Adjusted Cost Basis

[davem]

Hi there,

I did search for this so hopefully it hasn't been answered already but I had a question / suggestion regarding tax estimations. Please correct me if I make any wrong assumptions / assertions.

We have 3 buckets for tax types - taxable, deferred, and tax free. The tax deferred is essentially treated as income at the tax rate. Tax free just doesn't get taxed. But the taxable is taxed at the investment tax rate.

But when I will model my retirement, most of the money in the taxable account is not taxable since it was the original investment. Only the gains are taxed, and they are only taxed at a %, and then basically taxed as income.

Would it make sense to change the tax fields from

Investment Tax Rate / Income Tax Rate

to

Investment Inclusion Rate / Income Tax Rate, and Cost Basis?

So only the gains will be taxed, and they would be converted to income at the inclusion rate. Any new investments into the taxable account would increase the Cost Basis. Any withdrawals from the tax deferred (say, due to minimal withdrawals that are not used) and are reinvested into the taxable account would also go to the Cost Basis

As it is I have complex formulas to estimate a tax rate based on how the portfolio allocations are right now, and they are a decent approximation, but I have no control over how the investment income grows and still respects an estimate of the overall tax rate.

I do understand this is meant to be a simple tool to get a rough idea and I think it works great, but this simple addition seems like it would provide a easily understandable and easily modeled approach to get a better handle on taxation that could apply to many different tax codes.

Thanks!

[jimr]

But when I will model my retirement, most of the money in the taxable account is not taxable since it was the original investment. Only the gains are taxed, and they are only taxed at a %, and then basically taxed as income.

The model carries the taxable portfolio at a 100% cost basis throughout the simulation. This means that only the portfolio's gains are taxed (at the investment tax rate, in the year they happen). Withdrawals from the taxable portfolio are not taxed at all.

This simplification means the portfolio's actual cost basis isn't tracked and the model doesn't understand or account for any potential tax benefits from having unrealized capital gains not get taxed in the year the gains happen.

[davem]

Okay I guess I can see how it works ongoing using those assumptions.

Is there a way I can "fake" my balances between the 3 at the start to get a more accurate view of how the starting state would be taxed as opposed to starting with an assumption of 100%? If I move the cost basis portion into the tax free bucket then any growth in there is assumed to be tax free, which is incorrect. Or if I move half of the current growth into the tax deferred bucket that would be closer to reality maybe?

I think right now I am being undertaxed as a result but don't know what knobs to dial to get closer to reality.

I guess a more accurate cost basis modeling is too tricky? This is the one main omission that most retirement planners exclude. Frp is so great it would be nice if it could model this too since I think it plays a big role in how things evolve, but maybe I am overestimating it's significance

Thanks for the quick response in any case!

Dave

[jimr]

It might be interesting to use sensitivity analysis to vary the investment tax rate, perhaps between an unrealistically low rate and an unrealistically high rate.

This would give you at least a rough idea of how much the imprecision in investment tax handling might impact the plan's overall results.

[davem]

Okay I'll give that a try, thanks