Munis -> Treasuries: now, or after retirement? (~18mo)

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Topic Author
undid
Posts: 34
Joined: Tue Mar 12, 2019 11:35 am

Munis -> Treasuries: now, or after retirement? (~18mo)

Post by undid »

Hello fellow Bogleheads!

Here's the situation:

1. Currently in high marginal tax bracket (35%)
2. As a result of (1), the vast majority of my bonds funds are allocated to nation-wide Municipal Bonds to save on taxes (a mix of VWIUX, VWLUX and VTEB to be precise).
3. Retirement is hitting in 18 months, at that point, and with my current asset allocation (munis), I will pretty much fall to $0 taxable income (dividends are mostly qualified dividends, which will fall into the 0% bracket).
4. As a result of (3), once I'm retired, Treasuries, with their higher yields, will be much more profitable than munis. Currently they seem to be ~1-1.5% above munis (comparing YTMs eg. VTEB with BND or VWIUX with VGIT).
5. Original plan was to keep the munis until hitting the first year of low taxable income, and at that point convert Munis -> Treasuries for the higher yields.

Here's the dilemma:

Since [nominal] yields are still attracting right now (~4.3-4.5%) on Treasuries, and the market consensus (as well as the Fed's own prediction) is that they will go lower in the near future (~6-18mo), I'm wondering if I should wait or do the conversion now.

If I bought say $100k (example figure) in a ~4.3% Treasuries funds right now, I'd be "guaranteed" in a sense, that amount of income. Indeed, if rates fall, the theory goes that prices will move higher to adjust, and as such income will remain constant (lower yield, on a higher-valued bonds = constant absolute nominal $ received). So doing the conversion now would guarantee a $4.3k yearly income in average over a timeline equal to the avg duration of the fund. While if I wait, and do the conversion in say 18-24mo, rates may at that point be let's say at 3.5% (made-up figure), meaning I'd get $3.5k instead of $4.3k annually.

But at the same time... the theory I guess should go the same way for my current muni bonds holdings. So if rates do fall as the market expect, I guess my current say $100k muni holdings will rise in value meaning I wouldn't be converting $100k but something like say $110k, making it a non-issue to wait.

Except... is it really that simple? :) I've observed that muni bond funds, for the same duration as a Treasuries fund (of same duration), aren't quite as volatile as the Treasuries.

If you take VGIT vs VWIUX which are the same duration (and VWIUX has significantly longer maturities) from ~Oct 2021 to now (March 2024) you're looking at:

- Fed rate: +5.25% (0.08% --> 5.33%)
- Treasuries 5Y yield (should be equal to VGIT's): yields/rate: +3.1% (1.15% --> 4.25%) / price: -13.5%
- Munis (VWIUX): yields/rate: +0.9% (2.5% --> 3.4%) / price: -7.75%

I'm not too sure why ~5yr duration munis in Oct 2021 were yielding way more than the 5Y treasury, but fact is, munis only went up ~1% (~1.45% if you assume 35% tax bracket) in yields, treasuries went up ~3%, and price movements were vastly different (VGIT moved almost double that of VWIUX).

This makes me think that if the same happens in reverse when rates go down, I'd be better off converting now as VGIT/Treasuries would likely react more than munis in terms of yields movements, and as a result in terms of price movements. This should also make sense from a theoretical PoV: since munis generally target >30% tax people (with a 70% muni/treasury yield ratio), their yields would be expected to move only 70% of the treasuries' yields, meaning their prices also would move only 70% of that of treasuries... or am I mixing things up?


Another problem in executing the conversion is also that Treasuries funds seem to not have much of an offering between 5y duration (say VGIT/BND) and 15y duration (VGLT)*. While my munis are somewhere closer to ~7-8y in duration (overweight in VWLUX recently/since yields started climbing higher). I guess I could do some math to buy a mix say 80% VGIT and 20% VGLT but that seems like a pain to manage...



*: I had a look at iShares, and Schwab, and they also seem to be 2y - 5-6 - 15y - 25y, with nothing between intermediate-term (5-6) and quite long-term (15yr)
User avatar
grabiner
Advisory Board
Posts: 35431
Joined: Tue Feb 20, 2007 10:58 pm
Location: Columbia, MD

Re: Munis -> Treasuries: now, or after retirement? (~18mo)

Post by grabiner »

Timing the bond market doesn't work; if munis yield more after-tax than Treasuries, this is because investors are demanding a fair risk premium.

But you do know something that the bond market doesn't know: you can get a higher after-tax yield on munis than on Treasuries of comparable risk now, and a lower yield after you stop working. So you should sell at the start of your low-income year. (If you have capital gains, you should wait to sell until after January 1, so that the capital gains will be taxed at a lower rate. If you have capital losses, you may want to sell just before the end of the year so that you can use the losses one year earlier.)
Wiki David Grabiner
Topic Author
undid
Posts: 34
Joined: Tue Mar 12, 2019 11:35 am

Re: Munis -> Treasuries: now, or after retirement? (~18mo)

Post by undid »

grabiner wrote: Tue Mar 26, 2024 7:36 pm Timing the bond market doesn't work; if munis yield more after-tax than Treasuries, this is because investors are demanding a fair risk premium.
Oh? Risk premium for munis? For what? For corporate bonds I would understand. But at this point the US government is more likely to default on its debt than the entire market of munis all at once. The debt/income ratio of muni issues is something like 3x lower than that of the US government (granted the US government can print money, but that creates inflation which then generates higher rates which then affect bond prices negatively so it's not really a pros for bonds that the US gov can print money).

Also to be clear, this isn't really timing the bonds market in my view. Bonds do react to interest rates. So what I'm looking at is more "provided that there is a negative change (going lower) in interest rates, would I be in a better place in munis or in treasuries".

I've done more research and thinking into this since I posted and I think I'm leaning towards converting now. Reason is: the long-running muni/treasuries yields ratio is <100%, and it cannot really be above 100% for long since >100% means muni after-tax yields are way more interesting than treasuries (so efficient markets will rebalance). If the munis/treasury ratio stays <100%, then yields movements for treasuries will always be wider than yields movements for munis. As a result, price movements will also be wider.

Now if interest rates end up not going lower than that's fine by me. I don't think there is a great chance of rates going much higher than they currently are so worst-case scenario it's a draw. Also, if they go way higher, a bond holder is screwed regardless of what they're holding.
kenoryan
Posts: 267
Joined: Tue Sep 12, 2017 7:11 pm

Re: Munis -> Treasuries: now, or after retirement? (~18mo)

Post by kenoryan »

Would it make sense to take your loss in muni, deduct that from any realized gains, and then reinvest the $100k in treasuries?
Topic Author
undid
Posts: 34
Joined: Tue Mar 12, 2019 11:35 am

Re: Munis -> Treasuries: now, or after retirement? (~18mo)

Post by undid »

kenoryan wrote: Tue Mar 26, 2024 11:09 pm Would it make sense to take your loss in muni, deduct that from any realized gains, and then reinvest the $100k in treasuries?
I did not compute how much I could save tax-wise from selling at a loss, that's an interesting point. However, my losses are almost entirely long-term capital losses, and I actually have some short-term gains on this.

More importantly, I don't have any gains that need to be offset this year, and tax-gain harvesting now, just ~18mo ahead of falling into the 0% long-term-capital-gain bracket wouldn't really make much sense (since, if I want to harvest gains, I can just do so once I'm in the 0% long-term cap gain bracket)
Post Reply