Private Wealth Pathway · Wealth Planning
Retirement Planning and Sustainable Spending for CFA Level III
Updated 8 October 2026 · Fact-checked
Lifetime financial planning estimates whether a client's assets can fund their spending goals through retirement. You project spending, assets, returns, inflation and longevity, then test the plan with deterministic or Monte Carlo methods. Sustainable spending is the amount the portfolio can support at an acceptable probability of success.
Understand Lifetime Financial Planning, Retirement and Spending Needs
Lifetime financial planning starts with the client's goals. Some are needs, such as housing, food and healthcare. Others are wants, such as travel or gifts. Needs must be funded with high confidence. Wants can be funded with lower confidence. This is the core of goals-based planning: each goal has an amount, a time horizon and a required probability of success.
Spending in retirement is not flat. It usually rises with inflation, and healthcare costs can rise faster than general inflation. Some clients spend more early in retirement and less later. Always state whether your projection is in nominal or real terms and keep it consistent. Do not mix a nominal return with a real spending figure.
Longevity risk is the risk of outliving your assets. Plan to a horizon beyond life expectancy, because life expectancy is the midpoint, so about half of people live longer. For a couple, the relevant horizon is the survival of the last survivor, which is longer than for either person alone. Pension income, annuities and Social Security-type benefits transfer some longevity risk away from the portfolio.
A deterministic projection uses one fixed set of assumptions: one return, one inflation rate, one lifespan. It is simple and transparent, but it gives no sense of the range of outcomes and ignores the order of returns. Monte Carlo simulation runs thousands of random return paths using assumed means, volatilities and correlations. It outputs a probability of success, the share of paths where assets last to the end of the horizon. It captures sequence-of-returns risk: poor returns early in retirement, while withdrawing, hurt far more than the same returns later.
Monte Carlo has limits. Results depend on the inputs, and the model often assumes returns are independent through time and normally distributed, which understates fat tails and may miss mean reversion. It also tends to assume fixed spending, whereas real clients can cut spending in bad markets. Treat the output as a guide, not a guarantee. A sustainable spending rate is the withdrawal, as a percentage of the portfolio, that meets the target success probability over the horizon. Longer horizons, higher fees, lower expected returns and higher inflation all reduce it.
Key rules to remember
- Future value of a spending need
- FV = PV × (1 + inflation)^n
- Use the inflation rate specific to that expense, for example healthcare, if given.
- Real return
- (1 + nominal) ÷ (1 + inflation) − 1
- The approximation nominal − inflation is acceptable only if the question allows it.
- Required capital for a level real spending stream
- PV of annuity = Spending × [1 − (1 + r)^−n] ÷ r
- Use the real return r when spending is in today's money and grows with inflation. The formula assumes end-of-period payments; multiply by (1 + r) for payments at the start of the period.
- Spending rate
- Spending rate = Annual spending ÷ Portfolio value
- Compare with the sustainable rate. A higher rate means a lower probability of success.
- Probability of success (Monte Carlo)
- Successful paths ÷ Total paths
- Success means assets last to the end of the horizon. Shortfall probability = 1 − success.
- Funding gap
- Gap = PV of goals − current assets (and PV of future income)
- Discount all at the same rate basis. Gap above zero means the plan is underfunded.
How to solve Lifetime Financial Planning, Retirement and Spending Needs questions
Use this method for any retirement or spending-need question. Show every calculation, because a correct number earns credit.
- 1Read the client facts and list goals. Separate needs from wants and note each time horizon.
- 2Set the horizon using longevity, for a couple the last survivor, and add a margin beyond life expectancy.
- 3Fix the basis: nominal or real. Convert returns and spending to match.
- 4Inflate each spending goal to the needed date, or work in real terms with a real return.
- 5Value the goals: present value of the spending stream at retirement, then discount to today if asked.
- 6Subtract existing assets and the present value of reliable income such as pensions to find the gap or surplus.
- 7Test the plan: deterministic for a quick answer, Monte Carlo for the probability of success and sequence risk.
- 8Tie the recommendation to the client: needs get high confidence, wants lower. Say what to change if the plan fails: save more, spend less, work longer, or change the allocation.
Quickest way: Real-return shortcut for spending goals
When to use it: Use when spending is stated in today's money and rises with inflation, and you need the capital required.
- Compute the real return: (1 + nominal) ÷ (1 + inflation) − 1.
- Treat today's spending as a level annuity at the real return over the horizon.
- Use the calculator: N = years, I/Y = real return, PMT = spending, solve PV. Use BGN if spending is at the start of each year.
- Compare PV with the portfolio. For a probability question, read the Monte Carlo result and match it to the required confidence for the goal.
Common mistakes in Lifetime Financial Planning, Retirement and Spending Needs
Mixing nominal returns with real spending
Candidates take the stated return and the spending figure without checking their basis.
Fix: Write 'real' or 'nominal' beside each input before calculating, then convert one to match the other.
Planning only to life expectancy
It looks like the natural horizon.
Fix: Life expectancy is the midpoint. Use a longer horizon, and the last survivor for a couple.
Treating Monte Carlo output as certain
A precise percentage looks authoritative.
Fix: State that results depend on assumptions and that models may understate fat tails and ignore spending flexibility.
Ignoring sequence-of-returns risk in deterministic projections
An average return appears to give the same ending value in any order.
Fix: Remember that with withdrawals, early losses reduce the base that must recover. Say so when comparing methods.
Giving every goal the same required confidence
Candidates forget the goals-based distinction.
Fix: Assign high success probability to needs and lower to wants, and fund needs from the safest assets.
Using the wrong annuity timing
Spending at the start of the year needs an annuity due, but the default is end-of-period.
Fix: Check when spending occurs and set BGN or END accordingly.
Worked examples
Example 1
A client retires in 10 years and wants annual spending of 80,000 in today's money, rising with inflation of 3%. Assume this is a first-year need at retirement. Find the first-year spending in nominal terms at retirement.
Show the solution
- Use FV = PV × (1 + inflation)^n.
- FV = 80,000 × (1.03)^10.
- (1.03)^10 = 1.343916.
- FV = 80,000 × 1.343916 = 107,513.
Answer: About 107,513 in the first year of retirement.
Example 2
At retirement a client has a portfolio of 2,000,000 and needs 90,000 a year, in real terms, for 25 years, paid at year-end. The real return is 3%. Is the portfolio sufficient on a deterministic basis?
Show the solution
- Annuity factor = [1 − (1.03)^−25] ÷ 0.03.
- (1.03)^25 = 2.093778, so (1.03)^−25 = 0.477606.
- Factor = (1 − 0.477606) ÷ 0.03 = 17.4131.
- Required capital = 90,000 × 17.4131 = 1,567,179.
- Portfolio 2,000,000 − required 1,567,179 = surplus of 432,821.
Answer: Yes. Required capital is about 1,567,179, leaving a surplus of about 432,821. Because this is deterministic, say it ignores sequence risk, so a Monte Carlo test is still advisable.
Exam tips
- Read the command word. 'Calculate' needs a number with working. 'Justify' needs a reason tied to the client's facts.
- State the basis, real or nominal, in your working. It protects marks if you slip.
- When comparing deterministic and Monte Carlo, give one strength and one limit for each. Mention sequence risk and input dependence.
- For recommendations, link needs to high confidence and safe assets, and wants to lower confidence and growth assets.
- Answer only the number of points asked. Extra responses are not evaluated.
Lifetime Financial Planning, Retirement and Spending Needs in other exams
The same ground in other exams, if you are preparing for more than one or want another angle on it.
Lifetime Financial Planning, Retirement and Spending Needs: frequently asked questions
What is a sustainable spending rate?
It is the percentage of the portfolio a client can withdraw each year while meeting a target probability that assets last through the horizon. It falls with longer horizons, lower returns, higher fees and higher inflation.
Why use Monte Carlo instead of a deterministic projection?
Monte Carlo shows a range of outcomes and a probability of success, and captures sequence-of-returns risk. A deterministic projection uses one average path and cannot show the chance of failure.
What are the limits of Monte Carlo simulation?
Results depend on the assumed returns, volatilities and correlations. Many models assume independent, normally distributed returns and fixed spending, so they can understate extreme events and ignore a client's ability to adjust.
How does goals-based planning treat different goals?
Each goal gets its own amount, time horizon and required probability of success. Needs are funded with high confidence from safer assets, while wants can accept lower confidence and more risk.