ETFs and OEFs: Funds and Taxes

Back on April 3rd, The Wall Street Journal ran its quarterly section on funds (mutual and exchange-traded), one part of which was the now-standard "How Well Do You Know" feature. These quizzes are generally interesting and useful, but they always seem to include passages that don't ring quite right. Two such items stood out this time.

First, there was question five: "True or false: ETFs are always more tax-efficient than comparable index-oriented mutual funds." The answer, of course, is false. ("Always" is a high bar to clear!)

In their explanation, authors William Power and Leslie Scism point to ETFs' creation-redemption process as a key reason for their superior tax-efficiency. But then they quote Morningstar's Jeff Ptak, who's been mentioned in this space before:

"Generally speaking, one would expect an ETF's arbitrage mechanism to make it more tax-efficient than a rival open-end index mutual fund," says Morningstar's Jeff Ptak. "However, one wouldn't ordinarily expect these differences to be enormous, as [indexed mutual] funds are low-turnover portfolios to begin with," meaning that they aren't typically racking up big realized capital gains.

There's certainly some truth to that. More or less by definition, low-turnover portfolios should realize fewer capital gains (and offsetting losses) than high-turnover portfolios. But we don't think Ptak's point gives enough weight to the fundamentally different processes driving mutual funds and ETFs. In the case of index-based mutual funds, the portfolio itself will be relatively low-turnover. But that's not the only thing that matters, because the redemption activity of some shareholders can create tax implications for all.

A hypothetical: Even if an S&P 500 index fund doesn't change its constituents at all from one year to the next, an open-end fund based on that index can still distribute capital gains to all shareholders due to the redemption activity of some shareholders. This is one reason we prefer ETFs to open-end funds. Not only is the vehicle itself better designed for tax-efficiency per se; it also puts the investor in charge of his or her tax destiny. Bottom line: With open-end funds, tax implications are functions of the fund manager's buying and selling and of other shareholders' buying and selling.*

Second, there was question seven: "What is the best way to avoid finding yourself in a mutual fund that makes big taxable distributions?" Answer: "Buy a fund that has investors pouring in new money." Here's the explanation from Power and Scism:

If you want to find a fund that's tax-efficient, look for one with big cash flowing in. Those big inflows mean that, by year end, the dividends and realized capital gains earned earlier in the year are distributed over a larger number of shares. By contrast, funds closed to new investors don't get the dilution effect that comes with all that new cash sloshing into a fund, so they often are tax-inefficient.

This strikes us as wrong on two levels: (1) it puts the tax cart in front of the investment horse and (2) it encourages a generally counter-productive brand of self-dilutive performance-chasing. After all, which funds are likely to have "investors pouring in new money"? The hottest funds most prone to mean-reversion. This notion of chasing new money into hot funds doesn't strike us as particularly logical or helpful for rank-and-file investors.

~~~~~~~~~~~~~~~~

* Savvy handling of tax lots can help fund managers minimize distributions, but that has its limits.

Source

William Power and Leslie Scism, "How Well Do You Know...Funds and Taxes?" Wall Street Journal, April 3, 2008

Comments

[gilbertofco...:]

Thank you for your insights. Perhaps you can enlighten us on a related topic.

I've been told that the tax treatment of capital gains on gold shares, e.g. GLD, is different than for other ETF's. If so, how so?

What about gold coins? Are they considered "collectables", and treated differently 'tax-wise'? Thanks, GoC

2008 Apr 21 12:11 PM

[ETF Investor:]

Gold ETFs like GLD- you hold the actual gold. It is classified as a collectible and thus the tax rate is 40% something

2008 Apr 21 12:53 PM

[Jeffrey Ptak:]

Hi Kevin.

One of the questions-and-answers in the WSJ piece you cited reads as follows:

"6) True or false: Heavy redemptions will leave an index mutual fund saddled with hefty capital gains.

ANSWER: False. Again, it isn't necessarily so. Any index-fund manager "worth his salt routinely harvests losses that can be used to offset gains as realized," says Mr. Ptak."

So, it's not as if the piece didn't address the question of flows and their impact on tax efficiency.

If you see the potential for 'enormous' differences between the tax-efficiency of open-end index mutual funds and comparable ETFs, I'd like to understand how. Past isn't prologue, I understand. But it's not as if the largest indexers---most of whom I'd consider 'worth their salt'--have had problems with giant cap gains distributions in the past. Given that, while I would ordinarily expect an ETF to be more tax-efficient, I again wouldn't foresee a scenario where there'd be huge differences.

Regards,

Jeff Ptak
Morningstar, Inc.

2008 Apr 21 02:01 PM