There is no such Thing as a Constant Withdrawal Rate

Are you sitting down?  Yes, then good.  If not, you might want to pull up a chair.  I’ve got a confession to make.  You know that 4% safe withdrawal rate that me and other early retirement bloggers go on and on about, which is suppose to be the amount you can safely pull out each year and not run out of cash over a 30 year time frame. Well it turns out that most of the assumptions used to model that don’t really apply in real life (for full details, you can read this long, but excellent article).  The real truth of the matter is that a ‘safe withdrawal rate’ isn’t a constant at all but rather another variable.

What?!?! Then how do you model that into your retirement plans?  Simple, you can’t.  Depending on the situation your ‘safe withdrawal rate’ can range anywhere from 1.8% to 25%.  This all depends on several factors like the amount of fees you pay on your investments, the rate of return on your investments and the sequence of those returns, and your personal rate of inflation.  In a nut shell you can’t model it yourself because it becomes a circular reference, which you might be familiar with that error if you have ever had to do complex modeling in Excel.  In a nut shell you series of references to other variable results in your last object referring back to your first object, you end up with a closed loop that can’t be solved.

So if you can’t model it why are you telling me about it?  Ah, that is the right question.  I mention this fact because in reality, when you are actually living on your savings in your early retirement period you shouldn’t have a constant withdrawal rate.  Instead you should ramp it up and down depending on those factors I already mentioned.  So in today’s current context with low returns, low interest rates and slightly higher inflation you should consider lowering your withdrawal amount below 4%.  Then when you hit some good years like those leading up to 2008, you can take an extra vacation if you want.

This really isn’t that hard as people already do this in real life prior to retirement.  If you lose a job, your spending doesn’t keep going out at the same level.   You adjust your spending to your lower income as much as you can to ride out the bad times until your income level comes back up.   Yet doing this requires you to have some fat in your budget to cut back on, if you purely rely on cutting back spending.  The other alternative is to increase your income by getting some short term work or selling a non-income producing asset such as your vacation property.

So the lesson in all of this is you don’t want to retire early on the absolute lowest point of your spending, you want to in fact have a bit of fat or safety margin in your plans.  This is somewhat obvious risk planning, but you might be surprised how often the obvious isn’t really seen by people.  So please have a few backup plans when you retire early including some extras in your budget.  That way you can cut back during the lean times if you need to.

11 thoughts on “There is no such Thing as a Constant Withdrawal Rate”

  1. As I wrote in a comment to the previous post (“Risky?”), I have built into my ER budget a cushion or surplus I use to protect myself against unforeseen expenses. For example, I had to buy a new PC a few weeks ago and it cost me about $400 but it will have no measurable effect on my annual budget or lifestyle.

  2. It can certainly be a dangerous topic to blog about 🙂 But I have a feeling that many of the people who spend their time arguing about the precise safe withdrawal rate haven’t actually lived off of an investment portfolio. It’s too easy to argue that 1% can throw off your plans, when the 1% really does represent a large increase in spending and anyone who’s saved up that much will hopefully be wise enough to understand the consequences of spending that much more.

  3. It’s an interesting point. Personally, while retirement calculators are excellent at ‘ballpark’ figures, I would never trust one so completely as to base my final decision on when to retire based only on that information.

    I think (and hope) that anyone interested and dedicated enough to retire early would also see this. We tend to be a pretty spreadsheet and calculation happy bunch – I’ve never even used a retirement calculator before. It makes too many assumptions (as the link points out) and frankly, I trust my own calculations much more than anything I’d find on the internet. Something about the linked website doesn’t sit well with me; the commercial aspect of his advising business is too strong and I feel his information is partially aimed at business development.

  4. Pingback: - My Own Advisor
  5. Mr Money Mustache linked, and I’m looking forward to reading through your past posts.

    Seems to me that if you figure an average 4% withdrawal is pretty safe, you could just start taking out that amount each year. If it’s a bad year, you will touch some of your principle. If it’s a good year, your principle will go up. Over time, those ups and downs will balance out, especially when we are talking about 40 plus years of retirement. Obviously if your principle was trending down over time, you would adjust your spending to counter that trend. Thoughts?

    My thinking is that a retired person could take advantage of opportunities for discounted spending during “down” years. If the economy is doing poorly, chances are there is opportunities to be had at a discount. This might mean traveling with less expense, purchasing used items at a greater discount, etc. It might be worth living more leanly during boom times in order to be able to jump on the discounts during the busts.

  6. I prefer to think of investment returns as your ‘salary’ after you finish working. I think it’s dangerous, for the above mentioned reasons to simply assume that you can take our 4% each year, although I agree with James about the opportunities for discounted spending.

    In my mind, when when I retire I would continue budgeting and managing my money in much the same way as I do now. The only difference being that the source of the income is not my employer but my investments. I would still build up an emergency fund, and save much as I do now. In leaner times, this money would supplement my lessened income, or I would cut back in other areas.

  7. I stopped working in 1999 to live off my investment portfolio. I had read a lot about safe withdrawal amounts, especially of the 4% per year variety. I started at 4%, but with the intention of getting it down below the portfolio’s income, which hovers around 3.5%. I have currently mostly dividend paying Canadian blue chip stocks.

    My portfolio was half in a trading account and half in RRSP accounts. I got the percentage of my withdrawal down and it was 3.1% last year. However, there is also a movement of money from my RRSP accounts to my trading account (and now also a TFSA).

    To deal with withdrawals from my RRSP accounts, I am keeping in cash the difference between my projected withdrawals and my projected income over the next 5 years. (I do not want to be in a position where I have to sell a stock for current income.)

    My total withdrawals may be less than the income I make on my whole portfolio, but I am really taking out a higher percentage from my RRSP accounts than my trading accounts. I had not really anticipated this when I started, but it has worked out this way.

  8. I’ve been “retired” for just over a year now, and how I’ve worked out how much I can spend, is to calculate my investment income for the previous year, and then spend slightly less.
    By always being a year behind, I shouldn’t have any nasty suprises, and as long as inflation doesn’t get too high, I shouldn’t have to eat into my capital.

Comments are closed.