It's easy to get the odds your retirement will be ruined wrong
Retirees are often told the odds of running out of money during their lifetime.
Advisers will tell them, for instance, that if they withdraw 4% per year from their nest egg, there’s a good chance they’ll be able to fund their desired lifestyle over the course of 30 years. But if they withdraw more, the odds of outliving their assets grow; they might have only an 85% probability of meeting all their life’s financial goals.
But retirement researchers are beginning to question the value of what they refer to as "probabilities of ruin" in retirement planning, and say that such measures ought to be, well, retired. In the March/April 2016 edition of the CFA Institute’s Financial Analysts Journal, Moshe Milevsky, a York University professor, made that very case.
Q: What is the "probability of ruin," and why has it become a popular way to help retirees take stock of their retirement plan?
Milevsky: Think of the probability of ruin like a weather forecaster giving you the probability of rain. It’s a real easy and intuitive number to grasp, between zero (i.e., no rain) and 100 (i.e., rain for sure). The same idea has been applied by the financial planning industry to inform clients about the chances they will meet their retirement-income goals. It’s reported as a number between zero and 100. Zero means that you will run out of money for certain and 100 means that you will be perfectly OK. That is the basic idea.
Q: You expressed concern that these probabilities are being used (and abused) to both reassure and scare retirees about the viability of their plans. Why so?
Milevsky: Well, think back to the weather analogy. A 100% probability of rain could simply be a light drizzle (for which you might not even want to take an umbrella). Or, a 100% probability of rain could come with tornado-level winds and hail. The probability number doesn’t really mean much in and of itself. It must be accompanied with more information about magnitude.
Q: What are the problems associated with using failure probabilities?
Milevsky: I’ll mention them briefly, using the rain or weather analogy, and those interested can read the entire article here. Also read financial blogger and adviser Dirk Cotton’s take on the subject, too.
If the probability of precipitation is 40%, will you take an umbrella to work? How about 70%? Or what about 20%? What number is the point at which you say, “Jeez, it’s going to rain. I’m taking the umbrella.” It's personal. There really is no consensus in the financial-planning industry around an acceptable probability of failure.
As I mentioned, the number itself might not capture the magnitude of how bad the rain will be. Imagine a 15% probability of a rain (not bad), but if it does rain it will be 10 inches. (Wow, scary!)
The only way to be able to forecast these things is based on prior data or history. There simply isn’t enough history to be able to forecast these things with confidence in the context of financial markets. Yes, we have had rain for millennia, so the data is better. Not so with stock markets.
The models or computer algorithms being used to generate these forecasts aren’t very transparent or widely accepted. Unlike meteorology, there really isn’t a scientific consensus on how to build these very long-term forecasts. In fact, while the weatherman might feel quite comfortable forecasting the weather in the next few days, it would be ludicrous to do this for some date 10 years from today. Financial-planning models are trying to forecast the “financial weather” in 50 years. The financial world isn’t old enough.
Because people like the idea of 100% safety and security, investors often fall into the trap of buying insurance products that are marketed as 100% secure, when in fact there are risks there as well. Yes, if you stay inside all day the probability you will get wet from the rain is zero, but if the rain is really a hurricane and your home is made of hay, you will get soaked.
Q: What are some better ways to communicate the viability or sustainability of a retirement plan so that people understand it?
Milevsky: What I’m proposing is that financial advisers use simpler language to help individuals plan their retirement. I like the analogy to medical life expectancy. A doctor can tell you how long you can expect to live in retirement, somewhere between 15 or 30 years. Likewise, a financial adviser should be able to compute the “longevity” of your portfolio in years. So, next time you have a conversation with your financial adviser, ask them to compute the longevity of your portfolio — in years.
Robert Powell is editor of Retirement Weekly, contributes regularly to USA TODAY, The Wall Street Journal and MarketWatch. Got questions about money? Email Bob at rpowell@allthingsretirement.com.