LISTEN: mp3 audio (12:12 min)

In order to fully appreciate the impact of computers, you have to remember what life was like before them. In short, you have to be older than 60 – and remember a world where most receipts were handwritten, bankers calculated monthly loans from interest rate books in their desks, and complex mathematical calculations were performed on slide-rules.

It is impossible to overstate the degree that computers have revolutionized the use of numbers in everyday life in the past 50 years. Complex calculations have become commonplace, and lightning fast. Today, the only limit on the ability of computers to process mathematical information is the speed at which data can be entered.

However, for all their processing capacity, the usefulness of computer technology is dependent on a higher order of processing. Human intelligence is required to sift, sort and make sense of the vast array of information now available.

And while computer processing continues to expand exponentially (i.e., Moore’s Law), it can be argued that our ability to use it effectively is not keeping pace. Rather than providing insight and direction, the additional information is often making it harder to find the right answers.

An illustration of how human intelligence is trailing computer processing can be found in the use of computer programs to project retirement scenarios. Today, in just a few minutes, online calculators can deliver detailed reports that 20 or 30 years ago would have taken a team of math experts several days, or even weeks to produce.

But just because a lot of information can be generated in a hurry, doesn’t mean it will actually help you reach your financial objectives. It’s not that that the math doesn’t add up; sometimes the information simply isn’t relevant.

The Monte Carlo model: Great features, but…?
Ask a long-time financial professional to describe the earliest retirement calculators of 20 years ago, and the answer usually goes something like this: The advisor met with the client, and together they established several parameters for projecting the future. These variables typically included an annual deposit amount, an estimated rate of return, how many years until retirement, and how many years of estimated retirement. Using these four variables, it was possible to project an accumulation amount that would be available at the onset of retirement, and how long it would last. The next level of sophistication added projections for inflation and taxes, and perhaps included projected Social Security payments as well.

While these variables were believed by both the advisor and client to be realistic, they were nothing more than educated guesses about the future. Further, these early calculations were static – the amounts deposited remained the same each year, as did the rates of return and other factors, like taxes.

To reflect the fluctuating nature of investment performance, computer programmers began in the past decade to build uncertainty models into their calculations. Instead of one projected result, these retirement programs attempted to show a range of possible outcomes, and identify which outcomes were most likely to occur. These analytical programs, based on probabilities and incorporating future uncertainties, have infinite variations, but are commonly referred to as Monte Carlo programs.

If you type in the phrase “Monte Carlo calculators for retirement planning,” Google will deliver almost 50,000 entries. You can find free customer-friendly calculators provided by the largest financial institutions, or custom-designed models built by academics. Some deliver an answer by filling in a 5-question survey, others require more in-depth participation. Monte Carlo programs are everywhere, and working from the data that is inputted, Monte Carlos can tell you a lot of things.

A Monte Carlo can tell you the historical likelihood of achieving your objectives, usually expressed as a percentage (“historically, you have a 78% chance of reaching your objectives”). It can provide a “date of ruin,” i.e., when your money runs out. It can help you modify your results by changing your variables (adding more money, assuming less investment risk, estimating a lower inflation rate, etc.). In theory, this information should give you some reference points for making your financial decisions. Do you need to save more (or less)? Should you adjust your accumulations for more (or less) risky financial vehicles?
But while this information may be helpful in getting an individual to focus on the task at hand and take action (a good thing), there are several inherent flaws with Monte Carlo calculators.

Educated Guesses Are Still Guesses
Much of the data that generates a Monte Carlo calculation are guesses. The resulting projections must be guesses as well. The disclaimer at the bottom of a popular online Monte Carlo program candidly acknowledges this:

IMPORTANT: The projections or other information generated by the (company) Investment Analysis Tool regarding the likelihood of various investment outcomes are hypothetical in nature, do not reflect actual investment results, and are not guarantees of future results. The simulations are based on assumptions. There can be no assurance that the projected or simulated results will be achieved or sustained. The charts present only a range of possible outcomes. Actual results will vary with each use and over time, and such results may be better or worse than the simulated scenarios. Clients should be aware that the potential for loss (or gain) may be greater than demonstrated in the simulations.

It’s almost like a disclaimer from a psychic hotline: This service is for entertainment purposes only. The numbers generated by the Monte Carlo program might be interesting, but please don’t think they are accurate.

Individual Results Cannot Be Derived From Large Numbers
Monte Carlo projections are based on the statistical histories of large groups of people and investments. These piles of data generate things like probable life expectancies and average rates of return. But you are not a large group, and neither is your money. You are an individual that holds specific financial assets, and as an individual it is impossible for you to replicate the projections a Monte Carlo program might generate. Carl Berdie, CLU, author of “Insure or Invest: Which is Best?” in the November 2009 trade publication Life Insurance Selling says this distinction between individual results and group averages is another reason Monte Carlo projections miss the mark:

“It’s impossible to die exactly at your life expectancy. If you plan at age 60 for 20 years of additional life, when you are at age 80 (when 60-year old males are supposed to die), you have a life expectancy of about eight more years. At 88, you have almost five more years to live. Thus the life expectancy model is a moving target that will never be accurate. Life expectancy is based on the law of large numbers and can’t be brought down to the individual level.”

If the variables used to generate Monte Carlo projections are “moving targets” in constant flux, then constantly fluctuating projections aren’t really worth much. It’s like having a wobbly sight on a gun; you can’t shoot straight if your view of the target is constantly out of focus and shifting.

Answering The Wrong Questions
James S. Welch, Jr., is a “designer and implementer” of computer software programs with 50 years of experience. He is listed as the principal architect of a free, online, “alternative” retirement calculation program that claims to resolve the shortcomings he sees in Monte Carlo programs. In an article titled, “A critique of Monte Carlo Retirement Calculators,” last updated October 3, 2009, Welch offers this assessment:

The most serious problem is that conventional Monte Carlo retirement calculators answer the wrong question. The retirement question they attempt to answer is:

• When will the money run out?

The relevant question is:

• How much money can I spend each year so that my money will last a lifetime?

At first, Welch’s comments may seem like a simple matter of semantics. But it also reflects the “sift, sort, and make sense” function that must accompany the processing of information. If the data doesn’t answer the right questions, configuring it is a waste of time.

Making Technology Work For You
In the area of personal finance, computer technology can sometimes be the tail wagging the dog. We get so excited by all the new things we calculate, illustrate and collate that it makes us giddy. But step away from the distraction of eye-catching pie charts and one-page plan summaries updated in real time. Suppose someone told you the best way to prepare for retirement is to make some guesses, evaluate those guesses using averages derived from other peoples’ experience, then accept answers to questions you are not asking. Would that make sense? No.

But the problem isn’t with the numbers or the calculations. It’s the philosophies and assessment procedures that need to be fine-tuned.

You don’t want to feel like a number. You don’t want to project your future on mere guesses. You want strategies that will work for you, not ones that have a 75% success ratio with other people. And you want an approach that addresses your financial objectives, instead of a pre-determined list. To do those things, you need good sift-sort-and-make-sense intelligence – either from yourself or your trusted advisors.

Today, every financial professional has access to great computer programs. But who can handle the critical human elements that will ultimately make the difference in reaching your financial objectives? It’s the human element that determines how well technology works for you.

WANT TO MAKE TECHNOLOGY WORK FOR YOU?

GET THE INTELLIGENCE YOU NEED TO CONTROL THE PROGRAM BY TAPPING INTO OUR “HUMAN RESOURCES.” WE HAVE THE PERSPECTIVES TO MAKE SENSE OF THE MATH.