Thursday, March 1, 2012

Safety-First Retirements Are Expensive with Low Interest Rates!

I’ve made a table that pretty much speaks for itself. The table provides some idea about how expensive it can be to build a safe retirement portfolio with TIPS. But even this is not completely safe, since the TIPS ladder only lasts for 30 years.
I assume different constant TIPS yields. With no changes in TIPS yields and with TIPS yields the same for all different maturities, this is too simplifying. There is no reinvestment risk or interest rate risk. But it gives an idea. Those yields are shown horizontally.
Vertically, I show different desired withdrawal rates from one’s portfolio. Some relief is that Social Security and other income sources would be added on top. These withdrawal rates (providing inflation-adjusted withdrawal amounts) are only for the part coming from the portfolio.
What the table shows is the percentage of your portfolio assets you would need to dedicate to building your TIPS ladder for different yields and withdrawal rates. Red cells represent the impossible, you would need more than 100% of your portfolio to pull off that spending rate.
Also, you really should have some sort of emergency fund in case of unexpected necessary expenses. This could be at least 10% of your portfolio, making anything in the table above 90% infeasible.
So, for example, with a TIPS yield of 1%, even a 3.5% withdrawal rate is pushing the limits of feasibility, requiring 91% of the portfolio.
With a TIPS yield of 0%, a 3% withdrawal rate is about the best that can be hoped for.
The main message here, for whatever it is worth, is that building an inflation-protected spending floor in retirement can be incredibly expensive in a low interest rate environment!


Since this was a short blog post, for fun, here is the MATLAB code I used to make the table:
r=[-.02:.01:.05] % TIPS Yield
C=[.01:.005:.1]  % Withdrawal Rate
N=30             % Ladder Length

table1=zeros(length(C),length(r));

for i=1:length(r)
    for j=1:length(C)
        table1(j,i)=100*C(j)*sum((1/(1+r(i))).^[1-1:N-1]);
    end
end

9 comments:

  1. It's ironic (and disappointing) that Fed actions designed to push people into stocks in order to aid the overall economic recovery will result in many retirement portfolios being put at risk.

    ReplyDelete
  2. At a real interest rate of 0%, the real value of money is constant over time. Not surprisingly, you can draw your money out for 40 years at 2.5%, 30 years at 3.33%, 25 years at 4%, 20 years at 5%, etc. The most interesting part of the table is how little difference a real interest rate of 1% or so makes. I had not thought about that before, although a back-of-the-envelope calculation shows the same thing. For example, at 1% real, over 30 years, you collect approximately 15% of your principal as interest (because, on the average over the time period, about half of it remains), and over a 30-year period that would be about 0.5% per year, meaning that 3.5% withdrawals at 1% real will last just about as long as 3% withdrawals at 0% real, as indeed your table shows. Perhaps that's a simpler explanation for some readers.

    ReplyDelete
  3. Thanks to both of you!

    Anonymous, yes, it is always good to look at the same issue from many different perspectives to help strengthen the intuition. Thank you.

    I think another issue to look at is to then compare this table with the annuity payout rates I was looking at a couple of weeks ago.

    ReplyDelete
  4. Hi Wade: I am new to your site and this particular post.

    I having been looking at this a bit different, so tell me what I am missing please.

    I create a TIPs ladder going out 30 years and starting 10 years from now. Instead of looking at interest rates or withdrawal rates, I simply want to have $5,000 a year in real dollars per year available. My wife and I will both be 95 in 30 years so we are not worried about reinvestment. So a twenty year ladder x $5,000 is $100,000. I put it in and I am done. I hold the bonds to maturity so I should receive $5,000 real dollars a year (baring default).

    What am I missing in looking at it this way?

    ReplyDelete
    Replies
    1. Hi,

      Thanks for making a comment.

      I don't really think you are missing anything. The point of this post is just that the cost for doing this would be less than $100,000 if TIPS yields were higher than 0%. But there is nothing you can do about this. If you want to lock in that income floor, you have to pay what the market offers. And I think TIPS yields are just as likely to go further down than to go up, so there may not be any use in waiting.

      One small matter that you are missing, though, is just that you are exposed to the risk that you or your wife will still be alive after age 95. Well, that is great. But your money will be gone.

      I discussed some research about combining TIPS with a deferred income annuity here:

      http://wpfau.blogspot.com/2012/01/safe-retirement-income-with-tips-and.html

      Best wishes, Wade

      Delete
    2. Hi Wade: Thanks for taking the time to comment and the link.

      Delete
  5. Well, the stock market has had a slight negative REAL yield since 2000. It appears that the s and p 500 since 2000 has given a nominal yield of 3% or so which is slightly less than CPI inflation. So what do you suggest a person to do. Setting up a TIPS ladder since 2000 has been a much better play than the stock market.

    I wonder what the next 10 years will bring.

    ReplyDelete
  6. Somewhere there is a chart showing $100 per month in I savings bonds versus $100 per month in the S&P 500 index fund- starting in 1998. No need to guess at the results I suppose. Someday Boomers will be burning effigies of Jeremy Siegel.... assuming they can't get their hands on him.
    Unfortunately you can't put $30,000 in I bonds annually any more. Fred

    ReplyDelete
    Replies
    1. Thanks for the comments. I take it that you think the "expense" of the safety-first retirement is just a realistic assessment about the cost of retiring.

      Delete