Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
pointlesspart
Feb 26, 2011
How viable is it to use a Roth IRA as an emergency fund/normal brokerage account?

I already max out my traditional 401k, backdoor Roth IRA, and HSA every year and I don't intend to use any of them as an emergency fund. My company allows for the Mega Backdoor Roth in the 401k, which leaves me with about 34k (69k cap, 23k traditional contributions, 12k company match) dollars of potential tax advantaged space that I'm not using. I also have a taxable brokerage account which I will probably be liquidating in the next year or two for a home down payment.

I need to store emergency funds somewhere and I have that tax advantaged space I'm not using. On paper, a Roth IRA is definitely the best place to put the money, since it pays no taxes. Even if I just buy T-Bills in the thing, that's still coming out ahead. I can withdraw contributions, but not earnings, at any time, but has anyone actually done that and knows if the latency is a killer in an emergency? My credit cards have ridiculously high limits, so I should be able to float normal spending for a month before withdrawing from an IRA and paying them off in full.

It is also (on paper) the best place to put things like a car replacement fund, home repair funds, really any irregular, large purchase money. Those have less accessibility requirements, but I'd still need to get to them sometime.

Adbot
ADBOT LOVES YOU

pointlesspart
Feb 26, 2011

adnam posted:

I was wondering if this is a good place to ask this but I've got parents who want to contribute to little adnam's education fund. I was reading somewhere that grandparent held 529 have minimal effect on FAFSA applications, something about a 'grandparent loophole'. Is it worthwhile for me to help them set up a 529 account under their names or is this not worth the effort?

https://www.savingforcollege.com/intro-to-529s/does-a-529-plan-affect-financial-aid

The FASFA EFC counts 5.64% of student or parent owned 529 assets toward financial contributions. So keeping the money in the grandparents' hands reduces the amount you are expected to pay. This isn't really a loophole for grandparents, if anyone other than the parents or student holds the money, it doesn't count.

The CSS profile probably counts it, so if your kids wind up in the Ivy League, you're paying up regardless.

pointlesspart
Feb 26, 2011
I have a weird problem caused by my company's 401k.

My company allows the mega backdoor Roth in the 401k, but only inconveniently. You have mail all the forms to Vanguard and you can only do in service withdrawals 4 times per year. Vanguard customer service takes, on average, 4 weeks to handle this process, but sometimes they take more. The process is also buggy, Vanguard thought I quit the company last time and stopped adding my paycheck deductions to my account for a few weeks, which was a fun mess to sort out.

Contributions in excess of the 401k limit are automatically converted to post tax contributions. So the minimal headache solution is to set how much money I want to contribute at the start of the year and start mailing in forms after I put in more than the normal 401k limit. I have to contribute at least 8% of my salary per paycheck to get the company match, so front loading everything is not an option. My job security is somewhere around government employee and tenured professor, I am fine assuming that I will be employed the whole year.

The problem is that my start of year paychecks are larger than my end of year paychecks. This is because I get all of my tax deductible contributions done by summer, then I start paying taxes on mega backdoor Roth contributions. This year, I will go from ~$2900 a paycheck to ~$2500 and I hope to increase my contribution percentage next year, which will only increase the delta. Getting paid earlier is better than getting paid later, so I should be better off frontloading the tax advantaged part. Plus it minimizes paperwork and I can't really invest after tax contributions till they are converted, which would take a month and a half (on average) if I spread them evenly throughout the year. I am fine from a cash flow perspective, most months I spend under $3k. But I have some lumpy expenses which this makes it harder to plan for.

What should I do with the start of year tax savings? Just stick them in an extended emergency fund? T-Bills? Breakable CDs? I can't use it to max out my IRA and HSA, my end of year bonus already covers that.

pointlesspart fucked around with this message at 04:10 on Apr 18, 2024

pointlesspart
Feb 26, 2011

Leperflesh posted:


So the tl;dr of that is that it's harder, and IMO the way to do it is just discount your projected returns above inflation by some amount, like say 1%, and discount your projected withdrawals by some reasonable amount, like say 10% or so, compared to savings in an IRA.

Like say you project your roth ira to grow 8% above inflation annually for 20 years, you might project your identically-invested (asset balances) brokerage account to grow 7% above inflation instead. And then say you assume you'll be pulling X of the total balance of your IRA out annually in retirement and that's $Y, assume in order to get $Y from your identically invested brokerage account you'd have to withdraw 10% more than X. If X is 20k, assume you have to pull $22k from the brokerage account to get the same $20k of after-tax money.

Discounting your real returns doesn't cover the problem. If you have a real return 5% above inflation, but inflation is 4%, that is substantially worse for a taxable investor than a real rate of 5%, but inflation is 2%. This is because the inflation adjustments to the initial investment are taxable, so they eat into principal. To use an example consider $10,000 invested at 0% real for one decade, one with 4% inflation and one with 2%. They have inflation adjusted cost bases of

10*(10.04**10)=$14,802, $720 LTCG owed, 4.8% of total
10*(10.04**10)=$12,189, $328 LTCG owed, 2.6% of total

Assuming you have to pay 15% LTCG. The problem actually inverts if we have deflation, since the losses to your cost basis start making your returns tax free. But the odds of sustained deflation are substantially lower than inflation.

Over a long enough time period of positive inflation, the ratio of your cost basis to the current value of your investment tends toward zero. So you can analyze the worst case scenario by assuming you will have to pay LTCG on all your taxable investments. It can't get worse than that, unless you have to pay taxes on inflation adjustments along the way.

pointlesspart
Feb 26, 2011
I'm not sure you can make the assumption that retirement savings will follow the normal, gradually rising path for the OP. Windfalls and RSUs are highly variable and subject to feast and famine periods.

The way to properly do this is to specify exactly how much you will be investing in each time period (negatively for withdrawals), what the returns will be in each period, the inflation rate, and the tax rate. The cost basis issue means how you get to a certain investment amount matters, path dependence is a pain.

But you can avoid all of these cost basis issues and use a simpler model if you promise to only touch your taxable investments in the following ways:
1. Never dip into principal. You pay for taxes on dividends out of dividends, so multiply them by the appropriate tax rate. You will need two rates of return to model it, one on the principal and one on the dividends.
2. Never sell any taxable investments. Even if you have a loss, how much loss will vary based on your cost basis. So just never sell.
3. You may donate taxable investments to charity, you get the full tax write off from those regardless of the cost basis.
4. Die. If you have less than $13.61 Million, you won't pay estate tax and your investments get a step up in cost basis. If you have more than $13.61 Million, you should probably be paying someone to worry about this for you.

This will make it easier to model the problem, in exchange for making it harder to solve the problem.

Or you can accept some uncertainty in retirement planning and sell your assets as needed. The error bars for financial planning are fairly wide anyway, I'd project out a best, worst, and average case scenario. Maybe work out an average taxable cost basis for year X, add in some inflation scenarios (0%, 1%, 2%, 4%, 8%, 16%), and discount dividends stock and bond funds by LTCG rates in the historic range of 15%-25% (net of tax deductions). A simple script would be able to handle it.

pointlesspart
Feb 26, 2011

remembertorelax posted:

Wouldn't it just be subtracting 15% (or whatever LTCG rate you anticipate) from the gain, and adding that to the basis?

My approach is so simple and some of these responses so long that I'm probably wrong in some major way. My retirement savings in taxable/brokerage are all in a total stock market fund. The money I've transferred in is basically the same as the basis, and any change from that total is the gain/loss. So, I just multiply the investment gain/loss by 85% (to estimate 15% LTCG tax) and add to the basis. Then I use that estimate for things like monitoring net worth or estimating overall progress of retirement investments.

There are at least two problems with this approach
1. There are two kinds of return: capital gains and dividends. Both are taxed at the same rate (assuming the dividends are qualified), but one has to be paid every year.
As an example, consider two funds that return a nominal 7% every year, one of which is entirely capital gains and one of which is all dividends. After 10 years, if you invest $10k in both and pay 15% LTCG
Capital Gains = 10*(1.07**10)=$19.67k, $1.45k in deferred LTCG. So if you sold everything, you would get $18.22k
Dividends = 10*((1+.07*.85)**10)=$17.82k
Most funds are a mix of both Capital Gains and Dividends, so you have to know which percentage of each you are getting.

2. Your cost basis is tax exempt, but inflation changes the real value of those nominal dollars. This means you can't just plug in the long term real return of stocks and bonds, you have to guess inflation too. You end up paying a deferred LTCG tax on an increasing portion of your investments as the nominal amount of dollars you hold rise at the inflation rate.

The worst of all worlds is if the cost basis of your investments tanks to (for practical purposes) 0 and all your returns come from dividends. Think owning $SPY shortly after hyperinflation. $100,000 buys a hamburger and the DJIA is at $300 million, so you pay deferred LTCG on all of your investments. Stock buybacks were banned decades ago, companies have been paying out massive dividends and all your returns paid dividend taxes up to that point. Assuming the stock market returns 5% real over a 30 year time horizon, the real value of your taxable investments is 10*((1+.05*.85)**30)=$34.85k, with $5.23k in deferred LTCG.

Alternatively, say inflation was 0 for decades, Japan style, and all your returns were capital gains. Then you have 10*(1.05**30)=$43.21k, with $4.98 in deferred LTCG.

This gives the account a liquidation value lower bound of $29.62k and a upper bound of $38.23k. Nothing about the investment returns or tax rate changed, just inflation and the timing of the taxes. You could have 29% more money as a result of just those two. If tax rates are higher than 15%, which they have been, the inflation rate and dividends vs capital gains matter more.

None of this is hard to track for past investments. You can just look up what your actual basis is, actual returns, and actual inflation. But for retirement planning, you have to assume something about the future state of the world.

pointlesspart
Feb 26, 2011

Subvisual Haze posted:

But the original question was how to value money in a taxable account versus a tax-advantaged account, not whether it is advantageous to put more money into a tax advantaged account. Dollar for dollar money in a taxable account is more valuable than money in a traditional tax advantaged space. That is because every dollar inside a traditional tax advantaged space will be fully taxed at income rates when withdrawn, thus it should be discounted in value at your expected withdrawal tax rate. Roth dollars of course are maximum value. Taxable account dollars are valued somewhere in between trad and roth. Where exactly is difficult to define, but long-term capital gains will be taxed preferably to income at any tax level.

I agree that it is almost always worthwhile to maximize your tax advantaged space, even Trad. But this does not mean that a current dollar in Trad assets is worth more than a dollar in taxable.

That depends. If you have a low income, then trad dollars are worth as much as taxable dollars.

The standard deduction is $14,600. Then you have $11,600 at 10% and $35,550 at 12%, with double all these numbers for married couples. So if you have a low income, you will pay less than 15% tax on average on traditional account withdrawals.

Of course, LTCG tax only kicks in if your taxable income exceeds the 12% bracket. But for that first $14,600 (higher if you itemize), traditional and taxable dollars are taxed at the same rate.

You also get no tax liability for donating traditional dollars to charity as Qualified Charitable Distributions. So giving money away lets you count them both the same, provided you jump through the QCD hoops.

pointlesspart
Feb 26, 2011

Subvisual Haze posted:

Isn't that itself an argument in favor of taxable? Both taxable and trad are taxed at zero in the standard deduction range, but in the next $47k of single income, normal income is taxed at 10-12%, but LTCG is taxed at zero.

Maybe. Remember that the taxable dollars have already been taxed once, there are fewer of them because you have already paid taxes and have to pay taxes at LTCG on reinvestment. What those tax rates are matters. Both kinds of tax deferral offered by traditional accounts are hard to beat, but technically possible to. This is just noting that there are a few cases where taxable and traditional dollars are directly comparable instead of requiring a discount.

And if you go from a high taxable income to a low, you get a lot of savings on traditional accounts. That is true of a lot of retirees. People mostly spend those savings upfront instead of saving them, but that does not mean that the deal itself is not good.

Josh Lyman posted:

One thing that’s missing from this discussion is that there are income limits in order to deduct contributions to a traditional IRA if you’re eligible for an employer sponsored retirement plan.

So for the average person whose work offers a 401k, if their MAGI is more than $77k in 2024, they start to lose the ability to deduct contributions to a traditional IRA, phasing out entirely at $87k. So for those people, a traditional IRA doesn’t make any sense, unless it’s just to do a backdoor Roth.

I've always thought this showed that traditional IRAs are, on average, a good deal. Traditional IRAs would benefit higher earners more, but they are means tested out. And, of course, the 401k offers traditional contributions in whatever random plan that person's employer has.

pointlesspart
Feb 26, 2011

Leperflesh posted:

I've never heard of someone voluntarily not putting money into tax advantaged space that they completely understand, specifically because they have calculated that they'll pay less tax on that money in aggregate at the end. Maybe that's this edge case we're talking about and it turns up, but it's such a novel statement to me that I'm sort of reflexively super skeptical of it.

I think you're excluding the cases you would have heard about. People voluntarily don't take advantage of tax advantaged space they fully understand in spite of the tax advantages offered, not because of those advantages.

I can use myself as an example. I have some extra money saved outside of available tax advantaged space (Mega Backdoor Roth, but still). The optionality on the taxable savings to buy a house, car, etc is worth something over the tax advantaged space, especially since I would have to season the contributions for 5 years before I could withdraw it. The timelines just don't match up.

For a lot of lower income people, their savings have to function as extended emergency funds. Locking savings up in retirement funds means they can't be used in the event of job loss/car repairs/medical bills/family emergencies (at least without penalty). It isn't an ideal financial strategy, but it is a strategy people use. There are better ways to handle it, ignorance is certainly a problem.* But there aren't no reasons. Just bad or sad reasons.

*Also, strategies like the one described here https://www.bogleheads.org/wiki/Roth_IRA_as_an_emergency_fund, while financially superior, seem behaviorally suspect.

Adbot
ADBOT LOVES YOU

pointlesspart
Feb 26, 2011

Antillie posted:

I guess I can go into my situation a bit as well. While I do max out my wife and I's Roth IRAs each year I don't max out my Roth 401k and instead contribute a bit to my taxable account. (the 401k still gets *much* more than the taxable) I like knowing that if something happens I have a large amount of money readily available in addition to my emergency fund. For example if something were to happen to me my wife would have easy access to enough money to take care of herself and the kids for ~10 years before considering my life insurance payout (she doesn't work and she never fully recovered from aggressive chemo 9 years ago so getting a job would be hard for her). Likewise if I suddenly needed to reduce my monthly expenses by paying off my mortgage I could do that. Or I could slowly sell assets to cover the payment depending on the situation. Or maybe some crazy medical expense comes up that just blows past my emergency fund. My wife and I have both had cancer and that's not something you can ever really consider "cured".

I know that selling assets my taxable account would trigger long term capital gains taxes and depending on what the market is doing at the time it might be a very sub optimal thing to do but at least the option is there if I really need it.

That makes sense. You should see how much of projected 10 year expenses your wife could cover in a worse case scenario out of Roth contribution withdrawals and include that in worst case scenario planning. There is also SEPP, just paying the penalties, or hitting a penalty free scenario like a hardship withdrawal (medical expenses are a covered hardship in most plans, check yours) in the worst case scenario. Those could make you more comfortable using more of the Roth 401k space. But a Roth 401k is a kind of awkward fit for a 10 year, pre retirement savings needs, taxable is an easier fit.

Leperflesh posted:

My comma was probably confusing. What I meant was, I've never heard of people declining to use their tax advantaged space specifically because of having calculated that it would actually save them taxes to not use that space.

Thank you, I misunderstood.

pointlesspart fucked around with this message at 03:27 on Apr 24, 2024

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply