Two years of spending is $120,000. That is the whole cash bucket, and in most of the plans you have read it is also a countdown.
You already have bucket 1 funded in cash and bucket 2 funded in bonds, and the plan you followed says spend from the cash first. What it never says is when the growth side pays the cash side back. So the buffer quietly becomes a countdown, and what is missing is not your discipline. It is a number.
Size each bucket in years of spending instead of percentages, then write two rules down: refill bucket 1 whenever it drops under one year of spending, and skim any year bucket 3 finishes more than 20% above where it started. A refill trigger that fires against $60,000 does not move when the portfolio does, and that is the entire point of writing it in dollars.
By the end you will have the $120,000, $180,000 and $700,000 sizing to copy, the trigger and the skim written as rules with amounts and dates, and the two situations where this should not be built at all. The figures here are illustrative, not a promise of returns.
You are not missing willpower. You are missing a return line.
| Bucket | Job | Size in this example | What it holds | What triggers a move |
|---|---|---|---|---|
| 1 — Cash | Pay this year's and next year's bills | $120,000 (two years) | Treasury bills, money market fund, high-yield savings | Refill when it drops below $60,000 (one year) |
| 2 — Bonds | Cover years three through five and refill bucket 1 | $180,000 (three years) | Short- and intermediate-term bond funds, CD ladder | Skim from bucket 3 when stocks finish a year up more than 20% |
| 3 — Stocks | Growth for years six and beyond | $700,000 (no spending date) | Total US market index fund, total international index fund | Sell only when bucket 2 drops below $60,000 |
The Standard Version Only Drains
Most three-bucket plans only tell you which bucket to spend from next, and that is exactly why the cash keeps running out. You spend bucket 1 first, then bucket 2, then bucket 3. Somewhere in all those pages there is no sentence that says when money moves back the other way.
Run the example household through it. A $1,000,000 portfolio covering $60,000 a year, with a cash safety net of $120,000 sitting in bucket 1. Two bad years at $60,000 each spend that $120,000 down to zero, and now the plan has one instruction for you: sell from the next bucket. If bonds are down too, the next sale reaches stocks while stocks are down. That is not a strategy making a decision. That is whoever is holding the phone, in the worst month of the decade, letting the market set the price.
A rule works the other way around. You decide the move in a calm month, write it down, and the market only supplies the date it fires. That is the fix the rest of this piece builds: a written trigger for refilling bucket 1 out of bucket 2, and a written rule for skimming gains out of bucket 3 into bucket 2 before the bad years show up.
Here is the part worth sitting with. A cash bucket with no inflow rule is not two years of spending. It is 24 months on a clock, and you already know what happens when a clock runs out.
Size Each Bucket in Years, Not Percentages
A refill rule is only as good as the number it fires against, and a percentage of a moving portfolio gives it nothing steady to read. So fix each bucket at a count of years, then let the dollar figure follow.
On a $1,000,000 portfolio that has to produce $60,000 a year, bucket 1 holds two years, or $120,000. Bucket 2 holds three years, or $180,000. Bucket 3 is whatever is left: $700,000, with no spending date attached to it.
Compare that with the familiar 10/30/60 split. On the same million it hands you $100,000 in cash, which covers 20 months of $60,000 spending, not 24. After a bad year shrinks the portfolio, 10% buys even fewer months, while two years is two years no matter what the market did in March.
That steadiness is the whole point. Percentages drift all year and never tell you anything is wrong. The fixed pair, $120,000 and $180,000, is what the trigger and the skim read against whether the market is up, down or flat.
This does not fit every household. If a pension or Social Security already covers the bills, bucket 1 should be smaller, because the portfolio is not paying those months and two full years would just sit there losing ground to inflation.
The unit matters too: it moves from share of portfolio to months of spending. Buckets 1 and 2 together hold $300,000, and that is five years of $60,000 checks, a number you can feel.
The numbers this plan fires on
A $120,000 cash bucket, a $180,000 bond bucket, and three rules
Illustrative example. Returns are not guaranteed and your own spending figure is the one that matters.
What Actually Goes Inside Each Bucket
Each bucket takes only what will be spent or sold inside its own span of years, and nothing beyond that gets in.
Bucket 1 is $120,000 due to be spent within 24 months, so it holds Treasury bills, a money market fund, and a high-yield savings account. Nothing in there is allowed to fall in price. When the rent comes out in month 19, that $120,000 has to still be $120,000.
Bucket 2 is $180,000 spent across years three, four and five. That means short- and intermediate-term bond funds plus a CD ladder with a rung maturing in each of those years. The maturities matter because a refill of bucket 1 then never forces you to sell a bond fund at a bad price.
Bucket 3 holds the $700,000 with no spending date: a total US market index fund and a total international index fund, bought and then left alone. Its job is to grow, not to be raided on a bad afternoon.
Keep out of bucket 1 anything you might have to sell at a loss to raise cash, long-term bonds and dividend stocks included. And remember all three buckets sit inside one IRA plus one taxable brokerage, so the identical fund can act differently at tax time depending on which account holds it.
Bucket 1: sizing by years of spending, not by list
Refill and skim rules that touch bucket 1
- Refill when bucket 1 falls below $60,000 (one year of spending): sell $60,000 from bucket 2 to bring it back to $120,000
- Bucket 3 is sold only if bucket 2 drops below $60,000, so buckets 1 and 2 together are a five-year, $300,000 buffer
- If bucket 3 finishes more than 20% above its starting value in a calendar year, move the excess above that 20% gain into bucket 2
- Reset all three buckets in the first week of January
Skip this structure if
- Your spending is not roughly $60,000 a year or your portfolio is not roughly $1,000,000 — recalculate years of spending first
- You cannot hold a full $300,000 in buckets 1 and 2 without capping bucket 3 below $700,000, which changes the skim math
- You have no taxable brokerage alongside the IRA, so the tax location of each holding needs its own review before you copy these amounts
- You cannot commit to a written refill trigger, which is the actual mechanism that makes the cash bucket refill itself
The Refill Trigger and the Order of Sales
The number that fires a sale is one year of spending: once bucket 1 slips under $60,000, you sell from bucket 2 until it is back to $120,000. Say the account shows $58,000 on the first of the month. You sell $62,000 out of the bond funds and the CD rung that is maturing anyway, and that is the whole transaction.
The order comes next, and it is the part that keeps a bad year from reaching your stocks. Bucket 1 fills from bucket 2 every time, on any month the trigger trips. Bucket 2 only takes money from bucket 3 in a year the market finished up. And bucket 3 is sold only if bucket 2 itself has fallen below $60,000.
That sequence is what turns the first two buckets into a five-year, $300,000 buffer. Stocks stay untouched until five full years of spending have gone out the door. Two bad years is not a reason to sell a single share.
Write the rule down with a date and a dollar figure attached, in the same place you keep the account logins: "First business day of the month. If bucket 1 is under $60,000, sell from bucket 2 to restore $120,000." The decision gets made in a calm month, and the market only ever supplies the date. That is the entire difference between a rule and a panic.
🔁 The refill sequence, in order
Bucket 1 drops below $60,000
One year of spending is the line that fires the move.
Sell from bucket 2
Sell enough to bring bucket 1 back to $120,000, not to a round number.
Check bucket 2 against $60,000
If bucket 2 is still above $60,000, stop here and leave bucket 3 alone.
Sell from bucket 3 only if stocks are up
If bucket 2 is below $60,000 and stocks are up for the year, bucket 3 pays.
Log the date and the amount
One line in a notebook or spreadsheet, so January's check has something to read.
The Skim That Pays the Growth Bucket Back
A refill rule is only as good as the shelf it draws from, so bucket 2 needs its own inflow during the years bucket 3 runs up. That inflow is the skim, and it starts with one line: compare bucket 3 at the start of the calendar year to where it ends.
Suppose it opens at $700,000 and finishes at $875,000. That is a 25% year. The 20% line sits at $840,000, so $35,000 is the excess and $35,000 moves out of stocks and into bucket 2. The math takes five minutes on the last weekend of December.
The skimmed money stops at bucket 2, never at bucket 1. Cash held past two years of spending is just sitting there losing ground to inflation, so bucket 1 stays anchored at $120,000 and the skim tops up the middle shelf instead. The buffer grows where it can still earn something.
Most years the rule tells you to do nothing, and nothing is the correct action. At 8%, $700,000 becomes $756,000, well under the $840,000 line, and the transfer is zero. People read that quiet year as a missed chance; it is the rule working.
The reason it sits first in the plan: the five years of spending in buckets 1 and 2 has to be built during the up years, because that is the only time it is cheap to build. Skip the skim and bucket 2 becomes a countdown like everything else. It is the only inflow the whole structure has.
💡 PRO TIP
Skim before you need to
The skim exists to build the buffer during good years. If you wait until bucket 1 is low to think about refilling, you are selling whatever is available at whatever price the market sets that month.
Two Bad Years: When the Rules Conflict
Here is the shape of two down years in a row. Year one, $60,000 comes out of bucket 1 and it ends the year at $60,000. Early in year two the balance dips below the trigger, so bucket 2 sells $62,000 and refills the account back to $120,000. Bucket 3 is never touched, either year.
The number to hold onto is $300,000. Buckets 1 and 2 together hold five years of $60,000 spending, and five years is the length of downturn this design is built to sit through without selling a stock. That is not a prediction. It is the runway you bought.
If the market stays flat longer than the runway, the plan says cut spending or add a year to bucket 2 before you sell a share of stock. Dropping annual spending from $60,000 to $50,000 stretches the same $300,000 to six years. One budget change buys you twelve more months of not selling.
And the honest part: no set of rules makes a decade-long flat market painless. What the rules do is remove the decisions from the worst month of the worst year. They shrink the risk. They do not erase it.
The One January Check That Keeps It Honest
The rules hold only if someone re-reads the figures they fire on, on a fixed date, and that date is the first week of January. Then the work runs in a set order. First, top bucket 1 back up to $120,000, using the refill trigger and the order of sales rather than a gut call about the market. Second, bring bucket 3 back to its target weight, selling or buying only inside the growth side. Third, and easiest to forget, re-set the trigger number against the spending you actually did last year. Put all three in the calendar as one block of an hour, before the year gets loud.
Skipping that block because the market moved is the failure to watch for. The trigger then reads a number nobody has updated. Say spending has drifted up to $66,000 a year – two years of it is $132,000. A bucket still holding $120,000 looks full against the old figure and is short against the real one. It has quietly stopped being two years of anything, and nothing sounds an alarm.
All three buckets sit inside one IRA plus one taxable brokerage, and the two do not behave the same at withdrawal. Check the IRS's own publications on IRA and 401(k) distributions, then confirm the tax side with a preparer before January's moves. One hour, one order, one updated number.
Frequently Asked Questions
How much should be in each bucket?
Bucket 1 holds two years of spending, bucket 2 holds three years, and bucket 3 holds whatever is left. On a $1,000,000 portfolio covering $60,000 a year, that is $120,000, $180,000, and $700,000. If a pension or Social Security already covers some of your bills, size bucket 1 to the spending the portfolio actually pays.
Is it too late to switch to this version if I already have a percentage split?
No. The holdings do not need to change much; the numbers do. Convert your split to years of spending, write down the $120,000 and $180,000 targets, and set the refill trigger at one year. You can do that this week without selling anything.
Should bucket 1 be inside the IRA or the taxable brokerage?
That depends on your tax situation, which is why the plan puts all three buckets inside one IRA plus one taxable brokerage and leaves the account placement to your own numbers. A tax preparer or a fee-only fiduciary can tell you which account should hold the cash for your bracket.
What if the market is down when bucket 1 needs a refill?
You sell from bucket 2, not from stocks. That is the whole reason bucket 2 exists. Bucket 3 is only sold when bucket 2 drops below $60,000, so a bad year in stocks does not force a stock sale.
Do I have to sell on the first business day of the month?
Pick a date and keep it. The first business day is easy to remember and easy to automate. The point is that the decision was made in advance, so the month's market move does not get to make it for you.
What if I cannot build five years of spending in buckets 1 and 2?
Then this version is not the right fit yet. Build what you can, keep the refill trigger, and add a year to bucket 2 as you are able. The rules still work at a smaller buffer; they just buy you less time.
Move the Refill Line Into Bucket One
Write down one year of spending – $60,000 in this example – double it to $120,000, and move that amount into bucket 1 this week. Bucket 2, the skim and the order of sales all wait for January.
Two kinds of reader should not build this. If a pension or Social Security already covers your bills, two years of full spending sitting in cash is drag you do not need. And if you cannot split five years of spending – $300,000 on these figures – across the first two buckets, the buffer will not hold, and the simpler plan is the honest one.
For everyone else, you now have the two numbers that do what the plan never did: $60,000 fires the refill, and the 20% line fires the skim. Back to where this started – $120,000, two years of spending in cash. It refills itself only because you decided all of it in a calm month and wrote it down, so the market never gets to ask the question. Put the refill line on paper this week.