Pension profit sharing plan in new job vs non matching employee 401k

[apowers00]

I work in a high tax large urban environment. All jobs are either at academic hospitals or as an employee of private practices. I am currently in a private practice job as an employee with a 401k that gets no employer matching. I have an opportunity for a job that I could commute 45 min each way to a nearby county that would be about the same pay total but would allow me to become a partner and participate in a profit sharing plan where I could contribute up to $53,000 per year instead of the $18,000. In figuring out this decision, I was trying to think of a way to model this calculation of the benefit of putting addl $35,000 pretax into retirement vs investing the after tax difference (let's say $19,000) in a taxable account. Anyone can point me to a good spreadsheet or web site to figure this out?

[litovskyassetmanagement]

To do this right is not that simple because it requires making certain assumptions about the future and about a number of things such as returns and investment expense ratios among other things.  First order of magnitude estimate is simply the cost in taxes of keeping money after tax (tax owed on $35k, which comes off of your highest brackets).  Next order of magnitude is doing a side by side comparison.  Suppose you want to invest $54k in a 401k plan vs. $18k in a 401k plan plus an after-tax account for the $36k.  Without the after-tax account, the first order of magnitude assumption applies.  But if you were to do this correctly, you'd find that your results are significantly improved with the addition of the after-tax account.  While you still have to pay taxes on the $36k, you still get the benefit of compounding if you invest in a tax-efficient portfolio after-tax.  This is where it gets complex because the analysis depends on how long you plan to do this comparison for (so you have to account for ongoing contributions and compounding as well as cost).

While the following article compares a 401k vs. a SIMPLE, the same analysis can be used to determine the difference between 401k #1 and 401k #2, with appropriate amounts put in for expenses:

http://www.dentaltown.com/Dentaltown/Article.aspx?i=403&aid=5625

This will give you some ideas, so you can try the simplified version of this on a spreadsheet.  Compounding makes this a bit hard so you would need formulas for that, but if you have some time you will find this to be a lot of fun ;-)

 

[apowers00]

Thank you for the reference. The link did not work, but I was able to find the article here

http://www.dentaltown.com/images/Dentaltown/magimages/1115/RETIREpg78.pdf

My only criticism of the article is that it does not account for eventual tax that will need to be paid on withdrawals from the 401k at ordinary income rates whereas in the aftertax account, there is no tax that needs to be paid on principal, and withdrawals will have preferential lower capital gains rates. My suspicion is that with good planning using an aftertax brokerage account, one could come close to the 401k. The only problem is being disciplined enough to execute on the aftertax account -- ie, not selling anything and not touching it for any other expenses. The 401k penalties definitely dissuade from touching the money.

[litovskyassetmanagement]

apowers00 wrote:

Thank you for the reference. The link did not work, but I was able to find the article here

http://www.dentaltown.com/images/Dentaltown/magimages/1115/RETIREpg78.pdf

My only criticism of the article is that it does not account for eventual tax that will need to be paid on withdrawals from the 401k at ordinary income rates whereas in the aftertax account, there is no tax that needs to be paid on principal, and withdrawals will have preferential lower capital gains rates. My suspicion is that with good planning using an aftertax brokerage account, one could come close to the 401k. The only problem is being disciplined enough to execute on the aftertax account — ie, not selling anything and not touching it for any other expenses. The 401k penalties definitely dissuade from touching the money.

Click to expand...

But of course ;-).  Do you realize how complex that part would be given the preponderance of strategies available, including Roth conversion? The 2nd part is still on the drawing board.  I might publish something on that later when I have the time to work out the details - right now I have a general blueprint for the distribution strategy.  It require a lot of variables and assumptions though vs. the accumulation phase.

Yes, the 401k can come close to the after-tax account in terms of tax efficiency if you use Roth conversions, at least that's what I found out while playing with the numbers.  The math is significantly more complex, because it would require a Monte Carlo type of approach.

[The White Coat Investor]

Well, I hate commuting, so whatever the difference might be, it's not enough for me to do that sort of an additional commute. But maybe you don't hate commuting as much as I do and have a higher marginal utility for more money at this point in your life. 

I don't know of a spreadsheet or website that will do this calculation for you. But I'll give it a stab. Remember the 2017 401(k) limit is $54K. We'll hold that constant just to simplify things.

The first thing you've got to do is make some assumptions- let's assume the investment grows at 8% after-fees but before taxes and has a dividend yield of 2%. Let's assume you contribute money for 10 years (maybe you change jobs then or something), pull all the money out of the account in exactly 30 years and pay exactly 25% on it when it comes out. Let's assume you liquidate that taxable account at that time period as well. Let's assume 20% capital gains/dividend rate as it grows and 15% when you liquidate it. Does that all seem fair to you? If not, then make adjustments as needed. 

First, let's figure out what things look like in 10 year.

401(k): =FV(8%,10,-36000,0,1) = $563,237.55

Taxable: =FV(7.6%,10,-20000,0,1) = $305,891.06

Follow so far? Okay, let's do the next 20 years.

401(k): =FV(8%, 20,0,-563237.55,1) = $2,625,226.08

Taxable: =FV(7.6%,20,0,-305891.06,1) = $1,323,769.01

Still with me? Okay, now let's apply some taxes upon liquidating.

For the 401(k) we just multiply the balance by 75%, so $2,625,226.08 * 75% = $1,968,919.56

For taxable, it's more complicated. Remember the $200K in principal comes out tax-free. Plus all the reinvested dividends have also already been taxed. I guess we could calculate all that, but let's just fudge it and call the basis $400K. 

So we take $1,323,769.01 and subtract out $400K. That leaves $763,769.01. Multiply that by 85%, so $763,769.01 * 85% = $785,203.66. Now add back in the $400K, so $1,185,203.66. 

The advantage of the 401(k), not counting asset protection and estate planning benefits and all that could be something like $800K. But this calculation, like any other, is garbage in, garbage out. It could easily be more or less than that. But hopefully it gets you in the ballpark enough that you can make your decision. 

[litovskyassetmanagement]

Therein lies the problem.  I was able to get 401k plan taxed at an average rate of 15% or so with aggressive Roth conversion (similar to taxable), but this won't work for everyone.  It require a lot more work to do this calculation as we have to use life expectancy tables and RMD tables, as the problem is definitely non-linear, so we have to do year by year planning as well (and assume different rates of return in retirement, etc).

I did static calculations like the one above with a spreadsheet when I just started and I got very poor results because bad assumptions result in big errors. I actually did this analytically at first do develop the framework and I have an extensive write up, but when I did the numbers, they looked like crap, precisely because the problem is more complex. This is why I split the problem in half, and I'm still not done working out the distribution problem (though I have some interesting results that I might share at some point).  But I wouldn't make a decision on 401k vs. after-tax based on this type of analysis because long term projections are extremely inaccurate, so I would just stick with accumulation phase analysis to make the 401k vs. after-tax decision.