ETF providers last year got better at what they do, serving up funds that more closely tracked the performance of their underlying indexes, according to a Morgan Stanley study that found tracking error across the U.S. ETF market has improved by 30 percent in the last year alone.
The Wall Street bank said that year-on-year, nearly three-quarters of U.S.-listed ETFs had lower tracking errors in 2011 than the year before, and only one in 11 funds had tracking error of more than 1 percent. Tracking error is the difference in total return between an ETF’s net asset value (NAV) and its underlying index.
Tracking error across the 700-plus U.S. ETFs surveyed in 2011 averaged 52 basis points at the end of 2011—as much as 22 basis points below the 2010 average, and a far cry from the 1.25 percent average seen just two years ago. What’s more, more than a third of the ETFs surveyed kept tracking error below the 25-basis-point mark, an improvement from a 1-in-4 ratio seen in 2010.
Tracking error is a key measure of just how good an index ETF is at meeting its goal of replicating its benchmark’s performance. It’s become increasingly important as ETFs canvass areas of the market broad and narrow, providing a viable alternative to more expensive active management. The Morgan Stanley study excluded active ETFs, as well as leveraged and inverse funds.
“As a result of lower tracking error in 2011, 53 percent of ETFs exhibited tracking error less than or equal to their expense ratios,” the company said in the report. “This is a constructive development for ETFs.”
Morgan Stanley noted that factors like fees and expenses of a fund, optimization and diversification requirements, scale, access to capital markets around the world, as well as index turnover and dividend reinvestment policies can affect a fund’s NAV relative to its benchmark.
“A higher ratio of expenses to tracking error corresponds to more effective tracking of index performance,” the company added.
Expense Ratios Are Key
Investors may not realize it, but what they pay in management fees is counted as part of tracking error and, moreover, it’s often the single biggest source of tracking error.
It shouldn’t come as any surprise, then, that U.S.-equities-linked ETFs tapping into style and major markets consistently show some of the lowest tracking errors because those ETFs also have some of the cheapest price tags.
Most U.S.-equities-linked ETFs—from major market funds, to style funds, to specific sector strategies—ended the year on average with tracking errors of 35, 26 and 47 basis points, respectively.
AMLP:A Case Unto Itself
By comparison, U.S. equities ETFs with niche strategies that invest in everything from a buy-write portfolio to clean energy as well as long/short funds—had average tracking error of as much as 95 basis points, and were the only category of U.S. equities ETFs that saw tracking error performance deteriorate in the last year.
Behind that performance shortfall was the biggest offender in the tracking error game:the Alerian MLP ETF (NYSEArca:AMLP - News), an ETF that serves up a portfolio comprising master limited partnerships that also happens to represent more than 40 percent of total assets in the “U.S. Custom ETFs” space surveyed, the report said.
AMLP ended 2011 with a tracking error of 695 basis points—nearly 7 percent off its benchmark performance, a number that helped skew broader averages. AMLP’s index returned roughly 17 percent, meaning the tracking error knocked down actual returns to around 10 percent.
“The deviation in AMLP’s NAV performance versus its index is primarily the result of the fund’s structure as a C-corporation, which requires the fund to pay corporate income taxes on profits,” the report said.
“This accounting treatment may cause AMLP’s NAV to underperform its benchmark in rising markets, such as 2011, or outperform in declining markets.”
Optimization A Culprit
Another culprit behind tracking error is optimization, when an ETF opts to own some but not all of the securities in its index in an effort to keep portfolio costs down and turnover low.
Optimization is frequently used by ETFs canvassing relatively illiquid markets, which is why so many bond funds use sampling strategies.
Fixed-income ETFs had an average tracking error of 40 basis points in 2011, an 18-basis-point improvement from the year before and a five-year low.
That said, not all fixed-income funds are created equal.
Indeed, some bond funds, such as a lineup of Invesco PowerShares ETFs, showed just how tricky it can be to achieve that perfect track record in the fixed-income space. Bond funds like PowerShares’ PICB, PWZ, BAB and PHB each ended the year anywhere from 147 to 233 basis points off their benchmarks.
BAB, for instance, is highly optimized, owning less than 3 percent of the securities comprising its BofA Merrill Lynch benchmark.
It might come as no surprise that some of the widest tracking errors came from international equities ETFs that often rely on less liquid securities or turn to derivatives to access remote regions of the global economy.
On average, international equities ETFs ended the year with a tracking error of 68 basis points, according to data compiled by Morgan Stanley.
The iShares MSCI Belgium Index Fund (NYSEArca:EWK - News), for instance, ended the year 296 basis points off its index—EWK’s NAV dropped more than 15 percent in 2011 while its underlying benchmark dropped by 12.2 percent.
Many Asia-Pacific funds, as well as most of the China-linked ETFs surveyed, also struggled to keep their performance closely tied to their underlying strategies.
The Global X China Industrials ETF (NYSEArca:CHII - News), for instance, ended 2011 with a tracking error of 284 basis points, while the SPDR S'P China ETF (NYSEArca:GXC - News)’s tracking error hit 270 basis points.
As a group, funds that serve up exposure to Brazil also generally missed their indexes’ mark, many underperforming their benchmarks.
The NAV of the market’s largest Brazil-focused ETF, the iShares MSCI Brazil Index Fund (NYSEArca:EWZ - News), underperformed its index by 47 basis points last year, while a group of Global X Brazil ETFs came in roughly 1 percent below their underlying indexes.
Positive Tracking Error
But not all tracking error is necessarily a sign of underperformance.
In rare instances, some funds manage to outperform their benchmarks.
The Market Vectors Egypt Index ETF (NYSEArca:EGPT - News) was another fund where a nearly 2 percent tracking error average was in fact an outperformance of its index. EGPT’s NAV slid in 2011 by 49.8 percent while its benchmark dropped 51.7 percent.
Providers Under The Microscope
From a provider standpoint, just about every ETF firm in the U.S. market saw its tracking error average, in absolute terms, drop in 2011 from previous years.
Which types of ETFs they offer as well as the costs associated with those strategies are the biggest underlying factors determining a provider’s overall tracking performance, the report said.
Still, in 2011, the leading exception, as noted, was ALPS, which averaged a tracking error of 216 basis points across its five ETFs surveyed, the report said. Just a year earlier, the company’s ETFs averaged a tracking error of under 1 percent.
Other companies such as Pimco and U.S. Commodity Funds also ended the year with slightly higher tracking error averages than they did in 2010.
But companies like Emerging Global Shares narrowed the performance gap last year, ending 2011 with an average absolute tracking error of 77 basis points, compared with a 221-basis-point average a year earlier. EGShares is behind 19 ETFs.
Some of the largest players in the ETF industry, such as BlackRock and State Street Global Advisors, also showed consistently improving tracking performance.
BlackRock, parent of the world’s largest ETF provider iShares, averaged 40 basis points in tracking error last year, down from 49 basis points a year earlier and 85 basis points in 2009. The company has more than 250 ETFs on the market.
SSgA meanwhile ended the year with an average tracking error of 46 basis points across its 85-plus ETF platform—significantly lower than its 142-basis-point average just two years ago.
Vanguard, the quickly growing low-cost ETF provider behind more than 60 ETFs, came in with an average tracking error of 17 basis points, unchanged from 2010, but a significant improvement from its average of 83 basis points in 2009.
Permalink | ' Copyright 2012 IndexUniverse LLC. All rights reserved