Last week, my family and I took a trip to Halifax, Nova Scotia, Canada. We drove. (This, by the way, accounts for the missed blog last week and the delayed one this week). Many stretches of the interstate (and interprovincial) highways were lightly populated, and rest stops and “oases” were often far apart. It was tricky—particularly with a restless child on board—to predict our stops. On the one hand, we did not want a stop to come too quickly after the previous one because more stops altogether would stretch out the already-long 26 hours of straight driving that it takes to get to that Canadian city from Chicago’s southwestern suburbs. On the other hand, what can you do when you pass a possible stop only to hear 10 minutes later that your child has to go and then discover that there won’t be another opportunity for another 80 miles (120 km)? After a while, we got better at running these two calculations simultaneously and interpreting them together to find the sweet spots for periodic stops, if possible, with a Tim Horton’s at them. By the time we returned (by way of Montreal, adding three more hours), we became skilled at this art.
Using most financial planning software requires a somewhat analogous capacity to approach a question via two different methods of calculating, each with its own emphasis and bias. Because of this, two methods may indicate very different measures to take—from very modest to very drastic—to address the same financial questions. It is especially important to be clear on both the emphasis and the bias when, as is typical today, the client may also access the software and modify the input and the results or certain functions. It is up to the planner to help the client understand how one calculation may overstate an issue and another, simultaneously and using the same raw data, understate the same issue.
Here’s what I mean. A key objective for pretty much every financial planning client, including those who come to me for disability-related or for special needs planning, is to get through retirement without the money saved for retirement running out. Some clients want to time it just right so that they are writing their last check on their deathbed, and some want to make sure that they can leave a legacy for future generations. But no one wants to see their accounts drained by their age 75 when their life expectancy is 87 at least. Most people also realize that unexpected expenses may arise and want to have a bit of a cushion.
Right Capital, which is the planning software I use, looks at retirement security from two perspectives. The first uses cash flow. Based on client and planner data inputs for expected income and expenses, the software calculates cash inflow, outflow and the net amounts for each year until the end date of the plan. The net inflow or outflow each year serves to increase or decrease the client’s projected asset value from the actual figures input at the start of the planning process. These cash flow calculations are “straight line” meaning that the same increase factor is applied each year. Perhaps the client’s salary and general expenses increase at 2%. That’s 2% every year, not 1% one year and 3% another year for an average of 2%. Likewise, investments may be projected to return 6.5%, and while there is some ability to let that rate vary over time, it can only vary as a rate of increase. The investment return projection cannot incorporate an occasional year to year decrease within the cash flow calculations. If the client takes a look at the projected final year of the plan, s/he will either see that there are assets remaining, in which case the plan would appear to be successful, or see that there are none, in which case it would seem a probable failure.
But you don’t have to be a professional financial planner to know that that reality does not run in a straight line. It is highly improbable that even the same company will give exactly the same level of raise each year. It is clear that the rate of inflation varies from year to year for both general expenses and specific categories of expense like health care and tuition. We were reminded rather spectacularly this past week that the stock market and hence equity investment returns can go down as well as up, and even rather less voluble bond prices and yields can move in unexpected ways. To take that into account, Right Capital, like many planning software runs what is called a “Monte Carlo” set of simulations. Basically, the software runs 1000 scenarios in which it randomly changes variables from what the planner and the client have input as expectations. For example, a scenario might incorporate 10% inflation. Another might look at 0.5% inflation. One scenario might have the stock market blasting ahead with a spike of 20% growth, while another might include a steep drop in the manner of 2008. All of these calculations are behind the scenes. Unlike the cash flow projection, in which the client and the planner can see all the figures each year, one can only see the results of the Monte Carlo simulations expressed as a probability. If the client is expected not to run out of money in 650 of the 1000 scenarios, then s/he has a 65% probability of success.
The Monte Carlo is a much more conservative approach, since it can account for many negative scenarios, unlike the cash flow, which can account for only a few, such as the cessation of income or the addition of long-term care expenses. Ideally, then, one could aim for 100% success according to the Monte Carlo probability calculations, and it is these calculations that should take precedence when a client has adequate income and assets to make getting to that 100% realistic. On the other hand, if the client has less to work with, s/he may be discouraged by focusing on a less-than-100% probability, which is presumed to indicate failure or the fact that s/he would have to increase income or decrease expenses by 50% to get to that 100% target. I have seen plans in which the straight-line cash flow predictions show the client dying with $2 million, but the Monte Carlo shows the probability of success at only 40%. This happens when a client’s cashflow is strongly dependent on one particular aspect of income or expense that that piece of income, or more likely expense, is very sensitive to a small change of its rate of increase.
It is important to remember that real life does not move in a straight line. It is likewise important to remember that the Monte Carlo scenarios are only hypothetical, and some of them will not happen, and some will happen with lesser impact. It is finally important to remember that the plan itself and the strategies derived from it will be updated every year as both inflows and outflows go from projected to actual to historical. These periodic adjustments will help the client hone in on the happy medium of how much to save and how much to spend in the same way that our family was able to find a happy medium between stopping every time we saw a Tim Horton’s and condemning ourselves to a painful 40 miles (55 km) sitting with tightly crossed legs in the car.