Retirement Tax Planning

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
User avatar
Posts: 727
Joined: Fri Mar 14, 2008 9:34 am
Location: Chicagoland

Retirement Tax Planning

Post by Midpack » Fri Mar 14, 2008 10:54 am

I'm a newbie here, and I did try search first, so go easy on me.

I can (and have) develop spreadsheets all day to project spending, investment returns (incl probabilities), etc. factoring in inflation for retirement planning - but I never know how best to project taxes. What do you use to do so, current tax table calculations? Seems unlikely of course. And how do you calculate taxes for future short/long term gains, tax deferred acct drawdowns (trad & rollover IRA's), pension (I have a small one), SS, etc.? Thanks in advance.
You only live once...

Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Fri Mar 14, 2008 11:45 am

If you are a spreadsheet enthusiast you can certainly construct an income tax model that captures at least the critical components. Clearly one part of the model is to anticipate indexing of tax bracket break points, deductions and exemptions. If nothing else the exercise will cause you to learn tax computation to a degree you might never have imagined was possible.

There is no way to anticipate changes to the tax code other than to run parallel computations assuming different possibilities (eg. loss of preferred status for qualified dividends) and compare the results.

Certainly you will be able to see the impact of significant financial changes, such as SS election, roll-over strategy, etc., within certain assumptions.

I too would be curious who has managed a plausible approach to this problem.

Posts: 1362
Joined: Tue Feb 20, 2007 5:24 pm

Fortune Telling

Post by earlyout » Fri Mar 14, 2008 11:59 am

I don't think it possible to make accurate predictions about taxes and many people think it is a complete waste of time. Maybe one can make reasonable guesses about taxes for 3-4 years in the future but beyond that it is speculation.

For retirement projections, I assume:

1. US income tax will keep a bracket structure similar to what we have now. Thresholds for each bracket will grow with inflation.
2. Tax rates in each bracket will be higher in the future
3. Dividends and short term capital gain (< 1 year) will be treated as regular income.
4. Long term CG rates will be 25%.
5. Taxation of Social Security benefits will keep a similar structure and the thresholds for taxing SS benefits will not increase.


User avatar
Posts: 897
Joined: Fri Jun 01, 2007 12:03 pm

Post by Rager1 » Fri Mar 14, 2008 12:27 pm

We found it easier to do our retirement projections without considering taxes at the first cut. Once we determined our pre-tax needs, we then estimated a tax provision using an average tax rate that we thought was appropriate.

To determine the total required including the tax provision (using a 35% tax rate estimate as an example), we divided the pre-tax amount of our retirement projections by the factor (100% - 35% = 65%). So, for $80,000 pre-tax needs, the amount required including taxes is $123,100 ($80,000/.65 = $123,100).

We found this method much easier to use since taxes are out of our control. It also provided an easy method to use various tax rates to establish worse case scenarios.


Posts: 62862
Joined: Thu Mar 01, 2007 8:00 pm

Post by livesoft » Fri Mar 14, 2008 12:30 pm

I use my current version of TurboTax to run "what-if" scenarios. I am not concerned with tax rates changing because my retirement should not (and does not) depend on whether capital gains are taxed at 0%, 10%, 15%, 20% or 25%. Or what dividends will be taxed at.

In other words, I am not cutting it so close that I need to worry that much about taxes.

User avatar
Posts: 3469
Joined: Fri Mar 09, 2007 4:12 pm

Post by magellan » Fri Mar 14, 2008 12:58 pm

Hello Midpack, and welcome to the Bogleheads forum!

As an early retiree, I've spent lots of time wrestling with this issue myself. Of course, as other posters have pointed out, the possibilities for future tax rates are endless and it's probably folly to try to guess the future.

OTOH, there's no folly in understanding how "possible" futures might affect your retirement plan. What I try to do in my planning is to run "sensitivity analysis" on my plan in order to see how possible tax changes might impact it.

I noticed from the thread on the early retirement forum that you've been playing around with online retirement planners, including the one that I wrote. If you create a baseline with one of these tools, then adjust the cap-gains and/or income tax rate upwards by 5% at a time, you can get an idea of how bad the impact of higher taxes might be. Start with a working plan and keep bumping tax rates up to see how high they have to go before your plan starts to fall apart. Then ask yourself how likely that tax regime would be, based on history and on what tax rates are like in other developed countries.

IMHO, there's a huge amount of uncertainty built into this type of planning. As a result, all of the planning tools have an accuracy of +-10-20% under the best of circumstances. So that means that a plan with a 90% probability of success isn't statistically much better than a plan with an 80% probability of success. Even though it's important to experiment with possible future outcomes and try to get a handle on what things might sabotage your plan, at some point you have to throw your hands up and admit that some of this uncertainty just can't be tamed.

Hope that makes sense and doesn't dissuade you from continued experimentation...



Posts: 61
Joined: Sun Mar 04, 2007 6:23 pm
Location: Portland, OR

Post by baldeagle » Fri Mar 14, 2008 2:57 pm


I think you can only use the existing tax code in your spreadsheets and then try sensitivity analysis as noted by Magellan. When the code changes you change your spreadsheet and run it again.

In my post on "Best" withdrawal decisions I had to face the same issue as you. I built columns for AGI, tax bracket tops, deductions, and exemptions, all annually inflated, and finally taxable income computed from the preceding columns.

Post Reply