Page 1 of 1

Flexible spending policy vs. Guyton Decision rules

Posted: Fri Feb 29, 2008 2:32 pm
by frp user
Hi Jim,
I've been looking at the Guyton & Klinger paper on this subject and
appreciate being able to try some runs with your program. In doing so, I
noticed that the default inputs in the space entitled "Spending Policy
Configuration" are listed with dollar signs in front of them? For
instance, the input called "Minimum percent of expenses to fund" is
shown as "$75". Since that input is to be a percentage, does this input
mean 75%? Is the $ an error or just some Excel thing I'm not familiar
Thanks for your help.

Re: Flexible spending policy vs. Guyton Decision rules

Posted: Fri Feb 29, 2008 2:32 pm
by admin
Thanks for pointing out this bug in the Spending Policy Config screen. The '$' instead of the '%' is only a display bug and calculations aren't affected. As you assumed, the floor and ceiling are converted to percentages automatically and the conversion is done correctly. In any case, it should be pretty easy to fix and I should have it taken care of in the next few days.

You'll notice that if you go into the config window right after refreshing the screen (and reloading the program), the % is displayed correctly. Any subsequent visits to that screen will show the $ instead of the percent. It's a long story, but a while ago I added a feature to reformat the inputs (so if you type 50000 it gets rewritten as $50,000 once you hit run). In doing so I introduced several display bugs that I guess I'm didn't get them all.

Thanks again for the bug report. I appreciate the feedback.

Best Regards,


Re: Flexible spending policy vs. Guyton Decision rules

Posted: Fri Feb 29, 2008 2:33 pm
by admin
The bug you reported has been fixed and the planner program has been updated to version 1.0.5. Since the program is usually cached by the web browser, it's best to exit your web browser and reload the page. Clicking on the help button shows the program's version history.

One last thing I wanted to mention is that the planner isn't an exact implementation of the decision rules in the Guyton & Klinger paper. There are a handful of differences, mostly stemming from my attempts to build a tool that can handle variations in retirement income and in spending. Instead of modelling a withdrawal rate, I model a percent of expenses to fund. That makes it easier to handle changes in retirement income and in yearly funding needs.

I tried to document the algorithms that I used in for the spending policies and I also published the Java source code. Also, the spending percentage floor and ceiling values were not part of the Guyton work but were added after I read the Stout paper (see further reading).

If you have any other questions, please don't hesitate to ask.