In Defense of Direct Indexing

Direct indexing is technology that empowers advisors to replicate an index via a quantitative, rules-based approach to portfolio construction. It replicates a target index with a high R-squared or correlation. That replication exercise might incorporate ESG considerations, capital-gains budgets, legacy holdings, or limits on the number of underlying holdings.

Do these decisions constitute “active” management?

Critics allege that direct indexing could easily be abused by advisors to engage in active management. While any technology can be abused, the implication that ETFs deter advisors from engaging in active management is laughable. For example, having reviewed the portfolios of literally hundreds of RIAs over the years, I have yet to find a single firm that takes a market weight to non-US equities – a clearly subjective departure from the market. Many advisor firms make heavy sector bets or overweight growth stocks despite the overwhelming academic and empirical evidence against doing so. They’re not doing any of this with direct indexing technology, but with ETFs.

ETFs use is widespread among advisor firms. A significant number of advisors use ETFs to actively trade in and out of entire asset classes. Michael Kitces, in his article, “The Passive Investing Mirage and the Disintermediation of Active Mutual Fund Managers,” suggested that active management by advisors may be increasing – despite the widespread adoption of supposedly passive ETFs. Ironically, if there’s one market force that might delay the widespread adoption of direct indexing, it’ll likely be the resistance by some advisors to give up actively trading client portfolios using ETFs.

Direct indexing and tax management

For a plethora of reasons, direct indexing makes sense for taxable accounts. I won’t rehash the many studies that have attempted to quantify the tax alpha associated with portfolio tax management; most of us would agree that tax alpha exists. Tax alpha is significant and it’s much easier to harvest with direct indexing than other approaches to portfolio management.

Why? Because it’s better for taxpayers to have their own cost basis in their portfolios’ underlying holdings. With direct indexing, investors own individual securities; they subsequently have their own cost basis in those positions – not someone else’s. In doing so, they can harvest gains and losses at the individual position level, a benefit that’s impossible to realize with pooled investment vehicles, such as open-end mutual funds, where cost bases are socialized. With ownership of your own cost basis comes obvious benefits; investors in direct index portfolios can rebalance by gifting appreciated securities to charity (replacing them with cash). This of course isn’t the case with mutual funds, where shareholders inherit the fund’s bases in its underlying holding – a situation that often results in investors having to pay taxes on capital gains they themselves never earned. While this problem is less common with ETFs, it’s not entirely absent – especially if Congress does away with the preferential tax treatment of ETFs relative to their mutual fund counterparts.

Some have argued there are limits to tax management via direct indexing, since at some point the portfolio can become “locked up,” a situation where there are no more assets left to sell to realize losses to offset capital gains. All portfolios can become “locked up” after a long period of harvesting only capital losses, but this problem is not unique to direct index portfolios; it’s associated with tax management in general. And if we’re being precise, this is less of a problem for direct index portfolios holding many hundreds of securities than it is for portfolios consisting of perhaps only a dozen mutual funds and ETFs, since direct indexes have significantly more opportunities for ongoing tax management. And with direct index portfolios, it’s easier to surgically remove appreciated positions from the portfolio (thereby raising cash to facilitate rebalancing).

But while direct indexing is indeed attractive to taxable investors, it appeals to tax-deferred investors as well. There’s no reason an advisor couldn’t use direct indexing to build a customized ESG-focused portfolio for IRAs or non-taxable institutions. Other investors, regardless of the tax status of their wealth, may have a behavioral preference to “see” the individual holdings within their portfolios. Further, corporate executives or public accountants often face regulatory constraints on which companies or funds they can own in their portfolios – regardless of tax status. In such instances direct indexing could keep those clients compliant.

Direct indexing and diversification

Are direct indexes less diversified than their ETF counterparts? The proper way to answer this question isn’t based on the number of holdings but to instead focus on the portfolio’s diversification benefits, as measured by the portfolio’s R-squared relative to its benchmark. Let’s answer this statistically. To what extent do the largest 100 stocks in the S&P 500 Index explain the returns of the S&P 500? The answer: 99.7%. Said differently, the top 100 stocks in the S&P 500 explain 99.7% of the return variance we observe in the index.¹ The takeaway here for direct indexing is that you can capture the vast majority of diversification benefits of the S&P 500 by owning the largest 100 stocks in the S&P 500 Index.

To be clear, I’m not suggesting direct index portfolios won’t have tracking error relative to their respective indexes; nor am I suggesting that direct index portfolios should outperform their benchmarks. But so do ETFs and mutual funds. Sometimes tracking error helps; sometimes it hurts. The degree of tracking error in any portfolio comes down to the number of holdings and the various constraints advisors and clients put in place when constructing the portfolio. But this is an opportunity to educate advisors and investors, not a reason to avoid new technology.

Finally, direct index portfolios can be globally diversified within and across multiple asset classes; we do this quite easily via our unified managed account (UMA) provider whereby we own multiple direct index strategies, consisting of many hundreds of underlying securities across multiple global asset classes, and all within a single account. I’m not suggesting that every sleeve in an asset allocation should be implemented via a direct index solution. In our firm, it’s quite common for advisors to utilize direct index solutions for the largest sleeves in an allocation while using ETFs or mutual funds for things like small cap, fixed income, and non-US equities. Subsequently, arguments that direct index portfolios are “less diversified” than underlying ETF or mutual fund portfolios, whether within or across asset classes, are more academic than they are practical.

The takeaway for advisors

Advisor firms have two choices: They can compete on fees or on value. For those firms choosing to compete on value, direct indexing is a powerful value-added tool. Advisors can better justify their fees when they create, capture, and deliver greater value to clients; direct indexing helps them do that. Further, whether insourced or outsourced, direct indexing positions advisor firms to benefit from fee compression since it disintermediates the more expensive (“active”) SMA providers that have until recently dominated in the space.

Direct indexing provides more choice, greater transparency, better tax management, and lower fees. All of that is great news for clients.

Donald Calcagni is chief investment officer of Mercer Advisors. Mercer Advisors Inc. is the parent company of Mercer Global Advisors Inc. and is not involved with investment services. Mercer Global Advisors Inc. (“Mercer Advisors”) is registered as an investment advisor with the SEC. Content, research, tools, and stock or option symbols are for educational and illustrative purposes only and do not imply a recommendation or solicitation to buy or sell a particular security or to engage in any particular investment strategy. Past performance may not be indicative of future results.

¹Regression performed by the author using index data from YCharts, Inc. for the S&P 100 and S&P 500 Indexes.

Read more articles by Donald Calcagni