Monday, December 13, 2010

The Limits of Retirement Calculators

Figuring out how much money you need to save for retirement isn’t easy. It’s no wonder that so many people turn to experts for help. For the do-it-yourself crowd there are online retirement calculators to help. Unfortunately, the precise answers we get from most of these calculators just give us the illusion of certainty.

For fun I imagined my 25-year old self using one of these calculators. A quick online search for retirement calculators landed me at Mackenzie’s RRSP Calculator. This calculator is a common type where you punch in some numbers including assumptions about inflation and returns and the calculator gives unrealistically precise answers about how much you need to save.

So now I’ll conjure up the 25-year old me to answer the calculator’s questions:

Current value of RRSP

That’s an easy one: $0.

Current RRSP contributions

I haven’t really started yet, but let’s say that I start saving $100 per month.

Years to retirement and number of years for funds to last

I can’t imagine being 65 and having worked all my life. Let’s say I retire at 55 and the money has to last 40 years.

Income required in today’s dollars

I’d like to have a comfortable retirement. On top of CPP and old age security, let’s say I’ll need $40,000 per year from my RRSP.

Rate of return

I hear plenty of people talk about making 20% or more on their stocks each year. I’d better be more conservative and just assume 15%.

Inflation rate

All the reports I hear say that inflation is around 2%.

That was easy. Now click on “Next” to see what we get. Mackenzie’s calculator comes back with a few numbers in addition to the following message:

“Congratulations! Your savings plan will provide sufficient assets to meet your needs through retirement.”

Great news! It turns out that my $100 per month savings are enough to retire comfortably at 55. My savings plan is right on target.

Of course this is all nonsense. The most important element of this “savings plan” is its hopelessly unrealistic return expectations. I assumed that I’d make about 13% above inflation each year for the rest of my life. This won’t happen. Based on these same assumptions, saving $1000 per month would allow me to retire at age 38. This is just silly.

The lesson here is to be wary of any tool that gives the illusion of precision. Decisions about retirement planning are inherently fuzzy because future returns and inflation are not known. To use these calculators correctly, you have to make conservative assumptions about investment returns.

An entirely different way to think about retirement is to invest as much as you can in the best way you can and be somewhat flexible about your retirement age and retirement income.

11 comments:

  1. Aren't you being a bit harsh on the calculators? I mean I agree that they can be a bit simplistic but still it goes without saying that you need to make realistic assumptions. If you plug 100 calories/day in a diet calculator it will show you losing a lot of weight. This isn't a shortcoming of the calculator.

    ReplyDelete
  2. @DIY Investor: My concern is for the people who use these calculators. Those who struggle with long-term financial planning (which is almost everyone) need to be careful when using these calculators. Even when they input seemingly realistic numbers the answers they get will be too optimistic. I don't care whether this is a shortcoming of the calculator or the user -- either way the user gets bad information.

    ReplyDelete
  3. good post,.... and the rates of return need to be even more conservative to be realistic since the variability of returns year to year reduces the net equivalent compound constant rate that the simple calculators assume.

    ReplyDelete
  4. The calculator is a naive tool, and thus is dangerous into fooling people they are doing enough, in this case, hate the tool, not the game.

    ReplyDelete
  5. It's the same as the old computer axiom - garbage in / garbage out.

    Calculators are just a tool, same as any other. DIY home handymen can do immense damage to their homes if they don't know what they're doing - it's not DeWalt's fault someone with more confidence than sense cut a supporting beam.

    If someone is unwilling or unable to do get the proper know-how to do their own financial planning they should be going out to look for professional help - just like someone who has never picked up a nail gun and cares not to get properly prepared should be looking for a contractor to finish their basement.

    ReplyDelete
  6. @Canadian Investor: Thanks. It's definitely important to take into account volatility. So, compound rates are preferable to average rates. But it makes sense to be a little more conservative than even long-term compound return rates because a given investor may be unlucky and run into a period with below average returns.

    @Big Cajun Man, @Anonymous: To draw an analogy, hammers are useful tools but can be dangerous. I'm not trying to ban hammers. I'm warning novices to avoid smashing their thumbs.

    @Dale: Thanks for the pointer. I'll have to check out that retirement calculator.

    ReplyDelete
    Replies
    1. The last reply above is to Dale Rathgeber's comment:

      A better free calculator is available at www.retirementadvisor.ca. It is true that the data and assumptions inputted into all calculators needs to be realistic/conservative.

      However, for those considering early retirement, this type of tool is invaluable.

      Delete
  7. This is an interesting issue. With people's tendency to over-confidence, such calculators are bound to be abused.

    An interesting feature for a retirement calculator would be for it to have statistics for what the average historical return for investors has been (perhaps 5% nominal) and the average that people input into the calculator for projections (perhaps 12% nominal). Then it could compare the user's projection with how a typical portfolio would perform. The difference would be eye-opening, I'm sure.

    ReplyDelete
  8. @Gene: Perhaps even giving several outcomes based on different historical periods in some sort of chart would be helpful. When people see the range of possible outcomes, it at least gives them a chance to see that a bad outcome is possible.

    ReplyDelete
  9. Actually, I just stumbled across a book by Jim Otar related to retirement planning. I read his work years ago in "Canadian Moneysaver" magazine, and I was always impressed.

    In the introduction to his book, he talks about using historical data to determine future projections using engineering factor of safety methods.

    The book is available free for readers of the Wealthy Boomer blog, as highlighted by Canadian Capitalist in his Friday "This and That" roundup. Here's the direct link though:

    http://www.retirementoptimizer.com/downloads/URMG/URMGreem.pdf

    ReplyDelete
  10. I guess the people managing Canada Pension Plan are using the same calculators. If they invest in something and rate of return varied between 3-5%, they will automatically assume that the investment will return 7% :)

    ReplyDelete