Tax Alchemy: Turning Interest Income Into Capital Gains

I recently learned of two brand new ETFs (launched just last week) that purport to provide only price appreciation rather than interest income, despite being bond funds. The benefit of such would be that all of the return that shareholders receive would be capital gains — and thus potentially taxed at the favorable long-term capital gain tax rates, rather than being taxed as ordinary income.

• F/m Compoundr U.S. Aggregate Bond ETF (CPAG)
• F/m Compoundr High Yield Bond ETF (CPHY)

The strategy is very simple. Each of the funds is a “fund of funds.” But at most times, each fund will own only one or two underlying ETFs.

• The aggregate bond ETF (CPAG) expects to own only iShares Core US Aggregate Bond ETF (AGG).
• And the high-yield bond ETF (CPHY) expects to own iShares Broad USD High Yield Corp Bond ETF (USHY) and SPDR Portfolio High Yield Bond ETF (SPHY).

And, on the day before the ex-dividend date for the underlying fund, these new ETFs will swap that underlying fund for a substitute fund which 1) has broadly similar holdings and 2) which will not be paying a dividend on that day. And then on the following day, the new ETF swaps back to the “normal” underlying fund.

So in short the idea is to just own a boring/normal bond ETF in the category in question, and temporarily swap it out for some other boring/normal bond ETF at the appropriate time in order to avoid receiving any dividend distributions.

(Terminology note: the distributions from a bond fund are still known as “dividends,” but they’re taxed as interest when it’s ultimately interest income from the underlying bonds that is being distributed.)

So, in theory, investors would get approximately the same total return and risk characteristics as the underlying fund, but with better tax-efficiency.

So what’s the catch?

Costs

The first catch is a simple and obvious one: the new ETFs add a layer of expenses. CPAG will charge a management fee of 0.39%, and CPHY will charge a management fee of 0.49%. Those costs are in addition to the fees of the underlying ETFs.

So any tax-efficiency that you gain would have to overcome that additional cost every year. The higher that interest rates are and the higher your marginal tax rate, the more likely it is that that the tax savings would overcome the costs. The lower that interest rates are and the lower your marginal tax rate, the less likely it is that the tax savings would overcome the additional costs.

Tracking Error

The next potential concern is that, even ignoring the costs, the new ETFs might not achieve the same performance as their primary underlying holdings, due to the periodic swapping of those primary holdings for substitute funds. Nasdaq (which is operating the new indexes that these new funds will track) published a document showing the most likely substitute funds.

For instance, as substitutes for iShares Core US Aggregate Bond ETF (AGG), the best partners appear to be Hartford Core Bond ETF (HCRB) and Fidelity Total Bond ETF (FBND). Here’s a chart from testfol.io showing the performance of those three funds for as long as they’ve all been around:

They’re definitely very similar, but they’re not identical. Swapping one out for another for just a handful of days per year shouldn’t make a big difference. But it’s possible that it might.

What about IRC § 1258?

Another broad category of concerns can be described as, “are the tax code and Treasury department okay with this?” In other words, does the proposed strategy not run afoul of any rules?

My answer to that would be: not that I can think of, but it’s always possible I’m missing something.

Another fund company (Alpha Architect) has sought to achieve interest-to-capital-gains alchemy in another way with their Alpha Architect 1-3 Month Box ETF (BOXX). There’s an excellent article by Daniel Hemel that raises substantial doubt as to the validity of their strategy though.

But I don’t think the concerns raised in that article apply here. The code section in question (IRC § 1258) becomes an issue when “substantially all of the taxpayer’s expected return” is attributable to time value of money. With a box spread (the underlying investment strategy employed by BOXX), time value of money is the source of substantially all of the expected return. With an intermediate-term bond fund, there’s also interest rate risk at play.

But again, maybe there’s some other regulatory issue that I’m not thinking of.

Is This Really Necessary?

It often makes sense to try not to own bonds in a taxable accounts (i.e., own them in tax-deferred accounts instead, where you won’t have to pay tax on the interest every year anyway). If you don’t have to own bonds in a taxable account, then you definitely don’t need something like these new funds.

And if you do have to own bonds in a taxable account, you don’t have to own high-yield bonds or even a total-bond fund. Within the bond portion of a portfolio, we don’t have to diversify in the way that we do with the stock side of the portfolio. It’s not crazy for the fixed-income side of the portfolio to consist of nothing but Treasury bonds (i.e., omitting investment grade corporate bonds as well as high-yield corporate bonds). It’s perfectly fine to stick with something that’s reasonably tax-efficient to begin with, such as a short-term Treasury fund (or a tax-exempt bond fund, depending on your marginal tax rate and how municipal bond yields compare to yields on other bonds).

There’s always a risk to being a guinea pig. Personally, even if I were financially in the target market (i.e., needing to hold bonds in a taxable account), I would be inclined to take a “wait and watch” approach for at least a few years. I really don’t like my investments to be exciting. I’d wait until these are old and boring rather than new and exciting.