The Assumptions Behind Timeline’s Financial Plan

Summary

We don’t assume future returns or inflation — we use real historical data going back to 1915
Client funds are mapped to long-term asset classes
Returns are shown gross of fees — advisers add all fees manually
Surplus cash is carried forward by default (can be assumed spent)
UK tax system is fully modelled
Tax bands can rise with inflation (optional)
Default legacy goal is £1 (fully adjustable)
Default retirement age runs until survival probability falls to 10%
Annual rebalancing by default (customisable)
Neutral withdrawal order by default (fully configurable)
Withdrawal amounts are calculated automatically
The plan is built to evolve over time

Description

This article explains the key assumptions behind a Timeline financial plan and why they matter in practice.

Assessing the sustainability of a financial plan requires assumptions about uncertain variables such as returns and inflation. The critical issue is whether those assumptions are explicit and grounded in evidence, rather than implicit or arbitrary.

Our objective is not to create an illusion of certainty, but to provide a structured way to model uncertainty, so advisers can have better conversations and make more resilient decisions.

1. No assumed growth or inflation rates

We do not start by selecting a assumed returns or inflations. There is no single growth rate hardcoded into the system.

Instead, we rely on historical economic data going back to 1915. A unique aspect of Timeline Planning's simulation models is the integration of UK Consumer Price Index (CPI) rates. These rates are crucial in stress-testing each historical scenario. That means real returns, real inflation, real periods of stress, real recoveries, and long stretches of stagnation. Every scenario reflects something that has actually happened in markets over the last century.

This approach avoids guesswork. It removes subjective forecasts and replaces them with a long-term evidence base. Advisers are not betting on one prediction of the future. They are testing the plan against a wide range of historically observed environments.

2. Real investments mapped to long-term history

Most funds and model portfolios do not have 100 years of live track record. That creates a challenge when modelling long-term outcomes.

To solve this, we map each client holding to the most appropriate historical asset class in our dataset. This preserves the client’s real asset allocation while allowing us to extend the analysis back to 1915.

We use total return indices, meaning both price appreciation and reinvested income are included. This ensures compounding is captured properly and the modelling reflects how portfolios actually grow over time.

3. The plan evolves

A financial plan is not a one-off document. Markets move. Inflation shifts. Governments change tax policy. Clients change behaviour and objectives.

Timeline is designed to be updated regularly. It supports ongoing advice, scenario testing, and strategic adjustment over time. The value is not in a static projection, but in the ability to revisit and adapt the plan as reality unfolds.

4. Returns are gross of fees

All historical returns used in Timeline are gross of fees. We deliberately do not embed any assumed adviser fee, platform charge, or fund cost into the dataset.

This ensures transparency and flexibility. Advisers must manually input:

Platform fees
OCF
Adviser fees

Both fixed and percentage-based fee structures are supported. This allows the plan to reflect the real charging structure applied to the client.

5. Full UK tax modelling

The engine incorporates the major UK tax regimes, including:

Income Tax
Capital Gains Tax
Dividend Tax
Interest Tax
Inheritance Tax

Each wrapper is treated according to UK tax rules. The net withdrawal amount required by the client via planned spending is used as the reference point. The system then calculates the corresponding gross withdrawal required from each wrapper, taking into account the tax treatment of the account and the client’s broader income position. As a result, the actual amount withdrawn may exceed the net income requirement in order to satisfy any tax liability generated by that withdrawal.

This ensures the cash flow shown in the plan reflects real-world tax outcomes, not simplified gross figures.

6. Tax bands and inflation

By default, tax bands and allowances are uprated with inflation. This reflects a long-term structural assumption that thresholds tend to adjust over time.

However, this is not fixed. If an adviser expects prolonged fiscal drag or frozen thresholds, the inflation adjustment can be switched off. The system allows you to model both structural adjustment and policy stagnation.

7. Legacy goal

The default legacy target is £1. This avoids accidentally constraining the plan with an arbitrary inheritance assumption.

If a client has a specific legacy objective, it can be set explicitly. The success rate will then depend on achieving that target.

For example, if the legacy goal is £100,000 and a scenario ends at £90,000, that scenario is classified as unsuccessful. The goal directly influences the probability of success calculation.

8. Longevity assumption

By default, the plan runs until survival probability falls to 10% or less. This creates a realistic planning horizon without forcing extreme longevity assumptions.

Advisers can override this and specify a fixed end age if required. The system provides flexibility while maintaining a defensible default.

9. Surplus is carried forward

If income exceeds spending in any given year, the default assumption is that the surplus accumulates and remains invested.

Advisers can choose to assume that surplus is spent instead. The modelling is flexible and can reflect the client’s behavioural reality.

10. Rebalancing

Annual rebalancing is the default assumption. Advisers can change the rebalancing frequency or disable it entirely.

Advisers can change the rebalancing frequency or disable it entirely. The modelling can reflect both systematic rebalancing and a more passive drift-based approach.

11. Withdrawal order

We do not assume there is a single correct withdrawal strategy.

The default is neutral: withdrawals occur proportionally across wrappers, based on their relative weight in the total portfolio.

Advisers can fully customise sequencing, optimise for tax efficiency, or create wrapper-specific strategies aligned with their planning philosophy.

12. Withdrawal amount

The system does not assume how much a client should withdraw.

Instead, it calculates what is required to meet spending needs after accounting for:

Income sources
Spending
Taxes
Contributions
Transfers
Fees

This ensures withdrawals are driven by actual cash flow requirements, not arbitrary percentages.

In addition the adviser can choose specific withdrawal amount from specific tax wrappers.

13. Withdrawal and inflation rules

By default, withdrawals are fully inflation-adjusted and there is no dynamic spending rule applied.

However, advisers can introduce spending flexibility, apply dynamic rules, or adjust how inflation impacts withdrawals. The framework allows for both rigid and flexible retirement income strategies.

14. Monte Carlo

Monte Carlo simulation generates thousands of potential return paths. Instead of showing one straight-line forecast, the plan is tested across a full distribution of possible outcomes.

At Timeline we use a Geometric Brownian Motion framework. This assumes asset prices are log-normally distributed and requires expected return and volatility as inputs. It is a widely accepted approach in finance.

However, it is important to understand its limitations. GBM assumes no serial correlation and no structural regime changes. In reality, markets exhibit short-term momentum and long-term mean reversion. No simple stochastic model can perfectly capture that complexity.

Monte Carlo is therefore a tool for structured probability modelling, not a perfect representation of reality. It introduces variability and sequence risk, but it still depends on the assumptions behind it.

From the historical dataset, we derive two core inputs: arithmetic mean returns and volatility for each asset class.

Portfolio expected return is calculated as the weighted average of its components. Portfolio volatility is calculated using modern portfolio theory, taking into account diversification effects.

Correlations are not assumed in isolation. We calculate rolling historical correlation matrices across monthly windows and then derive long-term averages. This allows the portfolio risk to reflect how assets have interacted historically, not just in a single snapshot period.

For advisers who prefer to use their own capital market assumptions, the system allows manual inputs for expected return and volatility at portfolio level.

These annual inputs are then used to generate simulated monthly returns inside the Monte Carlo engine.

One important technical point: if geometric returns are used as the expected return input, volatility drag can effectively be counted twice, because geometric returns already embed volatility effects. For this reason, we strongly recommend using arithmetic mean returns as the input to the Monte Carlo model.

Conclusion

Timeline is built around clarity, evidence, and flexibility. We rely on long-term data and transparent modelling, while allowing advisers to apply judgement where it truly matters.

We believe financial planning should be about being prepared and ready to adapt