California 529: go big, or go home?

[msilenus]

Hi all,

I've got California 529 plans for the kiddos (both small) with a goal of paying for most of their college when it comes time for that.  In the past, I've been dollar-cost averaging into these plans via a monthly transfer from our bank accounts.  Nothing wrong with that, but recently I've been thinking of ways to reduce tax exposure.

I don't think there's anything theoretically wrong with a large lump-sum transfer from existing investments over to the 529 plans.  The idea would be to extrapolate how much I think they should need when they turn 18, project backwards from that to today based on a portfolio performance assumption, and transfer the needful sum over all at once.  (We could go a little less extreme and do it in a few smaller chunks over a year or so, but that's the general idea.)  I call this "go big."  The idea would be to benefit from a tax shelter for those investment dollars immediately, rather than leaving them in taxable accounts.

However, California is a high-income-tax state, and we fall in a high bracket.  California is also one of a few lonely states with both income taxes and no tax benefit on 529 contributions for state taxes.  What if I "go big" and then next year the state implements a tax break?  I wouldn't have any dry powder to take advantage of it with.  This suggests that I should consider stopping my monthly transfers, and holding out for a better tax deal.  (Let's call this strategy "Go Home" though we could also call it "Plan Apple" or "Plan Microsoft.")

Curious what people think, and why.  (Go big, go home, or no change?)  I have no idea what to think of this quandary.  I would especially appreciate any insights into state politics or legislative priorities.  I've e-mailed my rep, but he hasn't gotten back.

[clutchy]

I live in California and am in a pretty high bracket.

I picked Nevada's plan....

[nawhite]

I hear Utah's plan is pretty sweet. But seriously, why don't you keep up the cost averaging? In the Go Home plan you lose any of the reduction in tax exposure that you currently have. Sounds like continuing to cost average gets you the best of both worlds.

[msilenus]

I hear Utah's plan is pretty sweet. But seriously, why don't you keep up the cost averaging? In the Go Home plan you lose any of the reduction in tax exposure that you currently have. Sounds like continuing to cost average gets you the best of both worlds.

Our overall investment picture doesn't really respond to this decision.  If I'm not buying into 529 plan funds, I'm buying more in other investments.  I'll get the dollar cost averaging either way.  The only reason I might break up a large transfer from taxable to the 529 would be to increase the opportunity for loss harvesting in the event of a big pullback.

I'm mostly okay with the California plan. I looked into Ohio's (also reputably sweet) plan and found we wouldn't save much on expenses.  There aren't any plans out there that both offer in-state tuition and have universities worth leaving California for.

[nawhite]

I'm mostly okay with the California plan. I looked into Ohio's (also reputably sweet) plan and found we wouldn't save much on expenses. There aren't any plans out there that both offer in-state tuition and have universities worth leaving California for.

I'm confused. No one is saying you shouldn't send your kids to college in California. But most state's 529s can be used to pay expenses in any other state. Are you "pre-paying" for tuition? Because if not, you can use your 529 on any school related expenses you want in any state. And having a 529 plan in a different state does not disqualify you from receiving in-state tuition in California. So what is the problem in using a different 529 plan? (Note: I'm not saying you should. Generally you can't get benefits in another state's plan as an out-of-state person that are better than what you can get in your own state. I'm just saying that there is no reason why most 529 shouldn't be an option).

[msilenus]

I was definitely unclear.  There are a few (very few) 529 plans that allow you to pay in-state tuition for their state universities if you invest in their plan.  You can be effectively living in two states at once for the purpose of in-state tuition.  You could imagine that being a really great if it were a feature on plans from North Carolina, California, Texas --states with reasonably prestigious public universities.  The states that do offer that benefit on their plans don't have great public universities AFAICT, so it doesn't seem to make sense to base plan selection on that if you already live in a state with a bunch of great public universities.

[beltim]

This is a slam dunk for going big.  Under most reasonable sets of assumptions, the tax break on investment gains from holding investments in a 529 will exceed the tax break from CA state taxes, if that ever happens (I know CA's budget is much better than it was a couple years ago, but it's still not great, so my guess is that they won't unveil any new programs that would reduce tax revenue).

Consider $10k invested in either a taxable account or a 529 plan.  I'll use 12 years as our investment timeline, and 7% annual returns to come up with an estimate.  I'll guess that you're in CA's 9.3% tax bracket, and that your federal capital gains tax rate is 15%, and all of your investment gains come in one lump sum at the end of 12 years (that's the most advantageous way to treat the taxable investments).  Using these conditions, we have:
1) invested in 529 plan, no change in CA law: $22,522 at the end of 12 years.
2) stay in taxable account: $20,643 at the end of 12 years
I the law changes after 5 years, you sell the investments, put the proceeds into a 529, and take a deduction you'd come out ahead by ~$700 that year, which if you put into the 529 would result in an advantage of about $1125.  This is close to the best case scenario for not putting money in a 529 now.

Now let's examine a more realistic scenario for the tax treatment of taxable accounts, and delay the introduction of deductibility of 529 plans to 8 years from now.  We'll say that 3% of the investment gains are dividends, taxable at 15% each year.  The remainder comes from capital appreciation, which still impressively has no turnover.  I just made these numbers up, but the final amount in the 529 plan comes to $22,500, just slightly less than putting the money into a 529 plan now.

You can run the calculations yourself if you want, but I don't think the odds of a small gain from CA changing its laws years down the line are worth the risk of giving up guaranteed tax-free gains of money you can put away now.

[msilenus]

Okay, you inspired me.  (Thanks for that!)  I just ran a simulation with 19 different scenarios.

• Baseline: never take money out of taxable
• GoBig: LumpSum payment
• GoHome(n):Keep money in taxable until year N, then add a one-time compounding of 9.3% to reflect a tax benefit in year N, then treat further years as subject to tax-free compounding.

GoHome(n) should come pretty close to modeling the scenario where the legislature passes a 529 income tax exemption n years out. 

Other assumptions:

• Rebalance 10% every year.  Treat this as a capital gains tax event on 10% of current compounded balance
• Total tax rate 28.1% (15 LT capital gais, 3.8% medicare, 9.3% California)
• 2% dividends
• 7% capital gains
• $10k investment
• 16 year investment duration

I think you had me convinced of your analysis, but the simulation is showing that one-time bump is really significant.  If 529 plans are made non-taxable anytime within the next 12 years, the "go home(n)" plan beat "go big."

I added an option to cap the tax benefit, because that might happen.  When I capped it at $6,000 of contributions, the breakeven point moved to about n=6 years out.  At $10,000 it was n=8 years out.

It's a really important consideration that if you FIRE before you can take advantage of the income tax benefit on a contribution, you lose more than you stand to gain by waiting.  The difference between the taxable baseline and GoBig was about 10%.  On a risk-adjusted basis, the inflection point might be much earlier than 12 years.

Here's what the end of the simulation looked like.  Spreadsheet attached.:

• Baseline: $32,802
• GoBig: $36,425
• GoHome(1)/GoBig: 109.30%
• GoHome(2)/GoBig: 107.84%
• GoHome(3)/GoBig: 107.09%
• GoHome(4)/GoBig: 106.34%
• GoHome(5)/GoBig: 105.60%
• GoHome(6)/GoBig: 104.86%
• GoHome(7)/GoBig: 104.13%
• GoHome(8)/GoBig: 103.41%
• GoHome(9)/GoBig: 102.69%
• GoHome(10)/GoBig: 101.98%
• GoHome(11)/GoBig: 101.27%
• GoHome(12)/GoBig: 100.56%
• GoHome(13)/GoBig: 99.86%
• GoHome(14)/GoBig: 99.17%
• GoHome(15)/GoBig: 98.48%
• GoHome(16)/GoBig: 97.79%
• GoHome(17)/GoBig: 90.05%

[beltim]

That's a nice spreadsheet!  And thanks for posting it, I'm a bit of an Excel geek and love to look at how other people do this sort of modeling.

I haven't fully perused it yet, but I think I've identified an error in your formulas.  Actually, an omission.  You never take out a final amount for taxes when you sell the appreciated taxable stock.  You do so for the rebalanced portion each year, as well as for dividends, but you never calculate the final tax bill once you sell the stock to actually pay for college.  This isn't trivial; since you've taken tax out at step your cost basis is no longer $10,000.  But this could quite easily negate the advantages you've calculated.

[msilenus]

Great catch!  And I think you're right.  When I simulate paying accrued capital gains in sync with transferring the money over to the 529, the spreadsheet says we need the law within four years to break even.  (I'm incorrectly double-taxing dividends in this version, but I think I'll let that slide.  I don't see fixing that making the decision any less of a no-brainer.  When I treat all earnings as capital gains to estimate the impact, the threshold shifts by only about 6 months.)

There's still about a 10% benefit if a law were passed tomorrow, but it drops off at ~3pp/yr after the first year.  That's really fast.  If we guesstimate that Jerry Brown isn't going to sign such a tax break in his first term, then we should abandon all hope of getting it in time to get any serious savings.

Go big, I guess. :D

ETA: Way to use "peruse" correctly, BTW.

[beltim]

Yeah, that looks about right now. 

This has been fun!