Understanding Monte Carlo

There's a fascinating link between financial planning and other technical professions—such as engineering, aerospace, manufacturing—and I bet most people don't know anything about it.

It's the fact that we sometimes use the same methodology to do statistical analysis, referred to as monte carlo simulation. It attempts to measure the probability of a certain outcome, and in financial planning that's typically whether you'll have enough money to live the life you expect in retirement. As it relates to your money, it's especially important to understand because the less accurate, straight-line assumption of returns is still common in our industry.

The difference? Monte carlo runs multiple versions of a scenario, adjusting the parameters each time to simulate the random nature of what might actually happen. The end result indicates how many of the runs turned out favorable. So, for an oversimplified example, let's say you need $3,000 every month in retirement, and you're retiring now with $1,000,000 across several different kinds of savings and retirement accounts. Along with estimated returns and volatility assumptions (and many others), the analysis tries playing out the rest of your life many different times (we use 3,000 simulations for each analysis) and reports back how many of those scenarios worked out in the end. The model chooses different kinds of market returns for each run, both positive and negative.

MonteCarloOddsofSuccess.png

You'll end up with a likelihood of success: Maybe 76%, for example. With that, you can then make informed decisions about how much to spend or how much longer to work. You might see on additional tests that spending $2,000 each month in retirement instead of $3,000 increases your odds of success to 95%. Or maybe retiring in 2 years provides similar results to a scenario where you retire in 5 years and you consider retiring earlier.

Additionally, you can see estimates of future wealth to assist in crafting a more informed plan. In this sample image, the analysis indicates a median portfolio value of $5.8 million in 2048. Depending on how confident the client is in their assumptions, they may find satisfaction in the plan and get to work.

And another great feature is the ability to think through the worst case scenario. The Shortfall analysis that our monte carlo software creates indicates the median outcomes for the bad simulations only so that you can put in context how extreme the poor outcomes are and whether you might need to make adjustments to reduce their chances.

Ultimately, the real goal in evaluating monte carlo results as it relates to financial planning is to know and understand your trade-offs: should your investment allocation be more or less aggressive, should your spending be higher or lower, what if social security is lower than expected, what if your pension goes away, and other considerations. There's no better tool to help you.