The Limits Of Indexing

Are ETFs any good at tracking emerging markets at all, and should investors be considering alternative ways of getting their exposure?

A couple of days ago a friend who works at the London office of mutual fund manager Capital Group sent me a table comparing his company’s actively managed emerging market fund’s performance with that of the MSCI Emerging Markets index. The fund has done well over the five years to end-June, outperforming the net total return version of the MSCI EM index by nearly 3% a year, after fees.

That’s the kind of performance comparison against a benchmark that you’d expect an active fund manager to be publicising. But what interested me in Capital Group’s marketing literature was the addition of two extra lines in the table to show how index-tracking products are failing to do their job. The two index trackers mentioned were Vanguard’s MSCI emerging markets index fund, which trailed its benchmark by 1.3% per annum over the same five-year period, and the iShares MSCI emerging markets ETF (NYSE Arca: EEM), which lagged by 1.8% a year.

Why are funds that are supposed simply to replicate an index return not doing what they say on the tin? EEM, for example, had a particularly poor 2009, lagging its benchmark by more than 6%. The fund’s tracking problems have been attributed to its management policy, which involves holding only a sample of the index constituents, rather than all of them. This, in turn, is effectively forced on the fund by its massive size (US$37 billion) and the nature of the underlying asset class. Emerging equity markets, as Cris Sholto Heaton’s latest piece on India reminds us, have liquidity and capital constraints that are easy to overlook.

It isn’t just an iShares/EEM problem, however. Although Vanguard has been one of the main beneficiaries of EEM’s tracking difficulties, Capital Group’s comparison shows that Vanguard’s index fund hasn’t done very well either in tracking the index.

If we broaden the comparison to include some European ETFs, the picture gets murkier still. In the table below I’ve included the performance figures over the last six months for EEM, Vanguard’s emerging markets ETF, VWO, and three European ETFs: iShares’ European version of its MSCI emerging markets fund (LSE: IEEM), Credit Suisse’s Luxembourg-based fund (SWX: CSEM), which also uses physical replication, and finally the largest swap-based ETF in Europe, run by db x-trackers (LSE: XMEM). Return figures are for US dollar daily NAVs, and the funds’ expense ratios are also given.

All the ETFs using physical replication trailed the benchmark by more than can be accounted for by management fees – in EEM’s case, by nearly 2.5% over the first six months of 2010, and that’s following last year’s dramatic, 6%-plus underperformance. The returns of iShares’ US- and European-listed versions of the same ETF are inexplicably different, as are their underlying individual stock weightings and their total holdings (653 for EEM, 328 for IEEM, according to the latest fund factsheets, compared with the index’s 767 stocks). But neither Vanguard’s VWO nor Credit Suisse’s CSEM has done a particularly good job of tracking, either.

Only the db x-trackers ETF has performed almost exactly as might be expected: in other words, the index return minus half the annual management fee for the six-month period.

Does this mean that index investors should ditch emerging market ETFs that use physical replication and all buy European swap-based funds? Possibly, and this superior tracking ability hasn’t gone unnoticed, since db x-trackers’ fund has now grown to be the biggest emerging markets ETF in Europe, while Source’s version has also picked up significant assets in recent months.

But before you rush to do so, consider that you could do even better, albeit at the expense of additional counterparty risk, by going to an investment bank and buying a swap directly. According to a trader I spoke to earlier today, investment banks may guarantee up to 60-80 basis points outperformance of the MSCI EM net total return benchmark via a swap.  This extra return can be earned by the bank through dividend tax arbitrage and stock lending, giving it the opportunity to offer some of the proceeds to clients. It’s hard to turn down guaranteed outperformance of that type and, if you’re worried about counterparty risk, aren’t most of those swap-trading banks now government-backed? And if that guarantee were to fail, how much would your emerging markets equity fund be worth, anyway?

What are the implications of all this? First, don’t assume that your index tracker is going to track the index, particularly in less liquid asset classes. Second, there are many ways of buying index exposure, and so be sure to compare different versions of ETFs, without forgetting futures and swaps.  Third, you might be better off with an active manager.

More importantly, though, are we reaching the limits of indexing? The difficulties that the world’s largest index-tracking firms are experiencing in emerging markets suggest that this may indeed be so.