ichard Ferri would like nothing better than for the ETF market to be efficient enough so that he could use all ETFs. But it’s not, says Ferri, founder and chief executive of Portfolio Solutions, a registered investment advisor in Troy, Mich. So the beta-driven portfolios the firm constructs for its high-net-worth clientele are asset allocations of various index funds–including ETFs– diversified among a variety of asset classes.

In its analysis of asset classes, the Portfolio Solutions team looks for unique risk factors in the marketplace, Ferri says, pointing out that U.S. markets have different risks than Treasury bond markets, for example, and the risks in Pacific markets differ from those of European markets. “The point is that in a ‘normal’ market–if one thinks such a thing exists–the correlations between Europe and Pacific and emerging markets and U.S. stocks tend to deviate; real estate deviates from U.S. and so forth,” says Ferri.

Using a rolling correlation analysis to look at the deviations that occur lets the team see “how the correlations are shifting between asset classes to make a determination whether a particular asset class we’re considering is actually a unique asset class or not, or whether it’s just leveraged beta of some sort,” Ferri explains. Mid-cap stocks, for example, are basically leveraged beta; their returns indicate there’s no real risk in mid-caps, he says. Micro-caps, on the other hand, do have unique risk; sometimes they have a high correlation with large-cap U.S. stocks, but sometimes they don’t. “Mid-caps almost always have a high correlation with large-cap U.S. stocks,” he says. “All you’ve got there is leveraged beta, so we wouldn’t use that, but we would use microcaps.”

Once the team settles on a particular area of the market it wants to invest in, it looks at how best to represent that market class in its portfolios, says Ferri. “After we make the decision that we want to have, say, real estate in the portfolio, then we look at all the product that’s available out there–both open-end mutual funds and ETFs–and based on the underlying index and cost and turnover and liquidity, [we] make a determination which product we’re going to go with. If it turns out it’s an ETF, then it’s an ETF. If not, then we have an open-end fund.”

After a portfolio has been created, says Ferri, “we just do rebalancing and tax management; we don’t try to do tactical asset allocation, any stuff that requires speculation on the part of the managers.”

The ETF advantage

Ferri says it’s convenient when the method of delivery of an asset class is an ETF because this makes rebalancing simpler. “You can buy and sell ETFs all in the same day, whereas when you use an open-end mutual fund, you’ve got to wait a day or guess what you’re going to get as an NAV; so it gets to be more complicated and time consuming to use open-end funds than to use ETFs.”

There are other advantages to using ETFs in client portfolios, he adds. “We custody at Charles Schwab, and the commission cost to buy an open-end fund is higher than for an ETF. The creations and redemptions are a big deal. The advantage to the advisor is to be able to efficiently manage the business on a daily basis. You rebalance during the day rather than have to have delay. The advantages to the clients are lower commission costs, lower internal fees generally and more tax efficiency.”

Still, while he would like the ETF market to be efficient enough so that Portfolio Solutions could use all ETFs, he finds it wanting in the area of fixed income, so Portfolio Solutions stays away. “There was a time we tried the one bond ETF and realized there could be the possibility of some high premiums or discounts on that bond ETF, so we stopped using it about two years ago and went back to using all open-end funds. The bond ETF market is not as robust as the open-end fund market, with the exception of maybe a Vanguard fund,” Ferri says.

Even when using ETFs, an advisor has to be selective in choosing products, he continues. “An ETF should have some assets in it; it should have some reason for being; it should have some following.”

There is a lot of ETF product on the market at present, but in Ferri’s opinion, many funds have little liquidity. Funds with less than $20 million in assets after a year in existence are likely to close, he says. The break-even point for a new ETF is about $50 million; Portfolio Solutions won’t invest in a fund with less than $100 million.

“A lot of ETFs are closing right now, and we don’t want to be in an ETF that’s closing because there’s not enough liquidity in it, not enough trading,” says Ferri. “You need to have creations and redemptions in an ETF in order to take advantage of the tax benefit. That also lowers the spread between the bid and the ask. These are all the things we analyze when we’re trying to determine whether we should use an open-end fund or an ETF.”

Michael S. Fischer is a New York-based financial writer and editor. He can be reached at msf7@columbia.edu.