Retirement spending patterns: the smile curve and why flat-budget assumptions miss the reality

Most retirement financial models assume spending is roughly constant from year one — the same real dollar amount, adjusted upward each year for inflation, until funds run out. This assumption is simple, conservative, and wrong for most retirees. Research across multiple jurisdictions, including Australian studies, consistently finds that retirement spending follows a different shape: high in the early active years, lower in the quieter middle years, and rising again in the late years as healthcare and aged care costs accumulate. This pattern — sometimes called the retirement spending smile — affects how much money is actually needed at different stages, and planning built around a flat assumption will be inaccurate in both directions: overstating the income needed in mid-retirement and potentially understating the reserves required for late-retirement care.

The three phases: go-go, slow-go, no-go

The go-go years of early retirement — roughly the first decade from retirement, often ages 65 to 75 — are typically the highest-spending phase. The bucket list items that were deferred during working life get realised: international travel, hobbies that require equipment or membership fees, family holidays with grandchildren, home improvements. Health and energy support active engagement. Some research suggests early-retirement spending can run 10 to 20 percent above a retiree's former working-age baseline in this phase, at least for the first few years. The 10-20% go-go-phase elevation above working-age baseline is consistent with international retirement spending research (Blanchett/Morningstar; EBRI) and is the working assumption used in Australian financial planning practice. ASFA's quarterly Retirement Standard publishes year-by-year couple/single budgets but does not separately quantify a "go-go phase" uplift — practitioners typically apply a stylised early-retirement uplift to the comfortable-standard baseline figure ($77,375 for couples / $54,837 for singles). Retirees who have not explicitly planned for this higher early spend often draw down super more quickly than their long-run models suggested.

The slow-go years — roughly ages 75 to 85 — are typically the quietest spending phase. Travel diminishes, partly through preference and partly through gradual health and energy changes. Discretionary activities reduce. Established routines replace novelty. For many retirees, total spending in this phase is noticeably lower than in the early years — and considerably lower than a flat inflation-adjusted model would have predicted. This is the phase where variable-drawdown plans capture the most benefit: the retiree needs less, draws less, and allows the portfolio more time to compound.

The no-go years — roughly 85 and beyond, though the transition varies substantially by individual — bring a different kind of expense growth. Discretionary spending remains low, but healthcare costs accumulate: specialist visits, medications, allied health services, and eventually the costs of formal support. Home care packages (which can run several hundred dollars per week at higher levels of support) and, for many retirees, eventual residential aged care create a late-stage cost profile that bears no resemblance to the modest middle-retirement phase. Total spending can return to or exceed early-retirement levels in this phase — but the composition has shifted entirely from lifestyle to care.

Why a flat model produces the wrong answers

A model that assumes constant inflation-adjusted spending misses both the early-period uplift and the late-period composition shift. Applied to sustainability analysis — "will this portfolio last?" — it tends to overstate how much income a retiree needs from age 75 to 85 (underestimating the middle-period savings opportunity) and understate the reserves needed for care costs after 85. The practical result is that a retiree running a flat-model plan may feel they are underspending relative to plan in the middle years, while being structurally under-reserved for the care costs they will face later.

The alternative is a model that explicitly accounts for the three phases. Higher projected spending in years one to ten, reduced in years ten to twenty, and a dedicated care reserve for years twenty and beyond — with the amounts in each phase calibrated to the retiree's specific circumstances — produces a more accurate sustainability picture and a more sensible drawdown strategy.

Variable drawdown: matching the model to reality

For retirees with super, pension drawdown amounts can in most cases be varied within the minimum pension requirements. Rather than drawing a constant percentage of the portfolio, a smile-curve-aware strategy draws more in the early active years, less in the quieter middle period, and maintains a reserve for late-stage care. The middle-period underdrawing serves two purposes: it extends portfolio longevity and it builds the care reserve that the late period will need.

The precise percentages depend on individual circumstances — super balance, Age Pension entitlement, other assets, health profile, and risk tolerance. But the general structure of higher-lower-higher drawdown aligns with actual spending patterns in a way that constant drawdown does not.

A worked illustration

Consider a couple retiring at 65 with $900,000 in superannuation and entitlement to a part Age Pension of approximately $30,000 per year combined (at current rates, a couple with substantial super assets may receive a reduced or part pension depending on their assets test position). Their target total income is $75,000 per year in early retirement.

In years one to ten (go-go), they draw approximately $45,000 per year from super to reach their target — full travel and active lifestyle, which runs above their pre-retirement baseline.

In years eleven to twenty (slow-go), they reduce super drawdown to $25,000 per year, reflecting lower discretionary needs. The Age Pension has likely grown with indexation. The portfolio continues to compound on the undrawn balance.

In years twenty-one and beyond (no-go), they draw more heavily from the portfolio — $40,000 to $50,000 per year — partly for care costs and partly because Age Pension eligibility may have increased as other assets reduce. If one partner has since died, the surviving partner operates on single-rate income and costs.

The total super drawn over 25 years under this variable approach is similar to or less than a constant-drawdown model, but the timing and composition more accurately match actual need — more when they can enjoy it, less when they naturally spend less, and a reserve available for care.

Late-stage care reserve: the deliberate provision

One specific planning action that follows from the smile curve is the explicit allocation of a late-stage care reserve. Rather than treating the aged care cost as an unpredictable contingency, smile-curve-aware planning treats it as a predictable phase with a predictable cost profile — and allocates resources accordingly, building the reserve during the middle-period when drawdowns are lower than the portfolio could sustain.

Personal variation matters

The smile curve is a generalisation, and individual patterns vary substantially. Some retirees maintain high activity well into their 80s; some experience significant health costs earlier than the model suggests; some genuinely spend consistently throughout retirement with little phase variation. The framework is a starting point, not a prescription. What it provides is a more realistic default assumption than flat spending — and, for the large majority of retirees, a planning structure that better matches the life they will actually live.

Sources

• Age Pension rates and income/assets tests — Services Australia
• Common pension rates (Social Security Guide 5.1.8.10) — DSS
• Income stream (pension) rules and minimum payments — ATO
• Account-based pensions — Moneysmart
• Support at Home costs and contributions — My Aged Care
• How the Support at Home program works — Dept of Health, Disability and Ageing