the right range.
- Making Bucket 2 too risky. Putting Bucket 2 in long-duration corporate bond funds or anything with equity-like volatility defeats the purpose. Bucket 2 should drop less than 5% in a really bad year.
- Refilling Bucket 1 from Bucket 3 in a bad year. The whole point of the structure is that you don't sell Bucket 3 at the bottom. Discipline matters.
- Not refilling Bucket 2 in good years. If Bucket 2 has been drawn down by three years of bills, it needs to be refilled when the market is up โ not allowed to stay short.
- Treating the buckets as physically separate accounts when they don't need to be. The buckets are an organizational concept. They can live in the same account custodian, the same brokerage, the same IRA โ what matters is that you tag and track them mentally and operationally.
What this looks like in practice
The three-bucket structure is not exotic. It is not proprietary. It is not exclusive to one advisor or one product line. It is a well-documented framework that goes back decades in retirement income research, refined by Wade Pfau, Bill Bengen, Harold Evensky, and others. What it does is impose a discipline โ a set of rules for what to spend from when โ that makes sequence-of-returns risk survivable.
The Sherpa frame I use at every Retirement 101 seminar applies here exactly. The danger isn't on the way up the mountain. It's on the way down, when you're spending what you saved. The three-bucket structure is the rope-and-harness for the descent. Built right, it lets you stop watching the market on bad days and trust the structure to do its job. Sleep at night, knowing the bills are paid no matter what the S&P does this month.
Walk through the bucket structure in a room of pre-retirees
The Retirement 101 seminar covers six modules โ health and long-term care, estate, investments, Social Security, income strategy, and taxes โ including the three-bucket framework. Free, ninety minutes, plain English, no products pitched in the room.
The four outcomes:
- I never see you again. We wave at Home Depot.
- You take what you learned to your existing advisor. Great.
- You do nothing. The one I hate the most.
- We're a fit and we work together.
The bottom line
Three buckets โ cash for 12 to 24 months, income floor for 5 to 10 years, growth for the long run. Spend from cash. Refill cash from the income floor when needed. Refill the income floor from growth when markets cooperate. Never sell from growth at the bottom of a bear market. The structure is simple. The discipline is what makes it work. Sequence-of-returns risk doesn't disappear, but it stops being able to ruin a thirty-year plan. That's the whole point.
The example numbers used are illustrative. This article is general educational information and is not investment advice or a recommendation of any specific allocation, withdrawal rate, or product. Bucket sizing should be calibrated to your specific spending, income sources, asset balance, and risk tolerance.