QLAC?

Tags: 

IRS regulatory changes toward holding QLACs in IRAs are fairly new. Does ESP accommodate them? It seems not, since there's only one annuity supported on the "Annuities" (sic) tab of Pensions and Annuities. Yes, it supports two annuities if you count both partners, but what if I have multiple annuities?

Since it doesn't seem to handle QLACs explicitly, can anyone suggest how to model one? Maybe fake it with a pension?

Thanks

Chris

Thanks

Comments

I was about to post the same question, but would like to bump this thread instead...
I'm in the evaluation stage, haven't invested in a QLAC. My first thought is to reduce my 401(k)/Trad IRA balance by the planned investment amount, then add a joint-payable "pension" to the model starting at the QLAC payout date.

I'm skeptical about the value but don't want to draw conclusions without objective analysis.

The simplest way is probably to use the "choice of annuity" tab in "retirement accounts" after selecting the percentage of assets to be annuitized in the "smooth withdrawals" tab.

If you have more than 1 annuity, you could add one for your spouse in the same manner. Then add up to 2 more in the "pensions and annuities" tab. You can add up to 6 pensions as well if you want to go with that option.

In the tests that I've run so far, deferred single life (longevity) annuities have typically increased the person's/family's standard of living. However, there are ~15 different variables in ESPlanner that can be used to model different annuity options so it can be complex to determine which options are beneficial to your specific profile.

Best,
Brian

This issue I see with that approach is that annuities start when smooth withdrawals do. A QLAC starts late in life at some point in the future. How do I make smooth withdrawals work normally for the part of my retirement assets that I don't use to buy the QLAC, but defer the start of the QLAC well past the start of smooth withdrawals?

This is an issue, but you have issues either way so have to choose what works best for you.

I'd set up a profile with no QLAC and find the smooth withdrawal level.

Then, create a new profile, and set up smooth withdrawals to start when you want the QLAC to start. Before then use special withdrawals at the level of the smooth withdrawals (or close) as per above. This should get you close, will take care of RMDs, and you can fine tune from there.

Best,
Brian

Uh, I’m new at this, and maybe I’m just not understanding what you guys are talking about, or maybe my understanding of how ESPlanner works is flawed.

I didn’t think you’d need to do anything special for this. I have an annuity due to start payments in 2018. It’s not a QLAC, just an ordinary immediate annuity purchased a year or two ago with non-qualified funds. I enter the payments to me as two income streams (one non-taxable, one taxable) in the special receipts tabs, beginning in 2018. ESPlanner gives me smooth withdrawals, same amount every year; once the annuity starts, less money is taken from my savings each year.

Am I doing something wrong? Or not getting the smooth withdrawals I think I’m getting? Or have I just misunderstood what you’re talking about?

There are a number of ways to handle annuities in ESPlanner. If you purchase one through "regular" (taxable) assets, tax-sheltered assets, qualifying longevity annuity contracts (QLACs), etc. there may be differences in how to best model these.

Special receipts are always an option so you are probably okay. However, you could also use the "pensions and annuities" tab or under the "retirement accounts" tab you can specify the percent of assets to be annuitized, then input the annuity.

From Chris' comments, he wants to take a percent of tax-sheltered assets and annuitize them so they start paying out late in life and still keep smooth retirement withdrawals prior to that point. There are probably other factors that aren't clear. For example, you can invest up to 25% of your account balance (or $125,000 maximum) in a QLAC without concern about noncompliance with the age 70½ minimum distribution requirements. My comments were directed specifically at his question. Since there is more than one way to handle it, the user can model it as they want.

Best,
Brian