Visit Website →
← Back to blog

3.5–4% Starting Safe Withdrawal Rate: Model Your Own Number

September 2, 2026
3.5–4% Starting Safe Withdrawal Rate: Model Your Own Number

A conservative starting point sits between 3.5% and 4% of your portfolio in the first year of retirement, adjusted for inflation after that. Your right number depends heavily on your time horizon, your asset allocation, and how much guaranteed income (like the Age Pension or an annuity) sits alongside your portfolio. The Trinity Study and William Bengen's original research back the 4% figure historically, while Morningstar's forward-looking modelling now points closer to 3.9%. The rest of this guide explains why, and how to personalise the number for your own retirement.

TL;DR:

  • Asset allocation significantly impacts safe withdrawal rates, with conservative portfolios supporting rates below 3.5%, while aggressive ones may sustain higher rates but with increased risk.
  • The first decade of retirement is critical, as poor market timing during this period can irreparably deplete savings, making a cash buffer essential for mitigating sequence-of-return risk.
  • Forward-looking Monte Carlo simulations suggest a current safe starting withdrawal rate around 3.9% for a 90% success probability over 30 years, slightly below the traditional 4% rule.
  • Personal factors such as extending the retirement horizon beyond 30 years and the presence of guaranteed income sources can justify higher withdrawal rates.
  • Flexibility in spending, via adaptive strategies like guardrails or bucketing, generally increases sustainable withdrawal levels more than attempting precise fixed percentages.

Table of Contents

What is a safe withdrawal rate, and where did the 4% rule come from?

A safe withdrawal rate (SWR) is the percentage of your retirement portfolio you can draw out in year one, then adjust for inflation every year after, without running out of money before you run out of time. It's a starting point, not a fixed instruction, and that distinction trips up more retirees than any other part of the concept.

The number everyone quotes trace back to financial planner William Bengen, who tested withdrawal rates against historical US market returns going back to 1926. He found that a 4% starting withdrawal, adjusted annually for inflation, survived every rolling 30-year period in his dataset without depleting the portfolio. The Trinity Study, published by three finance professors a few years later, extended that work using different stock and bond mixes and came to a similar conclusion. Both papers used a 50/50 to 75/25 stock and bond split, a 30-year retirement horizon, and US historical returns as their sandbox.

That's the part people skip. The 4% rule isn't a law of finance. It's the output of one particular backtest, run on one country's market history, over one set of decades. Change any of those inputs and the "safe" number moves.

A few assumptions worth checking before you anchor to 4%:

  • Your horizon might be longer than 30 years. Retire at 55 and you could need the portfolio to last 40 years or more.
  • Future returns may not mirror the past. Bond yields today look nothing like the 1980s, when Bengen's dataset benefited from a multi decade rally.
  • Inflation adjustments compound. A 4% rule assumes you raise your dollar withdrawal every year regardless of how markets performed, which can be aggressive during a downturn.
  • The rule ignores your actual spending pattern. Real retirees spend more in the early "go go" years and less late, not a flat inflation-linked line.

It's a genuinely useful anchor for a first estimate. It just was never designed to be the final word on your retirement withdrawal strategy.

How asset allocation and retirement horizon change what "safe" means

Two people can retire with the same balance and the same withdrawal rate and face completely different odds of running out of money, purely because of how their money is invested and how long it needs to last.

The mechanics are straightforward once you see them laid out. A higher allocation to equities gives you higher expected long-term returns, but also more volatility, which matters enormously in the years right after you stop working. A longer horizon means more years for compounding to work in your favour, but also more years exposed to inflation, market crashes, and simple bad luck. Ruin probability, the chance your portfolio hits zero before you do, rises when volatility spikes early or when you're drawing down for longer than the historical models assumed.

In practice, this plays out in identifiable bands:

  • Very conservative allocations (20 to 30% equities) tend to support lower sustainable withdrawal rates over long horizons because growth is too slow to outpace inflation-adjusted withdrawals.
  • Balanced allocations (50 to 70% equities) are the sweet spot most historical and forward-looking research settles on, generally supporting rates in the 3.5% to 4.5% range depending on horizon and success threshold.
  • Aggressive allocations (80%+ equities) can support higher average returns but introduce more sequence-of-return risk, meaning your safe starting rate might not rise as much as the higher average return would suggest.
  • Shorter horizons (20 years or less), common for those retiring later or with strong non-portfolio income, generally tolerate a higher starting withdrawal rate than a 35 or 40 year horizon.

There's a practical takeaway hiding in those bands: don't chase the allocation that maximises theoretical returns. If you're unsure where your current mix sits, it's worth reviewing common allocation mistakes that erode sustainable withdrawal rates before you lock in a number.

Sequence-of-return risk: why the first decade decides everything

Sequence-of-return risk is the danger that the order of your investment returns, not just their average, determines whether your money lasts. Two retirees can experience identical average returns over 30 years and end up with wildly different outcomes purely because one hit a market crash in year two and the other hit it in year twenty-eight.

Two retirement return sequences compared

Here's a simplified illustration. Imagine two retirees, each starting with $1,000,000 and withdrawing $40,000 a year, adjusted for inflation. Retiree B experiences the exact same sequence of returns in reverse, with the crash arriving in year twenty-nine instead. Retiree A is drawing a fixed dollar amount from a portfolio that just shrank by a fifth, so that withdrawal represents a much larger bite out of a smaller base, and the damage compounds every year after. Retiree B has had almost three decades of growth to absorb the same crash, and their portfolio barely notices it. Same average return. Vastly different outcomes. Analysis of early retiree outcomes consistently shows this early-decade sensitivity is the single biggest driver of whether a fixed withdrawal plan survives or fails.

A handful of practical tactics reduce this risk without requiring you to predict the market:

  1. Hold a cash buffer of one to three years' expenses. This lets you avoid selling shares at depressed prices during a downturn.
  2. Build a short bond ladder that matures in sequence, giving you a few years of predictable, non-market-dependent income to draw from first.
  3. Adopt a temporary spending rule, such as skipping the inflation adjustment (or cutting spending by a set percentage) in any year the portfolio falls below its starting value.
  4. Stay flexible on withdrawal timing, drawing more from cash and bonds in bad years and rebalancing back into equities when markets recover.

Pro Tip: Build your cash buffer before you retire, not after the first crash. Trying to raise two years of cash by selling shares mid-downturn defeats the entire purpose of holding the buffer in the first place.

For a deeper look at how this risk plays out across different market cycles, our piece on sequence-of-return risk walks through the mechanics in more detail.

Flexible withdrawal strategies that beat a fixed percentage

Flexible withdrawal methods let you spend more in good years and less in bad ones, which several strands of research suggest can raise your average sustainable withdrawal rate without meaningfully increasing your risk of running out.

Guyton guardrails, developed by planner Jonathan Guyton, set upper and lower bands around your withdrawal rate. If your portfolio grows enough that your withdrawal rate falls below the lower guardrail, you get a raise.

Flexible withdrawal guardrails concept

Dynamic or variable-percentage withdrawals take a simpler approach: you withdraw a fixed percentage of your current balance each year rather than an inflation-adjusted dollar figure. This method mathematically cannot deplete your portfolio to zero, since you're always taking a slice of whatever remains. The tradeoff is that your income fluctuates with the market, sometimes significantly, which suits retirees with flexible spending needs far better than those with fixed, non-negotiable costs.

Bucketing and floor-and-ceiling hybrids split your money into segments: a near-term bucket in cash and bonds to cover several years of essential spending, and a longer-term bucket in growth assets you don't touch until the near-term bucket needs replenishing. Some retirees combine this with a "floor" (a guaranteed minimum, often covered by pensions or annuities) and a "ceiling" (a cap on discretionary spending drawn from the growth bucket). Research from the Conexus Institute documents several of these hybrid approaches and their varying effects on sustainable spending.

Flexibility in spending, accepting temporary cuts or spending variance, can materially raise sustainable initial withdrawal rates for many retirees, according to independent analysis of withdrawal strategies.

Who suits which approach comes down to temperament as much as maths:

  • If your essential expenses are largely fixed (mortgage, healthcare, insurance) and you dislike income variability, a conservative fixed rule with a cash buffer probably suits you better than a dynamic method.
  • If you have flexible discretionary spending and can comfortably absorb a lean year, Guyton guardrails or dynamic percentage withdrawals will likely let you spend more on average over your retirement.
  • If you want the psychological comfort of "this money is guaranteed" alongside growth exposure, a bucketing approach with a floor tends to reduce anxiety even when the mathematics look similar to simpler methods.

What Monte Carlo modelling changes about the 4% rule

Modern retirement research has largely moved past simple historical backtesting toward Monte Carlo simulation, a method that runs thousands of randomised market scenarios based on statistical assumptions about returns and volatility, then reports the percentage of scenarios in which your money lasted the full horizon.

This shift matters because historical backtests are limited to the sequences that actually happened. Monte Carlo modelling, by contrast, can generate sequences that never occurred but remain statistically plausible, giving planners a wider and arguably more honest picture of risk. The Financial Planning Association's foundational work on withdrawal rates helped establish this probability-based framing as the industry standard.

Morningstar's research team applies this method using forward-looking asset-class assumptions rather than pure historical averages, since they judge (reasonably) that today's bond yields and equity valuations don't perfectly resemble the twentieth-century conditions Bengen's data was built on. Their most recent baseline lands at roughly a 3.9% starting withdrawal rate for a 90% success rate over a 30 year horizon with a balanced portfolio, modestly below the classic 4% figure.

  • Historical backtests show you what actually happened across real market cycles, which has the benefit of being grounded in reality but the drawback of being a small sample of one country's history.
  • Forward-looking Monte Carlo models adjust for current valuations and yields, which arguably better reflects today's starting conditions but relies on assumptions about the future that could themselves be wrong.
  • Your personal success threshold (90% versus 85% versus 95%) is a risk-policy decision, not a mathematical fact, and shifting it up or down moves your recommended starting rate immediately and predictably.

Neither method is "correct" in isolation.

How to choose your starting withdrawal rate: a step-by-step checklist

Picking a number isn't a one-time calculation. It's a short process, and it's worth working through deliberately rather than anchoring to whatever percentage you read somewhere online.

  1. Estimate your true horizon. Add a margin for longevity risk. Social Security's own longevity data shows a growing share of retirees living into their late 80s and 90s, so budgeting for 30 to 35 years is safer than assuming a shorter retirement.
  2. Tally your non-portfolio income. Age Pension entitlements, defined-benefit pensions, or annuity income reduce how much your portfolio needs to carry alone.
  3. Confirm your asset allocation matches your horizon and temperament, not just your target return.
  4. Pick a success probability you're comfortable with, understanding that 90% is a common default but not a universal rule.
  5. Decide whether you want a fixed rule or a flexible one, based on how much income variability you can tolerate year to year.
  6. Set aside a cash or bond buffer to protect against sequence risk in the first five to ten years.
  7. Choose concrete review triggers rather than reviewing on a vague "when it feels necessary" basis.

Any of these should prompt a genuine review of your withdrawal rate, not just a mental note to "keep an eye on it."

Pro Tip: The order you withdraw from matters almost as much as how much you withdraw. Drawing from taxable accounts first, then tax-deferred, then tax-free (or the reverse, depending on your bracket trajectory) can meaningfully change how long your money lasts, because unnecessary tax paid today is money that can't compound tomorrow. It's worth reading up on tax strategies for retirement income before you lock in a sequencing plan.

Running the plan: monitoring, triggers, and guaranteed income

A withdrawal rate isn't a decision you make once at retirement and then forget. It's a plan you monitor, and the monitoring matters almost as much as the initial number.

Check your withdrawal rate against your current portfolio value at least annually, and compare your actual spending inflation against the general Consumer Price Index, since healthcare and aged care costs often run hotter than headline inflation. Track whether your withdrawals are drifting up as a percentage of a shrinking portfolio, which is the earliest warning sign that a fixed-dollar approach is becoming unsustainable.

Useful triggers to react to, rather than ignore, include:

  • A portfolio drawdown of 20% or more from its peak, which should prompt a temporary spending reduction under most flexible rules.
  • A market recovery that pushes your withdrawal rate well below your guardrail, signalling room for a spending increase.
  • A change in health or living circumstances that shortens or extends your realistic horizon.
  • A move into aged care or a major one-off expense, which should be modelled separately rather than absorbed into your ongoing withdrawal rate.

Guaranteed income changes this calculation substantially. Retirees with a defined-benefit pension, a lifetime annuity, or a larger share of essential costs covered by the Age Pension can typically run a higher withdrawal rate on their remaining portfolio, since that guaranteed income acts as a floor against the worst-case scenarios. Morningstar's own research notes that guaranteed income reduces reliance on the portfolio itself, freeing up more flexibility in how aggressively the rest can be spent. If you're weighing how government benefits fit into your income mix, a Social Security calculator can help estimate that non-portfolio contribution before you set your final withdrawal rate, and retirees looking to maximise those benefits further may find value in strategies for maximising Social Security income.

AlphaIQ's take: modelling your own number instead of guessing

None of them knows your actual super balance, your actual asset mix, or how your spending really moves year to year. That's the gap AlphaIQ was built to close.

AlphaIQ's scenario simulation lets you stress-test a withdrawal rate against your own numbers rather than a textbook portfolio. You can model how a cash buffer changes your success probability during a downturn, how a debt recycling strategy shifts your tax position over time, or how franking credits from Australian shares affect your real, after-tax income in retirement.

A common pattern shows up repeatedly in scenario testing: adding a two year cash buffer to an otherwise identical plan tends to noticeably lift the success probability of a given withdrawal rate, because it removes the need to sell growth assets during the worst possible years. That's not a guarantee for every household, but it's the kind of trade-off that only becomes visible once you model your specific numbers rather than relying on a rule of thumb.

If you're weighing up your own starting rate, running the numbers through a proper retirement projection tool is a far sturdier foundation than anchoring to whichever percentage happens to be trending this year.

Primary sources and further reading

Why I think most retirees overthink the number and underthink the plan

That tenth of a percentage point is noise compared to the difference between having a cash buffer and not having one, or between a rigid rule and a flexible one you'll actually stick to under stress.

What the research keeps showing, across Bengen's original work, the Trinity Study, and Morningstar's newer modelling, is that the starting number matters less than the behaviour around it. Flexibility beats precision almost every time.

— Jonathan

This article is general information, not a substitute for advice from a qualified financial advisor. Consult a qualified financial professional about your own circumstances before acting on anything here.

Sources