Retirement Planning: Working Out What You'll Need

The commonly cited replacement ratio suggests needing 70 to 80 percent of pre-retirement income. It is a rough guide that assumes reduced commuting and work costs, no further retirement saving, and a paid-off mortgage. Where those assumptions do not hold, the ratio misleads.

Start from actual spending

A more reliable approach is to track current spending by category, then adjust each line for retirement. Commuting and work clothing fall. Healthcare usually rises. Travel and leisure often rise early in retirement and fall later.

Housing is the largest single variable. A mortgage cleared before retirement transforms the required income; one that continues does the opposite. Property taxes, insurance and maintenance continue regardless of whether the mortgage does.

Split spending into essential and discretionary. Essential spending is what must be covered by reliable income sources; discretionary spending can flex with portfolio performance, and that flexibility is itself a form of risk management.

The 4 percent rule and its caveats

The widely cited guideline, from research by William Bengen and later the Trinity study, suggested that withdrawing 4 percent of an initial portfolio value and adjusting for inflation had historically sustained a 30-year retirement in US market data.

The caveats matter. It was derived from historical US returns, assumes a particular stock and bond allocation, ignores fees and taxes, and applies to a 30-year horizon. Retirements beginning earlier or lasting longer need lower rates.

More recent analysis has produced both lower and higher estimates depending on assumptions about future returns and valuations. Treat 4 percent as a planning anchor rather than a guarantee, and expect to adjust in response to actual results.

Sequence of returns risk

The order of returns matters enormously when you are withdrawing. Poor returns in the first few years of retirement, while withdrawals continue, permanently reduce the capital available to recover.

Two retirees with identical average returns over thirty years can have completely different outcomes depending on when the bad years fell. This is why the years immediately before and after retirement warrant more conservative positioning than earlier accumulation years.

Mitigations include holding one to three years of spending in cash and short-term bonds, reducing withdrawals in poor years, and maintaining flexibility in discretionary spending.

Social Security timing

Claiming age significantly affects lifetime benefits. Claiming at 62 permanently reduces the monthly amount relative to full retirement age; delaying past full retirement age increases it by roughly 8 percent per year until 70.

For married couples, the higher earner's claiming decision affects the survivor benefit, which continues for the surviving spouse. Delaying the higher earner's claim therefore provides longevity insurance for the household, not just for the individual.

The right answer depends on health, other income, and whether you need the money earlier. The Social Security Administration's own calculators are the appropriate tool, and the decision is worth modelling carefully because it is largely irreversible.

Healthcare before and after 65

Medicare eligibility begins at 65. Retiring earlier creates a gap that must be covered by marketplace insurance, COBRA continuation or a spouse's plan, and marketplace premiums before subsidies can be substantial at older ages.

This gap is one of the most commonly underestimated costs of early retirement. Model it explicitly rather than assuming coverage will be inexpensive.

After 65, Medicare still involves premiums, deductibles and cost-sharing, and does not cover long-term custodial care. Supplemental coverage decisions matter and have enrolment windows with lasting consequences.

Reviewing the plan

Revisit annually and after any significant change. Adjust withdrawals in response to portfolio performance rather than mechanically applying an inflation increase after a poor year.