The new db X-trackers Harvest MSCI All China has a fund-of-funds element.
Last week, Deutsche Asset & Wealth Management launched its db X-trackers Harvest MSCI All China ETF (CN).
CN’s launch is significant because it’s the first of its kind in the world, finally giving investors access to China’s fragmented and complex equity markets in one ETF wrapper.
CN captures both “onshore” and “offshore” Chinese equities by tracking the MSCI All China Index, which includes all Chinese share classes regardless of where they're listed (see our 2014 China ETF Guide for share class details).
Here at ETF.com, we’re big fans of the MSCI All China Index. In fact, we’ve long been using this index as the benchmark in our China ETF reports.
That said, CN’s investment strategy comes with a slight twist—at least in my book—that’s worth pointing out for a couple of reasons.
CN Uses ASHR For A-shares
CN does not hold A-shares directly. Instead, it holds the db X-trackers Harvest CSI 300 China A-shares ETF (ASHR) for its A-share component of the fund. (For the record, ASHR’s RQFII quota is currently 2 billion RMB ($320 million), per DBX’s website.)
Last November, db X-trackers made major headlines with the launch of its “groundbreaking” ASHR because it was the first U.S.-listed RQFII ETF, which are funds physically backed by China A-shares, as opposed to using derivatives.
ASHR is able to directly hold A-shares through its subadvisor, Hong-Kong based Harvest Global Investments, which has an RQFII license. The fund launched with a massive $108 million in seed capital.
So, in a nutshell, CN indirectly gets A-share exposure through ASHR, which directly holds A-shares. Yes, it’s a bit of a mind-bender.
Granted, this doesn’t mean CN is a derivatives-based ETF (assuming you don’t consider ETFs to be derivatives). It simply means CN is not an RQFII ETF, and according to its prospectus, it doesn’t intend to become one at this time.
But the real twist here isn’t so much that CN is a partial “fund of funds.” It’s that ASHR tracks the CSI 300 Index, which is a different index from the one that rolls up into the larger MSCI All China Index, the MSCI China A Index.
CSI 300 Vs. MSCI China A
The cap-weighted CSI 300 Index holds the 300 largest A-shares. Meanwhile, the cap-weighted MSCI China A Index captures 85 percent of the mainland market, which consists of the 460 largest A-shares.
My first thought was that if CN is holding an ETF that tracks a different index from the subindex of the MSCI All China Index, couldn’t that cause slippage in tracking?
After all, A-shares currently make up roughly 50 percent of the MSCI All China Index, which is a significant chunk.
My colleague and fellow analyst Howard Lee and I did some research around this. Not surprisingly, both indexes are highly correlated with each other, since they’re both broad, A-share cap-weighted indexes.
Over the past three years, according to Bloomberg, they showed weekly correlations of 0.995. Over the past year, they showed daily correlations of 0.995. That’s about as tight as they come.
Still, it’s important to separate correlations from returns over longer periods.
So, we used the MSCI China A Index as the benchmark, and ran rolling 12-month snapshots in performance differences of the CSI 300 Index against the MSCI index, going back two years.
Here’s what we found (these are total returns, as of April 30):
Median performance difference (12 mo.): -1.86 percent
Max upside deviation (12 mo.): +0.52 percent
Max downside deviation (12 mo.): -4.78 percent
According to our data, over rolling 12-month periods in the last two years, the CSI index trailed the MSCI index by a median 186 bps. Over the best 12-month stretch, it outperformed the MSCI Index by 52 bps, but over the worst 12-month stretch, the CSI index trailed by 478 bps.
Looking at the past year, from a pure difference in returns perspective, the CSI 300 Index underperformed the MSCI China A Index by roughly 3.14 percentage points.
While the difference in returns is likely a combination of several factors, one component could come from sector differences.
Most notably, the CSI 300 Index has close to 5 percent more weighting in financials, which haven’t performed well in recent years. The MSCI Index also has about 160 more mid- to small-cap holdings, which could impact its returns.
So what’s the takeaway here?
It’s not that the CSI index will necessarily trail the MSCI index going forward, nor does it mean that CN will trail its underlying index due to these differences.
It simply shows that the returns between the two indexes can differ, for better or worse, and it could simply make tracking the MSCI All China Index a bit more challenging for the fund manager.
It’s worth pointing out that there are positives with using ASHR as a holding. In many respects, it simplifies the structure of CN.
For creation/redemption purposes, since ASHR trades in the U.S. with great liquidity, market makers can hedge their positions better, which in turn should trickle down to tighter spreads. From DB’s perspective, it’s likely cheaper to go this route as well.
For shareholders, this all leads to lower end costs. CN clocks in with an expense ratio of 71 bps. That’s only 12 bps more than the SPDR S&P China (GXC | B-40), whose 0.59 percent expense ratio is the lowest in the segment.
That said, when measuring costs, you have to incorporate “total holding costs,” which includes slippage from tracking. The real test for CN will be whether it can keep any slippage from its index in check.
Market Vectors currently has an ETF in the pipeline that’s slated to track this same MSCI All China Index. It’ll be interesting to see whether their new fund (ticker: ALCH) will launch as an RQFII ETF.
Don’t count out iShares either. BlackRock just got RQFII approval from The China Securities and Regulatory Commission. Since iShares has a long-standing relationship with MSCI, we’ll see if it decides to join db-X and Market Vectors with an “all China” approach.
In the meantime, I’m enjoying seeing the continued innovation in the space. It’s great to see issuers finally cracking the Chinese alphabet soup of share classes.
At the time this article was written, the author held a long position in ASHR. Contact Dennis Hudachek at [email protected], or follow him on Twitter @Dennis_Hudachek.