Wednesday, November 11, 2015

Retirement Spending Stages

It’s definitely true that most people’s retirement spending declines as they age. Financial Planners tell a story of how this reduction in spending is natural and that you should plan for it in your own retirement. Here I tell a different story that leads to a different conclusion.

Certified Financial Planner Roger Whitney captures the usual story of the three stages of retirement clearly:
“In the ‘go go’ years of retirement, your spending may be at its peak. This is the time for travel, activities, adventures and family.

In the ‘slow go’ years, your spending may slow as you become more settled.

In the ‘no go’ years, you may spend even less as you settle in even more.”
This sounds so logical that it’s easy to accept the advice to spend a lot in your early retirement years. But let’s analyze this a little further.

Let’s call these stages, the 60s, 70s, and 80s. Will you really want to start cutting spending when you’re only 70? It’s true that, on average, people do begin to spend less when they’re this young, but why? I can understand being less adventurous at 85, but why only 70?

Let’s try a completely different story to explain the drop in spending:
In the ‘dumb-dumb’ early years of retirement, people see their big pots of savings and spend too much. Even if they try to stick to a reasonable spending level, they tend to dip into principal for larger items like fixing a roof, replacing a car, or helping an adult child.

In the ‘uh-oh’ years, people realize they’re spending too fast and begin to cut back.

In the ‘oh well’ years, there is little money left and people live on their government benefits and any pension streams they may have.
Obviously, this narrative doesn’t apply to everyone, but it does apply to many, which skews all the spending statistics. If spending less were a common choice, then planning for it would make sense. But if it’s forced on retirees because of overspending, then it makes no sense to bake it into your plans.

The next time someone tells you to be like everyone else and plan for peak spending in early retirement, what you should hear is that most other people spend themselves poor in early retirement, so you should too.


  1. The other exciting thing I hear folks talking about is folks who don't factor in the tax they will pay on their RRSP, they look at it as a HUGE pile of CA$H which is MINE ALL MINE!!!! Um, yeh, that will cause you to spend a little too much until the CRA sends you your first bill, oops....

    1. @Big Cajun Man: Good point. When you need to make a withdrawal to pay taxes that creates more taxable income for the next year.

  2. My plan is to track my spending every year and keep control over that. When my investments can cover this @ 5% I will consider myself in good position. The big spending years you talked about is such a common behavior even retirees with little to nothing invested think they can afford it and go into debt to act this way!

    1. @Le Barbu: Depending on the age you retire, 5% could be quite aggressive. I don't plan to get to a 5% withdrawal rate until I'm about 70. My plans also involve adjusting my spending to my current portfolio size. My calculations have the withdrawal rate ramp up from about 4% in my mid-fifties. My exact starting withdrawal rate will depend on how old I am when I retire.

  3. Both stories make sense, but which one is more typical for retirees, the "settled down and not spending" story, or the "oh crap, I don't have as much as I thought" story? I would love to see a survey.

    1. @Kevin: You may be interested in an article I wrote some time ago where I looked at a paper that included survey data:

    2. Indeed, I was interested. Thanks for providing the link. I have some concerns about the paper: (1) it is American, and I think we know that American and Canadian spending patterns are often different, and (2) the sample size was only 591 households divided into 30 age cohorts. Don't get me wrong - it's a good paper, but I think that caveats are appropriate.

      About its conclusions: I am glad to see where the smile curve comes from. Figure 5, which shows the whole sample, gave me additional concerns: while there was a smile curve of spending growth in real term (high at the beginning, low in the middle years, and higher at the end), the curve was not actually a good fit for the date points. There is a lot of variability from the curve, so the curve oversimplifies things. Secondly, the range of data points is actually fairly narrow: from -5% to +1% in average change in spending in real terms for each cohort. That's not a huge difference year-over-year.

      Figure 6, as you point out, is much more interesting. But I draw a different conclusion from you, although not one that is inconsistent with yours -- glass half full instead of glass half empty.

      Figure 6 shows that those with low net worth and high spending reduce their consumption between 65 and 75 by 0.5% to 7% in real term every year. That would be a pretty harsh retirement.

      Those with high net worth and low spending increase their spending between a little over 0% and almost 8% every year in real terms. The rate of increase trends down over the decade, except in the last year. That is a pretty sweet retirement.

      Those whose spending lines up with their net worth (low/low or high/high) see a more pronounced smile curve, with the high/high reducing their spending between 0% and 6% per year.

      I don’t agree with your conclusion that “the only quadrant of retirees who have a choice as to whether they can increase spending are the low spenders with high net worth, and they choose to spend more each year. So, a reasonable hypothesis is that the majority of people who spend less as they age do so because they are forced to rather than just wanting to spend less.” I don’t think we have the evidence for that.

      Let tell you a story about the “low spenders/high net worth” people: it isn’t that they are the only ones who have a choice to increase spending, it is rather than they realize that they have saved more than they need, and either begin to spend extravagantly, or give it away. The latter is entirely likely since the US Inheritance and Gift Tax provides strong incentives for pre-bequest gifting. Also, years ago I examined charitable giving in Canada and noticed a lot of seniors whose donations were a large percentage of their income, and in some cases, exceeded their income. The conclusion was that they were donation from their wealth rather than from their income.

      I think you could also say that the people who are financially savvy, i.e., their spending lines up with their wealth, know what they are doing, so a smile curve of spending reflects what they actually want to spend. Again, I think they way to get the real answer is not through spending data, but through a qualitative survey that asks seniors if they want to spend more, if they regret not saving more, if they are content with their standard of living, etc. But I’m not in the industry, so I’m not planning to fund such a survey. I just want someone else to do it for me. ;-)

    3. * real terms; data points; real terms; were donating

    4. @Kevin: When we dig into the details of people's spending patterns, each of us is unique, so no single narrative will ever apply generally. However, in broad terms, I find the narrative of savvy people spending just the right amount in line with their savings and desires to be lacking to put it mildly. You may not find my conclusions to be fully supported by existing evidence, but the usual narrative has far less support than my conclusions.

      I doubt most people's ability to judge whether they have saved more than they need. Most of the people who think this are likely wrong. A far more plausible explanation than their being savvy is that they do dumb things with their money because they have no feel for how quickly it can be wasted. Obviously there are examples on both sides, but I'm willing to bet that my explanation dominates cases.

      It would be challenging to design a survey to get useful information. There is no point talking to younger retirees; they will likely say all is well (and believe it) whether it is true or not. Then the challenge is to get older retirees to be self-aware and honest. Old people have a reputation for honesty, but this usually applies to their opinions about the rest of us and not in admitting their own mistakes (not that younger people are any better at admitting their own mistakes).

  4. Interesting perspective Michael. You may be right. Although I seem to know more people that had a great retirement or are currently in fine shape.

    I retired about 18 months ago and my wife retired about 4 years ago. We're both mid 50's now.

    I've spent some time looking at numerous withdrawal strategies. I'm not using a withdrawal % to drive our income/spending. What I've chosen is a spreadsheet that I update with our actual asset base at year end, and other pension income streams built in as applicable; that we project to age 95, with a projected real return (discount) rate of 1%. It also incorporates a spending reduction of 25% at age 80. (We have seen a real slow down on both sides of our family at about ages 75-78.) So our actual annual spending/draw is adjusted by our investment returns/asset base ending the year previous. I also track our spending each month, and do some rough budgeting for larger items and discretionary spending plans for the coming year.

    A large block of our investments are RRSP and we have been withdrawing sizable amounts that allow for our planned "draw" and for reinvestment in other accounts. We are working to minimize taxes over time while also considering other current and future income/pension benefits. Our first year "draw" (actual spending) was 2.94% of our starting asset base FWIW. I expect it to range from about there to 4.5% depending on the stage of life.

    I can't claim this approach is without flaws but so far is working okay for us. We have a fair bit of discretionary spending and lifestyle flexibility built in, which is our cut back area if things go worse than planned.

    I guess time will tell if we fall victim to the "dumb,dumb", "uh-oh" and "oh well" you described. I am going to work to avoid this on what we can plan/control, while enjoying life as much as we can.

    1. @RBull: I think it's actually difficult to tell if particular retirees are in good shape unless you have one of two situations:

      1) They are already very old and the quality of they spending plan is now obvious.

      2) You have complete access to their financial picture and do an analysis.

      I doubt you can tell much by just asking retirees or watching them. Things can seem wonderful whether they are overspending or not.

      Based on what you've described of your own retirement plans, they seem reasonable and safe, but it's hard to tell without complete information. Here's hoping all goes well for you (and for me).

  5. Interesting post. Assuming you had an average salary for about 40 years, you'll expect at least $750 per month from CPP. Add in OAS, GIS and if you're married then double that, you're looking at ~$36K per year (after tax, no less!) just from these 3 things.

    Our expenses (excluding mortgage payments) are just under $36K per year. Meaning we don't really need to save anything and will still get $36K per year after the age of 65, provided that OAS and GIS payments increase with the rate of inflation (which is unlikely).

    For us, our plan is to have 30 times our annual expenditure, or just over $1M in investable assets. Once we're there, we will take out 3% per year. This might be very conservative, but I plan on leaving a rather large stash and give to charity.

    1. @R: It's true that people with fairly low spending needs have less to worry about with their savings. Of course, you need to plan for the unpleasant scenario of losing a spouse. But a survivor's pension may be sufficient for this. Overall, your plan looks quite safe. You're likely to be leaving money behind, which works out for you because that's your goal. Good luck.

  6. @michaeljames
    Pretty much agree with your points #1&2. Although even with an analysis it would be subject to opinion as one persons safe might be another persons risky.
    Regarding my anecdotal comment, I am talking about a small circle of people- older/deceased close relatives that I know intimately financial details that fortunately have no health issues and admittedly don't like to spend, and several other close couples retired approx 10 years and well known to us. People easily observable as frugal, live pretty well but below their means generally. All of the people I refer to have at least 1 govt pension (as we do) and a couple have 2, and all have some savings too. I acknowledge this is probably not representative of the general retired population.

    I'm somewhere in between with my view on your "conclusion". While I agree plenty of people will mismanage their finances in retirement (like in pre-retirement) and your cautions are very worthy; I'm less certain that "everyone" advocating spending more in the early years means "most other people spend themselves poor, so you should too" in retirement. There are a lot of factors to consider. I try to refrain from telling others how to spend or from following others on how to spend our money, as it's a very individual thing. I think much depends on health (some luck, some prevention), ratio of financial needs/ discretionary spending, as well as all the other typical retirement planning assumptions. We've had a dramatic change in income over the past 10-12 years and have had little issue adapting to it. My issue is mostly about the investment return assumption, a big reason why I try to be conservative and will adjust our spending down in the event of trending negatively.

    Thanks for the comments and yes, let's hope we both have good retirements.