Retirement Distribution Strategy: A Framework for Financial Advisors

Back to Resources

When a client stops adding to their portfolio and starts living on it, the industry calls it “distribution”, or “decumulation”. The plainer way to say it: the household has moved from the saving phase to the spending phase, and the job of the financial plan has changed with it. A retirement distribution strategy is the framework an advisor uses to turn a portfolio into reliable monthly income for an unknown number of years. This isn’t a one-time planning event. The modern approach to distribution is continuous. It continuously monitors the plan, updating values and recalculating risks, and calls for an adjustment when thresholds called guardrails are crossed. It treats the spending plan as something the advisor operates for decades, not something the client picks once on the day they retire.

That last point is where most writing on this subject goes wrong. The move from saving to spending is usually described as a moment: retire, reposition the portfolio, select a withdrawal rate, done. In practice, the spending phase is a 25-to-30 year operating problem. Markets move, inflation accumulates, Social Security starts, and the right answer to “what can we spend?” moves with all of it.

This article covers the strategy itself: what the plan should deliver, why the common alternatives fall short, which accounts the money should come from, and how the plan stays current. For the mechanics underneath it, see the complete guide to retirement income guardrails.

The job changes when the paychecks stop

During the saving phase, planning is forgiving. Wages keep arriving, a bad market year is a buying opportunity, and the advisor’s most powerful lever is a single number: the savings rate. In the spending phase, the levers multiply and the risks invert. There is no new money coming in to repair a mistake. The order of returns starts to matter, because withdrawals taken during a deep early drawdown lock in losses a saver would have ridden out. And the client’s question changes from “am I saving enough?” to the one they will keep asking for the rest of their lives: how much can we spend?

The risk of spending the wrong amount runs in both directions. This is the part the industry consistently misses: running out of money is the fear everyone names, but the most common negative outcome we actually see is the opposite: underspending and regret. The household that underspent for thirty years, died with far more money than they ever hoped to have, and skipped the trips and experiences with friends and family that cannot be rescheduled is a victim of underspending risk. For those who have been good savers, this is a real risk. A good spending strategy has to guard against both overspending and underspending at the same time.

Why a withdrawal rate is not a strategy

A well-known answer to the spending question comes from the 4% rule. William Bengen’s 1994 research asked what initial withdrawal rate, adjusted for inflation each year, would have survived every historical 30-year retirement without exhausting the portfolio. It was important work, but as a plan for a real household it just doesn’t work. Its primary failure mode is not depletion; depletion only shows up in scenarios worse than we have ever seen historically. Its primary failure mode is underspending. In most historical periods the household following the rule finished with multiples of their starting wealth. The 4% rule protects against overspending, but it does so at a huge cost.

A second problem with a focus on any sort of “sustainable withdrawal” rates is that in real life there is no such thing as a particular, constant withdrawal rate: in real plans, withdrawals change when Social Security starts, when the mortgage is paid off, when a roof needs replacing. For the full treatment of how the layering in of different income sources at different times makes withdrawals variable, see how much can a retiree actually spend.

A second, lesser-known answer to the spending question is withdrawal-rate guardrails. There have been many approaches to this concept over the years, but some of the best known were introduced by Jonathan Guyton in 2004 and formalized with William Klinger in 2006. Guyton and Klinger deserve real credit for bringing pre-planned spending adjustments into mainstream advisor practice. But their specific mechanics, fixed percentage triggers around a withdrawal rate, do not hold up under real-world conditions. Not only does the focus on withdrawal rates fail when planned withdrawals change due to things like Social Security claiming or planned extra expenses, but even without those common factors the system just doesn’t work. As research independently produced by Fitzpatrick and Tharp, Pfau, and Jeske has shown, running the Guyton-Klinger system through historical scenarios, or hypothetical scenarios produced with Monte Carlo simulations leads to a clear conclusion: Guyton-Klinger guardrails often lead to extreme spending cuts. What’s worse: these spending cuts are usually unnecessary. The reason for this poor performance in testing is that withdrawal rates and guardrails started in withdrawal rate terms are too crude a tool to capture how things change as someone proceeds through life. A 6% withdrawal-rate trigger (a common setting for the guardrail that would trigger a spending cut) means something very different at 60 than it does at 85. For a 60-year-old, 6% might be too much. But for an 85-year-old it might be too little! And a rate-based rule cannot see the rest of the plan: pension start dates, uneven expenses, longevity, taxes.

The concept of guardrails is sound, but it needs to be implemented in a way that integrates all parts of the plan: investment risk, inflation risk, mortality risk, longevity risk. The execution has to be “risk-based” rather than “rate-based”. For a full treatment of this issue, see risk-based guardrails vs Guyton-Klinger guardrails.

The third answer to the spending question is the “probability of success” score. Unlike the 4% rule, or even Guyton-Klinger guardrails, this statistic is very widely used in retirement planning. Our position on this misleading number is simple: it needs to be eliminated entirely from financial planning. A probability score answers a question no client has ever asked, pushes every household toward underspending, and turns normal market noise into “what did you do wrong?” phone calls. We have written the full case against probability of success separately; for this article it is enough to say that a percentage is not a spending strategy, because it never tells anyone what to spend. Clients want directions, not scores.

What a modern spending plan delivers

A distribution strategy built for the way retirement actually unfolds delivers three things, and every piece of the ongoing discipline hangs off them.

A retirement paycheck, in dollars. The plan’s headline is a monthly spending number, sometimes called spending capacity, calculated at a chosen risk posture. Income Lab’s default posture is quite balanced and reasonable: 20% overspending risk and 80% underspending risk. That means the retirement paycheck keeps people spending within their means. The paycheck includes every income source the household has. Consider Gary, 68, and Susan, 66, who retired earlier this year with $1.1 million across Gary’s IRA and a joint taxable account. Their plan supports $6,400 a month, funded by Gary’s Social Security check, withdrawals from the IRA and the taxable account, and, once Susan claims her own benefit at 70, a smaller portfolio draw. For another household the mix might include a pension or the rent from a duplex; whatever the sources, the paycheck is one number, it includes all of them, and the mix changes over time.

Guardrails, expressed as portfolio balances. Around the paycheck sit two thresholds. Gary and Susan’s upper guardrail is $1.35 million: if the portfolio grows to that balance, the plan calls for a raise, with spending moving to about $8,900 a month, the target for their risk posture at that portfolio level. Their lower guardrail is $820,000: if the portfolio falls to that balance, the plan calls for a measured slowdown toward a lower spending target, about $5,500 a month. Between the two sits the no-change range, a range of portfolio balances where nothing needs to change, which is where the household spends most of its time. Note that, unlike static Guyton-Klinger guardrails, these risk-based guardrails are not permanently stuck at a particular number. Instead, they are point-in-time estimates that change as the world changes. If Gary and Susan had the same guardrails ten years from now that they have today, something would be very wrong. In ten years, they will be ten years older, some unknown amount of inflation will have accumulated, tax laws will be different, returns will have happened, and Gary and Susan’s spending will have been different from what they had originally planned.

A monitoring cadence. The plan recalculates every month: balances, accumulated inflation, the household’s age and plan length, and whether either guardrail has been reached. A default 5% minimum-change threshold filters out noise in both directions, so the plan only calls for a change when a guardrail is hit or when inflation has cut purchasing power by 5% or more. The advisor reviews with the client at their normal pace (quarterly, semi-annual, annual), and the monthly engine runs in between.

One more element of shape matters: the spending line itself is not necessarily flat. Retirees reliably spend more in their active 60s and 70s than in their late 80s, the pattern David Blanchett’s research dubbed the “retirement spending smile”, and a plan can be built on an age-based path that reflects it. Withdrawals are lumpy too. Maybe spending is higher in early years while retirees pay off the last of their mortgage, travel more, or fund a child’s wedding or house down payment. Withdrawals aren’t steady either: a household bridging the years between retirement and a delayed Social Security claim will draw heavily from the portfolio early and far less later, the spike Derek Tharp and I called the retirement distribution hatchet in our risk-based guardrails research. A static withdrawal rule cannot plan around that shape. A plan recalculated against the household’s actual cash flows can.

Which accounts the money comes from

Once the plan knows how much the household can spend, the next question is where each dollar originates, and this is where a spending strategy becomes a tax strategy.

The baseline ordering is mechanical. Required minimum distributions (RMDs), which currently begin at age 73 or 75 depending on birth year, come out first because the law says so. Age-based restrictions are enforced next, and the remaining withdrawals are spread across eligible accounts according to the plan’s selected tax strategy. Income Lab models this across different account types spanning taxable, tax-deferred, tax-free, inherited, and specialty accounts, so the sequencing reflects what the household actually owns.

The leverage comes from treating that ordering as a decision rather than a default. Most plans simply assume that withdrawals should be ordered (1) taxable, then (2) tax-deferred, then (3) tax-free (Roth). Income Lab’s Tax Lab runs 20 different tax-aware distribution strategies in parallel against the same plan and automatically ranks them by total net income, net legacy, total taxes, and average effective tax rate, highlighting the strategies that provide real potential for measurable value. Several moves recur in the winning strategies:

• Bracket management Roth conversions. In years when required distributions and other income leave room in a targeted federal tax bracket, the plan converts traditional IRA and 401(k) dollars to Roth to fill that space, paying tax at today’s known rate. The payoff arrives later, as smaller RMDs and tax-free withdrawals in the years the household needs flexibility most.
• Medicare surcharge awareness. Medicare’s income-related monthly adjustment amount (IRMAA) sets Part B and Part D premium surcharges based on the household’s modified adjusted gross income (MAGI) from two years earlier. That lookback means a large Roth conversion at 63 shows up in Medicare premiums at 65, so conversion timing has to be managed against the IRMAA brackets, not just the tax brackets. Tax Lab’s IRMAA bracket management does exactly that, and the stakes are real: a couple in the top IRMAA tier pays $13,872 a year (adjusted for inflation) in combined Part B and Part D surcharges.
• Charitable dollars from the right account. From age 70 and a half, qualified charitable distributions (QCDs) let IRA dollars go to charity without ever hitting taxable income, which is almost always better than giving from a taxable account, and better than converting to a Roth (assuming the charitable giving is part of the plan).

All of this serves the number the client actually cares about. Gross withdrawals are an accounting figure; the retirement paycheck is what lands in the bank account. Income Lab models spending capacity net of tax, so when two strategies produce the same gross draw, the advisor can see which one buys more groceries. To see a tax-aware spending plan built on a real household, Book a walkthrough.

What happens when a guardrail is hit

Eighteen months into retirement, Gary and Susan get the bad version of sequence-of-returns risk: a year when stocks and bonds fall together. Between the market and their withdrawals, the portfolio crosses below $820,000. The plan calls for a change, and the change is much smaller than the cliff most clients imagine.

The plan does not reset risk completely, which would typically mean a large pay cut. Instead, the right move is typically to close about 10% of the gap between current spending and the true risk target. Often, this means a 5-10% pullback in spending, even if the portfolio is down 20-30%. After the adjustment, we keep monitoring. If conditions keep deteriorating, the plan may call for another measured step; if markets recover, spending never fell far in the first place. The gradualism is deliberate. It prevents the household from cutting deeply on a temporary drawdown and getting whipsawed when the recovery comes. (The asymmetry runs the other way at the top: an upper-guardrail increase goes 100% of the way back to target because there’s no real need to tiptoe into good news.)

And spending is not the only lever. Retirees have a superpower: their ability to adjust. A lower-guardrail breach starts a conversation about the whole plan:

• Change the size or cadence of an expense. A new car every 7 years instead of every 5, or a deferred kitchen remodel, can close much of a gap without touching the monthly budget.
• Revisit Social Security claiming timing. Susan has not claimed yet. Moving her claiming date changes how much the portfolio has to fund, and that decision should be re-run at the breach rather than assumed.

The guardrail tells the advisor it is time to have the conversation. It does not pre-decide where the conversation will conclude.

The upper guardrail deserves the same advance framing, because it is the half of the strategy clients have never heard before. If Gary and Susan’s portfolio reaches $1.35 million, the plan calls for a raise, and the raise goes all the way back to the target for their risk posture: roughly $8,900 a month. The annual increase is enough for the Portugal trip they keep postponing and a few of the bucket-list items they quietly stopped mentioning. The upper guardrail can be reached three ways: the portfolio grew faster than expected, spending ran lower than expected, or inflation came in lower than expected. However it happens, the plan’s job is to say so out loud. That is the structural defense against the most common bad outcome in retirement, which is not running out of money; it is underspending and regret.

How the plan stays current for thirty years

Everything above describes a single point in time. The discipline is keeping it true for three decades, and that is a job for software. The math has to be redone monthly for every household in the advisor’s client base. That’s simply not scalable without automation.

Income Lab is second to none in the full lifecycle of financial planning, and its Track and Monitor layer is built to make life-long planning scalable. Each month, the software updates account balances, adjusts the plan’s length as the household ages, accrues inflation, applies cost-of-living adjustments to Social Security and pensions, and tests the guardrails. In the retirement phase, a monitored plan has a useful inertia: most months the answer is “no change is needed”. The plans that do hit a trigger are flagged for the advisor’s attention with the specific change the plan is calling for. The advisor delivers value without redoing the work, and the client gets a commitment with teeth: if something changes, you will hear from me.

Helping clients understand their plan

When a client wants to understand how the plan will behave over decades, the most effective tool is not more math; it is history. The Retirement Stress Test runs the household’s actual plan through real historical sequences of returns and inflation, including the Great Depression, 1970s stagflation, the Dot-Com bubble, and the Global Financial Crisis, and shows where spending would have adjusted and recovered. Walking a client through how their plan would have navigated 2008 does more for their understanding than any explanation of the methodology. Humans understand stories better than math. A historical stress test isn’t a prediction that the past will repeat itself. It’s an understandable story that shows how a plan adjusts, and helps clients evaluate whether those adjustments are changes they could live with if needed.

If more is needed, Penny, Income Lab’s AI paraplanner, can draft the client-ready summary of the plan and its guardrails; the advisor reviews, adjusts, and signs off before anything reaches the client. How that conversation runs in the meeting itself is covered in how to present a guardrails-based retirement plan.

The same strategy at 76 and at 84

The framework does not change across the spending phase, but the focuses do. Two more example households show the arc that can happen through the years.

Barbara is 76 and widowed, with $850,000 and a $2,100 monthly Social Security survivor benefit. Her plan supports $5,600 a month: the $2,100 benefit plus about $3,500 from the portfolio. Her Social Security decision is behind her, so for Barbara the discipline is mostly the monitoring cadence itself: the monthly recalculation quietly shortens her plan horizon as she ages, which is part of why her spending capacity holds up even as the portfolio draws down.

Gene and Alice are 84, with $1.8 million and a plan that supports $11,000 a month they mostly do not spend. Their constraint stopped being sustainability years ago; it is now taxes and legacy. Their RMDs arrive whether they need the income or not, so the work is QCDs to their church from Gene’s IRA, conversion decisions weighed against what their heirs would keep after tax, and Income Lab’s Total Net Legacy view, which shows the after-tax value of each account type to the next generation. The lower guardrail still gets tested every month. It just is not where the planning value lives anymore.

That is what it means to call distribution planning a discipline rather than a moment: the same planning process, the same guardrails logic, the same monthly engine, doing different work in different decades.

FAQ

What is decumulation in retirement?

Decumulation, or “distribution”, is the phase of retirement when a household draws from its portfolio to fund living expenses, the spending phase that follows the saving phase. A distribution strategy is the framework for converting the portfolio into sustainable monthly income over an unknown lifespan. The modern approach is continuous: recalculate sustainable spending monthly, set risk-based guardrails as portfolio balances above and below the current balance, and adjust spending only when a guardrail is crossed or inflation accumulates past a set threshold.

When does the distribution phase start?

It starts when portfolio withdrawals begin funding household expenses, which is usually but not always the retirement date. Some households start earlier, drawing on the portfolio during a sabbatical or a phased retirement. Some start later, living on severance, part-time income, or a pension before touching the portfolio. The practical marker for the advisor is the first month the plan must answer “how much can we take out?” rather than “how much should we put in?”

What is the difference between accumulation and decumulation?

Different math and different risks. In accumulation, the dominant lever is the savings rate, time repairs mistakes, and market drops are buying opportunities. In decumulation, the order of returns matters, withdrawals during early drawdowns do permanent damage, and the risk runs both directions: depletion on one side, underspending and regret on the other. Accumulation rewards a saving discipline. The spending phase rewards a monitoring discipline, because the sustainable spending number genuinely changes as markets, inflation, and the plan’s horizon move.

What is the 4% rule in decumulation?

The 4% rule comes from William Bengen’s landmark 1994 study, which found that a 4% initial withdrawal rate, adjusted annually for inflation, would have survived every historical 30-year retirement for a 60/40 portfolio. It is a research finding, not a plan. In most historical periods it left retirees with far more money than they started with, so its primary real-world failure is underspending; depletion only appears in scenarios worse than we have ever seen historically. It also assumes a constant inflation-adjusted withdrawal, and real household withdrawals are not constant.

What triggers a guardrail adjustment?

Two things, both filtered by a minimum-change threshold (5% by default). The first is a guardrail breach: the portfolio balance crossing the upper or lower guardrail. The upper guardrail can be reached three ways: portfolio growth, lower-than-expected spending, or lower-than-expected inflation. The second trigger is inflation alone: when accumulated inflation has cut purchasing power by 5% or more, the plan calls for a raise even though no guardrail was hit. Adjustment sizing is asymmetric: increases go all the way back to target, decreases close about 10% of the gap between current and target spending.

Can software run a decumulation plan on its own?

The software runs the math; the advisor runs the plan. Monthly recalculation across every account, accrued inflation, actuarial plan-length updates, and guardrail testing is work no human does reliably for a full client base. Income Lab makes the process scale. But a lower-guardrail breach opens a decision with several levers, including expense changes, Social Security timing, and tax moves, and that decision belongs to the advisor and the client.

Recommended reading

• The complete guide to retirement income guardrails
• How much can a retiree actually spend
• Risk-based guardrails vs Guyton-Klinger
• Derek Tharp and Justin Fitzpatrick’s risk-based guardrails research
• Why Guyton-Klinger guardrails are too risky for most retirees
• Penny, Income Lab’s AI paraplanner

“Decumulation” is the industry’s word. The client’s word for it is the question they will ask at every review, every market drop, and every late-night worry for the rest of their lives: “How much can I spend?” A decumulation strategy worth the name answers in dollars, puts guardrails around the answer, sources the dollars tax-intelligently, and re-earns the answer every month for thirty years. Book a walkthrough to see Income Lab run that discipline on a household like the one in your next review meeting.