Cash Flow Modelling in Retirement: A Practical UK Guide for 2026/27

What Is Cash Flow Modelling?

Cash flow modelling in retirement is the process of projecting every known income stream and every expected expenditure forward over your entire retirement -- often to age 90 or 95 -- to see whether the money will last, and what risks or opportunities exist.

A good cash flow model is not a crystal ball. It cannot predict what markets will do, how long you will live, or what inflation will average over the next 30 years. What it can do is show you the range of plausible outcomes, identify the scenarios under which you run out of money, and quantify the impact of decisions you can make today.

For anyone approaching retirement or already in drawdown, cash flow modelling is arguably the most important financial planning tool available. It turns abstract numbers into answerable questions: Can I retire at 60? Can I afford a new kitchen in 2030? What happens if care costs eat GBP40,000 per year from age 82?

The Building Blocks of a Retirement Cash Flow Model

A comprehensive retirement cash flow model has several core components.

Income Sources

List every income source and when it starts:

• State Pension: In 2026/27, the full new State Pension is GBP241.30 per week (GBP12,548 per year). This is index-linked via the triple lock (higher of earnings growth, CPI, or 2.5%), making it one of the most valuable and reliable income streams in any model.
• Defined Benefit (DB) pension: If you have a final salary or career average pension, the scheme will provide a projected annual income. Include any inflation-linking, spouse's pension provision, and the earliest date you can draw it without reduction.
• Defined Contribution (DC) pension: The current fund value and an assumed growth rate project the pot size at retirement. From there, model the drawdown income.
• ISAs: ISA savings can be drawn tax-free and at any time. Include the current balance and any future contributions (up to GBP20,000 per year).
• Other savings and investments: Bank accounts, general investment accounts (taxable), property income (rental yield net of costs).
• Part-time or consultancy income: Many people work in some capacity for the first few years of retirement. Model when this tapers off.

Expenditure

Spending in retirement tends to follow a pattern often called the "smile" or "go-go, slow-go, no-go" model:

• Go-go years (roughly 60-75): Active retirement with higher discretionary spending on travel, hobbies, and home improvement.
• Slow-go years (roughly 75-85): Reduced activity spending but potentially rising health costs.
• No-go years (roughly 85+): Lower personal spending but potentially significant care costs.

Distinguish between essential spending (food, utilities, council tax, housing) and discretionary spending (holidays, eating out, gifts). A good model allows you to flex discretionary spending in scenarios where income is under pressure.

Tax

Tax is one of the most significant -- and most often overlooked -- cash flow variables. Pension income is taxable as earned income. In 2026/27:

• Personal Allowance: GBP12,570
• Basic rate (20%): GBP12,571 to GBP50,270
• Higher rate (40%): GBP50,271 to GBP125,140

ISA withdrawals are tax-free. The 25% tax-free element of pension withdrawals (up to the pension commencement lump sum of GBP268,275 over a lifetime) can be managed to minimise the tax bill.

A thorough cash flow model integrates the tax position year by year, not just as a percentage of gross income.

Life Expectancy and Longevity Risk

The biggest uncertainty in any retirement cash flow model is how long you will live. UK life expectancy at 65 is currently around 85 for men and 87 for women, but these are averages. Roughly one in four 65-year-olds today will live to 90. One in ten will reach 95.

Running the model to age 90 is a reasonable base case. Running it to 95 or 100 as a stress test is prudent, especially if you are in good health or have family history of longevity.

Deterministic vs Monte Carlo Modelling

The most fundamental methodological choice in cash flow modelling is between deterministic and stochastic (Monte Carlo) approaches.

Deterministic Modelling

A deterministic model uses fixed assumed rates for investment returns and inflation -- for example, 5% annual growth and 2.5% annual inflation. You put in your starting values and the model calculates a single projected outcome.

Advantages: Simple to understand, easy to adjust manually, useful for "what if" scenario comparisons.

Disadvantages: A single assumed return masks enormous uncertainty. The actual return in any given year might be -20% or +30%. A 30-year sequence of 5% average returns with high volatility can produce outcomes ranging from "money runs out at 75" to "estate of GBP1m at 90," even with the same average. Deterministic models also obscure sequence of returns risk.

Monte Carlo Modelling

A Monte Carlo model runs thousands (sometimes tens of thousands) of simulated retirement scenarios, each using a randomly generated sequence of returns drawn from an assumed statistical distribution. The output is a probability distribution -- for example, "there is a 90% probability that your money lasts to age 90, a 70% probability it lasts to 95."

Advantages: Captures sequence risk and return variability. Gives probability-based outcomes that are more meaningful than a single projection. Better for stress testing.

Disadvantages: The output depends heavily on the assumed distribution of returns (standard deviation, correlation between assets). Most Monte Carlo models assume returns are drawn from a normal distribution, which understates the probability of severe crashes.

Which Should You Use?

For retirement planning purposes, Monte Carlo is generally superior to deterministic modelling. However, deterministic modelling is useful for:

• Initial scoping and communication (simpler to explain to clients or family)
• Tax modelling, where the variability is in tax law rather than investment returns
• Scenario comparisons ("retire at 60 vs 65" or "buy an annuity vs stay in drawdown")

Many professional financial planners use both: Monte Carlo for headline probability of success, deterministic for detailed tax and income planning within a specific scenario.

Key Assumptions and How to Set Them

The inputs matter as much as the model itself. Garbage in, garbage out.

Investment Return Assumptions

The Chartered Institute for Securities and Investment (CISI) and the Financial Reporting Council both publish guidance on assumptions, but there is no universal standard. Reasonable assumptions for 2026/27 planning purposes might be:

Note that these are long-run assumptions and may not reflect near-term conditions. Stress test with lower numbers (e.g., 3% nominal for a balanced fund) to see how the model holds up.

Inflation Assumptions

The Bank of England targets 2% CPI. For planning purposes, 2.5% is a reasonable base case for general expenditure inflation. For care costs, research suggests care home fees inflate at closer to 4-5% per annum historically. Use a higher inflation rate for that component of spending.

The State Pension is protected by the triple lock, so it may grow faster than general inflation in years when earnings are strong. The pension annual allowance and other thresholds may or may not rise with inflation.

Spending Assumptions

Getting spending assumptions right is arguably more important than getting investment return assumptions right. Common mistakes:

• Underestimating spending in active early retirement
• Forgetting one-off large costs (a new car every 7-10 years, home repairs, helping children with house deposits)
• Ignoring care costs entirely

A useful starting point is your current non-work spending. Add discretionary costs you plan to increase (travel, hobbies). Subtract work-related costs (commuting, work wardrobe). Then stress test by adding a GBP50,000-100,000 care cost from a given age.

When to Review Your Cash Flow Model

A cash flow model is not a one-time exercise. It needs to be updated whenever material assumptions change.

Annual Review

At minimum, update the model every year to reflect:

• Actual portfolio performance versus assumption
• Changes to State Pension or benefit rates (2026/27 State Pension: GBP12,548 per year)
• Any changes to pension legislation (e.g., annual allowance, lump sum limits)
• Actual spending versus plan

Event-Triggered Reviews

Trigger a full model review after:

• A significant market event (e.g., portfolio down more than 15%)
• A health diagnosis affecting life expectancy or care needs
• Death or serious illness of a partner
• A large unexpected expense or windfall
• A change in housing (downsizing, moving to rented accommodation)
• Any change in tax law affecting pension withdrawals or ISAs

Tools for Cash Flow Modelling

Professional Software

Financial planners typically use specialised software such as Voyant, CashCalc, or Truth (part of Aviva's planning tools). These tools integrate Monte Carlo modelling, tax calculations, and scenario planning. They are not designed for consumers to use directly but your financial adviser should be using one.

Consumer Tools

Several online tools offer simplified retirement cash flow modelling:

• Pension Wise (government-backed, free): Useful for understanding options but limited modelling capability
• MoneyHelper: Government-backed, free, basic projection tools
• Pension provider calculators: Most major providers (Aviva, Standard Life, Royal London) have online drawdown calculators

These consumer tools are useful for initial orientation but lack the sophistication to capture sequence risk, tax complexity, or care costs adequately.

DIY Spreadsheets

A well-constructed spreadsheet can serve as a deterministic cash flow model. Include year-by-year columns for age, income sources (nominal), expenditure (nominal), tax, net cash flow, and cumulative portfolio value. Add an inflation factor to grow spending each year and a growth factor for the portfolio. This approach is time-consuming but gives you full transparency over the assumptions.

Summary

Cash flow modelling is the cornerstone of effective retirement planning. It transforms abstract numbers -- your pension pot, your State Pension, your ISA savings -- into a concrete picture of whether your retirement will be financially secure, and for how long.

The quality of the model depends on the quality of the assumptions. Spend time getting spending right, use realistic long-run return assumptions, account for tax year by year, and stress test against longevity. Monte Carlo modelling is superior to deterministic projection for capturing real-world risk, but both approaches have a role.

Review your model every year and after any significant life event. The sooner you identify a gap -- whether it is an underfunded pension, excessive withdrawal rates, or inadequate provision for care -- the more time you have to address it.