I've been thinking about the cash buffer problem for a while now, and I keep finding that most retirement advice either hand-waves through it ("keep some cash on hand!") or goes so deep into Monte Carlo simulations that the actual practical steps get buried. So here's my attempt to work through it concretely. The core idea is straightforward: if you hold enough cash to cover roughly two years of living expenses outside your investment portfolio, you can ride out a bad market without being forced to sell equities at depressed prices. That forced selling is, I think, one of the most underappreciated risks in early retirement — not the average return, but the sequence of returns in the first two or three years. This piece is specifically about the mechanics of building that buffer before you retire, choosing the right account types, and thinking through how to replenish it once you're actually drawing it down. I'm still working some of this out myself, but I'll share what I've found useful.
why two years specifically (and why that number isn't magic) ¶
The two-year figure comes from a rough heuristic rooted in historical market data: most equity bear markets, including the 2000-2002 crash and the 2008-2009 cycle, resolved enough within 18-24 months that a retiree who didn't sell could let their portfolio recover. I want to be careful here — I'm not saying two years is scientifically optimal. Some advisors use 12 months; others prefer three years. The honest answer is that it depends on your withdrawal rate and your spending flexibility. If you're drawing down at 3% of your portfolio per year and can trim discretionary spending by 20% in a downturn, one year of cash might be enough. If you're at 4.5% and your spending is mostly fixed (mortgage, healthcare, medications), two years gives you more runway. My starting framework: take your expected annual retirement spend, subtract any guaranteed income you'll receive (Social Security, pension, annuity), and the leftover — the gap you need your portfolio and cash to cover — is what you're sizing the buffer against. So if you expect to spend $72,000 a year and will get $28,000 from Social Security, your gap is $44,000. Two years of buffer is $88,000. That's the target number.
where to actually hold the cash (account types that make sense) ¶
This is where I think a lot of guides go vague, so let me be specific. For a cash buffer, I think in terms of a laddered approach across a few account types, roughly ordered by liquidity and yield. First tier: a high-yield savings account (HYSA) at somewhere like Marcus, Ally, or Fidelity's cash management account. As of early 2025, these are yielding around 4-4.5% — not nothing. I'd keep three to six months of the gap amount here. This is your truly liquid layer. Second tier: short-term Treasury bills or a Treasury money market fund (something like Vanguard's VMFXX or Fidelity's SPAXX). These are nearly as liquid, often yield slightly better, and carry the additional benefit of being exempt from state income tax — which matters if you live in a high-tax state like California or New York. Third tier: a short-term CD ladder or a 6-to-12-month Treasury ladder. The idea is to have tranches maturing every few months so you're not locked up if you need the cash earlier than expected. I'm intentionally keeping this buffer in taxable accounts or Roth accounts, not traditional IRAs, to avoid triggering unnecessary taxable income before I've optimized my Social Security timing and Roth conversion strategy.
the sequencing question: building the buffer without derailing your portfolio ¶
Here's the tension I find genuinely tricky: you want to build a cash buffer, but hoarding cash too early means dragging your portfolio's growth for years before retirement. I think the right move is to build the buffer gradually in the final two to three years before your retirement date, not all at once. One concrete approach: in the 36 months before you retire, redirect a portion of new savings — say, 30-40% of what you'd normally be adding to equities — into the cash buffer instead. If you're also rebalancing your portfolio to become more conservative as you approach retirement (reducing equity allocation from, say, 80% to 60%), the proceeds from selling equities during rebalancing can go directly into the cash layer rather than into bonds. This means you're not making a separate, additional sacrifice — you're just directing the natural rebalancing flow. Another approach I've seen work well for people closer to retirement with a concentrated year-end bonus: take one or two large bonuses and park them directly into the HYSA and T-bill ladder rather than contributing to a 401(k) beyond the employer match. The tax hit isn't ideal, but the behavioral benefit of watching that buffer grow is real.
the Roth account angle (this one is underused, I think) ¶
One thing I find underappreciated: Roth IRA contributions (not earnings, just the contributions) can be withdrawn at any age, at any time, with no taxes and no penalty. This makes your accumulated Roth contributions a kind of stealth cash buffer that doesn't feel like cash because it's inside an investment account. If you've been contributing to a Roth IRA for ten years and have, say, $60,000 in contributed principal (separate from any growth), that $60,000 is accessible without a 10% early withdrawal penalty, even if you're not yet 59½. I think of this as the emergency extension of the cash buffer — not the first thing I'd tap, but a real resource if markets are down 40% in year one of retirement and I've burned through the HYSA and T-bills. The key is tracking your Roth contribution basis carefully. Fidelity and Vanguard both maintain this in your account records, but you want to have your own log too. I keep a simple spreadsheet with annual contribution amounts going back to when I opened the account. Form 8606 filed with your taxes each year also records this formally.
replenishing the buffer once you're in retirement ¶
Building the buffer is one problem; knowing how to refill it after you've spent it is another. Here's how I think about the refill cadence: in a year when your portfolio is up meaningfully (say, up 15% or more), you pull slightly more from investments than you need for living expenses — maybe an extra $10,000-$20,000 — and route it back into the HYSA. In a flat or down year, you draw from the cash buffer and leave the investments alone. This is essentially a simplified version of what advisors call the 'bucket strategy,' but without the complexity of maintaining three separate buckets with distinct rebalancing rules. The trigger I'd use: if the buffer drops below one year of the gap amount, and the portfolio is up year-to-date, start replenishing. If the portfolio is flat or down, stay patient and let the buffer absorb the stress. I'd also note that part of the replenishment calculus is your guaranteed income picture changing over time — if you delay Social Security from 62 to 70, your monthly benefit grows substantially (roughly 8% per year between 62 and 70 in real terms), which shrinks the gap amount and therefore the buffer you need to maintain in later years. So the buffer is actually a dynamic number, not a fixed target forever.
a worked example to make this concrete ¶
Let me walk through a simple case. Suppose you're 63, planning to retire at 65, and you've decided to delay Social Security until 67. Your expected annual spending in retirement is $80,000. At 67 your Social Security benefit will be $30,000, so your gap is $50,000 per year. Your two-year buffer target is $100,000. Right now, at 63, you have $18,000 sitting in a regular savings account earning 0.5%. Step one: move that $18,000 into a HYSA today. Step two: over the next 24 months, redirect $2,500 per month from your investment contributions into a T-bill ladder — each month buying a 6-month T-bill so you have rolling maturities. After 24 months you'll have added roughly $60,000 to that layer (plus interest), bringing your total buffer to around $80,000. In year three — your final year before retirement — you sell some of the equity position you'd have been rebalancing away from anyway, netting the final $20,000 into the HYSA. You hit $100,000 just as you hand in your notice. That's the rough shape of it. (The numbers are illustrative, and your tax situation will affect the exact sequencing — especially whether it makes more sense to sell equities inside a tax-advantaged account or in taxable.)
things I'm still uncertain about (and you should be too) ¶
I want to be honest about where my thinking is incomplete. First, inflation erodes cash, and holding two years of expenses in low-yield instruments during a high-inflation period is a real cost. I don't have a clean answer here — TIPS (Treasury Inflation-Protected Securities) partially address this but aren't as liquid, and I-bonds are limited to $10,000 per year per person. Second, healthcare costs before Medicare eligibility at 65 are genuinely hard to model and can spike a cash buffer need significantly — I've seen people retire at 62 and spend $24,000-$28,000 in a single year on ACA premiums and out-of-pocket costs they didn't anticipate. Third, the right buffer size is deeply personal and interacts with your spending flexibility, your other income sources, and honestly your psychological relationship with volatility. I know people who sleep fine with six months of cash and people who need three years to feel calm. There's no universal correct answer, and anyone who gives you one with total confidence is probably oversimplifying.
If you're within five years of retirement and haven't thought explicitly about the cash buffer as a separate planning problem from your portfolio allocation, I'd suggest starting there — just with the arithmetic. Figure out your gap number, and see how far your current liquid assets get you toward a two-year target. That single calculation tends to clarify a lot. I'm continuing to think through the replenishment rules and inflation hedging side of this, and I'll write more as I work it out.