What rate of return should you assume for your retirement plan?

Every retirement plan requires making some key assumptions. One of those important assumptions is the rate of return you are going to get on your investment portfolio.

Over on MapleMoney, some of my fellow bloggers, Nelson and Robb, fostered a healthy debate about what the right rate of return should be. Both articles are a great read:

In fact, I liked the topic so much I thought I should give my two cents on the issue.

Using past performance as a benchmark for future performance.

It’s nearly impossible to predict the future so more often than not we use past performance to try and create some assumptions for the future. In Nelson’s article, he sites some data on the S&P 500 during some of the worst periods in history.











Despite that history, the worst return is only 8.67% which is why he likes to use 8 % to 10% for his personal projections.

The problem with using historical returns is you can come up with whatever number you want depending on what time frames you use. I call this snapshot performance. Snapshot performance can be a real problem in trying to use past returns to predict future returns. A good example of this is the 10 year period from 2000 to 2010 where stock market returns were not any better than low interest GICs.

Do people really hold for 20 or 30 years?

The other problem with these snapshot returns is who holds for 20 or 30 years anymore? Times change too much and too fast which makes it increasingly difficult for most investors to implement that the traditional buy and hold strategy. Buy and hold forever is just not as applicable as it once was.

A case of bad timing can ruin everything

The real problem with investing for retirement is the stock markets can do really unpredictable things when you may not want them to happen. I’ve often said the stock market is not predictable or controllable so the variability can be really destructive to the timing of people’s retirement. Just think back to the last 10 years and how the stock market delayed many people’s retirement. The stock market hit retirees hard during severe corrections.

End date bias can change a portfolio’s return dramatically within months, weeks and even days. The last thing anyone wants it to retire just as the stock market takes away 20%, 30%, 40% or more. Projecting rates of return is essential but the biggest problem is the risk of the markets can change that return very quickly – I call this the retirement risk zone. For more on stock market risk and retirement, check out an old article called 6 perspectives on why retirees need to be more conservative with their portfolios.

Project with multiple returns

In the end, whatever return you choose for your retirement projections will change. In other words, you are guaranteed to be wrong. Some people think it’s better to be conservative and hope for better. Other’s feel more confident about their investment abilities.

When I teach people about retirement planning, I always say that a retirement plan is dynamic and guaranteed to change. On that basis alone, I suggest people do their retirement projections using a minimum of 3 returns:

  1. Use a conservative number almost based on a worst case scenario. I often default to the risk free rate of return or the 3 year GIC rate.
  2. Then use an optimistic return. If we look at the stock market, it’s commonly thought that the stock market could produce 10% to 15% returns.
  3. Then use a balanced or realistic approach. Some call it average returns others call it the middle.

Be running three different projections, what you are going to see is the range of possibilities. For some people even running the conservative projection gets people to their goals. In that case, they have the comfort of knowing they do not have to take any risks.

Seeing multiple projections based on different rates of return helps prepare you for the future better because you can see different possible outcomes. The more variable your portfolio is, the more important it is to use a range or returns.



    A case of bad timing can ruin everything – A good case for segregated funds!

    • Jim Yih

      That’s an interesting comment because if you are invested in the stock market it will still affect your portfolio whether you are in segs, mutual funds, indexes, ETFs or stocks. Just because you are in segs does not make you immune to the risks of market volatility.

  2. Echo

    Jim, I like your approach to using three scenarios. We often do this in business when projecting future revenues and expenses. Why not apply the same principles with your own finances?

    • Jim Yih

      Thanks for inspiring to get this article done. I had something drafted in my ‘idea’ folder but when I read your article and Nelson’s it was the catalyst I needed.

  3. Balance Junkie

    Great article Jim. It’s so hard to know what rate of return to enter in those retirement calculators. There’s no way for any of us to know that in advance. Your idea of using a range of possibilities is a great approach. Just comparing the numbers is often a great exercise in really understanding how much a 1% difference in returns can affect your savings over different time periods.

    Thanks! 🙂

    • Jim Yih

      Thanks Balance Junkie. I think you are absolutely right. I think the best thing you can do with those calculators is ‘play around’ with if then scenarios to know which course of action will help you achieve your retirement and financial goals.

  4. Michael James

    Running multiple projections with different returns makes sense. I think it also makes sense to focus on real returns. If we make use of returns from history when inflation was higher but then use today’s inflation numbers we could be too optimistic about future real returns.

    • Jim Yih

      Another great point. I guess you could rally complicate things and project a range for inflation too although that range is probably less variable than market returns. None-the-less, I are correct in that real returns are probably better than nominal returns.

  5. Ken Faulkenberry

    An important question that most investors put little thought into. I too like your idea of using multiple return scenarios. It’s a great exercise.
    I prefer to use a real rate of return instead of nominal returns. This takes the inflation guess out of the equation. Inflation needs to a part of an investors asset allocation decision. This also makes you think in terms of today’s dollars and what your investment portoflio will buy when you are retired.

    • Jim Yih

      Thanks Ken … maybe that’s my next post – real vs nominal returns

  6. Financial Uproar

    Thanks for the mention Jim. Glad I inspired a post.

    As for returns, I have a simplicity is best mindset. I looked at returns over long periods of time, saw they were 8-10%, so assumed 8% was quite doable. So I assume 8% as my long term return. I think getting 8% over the next 40 years won’t be exceptionally hard. It wasn’t hard to do over the past 40 years.

    “The other problem with these snapshot returns is who holds for 20 or 30 years anymore?”

    Who holds for 20 or 30 years? Wealthy people.

    • Jim Yih

      Nelson, Thanks for stopping by! I remember in the 1990’s when the financial industry (me included) used 10% for equity returns and that was considered conservative because the markets seemed to delivery returns in excess of 15%.

      I think your point of how people may be underestimating future returns because the last 10 years were not the greatest years for the stock market. Eight might be a good middle ground. Although I am also a fan of simplicity, I think it’s important to be aware of the risk of being too simple in projecting future returns.

  7. My Own Advisor

    Good stuff Jim.

    I think folks could forecast 7% returns with ease.

    The challenge will be, will they get it given the mass-marketing of different products, chasing story stocks or sexy ETFs that come and go?

    Many won’t – for all the reasons 100s of books have been written on behavioural investing. Thoughts?

  8. Janet Hutchins

    I love the way you lay this out. It is so true that you can make statistics tell you anything you want them to, which is why it is so important to ask questions, especially with financial products where statistics are everything.

  9. simplewealthguy

    Your return will depend entirely on your risk profile. If you are very conservative expect a lot less in the longer term than if you are aggressive.

  10. Richard B

    Very good and interesting article, thanks Jim.

    I am age 59 and decided to retire about 18 months ago. My retirement nest egg, which is looked after by a large wealth management firm, has had an average annual rate of return of 6.3% (net of fees) over the past 7 years. My risk/investment profile is somewhat conservative, with an approximate 55/45 (equity/bonds) asset allocation.

    For my projection of the average annual rate of return going forward.. after much deliberation I have decided to use a conservative number of 5% in my excel spreadsheet, and (at least for now) plan to withdraw, as income, roughly 4.5% of the total value of my portfolio annually. As I grow older I will incrementally raise the withdrawal percentage to cover increases in the cost of living and the fact that I am getting older (i.e. draw down the nest egg as life expiration draws ever nearer).

    I’ve discovered it is VERY difficult to transition from being a “saver” to a “spender”, especially those times when an ‘off year’ occurs in the markets (such as in 2015).

  11. jorgensen

    I’m 65 and facing retirement soon. I did a spreadsheet assuming different real rates of return and different draw down periods. I am planning based on 2% real returns going forward from here and hoping for 3% . Anything much over 3% real is probably unrealistic.

    (As for who holds for long periods: my wife and I have Boeing, CP, and CNR shares we have held for over twenty years and may hold for another twenty years. We have XIU units that started out as TIPS units about thirty years ago.)

Leave a reply

Your email address will not be published. Required fields are marked*