How does Income Lab create its default capital market assumptions?

Firms and advisors specify their own capital market assumptions (CMAs) for use with the Traditional Monte Carlo and Regime-Based Monte Carlo analysis methods. To assist with this process, Income Lab provides default capital market assumptions that reflect historical averages for asset class indices. Default CMAs are updated at monthly at the same time that the application updates implemented plans (plans that are set to be monitored automatically) and other data. That means that plan results may change over time if the plan uses application-default CMAs, due in part to changes in CMAs. 

If you use your own custom CMAs, these values will remain the same over time unless you edit them. Application default CMAs are provided only as a convenience. They do not represent the opinions or financial advice of Income Lab. Users should use CMAs that represent their own analysis preferences.

Default CMAs are produced formulaically, as described below. They may be lower or higher than overall historical returns, as indicated in the software's historical analysis chart or the Retirement Stress Test. Default CMAs may also be lower or higher than your preferred assumptions and higher or lower than the returns that someone experiences in the real world.

Traditional Monte Carlo Default CMAs

The Traditional Monte Carlo analysis method uses one set of CMAs to model all periods. With two exceptions, default CMAs are the means, standard deviations, and correlations among asset class returns over the most recent 50 years. Emerging Markets Equity capital markets assumptions reflect data since 1987. Inflation assumptions reflect CPI-U inflation from the last 30 years.

Regime-Based Monte Carlo

The Regime-Based Monte Carlo analysis method uses two sets of CMAs: Near-Term and Long-Term. By default, Near-Term CMAs are used to model the first ten years of a plan, and Long-Term CMAs are used to model all other years. However, users can adjust the length of the near-term period that they use when applying Regime-Based Monte Carlo analysis.

Just as with the Traditional Monte Carlo analysis method, default Near-Term and Long-Term CMAs reflect historical means, standard deviations, and correlations among asset class returns. However, to produce distinct near-term and long-term defaults, we filter the set of 30-year rolling historical return periods to include only one-third of such periods that are most similar to the present in economic terms (e.g., those periods with the most similar inflation rates, etc.). From this filtered set of historical returns, we derive near-term assumptions based on the first 10 years of each period and long-term assumptions based on the final 20 years of each period. When producing default Regime-Based CMAs, we also apply a ceiling to ensure that no average return assumption exceeds 0.1 standard deviations above the Traditional Monte Carlo default average return for that asset class. For example, if the Traditional Monte Carlo return assumption for an asset class is 5% with a standard deviation of 10%, the Regime-Based default average return will be no more than 5% + 10% * 0.1 = 6%.

This process yields two sets of default CMAs: one for the Near-Term and the other for the Long-Term. It is possible, even likely, that neither of these sets of CMAs will match the default CMAs for the Traditional Monte Carlo analysis method.

Translating between Monthly and Annual

Wherever possible, Income Lab uses monthly data. (The exception is tax calculations, which are annual since taxes are based on annual income.) There is no perfect way to translate between annual and monthly return statistics. We use the following formulas for translation.

Keep in mind that though CMAs are entered and displayed as annual numbers, they are converted to monthly values for use in many calculations. If you have monthly capital market assumptions you would like to enter into Income Lab, please annualize them using these formulas.