Extended i-Orp help needed

[DawgMan]

As someone recently retired, I am trying to use your Extended i-orp tool at least on first pass without trying to add too many variables to make sure I understand how the results are being computed. I have also sent these questions to the i-Orp folks, but know many of you are proficient here. To start, I have used the following assumptions (otherwise, left blank)...

- Initial Balances for After Tax & Tax Deferred
- Retirement Age (current): 57/57 & Planning Horizon: 100/100
- Allocation at Retirement: 60/40 (both After Tax & Tax Deferred)
- Inflation: 3% (Spending & Income)
- Stock Investments: 7%/yr return; Stock Dividend: 2% yield; Fixed Income: 2% yield

Based on the referenced run, questions in reviewing the results…

- Does the model assume a rebalance once a year after withdrawals are taken?
- I was trying to initially manipulate the portfolio gross returns to 5%/year, inclusive of capital gains/interest/dividends. Is there away to manipulate the return and yield values? I noticed on my model my returns are growing less than 2%/year?
- How does the plan make assumptions for both Taxes (short term gains/interest/non-qualifying dividends) to produce a value for Yield? What tax assumptions are being used for the After Tax amount being withdrawn (i.e. return of capital, some % of capital gains)? I’m assuming once I add Cost Basis these numbers may change? If so, what adjustment to the Yield is made?
- I see DI = Yield – Taxes + After Tax/Tax Deferred, but what does DI really illustrate, how does it help? Once I get to RMDs I notice the RMD amount is not part of the calculation, but would affect the overall withdrawal amount?
- Should I assume the tax brackets being used are a constant, based on the current tax code only? Is there any growth on the tax brackets assumed year after year?

Thanks in advance. I really like the general output of this program (assuming I get it to work to where I understand it) and will explore Roth conversions as well, but want to make sure I understand how to use/interpret the data correctly.

 

[Out-to-Lunch]

One of the nice things about I-orp is that there is extensive documentation. Forgive me if you know this already, but I get the impression you don't: If you click the term (like "Fixed Income % Yield") it gives a lot of explanation in a pop-up box.

 

[DawgMan]

One of the nice things about I-orp is that there is extensive documentation. Forgive me if you know this already, but I get the impression you don't: If you click the term (like "Fixed Income % Yield") it gives a lot of explanation in a pop-up box.

Ahhh... I clicked some of the definitions on the actual model page, but not the input page. I believe I see the error of my ways on a few questions, but am still a little confused on the following, even after reading the definitions/explanations...

- Once RMDs hit, I notice I have distributions under RMD, Tax Def, After Tax, Yield, even after the deduct for taxes, all add up to a value significantly larger than the DI value provided? DI lines up for the years prior.
- So it appears by doing the Constant Spending, the output assumes you spend all of the DI as opposed any excess going back into investments?

Thanks

 

[Out-to-Lunch]

Ahhh... I clicked some of the definitions on the actual model page, but not the input page. I believe I see the error of my ways on a few questions, but am still a little confused on the following, even after reading the definitions/explanations...

- Once RMDs hit, I notice I have distributions under RMD, Tax Def, After Tax, Yield, even after the deduct for taxes, all add up to a value significantly larger than the DI value provided? DI lines up for the years prior.
- So it appears by doing the Constant Spending, the output assumes you spend all of the DI as opposed any excess going back into investments?

Thanks

I think the distributions of RMD and Tax-Deferred are NOT exclusive of one another. The RMDs are part of your tax-deferred distrubution, after all.

And, yes, my understanding is that "disposable income" really assumes that you actually dispose of it each year (for the constant spending model). If it meant only potentially disposable, well, then 100% of your funds (less only taxes) are disposable every year! (Only once, of course )

 

[DawgMan]

I think the distributions of RMD and Tax-Deferred are NOT exclusive of one another. The RMDs are part of your tax-deferred distrubution, after all.

And, yes, my understanding is that "disposable income" really assumes that you actually dispose of it each year (for the constant spending model). If it meant only potentially disposable, well, then 100% of your funds (less only taxes) are disposable every year! (Only once, of course )

Ok, I can see that now. I'm starting to dial this thing in. I now have 2 models both set to constant spending until age 100. Model 1 has no Roth conversions/no SS. Model 2 has Unlimited Roth Conversions and 75% of SS PIA starting at age 67. As I understand it, the constant spending model determines DI year 1 and then runs it out until age 100, bumping 3% in my case, with the goal of exhausting accounts? The Roth Conversion model somehow determines max benefit?? I noticed the DI on the Roth Conversion model was 17% higher year 1. In this case, should I be comparing the year 1 DI values between the 2 plans as an illustration (at least financially speaking) or how else should I determine the ideal Roth conversion plan? The reports don't appear to add up the lifetime taxes paid so not sure if that is the right measurement? I suppose in the end I am trying to measure the Roth conversion best strategy (relative to no strategy or something in between). If DI is the correct measure, right now it appears to be around 17%/year.

 

[sengsational]

It's not the lifetime taxes that's being minimized, it's the spending that's maximized. Although you might think those are the same, they are not; some moves can allow you to spend more and pay more taxes!

As to what you should compare, I'd say yes to comparing the the "year one" available spend amount. In the constant spend model, the spending power is the same for the entire plan, since it's the year one spend, inflated. And true, it doesn't have you plow unspent money back in...the model says spend it in year one, it's gone when year 2 of the model starts. There's no way to specify "save a certain amount"...that doesn't really make sense the way the program works.

The way I have done the Roth conversion decision is to keep everything the same except for the Roth limitation input. I might have understood that SS was different between two runs. I'd leave SS alone, and only manipulate Roth. Later on, you can iterate on the SS decision. Then back to Roth, if need be.

There are cases where a huge, early Roth conversion will increase the available spend by a small amount. In that case, the "bird in the hand" idea might have you make a smaller Roth conversion.

With respect to asset allocation through time, the "glide path" function can get you from X% equities early on, to Y% equities later, where, typically, Y is less than X.

One warning I usually give is that the optimization will preferentially spend from the lower earning tax category first. So if you have more bonds in Roth, for instance, it will put weight toward pull from Roth, for instance. I don't think this is realistic for most people (doesn't match what's likely to happen), especially those like me, that rebalance across all tax categories. So I make sure my asset allocation targets are the same across all three buckets. This allows the tax consequences to drive the optimization as opposed to which bucket has more bonds. It's a bit confusing, but I contend that if you rebalance across all tax buckets, then you should put the same percent equities in all three (Tax Deferred, Roth, Taxable), irrespective of what it looks like in reality.

 

[DawgMan]

It's not the lifetime taxes that's being minimized, it's the spending that's maximized. Although you might think those are the same, they are not; some moves can allow you to spend more and pay more taxes!

As to what you should compare, I'd say yes to comparing the the "year one" available spend amount. In the constant spend model, the spending power is the same for the entire plan, since it's the year one spend, inflated.

The way I have done the Roth conversion decision is to keep everything the same except for the Roth limitation input. I might have understood that SS was different between two runs. I'd leave SS alone, and only manipulate Roth.

There are cases where a huge, early Roth conversion will increase the available spend by a small amount. In that case, the "bird in the hand" idea might have you make a smaller Roth conversion.

With respect to asset allocation through time, the "glide path" function can get you from X% equities early on, to Y% equities later, where, typically, Y is less than X.

One warning I usually give is that the optimization will preferentially spend from the lower earning tax category first. So if you have more bonds in Roth, for instance, it will put weight toward pull from Roth, for instance. I don't think this is realistic for most people (doesn't match what's likely to happen), especially those like me, that rebalance across all tax categories. So I make sure my asset allocation targets are the same across all three buckets. This allows the tax consequences to drive the optimization as opposed to which bucket has more bonds. It's a bit confusing, but I contend that if you rebalance across all tax buckets, then you should put the same percent equities in all three (Tax Deferred, Roth, Taxable), irrespective of what it looks like in reality.

So I have left the starting and ending AAs the same making both my after tax and tax deferred accounts the same 60/40 (currently no Roth accounts). When I let i-Orp optimize Roth conversions, I noticed I am hitting the 35% tax bracket year 1 (57), then hit 33% at 61/62 and then 28% until age 72 and thereafter. It would seem the best conversion strategy would be to stay 28% or less, no??

 

[sengsational]

So I have left the starting and ending AAs the same making both my after tax and tax deferred accounts the same 60/40 (currently no Roth accounts). When I let i-Orp optimize Roth conversions, I noticed I am hitting the 35% tax bracket year 1 (57), then hit 33% at 61/62 and then 28% until age 72 and thereafter. It would seem the best conversion strategy would be to stay 28% or less, no??

It's pulling those conversions into year 1 for a reason. I think there might be some assumptions about current "temporarily low" tax rates going back up, or something. This is what I mean with bird in the hand. Change from "unlimited" to "limit to 28%" and see if the spend changes very much. If the program works as I expect it does, if that's the only change, you'll see the available spend decrease, but probably not by much.

The problem with models like this is that some seemingly small assumptions can make a big difference. But the way I think about it, if you model it at all, you're probably better off than 95% of the people who just fumble through it on their own or worse, pay an AUM fee annually (gasp!).

 

[DawgMan]

It's pulling those conversions into year 1 for a reason. I think there might be some assumptions about current "temporarily low" tax rates going back up, or something. This is what I mean with bird in the hand. Change from "unlimited" to "limit to 28%" and see if the spend changes very much. If the program works as I expect it does, if that's the only change, you'll see the available spend decrease, but probably not by much.

The problem with models like this is that some seemingly small assumptions can make a big difference. But the way I think about it, if you model it at all, you're probably better off than 95% of the people who just fumble through it on their own or worse, pay an AUM fee annually (gasp!).

Yep, changing the top bracket to 24% (model adjusts to 28% in 2026) has me in the 28% bracket for life... other than age 99/100 when I hit 33%! The DI only dropped $1 from the unlimited model.

The good news is I believe I understand how the model is used and how it can be used as an annual check & balance tool.

I appreciate all the help here getting my arms around this thing... thanks.

 

[Out-to-Lunch]

FWIW, I completely agree with sengsational's comments.

I think the following are generally true: (1) I-orp consistently recommends aggressive Roth conversions (early and big); (2) I-orp reports only modest improvements in disposable income for these aggressive conversions.

I plan to chart a middle ground, of conversions to (in my case), the top of the 22% bracket until the brackets are set to revert in 2026.

 

[walkinwood]

An enlightening discussion of i-orp's modeling, Sengsational and Dawgman. Thank you.