iShares Core Suite Dominates July ETF Flows

BlackRock’s iShares Core suite won the battle for inflows in July. It’s not even close. Seven out of the top 10 July inflows went to iShares funds—five in the Core suite. iShares Core S&P 500 ETF (IVV-US) took in $4.10 billion vs. $1.06 billion for its direct competitor, AUM giant SPDR S&P 500 ETF Trust (SPY-US). Never mind that SPY-US ended June with twice the assets of IVV-US.

Here are the top 10 asset gatherers by dollar amount among U.S.-domiciled ETFs for July 2017:

The 25-fund iShares core suite captured $12.6 billion, or 51.4% of the net inflows for all 2060 ETFs that traded in the U.S. during July. Not too shabby for a fund set that comprised only 11.2% of the ETF landscape by AUM at the start of the month. The suite gained $2.57 billion of market share over competing funds.

What is iShares Doing Right?

BlackRock’s core suite is hitting the sweet spot of client demand. The thirst for cheap, broad-based, cap-weighted, dirt-simple ETFs is the major theme of 2017. That’s exactly what iShares core offers, with the addition of two value/growth funds, and four target risk funds that offer a one-fund complete stock and bond portfolio.

In some cases, iShares Core funds won the battle for flows because they have no competition. But in the majority of the 15 hotly contested segments where iShares Core battles Vanguard, Charles Schwab, and State Street Global Advisors for every dollar, the Core suite simply offered the best combination of exposure and cost. In fact, in five market segments, iShares Core edged out Schwab despite a .01% or .02% higher expense ratio.

The table below highlights a subset of funds from each of the segments: those cheap vanilla funds that are direct competitors to the iShares Core suite, plus iShares’ legacy funds, which are vanilla, but not cheap.

The extent of iShares’ core suite’s rout within the cheap vanilla space becomes clear when we look at flows vs. AUM-based flows expectations. The flows gap—the difference, in millions of dollars, between July’s actual inflows, and the fund’s pro-rata share of the net segment flows—tilts in iShares Core’s favor in 11 out of the 15 segments.

Flows Gap by Segment and Issuer/Suite, July 2017, in $ Millions

*Vanguard fund is VXUS, not VEU
** US Mid Caps experienced net outflows in July.  Although SSgA's MDY's outflows were lower than its pro-rate share, leading to the $232 million positive flows gap.
Schwab's SCHM-US saw atual net inflows of $19.1 million

The July flows data presents a puzzle: iShares’ Core suite pretty much ran the tables in July, even though Vanguard and Schwab funds are often just as cheap —sometimes cheaper—and mostly offer excellent liquidity. Clearly, investors were looking at something beyond costs. The key to the puzzle may be this: in 12 of the 15 segments where iShares Core ETFs compete, the competitors’ cheap vanilla funds don’t have the same portfolios. They track different indices. And that matters more often than we might think, because index composition drives returns. 

Know Your Index

Take the example of two well-known U.S. large cap indexes. The Russell 1000 has twice the constituent count of the S&P 500. The additional 500 constituents matter. Over the 12 months through July 31, iShares Russell 1000 ETF (IWB-US) returned 15.79%, while iShares Core S&P 500 ETF (IVV-US) returned 15.99%. Returns are based on NAVs and include re-invested dividends.

The 0.20% returns difference is nearly twice the 0.11% difference in expense ratios.

Sometimes, differences in index methodology can really smack an investor’s bottom line. Take a look at the one-year market total returns (to July 31, 2017) for the five emerging market equity funds compared below. You’ll find a 3.87% difference between MSCI and FTSE’s offerings. 

MSCI classifies South Korea as emerging, but FTSE (tracked by Vanguard and Schwab ETFs) and S&P consider it to be developed. S&P and FTSE also differ on which Chinese securities to include. The takeaway: indexes with similar names can have quite divergent portfolios, so it pays to look under the hood.

And that’s just what July’s investors did.

Suddenly, Investors are Index Nerds

It’s not that investors disregarded costs. Expense ratios and spreads still mattered. But, where cost differences were slight—one basis point, maybe two—it seems that investors chose funds based on the index they track. Investors sought full market cap coverage, and well-researched groupings (such as large/mid/small)—the same aspects that are emphasized in the segment benchmarks behind FactSet’s ETF Fit score.

Here’s a table showing several common ETF selection criteria, with indications of which ETF issuer’s fund has the advantage in each segment. Green highlights indicate that the advantage fell to the winning fund in July.

* iShares Core, Vanguard, and SSgA all offer S&P 500/400/600 funds. These funds have identical Fit scores. Large and small cap investors chose the cheapest S&P-based funds; mid-cap investors chose the cheapest fund overall.  

The winning fund—the one that garnered the most dollars compared to its initial market share—had rock bottom costs just under half the time (seven out of 15). The winner had the absolute lowest trading spreads in only a third of the cases. And size didn’t seem to matter that much either, as only 40% of the winners were also the largest fund. Only real estate investors seemed willing to bet on funds that took on active risk against FactSet’s segment benchmark.

Choosing the Best Benchmarks

The FactSet ETF Fit score measures the degree of similarity, in both portfolio composition and historical returns, between a fund and our segment benchmark.  The segment benchmark is ETF Analytics’ bogey for every fund classified into a segment—a particular slice of the securities markets, such as small cap Japan equities or short-term high-yield U.S. fixed income. A high fit score indicates that the fund is a close match to the segment benchmark, while a low score points to significant bets against the market. Fit is an inverse measure of active risk: low active risk against our segment benchmark generates high fit scores, and vice versa.

FactSet’s ETF Analytics selects segment benchmarks the hard way, by reading the entire index construction methodology document from every major index provider, in order to identify the most comprehensive, representative, well-researched index suite available. We look at the breadth of the universe (in terms of market cap and countries covered)—broader is better, because it captures more of the global opportunity set, by country and by market cap. We also look at how the indexer draws divisions, such as between large, mid, and small caps, value and growth, sectors/industries, emerging, and developed markets. Our preference is to build these dividing lines using well-researched principles rather than drawing arbitrary lines or relying on committees.

It’s not an exaggeration to say that a high ETF Fit score means the fund closely conforms to FactSet ETF Analytics’ index nerd standards.

In July, in 11 of the 12 segments where ultra-cheap vanilla funds compete with different underlying indexes as well as on costs, investors bet on the highest-fitting funds. No other metric came close to explaining investor behavior this month.

Tellingly, where investors chose among funds tracking the S&P 500/400/600 suite, where portfolios are virtually identical, cost drove the July flows decisions in large caps and small caps. Mid cap investors went with the cheapest expense ratio overall, withdrawing assets from almost every vanilla midcap while adding a mere $19 million to Schwab’s U.S. Mid-Cap ETF (SCHM-US).

iShares core offers low expenses, low spreads, and high-quality indexes. July’s ETF investors voted for all three. Welcome to the ETF nerd club.