How to Build a Retirement Paycheck (the Two-Bucket Way)

From paycheck to portfolio: the psychological cliff

Most of the early-retirement playbook focuses on the accumulation problem. Save, invest, compound, repeat. That's the fun part of FIRE. The math is clean and the line on the chart goes up.

What almost nobody talks about is what happens the day after you hit your number. For thirty years you've had money show up in your checking account every two weeks like clockwork, with a number you didn't have to choose and a cadence you didn't have to manage. The day you stop working, that pipeline shuts off. You're now the operator of your own income. You decide when, how much, from where, and at what tax cost. Most people approach this with no system at all and end up either underspending out of fear or overspending out of overcorrection. Neither is what they wanted.

The fix is not a new investment strategy. It's a paycheck system that hides the complexity from you and just deposits money into your checking account on a predictable schedule. The simpler the system, the more likely you are to actually run it for thirty years.

The two-bucket system, end to end

Most online versions of the bucket strategy use three buckets: cash, income, and long-term growth. I think that's one bucket too many for someone with a simple index-fund portfolio. Two buckets is enough, and the simpler version is the one I actually run.

The flow is the entire system. Cash builds up in the investment portfolio from dividends. A quarterly transfer pushes that cash into a money market account. A biweekly transfer pushes a constant amount from the money market into checking. Bills get paid from checking. The biweekly transfer feels like a paycheck because mechanically, it is one. The brain doesn't know the difference.

Bucket one: cash, two layers deep

The first bucket is just cash. Two layers, two specific jobs.

Total cash on hand: about seven months of expenses across the two accounts. That's larger than the standard "three to six months emergency fund" because it's serving a different job. It's not insurance; it's the operating reservoir for the paycheck system. Seven months of cash gives me enough runway to ride out a market downturn without being forced to sell from Bucket 2 at a bad price.

Bucket two: the portfolio that does all the work

This is the actual investment account. A simple index-fund portfolio split between stocks and bonds.

The whole portfolio is in low-cost index funds and bond ETFs. Nothing exotic. The 30% bond allocation has a specific job: roughly eighteen months of expenses sit in short-term Treasuries inside that bond bucket. That's the portion I can liquidate quickly and at relatively stable prices if a recession hits and the dividend stream temporarily dips.

The stock side throws off dividends. The bond side throws off interest. I have dividend reinvestment turned off on this account. Cash piles up inside the brokerage account every month from the underlying ETFs. At the end of every quarter, a recurring transfer sweeps that cash to the money market account in Bucket 1A. That's the only manual financial decision in the whole system, and it's automated, so it's not really a decision at all.

The dividend yield on this portfolio currently runs at about 140% of my monthly expenses. That overshoot is intentional. The extra 40% is buffer: it covers taxes, funds occasional rebalancing, and gives me cash flow margin so a bad dividend year doesn't immediately force me to sell shares.

What happens during a recession

The system has a built-in failure mode for bad markets, which is the only kind of failure mode that matters. In a recession, two things tend to happen at once: dividends temporarily decline (companies cut payouts to preserve cash) and stock prices drop (you don't want to be selling).

The system handles this in three layers, in order.

Layer one: spend the buffer. The 140% dividend coverage means I can absorb a roughly 30% dividend cut and still meet expenses without touching the portfolio. Dividend cuts in past recessions averaged 20 to 25% from peak. The buffer covers most realistic scenarios.

Layer two: trim variable spending. If the cut is deeper or longer than the buffer can handle, I shift to a leaner spending profile. Same principle as Lean FIRE as a safety budget: the cuts come from discretionary categories first, the necessities stay funded.

Layer three: sell from short-term Treasuries. If the downturn drags on and the lean budget still leaves a gap, I sell from the eighteen months of short-term Treasuries inside the bond bucket. Those holdings barely move in price during equity downturns, so I'm not selling at a bad time. By the time I'd burn through eighteen months of Treasuries, the market has historically recovered enough that the equity side can do its job again.

The order matters. Most failure scenarios resolve at layer one. Layer two handles almost everything else. Layer three is the deep-failure backstop. I've never had to use layers two or three, and the system is designed so that I never have to use them in any single year.

Taxes: why this works so well at this stage of life

The tax efficiency of this system is the part most people miss when they first see it. When your only income is qualified dividends and long-term capital gains, you get treated dramatically better by the federal tax code than someone earning the same amount in wages.

The 2026 numbers are straightforward and they tilt heavily in favor of an early retiree drawing from a taxable brokerage. A married couple filing jointly can have up to $98,900 of taxable income from qualified dividends and long-term capital gains and pay zero federal tax on it. Add the standard deduction of $32,200, and the practical ceiling is roughly $131,000 of gross qualified investment income with no federal income tax liability. (Single filers see the same shape with thresholds around half: $49,450 in the 0% bracket plus a $16,100 standard deduction.)

That bracket structure is the quiet financial superpower of being retired and having no wages. The same $98,900 earned as W-2 income would land squarely in the 22% federal bracket plus payroll taxes. The same $98,900 as qualified dividends pays nothing. That's not a loophole; it's literally how the code is written. It rewards capital income at lower rates than labor income, which is a long-running policy choice that benefits people whose income stream looks like the system above.

What it actually takes to maintain

The total ongoing time cost of this system, after setup, is roughly an hour a month and an extra few hours at year end. The monthly hour is mostly spreadsheet work: I track expenses by category, log account balances, note any one-time inflows, and confirm the cash flows looked the way they should. Once the system is running, very little ever surprises me, which is the entire point.

The year-end work is the real maintenance moment. Three things happen.

I rebalance the portfolio back to 70/30. If the market has been up, that means selling some equities and buying bonds. If it's been down, the opposite. The rebalance also tops up the eighteen-month Treasury allocation if it's drifted.

I set aside cash for taxes. Once a year I pull cash from the buffer to cover whatever federal and state tax bill the dividends produced. In years when I sit fully inside the 0% federal bracket, the tax bill is small (state taxes plus any non-qualified dividends). In other years it's bigger. Either way, it's predictable.

I do a fresh look at the spending baseline. Costs drift. Inflation is real. The check is whether the 140% buffer is still 140% (or whether it's quietly become 110% as expenses crept up). If the buffer is shrinking, that's the early warning that something needs adjusting before it becomes a problem.

That's the whole job. Set up the two transfers once, run the spreadsheet monthly, do the year-end ritual once a year. The portfolio does the actual work of generating income. The system just routes the cash to the right place at the right cadence.

Sizing the system to your number

The whole structure scales with your portfolio. If $100,000 a year is the paycheck you want, your portfolio needs to be at least $3,000,000 to support that at a comfortable safe withdrawal rate. (The 4% rule says $2,500,000; I prefer the cushion of a 3% to 3.5% withdrawal rate, which lands closer to $3,000,000 plus.) If you want $48,000 a year, you need around $1,500,000 to $1,600,000 with the same conservatism. Halving the paycheck halves the portfolio target.

The bucket sizing scales the same way. Two months of expenses in checking and five months in the money market means the cash buffer grows or shrinks proportionally with your spending. That's it. The system doesn't change. Only the dollar amounts do.

The bottom line

The transition from earning a paycheck to creating one is the part of early retirement that almost nobody prepares for. The two-bucket paycheck system is the simplest way I've found to handle it. One investment portfolio. One cash buffer. One checking account. Two recurring transfers. About an hour a month of upkeep, ten hours a year, in exchange for a stable income stream that funds your life on autopilot. The goal of financial independence was never to spend the rest of your life managing your money. It was to stop having to. A clean paycheck system is what makes that real.