How Investment Management Changes When You Retire

What actually changes

During accumulation the arithmetic is forgiving. Contributions arrive on a schedule, a decline reduces the value of what you already hold but also lowers the price of what you buy next, and time is on the side of the plan. Average return over the whole period is a reasonable summary of how things went.

Once withdrawals start, three things change at once. Contributions stop, so a decline is no longer partly offset by buying in cheaper. Money leaves on a schedule regardless of what markets did that year, which means a decline is realised rather than merely observed. And the time horizon, while still long, is no longer indefinite.

The consequence is that average return stops being a sufficient description. A plan can hit its assumed average and still run into trouble, because the same average arranged in a different order produces a different result once you are withdrawing from it.

Sequence-of-returns risk

Consider two retirements with the same set of annual returns in different orders. In the first, the weak years land at the beginning. In the second, they land at the end. Both average the same. The first retirement takes its withdrawals out of a shrinking balance in the early years, so each withdrawal removes a larger share of what remains, and there is less capital left to participate when returns recover. The second takes early withdrawals from a growing balance and meets the weak stretch later, with a larger base and fewer years of withdrawals still to fund.

Two features of this risk shape how it is handled. It is front-loaded: the exposure is concentrated in the first several years after withdrawals begin, and declines as the remaining number of withdrawals falls. And it is not a forecasting problem. You cannot know in advance which kind of sequence you will get, which means the response has to be structural rather than predictive.

Why diversification is not the answer

This is worth stating plainly, because the two ideas get conflated constantly and the conflation is comfortable.

Diversification addresses the risk that any single holding, sector, or region performs badly. It works by making sure your outcome does not depend on one thing. It is genuinely important and it remains important after you retire.

What it does not do is change the order in which a diversified portfolio's returns arrive. A well-diversified portfolio can still have a weak three-year stretch, and if that stretch lands in your first three years of withdrawals, diversification has not protected you from the sequence problem. It has only ensured the weak stretch was not caused by one bad holding.

If a prospective advisor answers a question about sequence risk by describing their diversification approach, the question has not been answered. That is worth noticing rather than letting pass.

The structural responses

Because the risk cannot be forecast, the responses all work by reducing how much of an early decline has to be absorbed by selling into it.

• A stated cash reserve. Holding a defined number of years of planned withdrawals outside the growth portfolio means an early decline does not have to be met by selling depreciated assets. The number of years should be written down, along with what refills it and when.
• A rule for down years. A plan that specifies in advance what happens to the withdrawal when the portfolio falls by a stated amount is easier to follow than a plan that leaves it to judgment in the moment. The point is as much behavioural as mathematical.
• A glide in the allocation. One approach holds more in bonds at the start of retirement and lets equity exposure rise as the front-loaded risk recedes, which is the reverse of the conventional path. It is not the only defensible choice, and it involves a trade-off against long-term growth that should be stated rather than assumed.
• A floor under the essentials. Covering non-negotiable expenses from sources that do not depend on market levels, such as Social Security timing decisions, reduces how much of the withdrawal is exposed to sequence at all.
• Withdrawal sourcing. Which account a withdrawal comes from in a given year is both a tax decision and a sequence decision. We cover the tax side in tax-aware withdrawal sequencing.

None of these eliminates the risk. Each of them reduces how much of an early decline is converted into permanent damage, and each has a cost. A large reserve gives up growth. A bond-heavy start gives up growth. A spending rule gives up some flexibility. A plan that claims one of these without naming its cost has not finished thinking about it.

Stress testing, and its limits

Because averages hide the problem, testing a plan means testing sequences.

Two methods are common. Historical simulation runs your plan through actual past return sequences, starting in many different years, and asks how it would have fared in each. Monte Carlo simulation generates a large number of randomised sequences from assumed return and volatility inputs and reports the distribution of outcomes.

A probability figure from either method is a property of the model, not a promise about your retirement. Ask what assumptions produced it.

The useful output is not a success percentage. It is an understanding of which inputs the plan is most sensitive to, and what the plan does when one of them goes the wrong way. A plan that survives every scenario in the model and has no stated response to a bad one has been tested but not designed.

Spending rules

A fixed withdrawal, adjusted only for inflation, ignores what the portfolio is doing. A fully flexible withdrawal responds to the portfolio but makes household budgeting impossible. Most practical approaches sit between the two.

The common shape is a rule with guardrails: the withdrawal adjusts within stated bounds when the portfolio moves past defined thresholds, so there is a response to a decline without the income swinging with the market. Several formalised versions of this exist, and published research on sustainable withdrawal rates is revised regularly as conditions change.

What matters more than which rule you use is that the rule exists in writing before it is needed. The hardest moment in a retirement plan is the first significant decline, and the value of a written rule is that the decision was made when nobody was frightened.

What to ask

Four questions separate a practice that manages retirement portfolios from one that manages portfolios.

• What happens to my withdrawal if the portfolio falls by a fifth in my first two years, and where is that written down?
• How many years of withdrawals sit outside the growth portfolio, and what refills that reserve?
• How do you test the plan against bad sequences rather than averages, and what assumptions go into the test?
• How many households are you currently taking withdrawals for?

This practice is founder-led and based in Palos Heights. Richard Casolari has spent roughly fifty years in financial services, and portfolios are built with a documented buy and sell discipline rather than preset risk buckets. The work coordinates investment management, retirement income planning, tax-aware withdrawal sequencing in coordination with your CPA, and Medicare enrollment timing into one written plan that is revisited on a schedule.