Methodology

Monte Carlo Simulation

Foresight Planner runs thousands of independent simulations of your retirement portfolio. Each simulation draws historical market returns in random order, applies your spending, taxes, and benefits, then tracks your portfolio month by month until the end of your horizon.

The result is a probability distribution, not a single forecast. You see the percentage of scenarios where your money lasted, along with the full range of outcomes.

Block Bootstrap Resampling

Unlike simple random sampling, we use a stationary block bootstrap to preserve the autocorrelation found in real market data. Returns are resampled in contiguous blocks of random lengths, maintaining the serial dependence between consecutive months.

Geographic Sampling & International Diversification

Each block also draws a random "domestic" country, the market whose returns stand in for your home equity over that stretch of history. Countries are chosen in proportion to how much usable history each one has, matching the pooled sampling in the research we build on. This removes home-country bias, so aggregate results aren't anchored to any single market's historical luck.

Your stock allocation is then split between that domestic market and a market-cap-weighted international basket of the remaining countries. Reaching beyond a single home market is the change that most reduces the chance of running out of money. International diversification matters more for capital preservation than for growth.

Historical Data Sources

Long-horizon returns come from the Dimson-Marsh-Staunton (DMS) Global Investment Returns dataset, which provides survivorship-bias-corrected total returns across global markets, with many series reaching back to 1900.

• Domestic and International Equities: Global broad-market total return indices (dividends reinvested) spanning up to a century of historical data.
• Bonds: 10-year government bond indices for each country, adjusted for coupon reinvestment.
• Inflation: CPI data for each country, used to inflation-adjust spending and benefit amounts.

Data Attribution & Copyright

DMS Global Investment Returns data covering 1900–2025.

© 2026 Dimson-Marsh-Staunton. All Rights Reserved. The Dimson-Marsh-Staunton information contained herein is proprietary to Dimson-Marsh-Staunton and/or its content providers, may not be copied or distributed, and is not warranted to be accurate, complete or timely.

E. Dimson, P. Marsh & M. Staunton, Triumph of the Optimists: 101 Years of Global Investment Returns (Princeton University Press, 2002).

E. Dimson, P. Marsh & M. Staunton, Global Investment Returns Yearbook 2026 (Zurich: UBS, 2026).

Tax Model

The built-in tax engine calculates federal and provincial/state taxes for each simulation year. It includes:

• Progressive marginal and state/provincial tax brackets specific to your region
• Treatment of country-specific tax-advantaged accounts (e.g. 401(k), IRA, RRSP, TFSA, ISA)
• Social security or public pension benefit clawbacks when net income exceeds thresholds
• Capital gains inclusion rates and cost basis tracking
• Dividend tax credit and gross-up mechanisms

Government Benefits & Social Security

Foresight Planner models public pension systems (such as Social Security, CPP/OAS, and State Pensions) complete with claim-age adjustments, incorporating:

• Reductions for early claiming and bonuses for delaying benefits.
• Pro-rated amounts based on residency or contribution years.
• Means-tested clawbacks on the highest-income brackets.
• Inflation indexing applied throughout the duration of retirement.

Withdrawal Strategies

The withdrawal strategy decides how much you spend each year in every simulated scenario. It sets your income in the first year and the rule that adjusts it thereafter. All three strategies below set that first-year amount from a withdrawal rate applied to your starting portfolio (the classic 4%), or, for Constant Dollar, a fixed dollar amount. They differ only in how spending moves over time.

Constant Dollar (the "4% rule")

This is the simplest rule. You spend the same amount every year, increased only by inflation, no matter how markets perform. Income is completely predictable, but it never adapts. A deep downturn early in retirement can drain the portfolio, while strong markets leave money unspent.

Floor & Ceiling (Vanguard Dynamic Spending)

Each year's spending starts from last year's and moves with your portfolio's investment return, but the change is capped. A ceiling limits how much spending can rise after a good year; a floor limits how far it can fall after a bad one. Both bounds are anchored to the inflation-adjusted prior year, so they constrain changes in real (purchasing power) terms. The result responds to markets while smoothing the swings.

Yale Endowment

This model comes from university endowments. Each year's spending is a weighted blend of two numbers: last year's spending grown by inflation, and a target percentage of the current portfolio value. A smoothing weight sets the mix. At 100% it behaves like Constant Dollar (pure inflation growth); at 0% it is pure percentage-of-portfolio spending. In between, income follows the portfolio but changes gradually, which makes it the smoothest of the dynamic strategies.

Interpreting Your Results

The success probability is the fraction of simulations where your portfolio never reached $0. Here is a general guide:

• > 90%: The portfolio lasted the full horizon in more than 90% of simulated scenarios.
• 75–90%: The portfolio ran out in roughly 1 in 4 to 1 in 10 simulated scenarios. Many people explore how flexible spending or a longer horizon changes this number.
• < 75%: The portfolio ran out in more than 1 in 4 simulated scenarios. You can try different savings, spending, or timing inputs to see how the projection responds.

References