Thursday, August 26, 2010

Unrealistic RRSP Contribution Expectations

We hear frequently that young people should start contributing to an RRSP early in life. Recently, I encountered yet another of these arguments. However, there are some unrealistic expectations buried in the assumptions used.

Here is a typical version. If you start contributing $500 per month to an RRSP at age 25 and make an 8% return each year, you'll have about $1.7 million by the time you're 65. But, if you delay making contributions until you're 35, you'll only have about $800,000 at age 65. This is less than half as much.

The math is right. The contributions during the first decade really do count for more than the remaining three decades because of the magic of compound interest. However, there is a serious problem with the built-in assumptions.

Let's look at this from the point of view of a 65-year old who is just making his last $500 RRSP contribution and is about to retire with $1.7 million. Supposedly, he made $500 RRSP contributions each month starting 40 years ago. But what about inflation?

If we assume inflation has averaged 4% per year over those 40 years, that first $500 contribution when our retiree was 25 years old is the equivalent of $2400 today! How many 25-year olds do you know who can contribute $2400 per month (or $28,800 per year) to their RRSPs?

I'm a believer in saving money from a young age, but the typical justification of the type I described here is hopelessly flawed. Your early years of saving are important, but not quite as important as they can be made to seem.


  1. Michael,

    You have it backwards and are building towards a weak proposition. The critical piece that you're missing is the diminutive effect of inflation over time. The $2400 amount in current dollars will have a purchasing power equivalent to $500 in 40 years ($2400/(1.04)^40).

    The $1.7M end state has the equivalent purchasing power of $354,091 under 4% inflation. If you draw that down at a conservative rate of 3% or 4% annually, that provides an annualized income of $10,622 or $14,164.

    From this baseline, subtract 100 (bonds) or 200 (equities) basis points annually for mutual fund fees and then add on the mutual fund annualized returns net of fees (3-5% for bonds, and 5-7% for equities). After inflation and fees, there's some measly winnings after 40 years.

    The $500 amount monthly makes more sense than you're concluding in the article. Normally, your writing is quite sensible, but I would suggest you reconsider this one.

  2. @Anonymous: You have looked at it from the point of view of a 25-year old today. I looked at it from the point of view of a 65-year old today. If you want to look at it from the point of view of a 25-year old, then my point is that he cannot make $2400 per month contributions for 40 years in order to get $1.7 million in purchasing power at age 65.

    I believe you have made the same mistake that these analyses usually make. It makes no sense to assume a constant level of contributions over a lifetime. A 25-year might start at $500 per month today, but in 40 years, that amount will have the purchasing power of only $104. It makes no sense to assume that a person would make ever smaller contributions (in purchasing power) as he enters his higher-earning years. A 25-year old who can start at $500 per month today should be planning to increase this amount over time so that he will have more than $354,091 (in today's dollars) at age 65.

  3. @Michael: You make a great point. Really, any retirement planning calculator that assumes a person will make a consistent monthly contribution for decades is rather useless. It makes more sense to think in terms of saving a fixed percentage of your income. This accounts for both inflation and career path.

    And let's face it, as great as it is to contribute early, most of us will focus on other financial goals in our 20s and 30s and then ramp up our RRSP savings after the kids are gone and the house is paid for.

  4. Excellent analysis. I've never thought of this before but I agree completely.

    Another way to look at it from the 25 yr. old's point of view is to say that instead of earning 8%, he'll really only earn 4% (subtract inflation from the gains). If he contributes $500 per month for 40 years, he'll end up with around $590k (in the 25 yr. old's dollars).

    But if the 25 yr. old waits 10 years, and then starts contributing $500 per month for 30 years at a real gain of 4%, he ends up with about $350k (in dollars from the point in time where he was 25).

    So the difference in the methods of analysis is noteworthy. Looking at it one way, you end up with about 130% more money by starting 10 years earlier. Looking at it the other way, there's only a 70% advantage.

    I guess the key points for this type of analysis are:
    a) coming up with a realistic rate of return you can expect from your portfolio
    b) use expected real rates of return, since purchasing power is all that matters in retirement, and you'd expect contributions to (at least) keep pace with inflation.

    In fact, it's probably realistic (as you suggest) to expect contributions to out-pace inflation. As an individual gets a mortgage paid off and/or children leave home, they are likely to have more money available to save.

    As a counterpoint though, for most people out there I expect the last thing they need to hear is a good argument why it isn't such a big deal to save much in your early adult years. :-)

  5. @Michael,

    Great reply. Hopefully, we can shift to the critical parts for anyone else reading this discussion. [BTW, not in the industry; just have a math and finance interest]

    Whether you look forward or backward is relatively moot; that's just time-shifting. Your assertion of lowering contribution start points is, to reiterate a point, a weak proposition.

    For the past 20 years, I have had a modest income and started putting away a minimum of $5000/annum which has varied upward depending on employer contributions. I lived independently since age 18 and bought a first home at age 30, and had a family with my wife staying home for most of the child-raising period. Your proposition is now 20 years in the future of my starting point, but advocates delaying cumulative contributions and advocates a lower start point than my personal example. At a 4% annualized inflation rate, good luck. I've lived the counter-example to your original article advocating 1) delay and 2) reduced contributions at an early age. Initial starting points were extremely important and set the conditions for success. I have the first million in common stock securities along with a furnished 4500-square foot house with a very modest mortgage, with a substantial 20-year road ahead.

    To recap, the strong proposition is based on setting good/better/best starting conditions (larger initial contributions), paying attention to things that diminish returns (e.g. mutual fund fees), and sticking to your knitting (contribute more if you can annually) through all the ups and downs of each year. The start point is the most critical of the three since it sets habits, shapes the end-vision, and rewards with success.

  6. @Canadian Couch Potato: You make a good point about basing the saving rate on income. The actual percentage may vary a little over time, but trying to keep this percentage stable is a sensible goal.

    @Returns Reaper: You're right that there is an advantage to starting early. It's just that this advantage isn't as large as the flawed analysis makes it out to be. I definitely don't want to send the message that young people don't need to save.

    @Mark: In today's low inflation environment, you're right that expecting 8% returns is likely unrealistic. However, a real return of 4% over an extended period seems more reasonable.

    1. The last reply above is to Mark Wolfinger's comment:

      And don't ignore the fact that an 8% return is extremely optimistic.

      It's numbers such as those that placed local governments into such debt.

      Pension obligations can never be met due to underfunding and optimistic expectations.

  7. @Anonymous: I think we are arguing about different things. My claim is that most people are in a position to make larger RRSP contributions through the middle of their careers than they are at age 25 due to higher real incomes and inflation. This may not be true for a small minority (possibly including you personally), but it is true for most people. The typical simple analysis of a lifetime of RRSP contributions has the hidden assumption that contributions remain constant in nominal dollars for decades. I don't think this makes sense.

    If your point is that it is possible for young people to save substantial amounts, I agree. My wife and I paid off our first house in 4 years. Paying off a mortgage isn't exactly the same as saving, but it is similar.

  8. Great analysis Michael. It seems like a 25 year old would be better off paying down debt and starting a TFSA, especially if he/she has a lower marginal tax rate. Save the RRSP room for when you're older and (hopefully) have a higher income and the tax rate to match.

  9. @Balance Junkie: You're right that a TFSA is often a better option when you income is low.

    @CC: Your comment reminds me of when I was a kid and I asked my grandfather if he had earned a million dollars in his life. I reasoned that if he had earned $20,000 per year for 50 years, that would make a million. He tried to be nice about it, but he basically laughed himself silly. I don't think he earned $20,000 total in his first decade of work. Dollars were much bigger and people earned fewer of them back then.

    1. The second reply above is to Canadian Capitalist's comment:

      These popular articles that ignore the time value of money drive me nuts. Yes, investment returns grow over time but over that same time period inflation is relentlessly eating away its value. Also, investment returns are lumpy and luck plays a large role. That's not to say one shouldn't save but one has to have realistic expectations too.

  10. I would argue that saving a constant percentage of your income will only be effective if the rate is high enough ie 10%.

    I think it's more realistic that younger people will save a smaller percentage than older people - they make less money and have more demands (ie kids, first house etc).

  11. Good point, Michael.

    I especially enjoyed the comment you wrote about your grandfather. It reminded me of something I read (I think it was part of Zweig's commentary in the Intelligent Investor.

    It was a quote from a stand-up comic, I think, who said "Kids are getting stronger these days. 30 years ago it would take two adults to carry home $25 worth of groceries... now any 12 year old can do it."

    I paraphrased from memory, so am not sure on the numbers, but the point is the same.

  12. @Money Smarts: I agree that the saving percentage has to be high enough (particularly through higher income years) to make a difference. Saving 1% won't do much good unless you make millions each year.

    @Myke: That quote from a comic is a good one. I'll have to use that sometime.

  13. This is an interesting post, and the comments add a lot to the argument and counter-argument as well. Anonymous's comments helped to flesh out some salient points, and I admire his savings prowess.

    It's an intriguing question as to what a 25-year-old investor should be able to save. In my experience, my savings rate was high in my pre-marriage years because I had a decent income, cheap rent, and not much to spend money on. This was a great time to start building a nest egg.

    I would be tempted, if I had a blog, to write a post along the lines of "The 40-year-old Virgin: The Path to Investment Success".

  14. @Gene: Your "40-year old virgin" idea sounds like a conspiracy to keep others from saving money :-)

  15. I also have to disagree with your conclusions and think you are looking at it backwards. It's not that the investor would need to contribute $2400 a month at 25, but rather his $500 a month contribution would have to at least keep up with inflation during the next 40 years.

    You also assume that the investor, although maintaining an absolute dollar amount ($500), would actually be decreasing the value of their contributions with time due to inflation. If you are going to assume inflation, then you should realistically also assume that the contributions per month increase by the rate of inflation each year, e.g.: assuming 4% inflation:
    25 - $500 a month
    26 - $520 a month
    27 - $540 a month etc.

    Now starting at 25, with 4% inflation and 8% return, the portfolio at 65 will be worth ~$2.8 million.

    If the investor started at 35, they would begin contributing at $740 a month ($500 at 4% inflation for 10 years). At 65 his portfolio would be worth ~$1.7 million.

    A difference of ~$1.1 million in the final portfolio value, with the difference in contributions being ~$72,000 in absolute dollars (~$315,000 in actual value at age 65).

  16. @Xenko: As the saying goes, I think we are in violent agreement. My point is that it makes no sense to assume that contributions will remain constant for a lifetime. This is the same point you are making. It doesn't matter whether you look at it from age 25 looking forward or age 65 looking backward, the conclusion is the same: contributions should increase over time.

  17. I agree that it makes no sense to assume that contributions will remain constant for a lifetime, but I still think these illustrations are helpful. The point is that they demonstrate the importance of compound interest and starting early.

    Reading similar material in the Wealthy Barber at 15 the lesson I took was contribute as much as possible as early as possible. I started from $50/month into an RRSP in high school to $1000/month into an RRSP and TFSA by 22. Once you understand the importance of compound interest and starting early it follows that you should continue to maximize your investments by increasing your contribution.

  18. Michael,

    Make it even simpler so it can be more easily understood.

    To begin with, assume returns from investment are exactly the same as the inflation rate - I know this is generally not true but it helps illustrate the point I would like to make. $500 now might grow to $1000 in x years, but with the investment returns exactly matching the inflation rate, then the $1000 in x years will have EXACTLY the same purchasing power as $500 now. i.e. they cancel out.

    So if they cancel out, then you may as well just talk in 2010 dollars for the entire exercise by assuming that the inflation rate and the return from investment are 0 for the 40 year period. So how much will 500 a month for 40 years accumulate to? A very simple calculation: only $240,000 in 2010 purchasing power. That's all.

    But that is only the starting point since investment returns GENERALLY do exceed the rate of inflation. By how much? 2%? 3%? And what about taxes? And, of course, there have been long long periods in history when investment returns have been LESS than the inflation rate so what happens if the 25 year old has the misfortune to live in one of those eras? It all gets complicated doesn't it but I'll leave it to others to work out the required adjustment under whatever assumptions they choose.

    but put another way, the point is that the $1.7Million in 40 years time has to be discounted back to 2010 dollars. And when you do that, it ain't going to be able to buy you anything near $1.7M worth of goods in today's money. Something over $240,000? Almost certainly. But nothing like the fanciful $1.7M quoted.

  19. The canonical example of "saving $500 per month" just needs to be refined to "saving $500 in todays dollars per month".

    That way, whether you projecting into the future (current 25 year old), or looking back at the road already travelled (current 65yo), it's clearly implied that you need to index that $500 (and the accumulated total) by the inflation rate.

  20. @Mark: I agree with your conclusion that the $1.7 million is misleading if it is in 40-year old dollars. If it is in present-day dollars for the 65-year old, then the contribution amount for the 25-year old is unrealistic.

    @Young and @Anonymous: I agree that these illustrations can me motivating. Perhaps they should take the idea of Anonymous and do the calculation in present-day dollars. The result would be less dramatic, but it would still show that saving when young is a good thing.

  21. I also think there is a lot of marketing of the pipe dream that the average middle class Joe, contributing $500 a month for 40 years is going to end up with millions of dollars for retirement; marketing done by the Government, who are undoubtedly crapping their pants about paying Old Age Pensions out to baby boomers and do not want the public to panic, and by financial institutions and investment specialists, who want to reap a hefty cut off the mutual funds fees etc. By making the monthly contribution sound reasonable, and the reward mind bogglingly good, lots of people hop on the RRSP bandwagon only to be hit with the realities of all the previous posts- contributions are not constant because of inflation, rates of returns are not guaranteed, and half way through the 40 years, they do not have anywhere near half the money promised.
    Investing young is undoubtedly the smart move, but for most people they will never achieve the promised nest egg or anything close to it.