roth vs pre-tax and tax efficency

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
babystep
Posts: 432
Joined: Tue Apr 09, 2019 9:44 am

roth vs pre-tax and tax efficency

Post by babystep »

Hi,

It is mostly accepted to put bonds in the pre-tax and stocks in the Roth IRA. I am wondering if there are other ways to expand the idea.

e.g. 2x long S&P in Roth IRA and 2x short S&P 500 in pre-tax? This is just to explain the idea. This is kind of odd that you are mostly neutral and not long at all. I am not sure how this would behave for a long period of time like 10 years.

Other options
LEAPs such that you are more likely to have gains in the Roth at the expense of equivalent loss in the pre-tax?
2x leveraged ETF and long dated bonds?

Any thoughts on this?
ckangas
Posts: 54
Joined: Fri May 21, 2021 8:08 pm

Re: roth vs pre-tax and tax efficency

Post by ckangas »

You can typically rebalance a portfolio within a Roth IRA or within a pre-tax 401K, but not across them.

I might be missing your question, but you'll likely need all relevant assets in each account to rebalance between them. A new investor can likely keep things balanced when they're adding a high portion in each year. But eventually, it's likely to become unbalanced. And you're capped (excluding something like a mega Roth backdoor?) at Roth IRA contributions each year.

And whether a Roth IRA or pre-tax 401K space is more valuable comes down to your specifics; obviously dollar for dollar Roth IRA is more valuable, and there's some added flexibility such as being able to withdraw deposits without penalty. But it's mostly just whether you are paying those taxes now or later.
User avatar
grabiner
Advisory Board
Posts: 29866
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: roth vs pre-tax and tax efficency

Post by grabiner »

The problem with going long in the Roth and short in the traditional account is that it isn't market-neutral.

If you are going to retire in a 25% tax bracket, you can view the traditional account as being 25% owned by the IRS, and 75% owned by you tax-free. $4000 in the traditional account and $3000 in the Roth account will have the same after-tax value if invested the same way. Thus, if you go long $4000 in the Roth and short $4000 in the traditional, you have $1000 of net exposure to the stock market. If you go long $3000 in the Roth and short $4000 in the traditional, you are market-neutral, but you aren't getting any benefit from it either; your return will be the same as cash less expenses.

There are still minor advantages for holding more volatile assets in a Roth. If your traditional account grows too large, you might have to withdraw in a higher tax bracket than you expected, or take RMDs you don't need, so you break the equivalence. Thus, if all else is equal, it is better to hold bonds in a traditional account and stocks in a Roth account. (This is subject to your asset allocation; if the traditional account isn't large enough to hold all your bonds, and your tax situation makes holding bonds in tax-advantaged accounts desirable, you should hold some bonds in the Roth account.)
Wiki David Grabiner
tomsense76
Posts: 407
Joined: Wed Oct 14, 2020 1:52 am

Re: roth vs pre-tax and tax efficency

Post by tomsense76 »

Have also been wondering about this

Though what grabiner says makes sense. If your net exposure to stocks over all accounts is zero, that doesn't help much

Maybe one can use an S&P 2x fund in tax-free and only bonds in pre-tax to achieve the desired exposure? There may be issues with that (levered ETFs suffer from volatility decay for example). That said, am not sure what the other issues are here, maybe someone can provide more guidance
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
User avatar
grabiner
Advisory Board
Posts: 29866
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: roth vs pre-tax and tax efficency

Post by grabiner »

tomsense76 wrote: Mon Jun 07, 2021 11:47 pm Have also been wondering about this

Though what grabiner says makes sense. If your net exposure to stocks over all accounts is zero, that doesn't help much

Maybe one can use an S&P 2x fund in tax-free and only bonds in pre-tax to achieve the desired exposure? There may be issues with that (levered ETFs suffer from volatility decay for example). That said, am not sure what the other issues are here, maybe someone can provide more guidance
In theory, this would work. However, the higher expenses and the difficulty of rebalancing (to eliminate volatility decay) make this probably not worthwhile. If you have $3000 in stock in a Roth and $4000 in a traditional account, you are 100% stock after tax-adjustment no matter what happens. But if you have $3000 in 2x-levered stock in a Roth and $4000 in bonds in a traditional account, you are 100% stock now.

If the market falls so that you have $2000 in 2x-levered stock in the Roth, you now have only 80% stock in after-tax value and need to buy more stock in the traditional account to keep the same allocation. This is the same phenomenon as volatility decay. If you don't rebalance and the market recovers ($6000 in stock fell 16.7% to $5000 and then rose 20% to $6000), you will have $2800 in 2x-levered stock in the Roth, so you lost $200 even though the market broke even.
Wiki David Grabiner
pattiwack
Posts: 15
Joined: Wed Jan 19, 2011 8:34 pm

Re: roth vs pre-tax and tax efficency

Post by pattiwack »

grabiner wrote: Mon Jun 07, 2021 11:39 pm If you are going to retire in a 25% tax bracket, you can view the traditional account as being 25% owned by the IRS, and 75% owned by you tax-free.
The only problem I have with this statement is it would only seem accurate if you already have an income stream pushing you right up to the beginning of the 25% bracket. If your only income stream is the tIRA then it's really the effective tax rate.

For example, if you wanted to withdraw 80k with no other income (say before you start drawing SS), then the tax bracket (for MFJ) would be 12%, but the effective tax rate would only be 7.8%.

I think for large portfolios ignoring effective rates probably makes sense because there are usually other sizeable income streams also present. For more modest portfolios that are primarily a 401k/tIRA, it would appear that a tIRA is pretty close to tax-free.
tomsense76
Posts: 407
Joined: Wed Oct 14, 2020 1:52 am

Re: roth vs pre-tax and tax efficency

Post by tomsense76 »

grabiner wrote: Tue Jun 08, 2021 8:52 am
tomsense76 wrote: Mon Jun 07, 2021 11:47 pm Have also been wondering about this

Though what grabiner says makes sense. If your net exposure to stocks over all accounts is zero, that doesn't help much

Maybe one can use an S&P 2x fund in tax-free and only bonds in pre-tax to achieve the desired exposure? There may be issues with that (levered ETFs suffer from volatility decay for example). That said, am not sure what the other issues are here, maybe someone can provide more guidance
In theory, this would work. However, the higher expenses and the difficulty of rebalancing (to eliminate volatility decay) make this probably not worthwhile. If you have $3000 in stock in a Roth and $4000 in a traditional account, you are 100% stock after tax-adjustment no matter what happens. But if you have $3000 in 2x-levered stock in a Roth and $4000 in bonds in a traditional account, you are 100% stock now.

If the market falls so that you have $2000 in 2x-levered stock in the Roth, you now have only 80% stock in after-tax value and need to buy more stock in the traditional account to keep the same allocation. This is the same phenomenon as volatility decay. If you don't rebalance and the market recovers ($6000 in stock fell 16.7% to $5000 and then rose 20% to $6000), you will have $2800 in 2x-levered stock in the Roth, so you lost $200 even though the market broke even.
Yeah that's a good point. Agree leveraged ETFs are probably not a good solution. Was re-reading this thread ( viewtopic.php?t=344121 ) (had seen it before, but not read in detail). There are unfortunately just a lot of issues with levered ETFs. There's something nice about the more hands-off nature, but yeah it's just not a good long term strategy.

Perhaps rolling futures would work better? Though I think this runs afoul of the same issue you have identified here about significant crashes resulting in one's portfolio getting out of wack without a way to fix it.

One thing related to this that had occurred to me is doing some Roth conversions at that point. It does mean paying taxes, but one still might come out ahead (for example a 50% decline means the taxes one would have paid before the decline are now halved). Also one is effectively using bonds in pre-tax to buy stock in tax-free when it is down, which seems like a good idea generally. Not sure if there are other wrinkles there though.

That all being said, I think the real issue with all of this is it could just be too hard behaviorally. It relies on someone being rather involved in their portfolio (particularly during major down days) and staying rational. It reminds me of something Morgan Housel said (which I'm probably going to butcher as I can't find the quote), "it's not about how well your portfolio performs, but about how well you perform."
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
Lee_WSP
Posts: 4712
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: roth vs pre-tax and tax efficency

Post by Lee_WSP »

tomsense76 wrote: Tue Jun 08, 2021 3:42 pm
Perhaps rolling futures would work better? Though I think this runs afoul of the same issue you have identified here about significant crashes resulting in one's portfolio getting out of wack without a way to fix it.
It's been a while since I ran the numbers, but I believe the end result of rolling futures is approximately the same as a 2x LETF since there's a tax drag more or less equivalent to the expense ratio of the ETF. LEAPS was the other popular alternative, but the hidden "fees" from the spread and any execution risk may also end up being more or less similar to an ETF.

If I were going to do the supercharged stocks in Roth strategy, I'd just rebalance the accounts within each account. Ie, rebalance just the Roth and then all the trad accounts.
User avatar
grabiner
Advisory Board
Posts: 29866
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: roth vs pre-tax and tax efficency

Post by grabiner »

pattiwack wrote: Tue Jun 08, 2021 11:29 am
grabiner wrote: Mon Jun 07, 2021 11:39 pm If you are going to retire in a 25% tax bracket, you can view the traditional account as being 25% owned by the IRS, and 75% owned by you tax-free.
The only problem I have with this statement is it would only seem accurate if you already have an income stream pushing you right up to the beginning of the 25% bracket. If your only income stream is the tIRA then it's really the effective tax rate.
This is the advantage of having enough in the traditional account to use up the lower brackets; the IRS owns less than 25% of the total dollars. However, the decision of how to invest traditional versus Roth is based on the marginal rate. If your marginal rate is 25%, then a $4000 investment in the traditional IRA which gains or loses 50% results in a gain or loss of $2000 pre-tax, which is $1500 after-tax. Therefore, it should be invested as if it were $3000 tax-free.
Wiki David Grabiner
Topic Author
babystep
Posts: 432
Joined: Tue Apr 09, 2019 9:44 am

Re: roth vs pre-tax and tax efficency

Post by babystep »

Thanks everyone for the responses. I understand that 2x leveraged ETF are daily reset so it will not work as desired.

Let us take one specific example for LEAP to analyze. Assume pre-tax is all bonds.

SPY current price is 422 and Dec 2023 LEAP, strike price $210, premium is $212. If you buy this call in Roth and hold it for the next 2 or more years. Would you expect 2x or close to 2x returns assuming market is up from here in the next 2 or more years?

Similarly, if the market is down in the next 2 or more years then would you expect that -2x or close to -2x return?

If the market is up then that is good scenario. But if the market is down then what choice do you have? Do you take the loss and buy another LEAP? If it goes below $210 then you could be wiped out as well.

I was hoping if there is some approach to leverage the stocks holding in Roth but looks like just holding the stocks in Roth and bonds or bonds + stocks in pre-tax is still the best option?
Lee_WSP
Posts: 4712
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: roth vs pre-tax and tax efficency

Post by Lee_WSP »

deleted, math incorrect
Last edited by Lee_WSP on Wed Jun 09, 2021 5:31 pm, edited 1 time in total.
User avatar
grabiner
Advisory Board
Posts: 29866
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: roth vs pre-tax and tax efficency

Post by grabiner »

babystep wrote: Wed Jun 09, 2021 2:15 am Thanks everyone for the responses. I understand that 2x leveraged ETF are daily reset so it will not work as desired.

Let us take one specific example for LEAP to analyze. Assume pre-tax is all bonds.

SPY current price is 422 and Dec 2023 LEAP, strike price $210, premium is $212. If you buy this call in Roth and hold it for the next 2 or more years. Would you expect 2x or close to 2x returns assuming market is up from here in the next 2 or more years?

Similarly, if the market is down in the next 2 or more years then would you expect that -2x or close to -2x return?
Yes. If you hold one option to expiration, and then sell it, you will have one share of SPY (one contract is actually 100 options), so you are trading the option price today for something worth $210 less than a share of SPY in two and a half years. You can buy this option for $212 Thus, if SPY doesn't move, your $212 will become $212 in two years, breaking even. If SPY rises by 10% to $464, your $212 will become $254, a 20% gain. If SPY falls by 10% to $380, your $212 will become $170, a 20% loss. Only if SPY falls below $210 does the equation break down, as you can't lose more than 100% of your investment.

So the combination of one option, and a bond worth $212 maturing on the same date, is almost equivalent to buying SPY itself, less the frictional costs. If you buy the bond and the option, and then exercise the option when it expires, you will have one share of SPY and the interest on the bond (unless SPY falls below $210). If you buy one share of SPY today, you will have one share of SPY and the dividends paid by SPY.
Wiki David Grabiner
tomsense76
Posts: 407
Joined: Wed Oct 14, 2020 1:52 am

Re: roth vs pre-tax and tax efficency

Post by tomsense76 »

Lee_WSP wrote: Tue Jun 08, 2021 6:30 pm
tomsense76 wrote: Tue Jun 08, 2021 3:42 pm
Perhaps rolling futures would work better? Though I think this runs afoul of the same issue you have identified here about significant crashes resulting in one's portfolio getting out of wack without a way to fix it.
It's been a while since I ran the numbers, but I believe the end result of rolling futures is approximately the same as a 2x LETF since there's a tax drag more or less equivalent to the expense ratio of the ETF. LEAPS was the other popular alternative, but the hidden "fees" from the spread and any execution risk may also end up being more or less similar to an ETF.

If I were going to do the supercharged stocks in Roth strategy, I'd just rebalance the accounts within each account. Ie, rebalance just the Roth and then all the trad accounts.
Interesting where's the tax drag coming from if this is held in a Roth?
"Anyone who claims to understand quantum theory is either lying or crazy" -- Richard Feynman
Lee_WSP
Posts: 4712
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: roth vs pre-tax and tax efficency

Post by Lee_WSP »

tomsense76 wrote: Thu Jun 10, 2021 2:31 pm
Lee_WSP wrote: Tue Jun 08, 2021 6:30 pm
tomsense76 wrote: Tue Jun 08, 2021 3:42 pm
Perhaps rolling futures would work better? Though I think this runs afoul of the same issue you have identified here about significant crashes resulting in one's portfolio getting out of wack without a way to fix it.
It's been a while since I ran the numbers, but I believe the end result of rolling futures is approximately the same as a 2x LETF since there's a tax drag more or less equivalent to the expense ratio of the ETF. LEAPS was the other popular alternative, but the hidden "fees" from the spread and any execution risk may also end up being more or less similar to an ETF.

If I were going to do the supercharged stocks in Roth strategy, I'd just rebalance the accounts within each account. Ie, rebalance just the Roth and then all the trad accounts.
Interesting where's the tax drag coming from if this is held in a Roth?
Most providers don't let you buy futures, and I forgot which thread this was.
Post Reply