Life After FI: Making the Money Last

Reaching independence ends the saving phase and starts a drawdown that may last fifty years or more. That is longer than most investment plans are ever tested over. This chapter covers what keeps a long drawdown on track: understanding sequence-of-returns risk, spending flexibly instead of mechanically, holding a buffer, fitting Social Security into the plan, and reviewing it once a year. It closes with what people tend to do with the freedom, including giving.

Sequence of returns: the risk that matters most

Two retirees can earn the same average return over thirty years and end in very different places. The difference is the order of the returns. A large fall in the first few years, while you are withdrawing, forces you to sell more shares at low prices to fund the same spending. Those shares are gone when the recovery comes. The same fall twenty years in does much less harm, because by then the portfolio has had time to grow and the remaining horizon is shorter.

This is sequence-of-returns risk, and it is the main reason early retirements fail when they do. A long retirement makes it worse, because it gives a bad early stretch more years to compound.

The engine below cannot reorder returns: it uses one steady rate a year. But it shows how sensitive a long retirement is to the level of returns, which is the same lesson from a different angle.

With $1,250,000 and a first-year withdrawal of $50,000 raised with inflation, a steady 7.0% return carries the plan through all 50 years (lasts: yes), and the highest rate that lasts is 4.4%. At 5.0% it does not (lasts: no), and the highest rate that lasts falls to 3.1%. Real markets add the ordering problem on top. To see it, the sequence of returns calculator runs the same average return in different orders, and the Monte Carlo simulator tests thousands of possible paths.

Ways to manage it

No single method removes the risk. Each of these reduces it at some cost.

• Spend flexibly. The 4% rule assumes you raise spending with inflation no matter what. Real retirees can cut back after a bad year. Rules that formalise this, such as the guardrails tested by Guyton and Klinger (2006), allow a higher starting withdrawal in exchange for trimming spending when the withdrawal rate drifts too high. Variable percentage methods withdraw a share of the current balance each year, which can never run out but makes income move with the market. The cost is a less predictable budget.
• Hold a buffer. One to two years of spending in cash or short-term bonds lets you stop selling stocks during a fall and draw from the buffer instead. The cost is lower expected growth on that money.
• Use a bond tent. Pfau and Kitces (2014) suggested raising the bond share in the years around retirement, when sequence risk is highest, then gradually moving back toward stocks. The cost is giving up some growth in those years.
• Keep some income. Part-time earnings in the first years lower the withdrawal exactly when it matters most (Chapter 7).
• Start lower. A starting rate below 4% leaves more margin. The cost is a larger target or a smaller budget.

Spending in real life

Retirement spending is rarely the flat, inflation-adjusted line the models use. Research such as Blanchett (2014) found that many retirees' real spending tends to fall gradually as they age, with health costs rising later on. Early retirees often spend more in the first active years on travel and projects. Large one-off costs, such as a car, a roof or help for family, arrive unevenly.

Two habits help. Track spending for the first year or two after leaving work and compare it with the plan, because your pre-retirement budget was an estimate. And keep a separate line for irregular costs, so a new roof does not look like a spending problem.

Social Security

Retiring early does not remove Social Security, but it changes it. You need 40 work credits, roughly ten years of covered work, to qualify for a retirement benefit. The benefit is built from your highest 35 years of indexed earnings; if you worked fewer years than that, the missing years count as zero, which lowers the benefit. Someone who stops at 40 after twenty years of work will usually get a smaller benefit than their statement shows, because that statement may assume they keep working.

You can claim from 62, with a permanently reduced benefit, or wait. Waiting past full retirement age earns delayed credits until 70, after which there is no gain from waiting. For an early retiree, a later claim means the portfolio carries more years alone, but a larger, inflation-adjusted benefit for life afterwards. The right age depends on health, other income, and a spouse's benefits. Your own record is on your my Social Security account at ssa.gov, and the Social Security break-even calculator compares claiming ages.

The annual review

Once a year, on a fixed date, check the plan:

1. Compare last year's spending with the budget.
2. Work out this year's withdrawal under your chosen rule and from which accounts.
3. Rebalance to your target mix.
4. Decide this year's Roth conversions or capital gains harvesting (Chapter 5), keeping an eye on your health insurance credit (Chapter 6).
5. Rerun the plan with current balances in the safe withdrawal rate calculator or the Monte Carlo simulator.

A review once a year is enough to catch drift without reacting to every market move.

What the freedom is for

People who reach independence describe using it in very different ways: time with children or ageing parents, travel, creative work, a business with no pressure to grow, or service. Many give. A donor-advised fund lets you make a tax-deductible gift in a high-income year, such as your last year of work, and recommend grants to charities over the following years. Gifts of appreciated shares held for over a year can avoid capital gains tax entirely. From 70½, qualified charitable distributions can go straight from an IRA to a charity without counting as income. The tax details depend on whether you itemise, so check before a large gift.