Monte Carlo Simulation - Spending Policies

A retirement planner that’s a little more flexible

flexibleRetirementPlanner.com

DocumenTation — Spending Policies

Traditional retirement planning tools typically fix annual retirement spending, adjusting it only for inflation. A significant difference with this tool is that it allows the user to try other approaches where the program “adapts” and adjusts spending as the retirement plan unfolds (rather than waiting until the portfolio is almost gone). The next few paragraphs attempt to provide very specific details about what the simulation does when each of the supported spending policies are selected.

The retirement planner supports three retirement spending policies. These are Stable, Flexible, and Conservative spending policies. When the simulation is run, the selected policy controls how the simulation attempts to fund the retiree’s requested spending each year as the plan unfolds.

The “Stable” spending policy causes the simulation to attempt to fund 100% of the requested annual retirement spending. This is done by withdrawing the full amount of expenses that are not covered by retirement income from the retiree’s portfolio. The annual spending amount is automatically increased each year by the inflation rate so that the retiree’s spending level keeps up with inflation. This spending policy most closely matches the spending policy that’s implicit in most other retirement planning tools. Note that even with the Stable spending policy, it is still possible for the percent of expenses funded to fall below 100% in a given year. This can occur as a result of the portfolio running out of money in some trials of the simulation, thus reducing the percent of expenses that are able to be funded. When the data for all runs is averaged together, these failed years cause the average spending to fall below 100%.

The “Conservative” spending policy attempts to model what a conservative or cautious retiree might do as their retirement unfolds. Basically, in any year where the portfolio shrinks and the portfolio balance is smaller than it was at the start of retirement, this policy withholds the cost of living increase (cola) for that year. This means that the retiree must cut back their spending. Otherwise, if the portfolio is bigger than it was at the start of retirement, some extra purchasing power that was lost in previous years is restored. However, the retiree never gets more than 100% of the expenses (adjusted for inflation) they requested, no matter how well the portfolio does. Finally, if the portfolio is growing, but has a smaller value than at the start of retirement, the percent of expenses funded is held steady (retiree gets a cola, but no extra increase).

Next, the “Flexible” spending policy extends the conservative policy by allowing the retiree to spend more than originally specified in the plan if their portfolio does well. In lean years, this policy mimics the conservative policy by withholding the COLA when the portfolio is shrinking and the balance is smaller than it was at the start of retirement. If the portfolio is growing, but is smaller than it was at the start of retirement, the percent of expenses funded remains constant (retiree gets a cola, but that’s it). However, following good years when the portfolio has a balance that’s greater than the starting balance, the flexible policy allows the percentage of expenses funded to grow well above 100%. The amount of spending growth depends on the relative size of the portfolio compared to the balance at the start of retirement. If the portfolio is at least two times the size it was when retirement started, spending percentage is increased by the inflation rate. If the portfolio is between one and two times the original size (at retirement start), an increase of 1/4 of the inflation rate is given. These thresholds were chosen somewhat arbitrarily and is still subject to further adjustment. However, simulation results aren’t hugely sensitive to small changes in these variables in otherwise successful plans.

The spending policy does not begin to affect the amount of spending that gets funding until the first year of retirement. Any spending that is specified prior to the first year of retirement will be subtracted from the portfolio without any spending policy adjustments.

There are some additional spending policy and withdrawal settings that you can adjust by clicking on the “Config Button”. When you click this button, a popup window appears that allows you to modify the additional settings.

In the first section of the popup window, there’s a Spending Policy Multiplier, this is a parameter that is multiplied by the annual inflation rate whenever the program makes annual adjustments to the percent of annual spending that gets funded. This has the effect of magnifying the results of applying the flexible or conservative spending policy. When the sensitivity is set to 1, only the annual inflation rate is used in making adjustments. The next two parameters are the floor and ceiling values for the percent of expenses to fund. With these settings, you can control the upper and lower bounds for the percentage of expenses to fund. This will constrain how much the spending policies can deviate from your requested annual spending. For example, setting the floor to 75% will assure that in all simulation runs, the minimum amount of spending that will be funded is 75%. The only reason the funding can drop below this floor is if the portfolio runs out of money.

The next section of the Spending Policy Configuration window has what are called the “Back to work parameters”. These parameters control a capability in the planner to model what would happen if you return to work when a dramatic drop in your portfolio’s value occurs. Some research indicates that returning to work for a few years following extended periods of portfolio underperformance can have a big impact on your retirement plan’s chances for success. To activate this capability, click on the checkbox. The Portfolio Value Trigger is a threshold value that controls when the back-to-work scenario is invoked. In any year following retirement, if the portfolio value falls below this percentage of its value at retirement start, withdrawals from the portfolio are suspended for the number of years specified in the Return to work duration setting. Next, there’s a Max age to work setting to limit the age you might have to work until. The results table at the bottom of the main display shows the probability of having to return to work based on this capability.

There are two caveats about how this back-to-work feature works. The first is that only one back-to-work event will occur in each simulation iteration (for the duration you specify) no matter how many times the portfolio dips below the threshold. The second thing to be aware of is that if the back-to-work portfolio threshold is crossed in a year when your age + the back-to-work duration is greater than the Max age to work, the back-to-work logic will not be invoked.

Next, There are controls to configure the order that portfolio withdrawals are taken from various account types. Withdrawals are always attempted from taxable funds first, thus letting tax deferred or tax-free funds grow unhindered. However, you can select whether the simulation attempts to withdraw from tax-free funds or tax-deferred funds next.

A final option can be configured to cause the simulation to take required minimum distributions from IRAs and other tax-deferred accounts. When this option is selected, the simulation will withdraw the minimum IRS specified withdrawal percent each year after you reach the age of 70 (in the simulation), based on IRS published life expectancy tables. The funds from the withdrawals are used to pay expenses (and taxes) and any remaining funds are transferred to the taxable portfolio, after taxes are deducted.

Next Page >