# Change of Procedures

Over the last few posts I’ve explained how you can use FIREcalc, a free retirement calculator at http://www.firecalc.com/, to estimate the risk of coming up short of any savings goal you might have. The “trick” is to use the detailed results in one of the Excel spreadsheets that can (or could) be obtained from FIREcalc.

Unfortunately, I discovered last week that I can no longer access the FIREcalc spreadsheet. I’m not sure if the problem is on FIREcalc’s end, or on my end. But either way, this means I can no longer use the procedure I described in the June post to estimate shortfall risk.

So, having given myself a headache last weekend trying to think up a workaround, I finally was able to come up with an alternative way to use FIREcalc that doesn’t require the Excel spreadsheet (and is actually simpler for that reason). In this month’s post I’ll explain this new procedure, and then beginning next month I’ll continue with the series of posts I had planned on how to optimize a savings plan using FIREcalc. The procedure I explain below should be used in place of the one I described in the June post.

I’ll use the same example I used back in June to illustrate the new procedure. Let’s suppose that you would like to save $50,000 over the next ten years, in order to be able to fund something you want. It doesn’t matter what the “something” is—it could be a down payment for a house, it could be a trip around the world—whatever. We’ll also suppose you’ve already saved $10,000 towards your $50,000 goal. You’ve invested the $10,000 in a portfolio split 60 percent stocks and 40 percent bonds. You plan to save $3,000 per year towards your $50,000 goal. What are the odds that you will come up short of your $50,000 goal in 10 years?

To answer this question, make sure you are on the “Start Here” page of the FIREcalc website (the tabs for the different pages appear just below the green-colored banners at the top of each page). Scroll down about a quarter of the way, and along the right hand side of the page you will find a small box labelled “Start Here.” That box asks for three inputs from you: “Spending,” “Portfolio,” and “Years.” Normally, if you were using FIREcalc to assess your risk of going broke in retirement, for “Spending” you would enter the amount of money you plan to spend in the first year of retirement to cover your living expenses. However, since in our example we are using FIREcalc to estimate the risk of coming up short of a savings goal, enter 50,000—the amount of money you plan to spend on your house down payment, your trip around the world, or whatever else your ultimate goal might be. Next, for “Portfolio” enter 10,000—the amount of money you’ve saved so far towards your $50,000 goal (normally, for “Portfolio” you would enter the amount of money you hope to have saved for retirement by the time you retire). Finally, since you plan to have reached your $50,000 goal in ten years, enter 10 plus 1—eleven in the “Years” box. You must add one additional year, to cover the year you plan to spend the money you’ve saved.

Next, click on the “Not Retired?” tab. Here the first question asks: What year will you retire? Notice that in parantheses this question is asked in a different form: “how many years before you retire?” Enter 10 as your answer to this parenthetical question. By doing so, you are telling FIREcalc to treat the first 10 years as the period over which you will be saving towards your $50,000 goal, and that you will spend the money in the eleventh year.

Next, FIREcalc asks how much you will add to your portfolio in each of these ten years. Enter 3,000, the amount you plan to save each year. Then, click on the tab “Spending Models.” On the first line that requests user input on this tab, you are able to specify the inflation assumption FIREcalc should use. Since in our example we are making the simple assumptions that neither the $50,000 savings goal nor the $3,000 annual savings will be changed to reflect inflation over the next ten years, we want to set FIREcalc’s inflation assumption to zero. To do this, click on the third option (the one that allows you to specify a fixed inflation percentage in the accompanying box). Then, enter a zero in the box. (In future posts we will return to the issue of inflation. We will learn that when you do away with the simplifying assumption of zero inflation, shortfall risk skyrockets! But that’s a subject for the future; for now let’s keep things simple and assume no inflation.)

Finally, go to the “Your Portfolio” tab. In the check boxes along the left-hand side, you should see that the “Total Market” option has been checked. Inside this option (bottom center) you will find a box showing 75 (indicating that 75% of the portfolio will be invested in stocks). Since in our example we are using a 60% stock portfolio, change the 75 to 60. Then, scroll down to the bottom of the page and click the “Submit” button.

You should then be taken to a page that shows your results (or, if you aren’t automatically redirected to the Results page you should get a flashing tab that will redirect you to that page if you click on it). Go to the fifth paragraph below the heading “FIREcalc Results.” The second sentence should read “FIREcalc found that 10 cycles failed, for a success rate of 92.6 percent.” What FIREcalc has done is calculate the growth of your savings over each rolling ten-year period starting in 1871 up to the present (for example, 1871 through 1880, 1872 through 1881, 1873 through 1882, etc.). Then, at the end of each of these ten-year periods—that is, in the eleventh year—FIREcalc spent $50,000 of the money you saved in the previous ten years. If the total savings was less than $50,000, FIREcalc counted that eleven-year cycle as a failed cycle—that is, you failed to reach your $50,000 goal in that particular cycle. Overall, FIREcalc determined that 92.6 percent of the cycles were successes. By subtracting this 92.6 percent success rate from 100, you can estimate your shortfall risk.

The result is 7.4 percent. If you compare this result with the 5.9 percent result we obtained in the June post using the old procedure (the one that relied on the Excel spreadsheet of detailed results that is no longer available), you will find that the new result is slightly larger. There are two reasons for the difference. First, because in the new procedure we had to add an eleventh year to represent the year the money is spent, the total number of historical cycles available to FIREcalc is one less (135 vs. 136) for the new procedure *vis a vis* the old procedure. Second, and more importantly, the detailed results available using the old procedure enabled us to check whether, in any of the ten failed cycles, we actually reached the $50,000 goal *before *the tenth year, only to fall back below the goal due to stock market losses incurred after reaching the goal. As is described in the June post, the $50,000 goal was in fact reached prior to the tenth year in 2 of the 10 failed cycles. Assuming that one would pull one’s savings out of the stock market before the tenth year if the goal was reached sooner, we treated these 2 failed cycles as successes.

Unfortunately using the new procedure, it is not possible to adjust the FIREcalc results to reflect the possibility that some cycles might reach the goal prior to year ten, only to fall back below the goal. However, this is an unusual situation and hence the resulting error should be small (5.9 vs. 7.4 in our example). Furthermore, our inability to adjust the results using the new procedure will cause us to err on the conservative side. A procedure that yields a conservative estimate of shortfall risk is by no means a bad thing.

Now that we have an alternative approach for estimating the risk of coming up short of a savings goal, I can show you how to use FIREcalc to optimize a savings plan for any goal you might have. Check back on September 6, when you’ll find the first in a series of posts explaining this plan optimization procedure.