Page 1 of 1

Annual Transfer From IRA to Savings?

Posted: Fri Aug 24, 2018 5:54 pm
by mountainsoft
I was experimenting with FRP today and stumbled across something I don't know if I'm doing right, because the results seemed surprising.

I defined my tax deferred savings (my IRA) with a 6% return, 6% deviation.
I defined my regular taxable savings with a 1.6% return, 0% deviation (on the top half of the additional inputs screen).
Then I added my wife's pension and both of our social security incomes on the additional inputs screen.
Everything seems fine when I run the simulations.

Today I wondered what would happen if I withdrew 15K annually from the tax deferred portfolio (the IRA), and deposited that 15K to the taxable portfolio (our regular savings) on the additional inputs screen. I was surprised to see the overall portfolio values and the probability of success increase. That didn't quite make sense to me since the money was going from a 6% return to a 1.6% return, albeit with less deviation. The IRA ran out of money sooner, but the regular savings went on to grow larger with less risk. Somehow this seems too good to be true.

What am I missing?

Re: Annual Transfer From IRA to Savings?

Posted: Fri Aug 24, 2018 8:04 pm
by mountainsoft
I see now why this was too good to be true. I modeled this with two separate additional inputs, one deducted 15K from the IRA, the other added 15K to Savings. Of course, the second input kept adding 15K to savings even when there was no more money in the IRA to get it from. Free money! :)

Doing this with smaller amounts (5K) seems to help eliminate funding failures as long as the IRA balance doesn't reach zero. Makes sense as it's just shifting slowly to a less risky asset allocation. Is there an easier way to model a slow shift in asset allocation over time?

Re: Annual Transfer From IRA to Savings?

Posted: Mon Aug 27, 2018 4:56 pm
by jimr
Sorry I missed this post originally. While I was recreating your setup to help answer your other post, I noticed a potential error in the planner logic and re-released to code.

The error is that setting up regular savings deposits in additional inputs that continue after retirement starts can cause the planner to incorrectly report the probability of success as 100%. All the other results that get reported are still correct, but by replenishing the portfolio after it's been depleted, the logic that detects that the simulation ran out of money can get tricked into not noticing that the funds were depleted.

You can download the latest software version at this link: