- Comments (18)
- Text Size: Down Up
moneysense.ca, 8/01/09
Currency Neutral Funds Performed Poorly (Again) in 2008
Current issue:
When the Canadian dollar rapidly appreciated against the greenback, currency-neutral funds became popular. But, as noted in an earlier post, the short performance history showed that these funds exhibited large tracking errors. For instance, the iShares CDN S&P 500 Index Fund (XSP) is a currency-neutral version of the iShares S&P 500 Index Fund (IVV). In 2006 and 2007, XSP trailed IVV’s return by 1.72% and 2.29%, respectively. Another example is the performance difference between the TD US Index (US$) e-Series Fund (TDB952) and the TD US Index Currency Neutral (TDB904). TDB904 trailed the returns of TDB952 by 1.8% in 2007 and 1.1% in 2006.
In 2008, the currency-neutral funds continued to trail the performance of comparable US dollar denominated funds. The total returns of XSP and IVV was -40.08% and -36.68%. In other words, XSP trailed IVV by a stunning 3.4% in 2008. The tracking error of TDB904 over TDB952 was better but still very high at 1.63%.
Investors interested in hedging their currency exposure should pay attention to the large tracking errors exhibited by currency-neutral funds, which are much greater than the MER differential between these funds.
moneysense.ca, 8/01/09
Do you think XSP makes sense for a Canadian investor who plans to live/retire in Canada and who is very bearish about the US dollar relative to the Canadian dollar over the mid to long term?
That’s a pretty high cost that will damage returns quite a bit. I haven’t heard of it before but it’s not suprising that it isn’t advertised. It’s not a necessary cost for long-term investors either since stocks are tied to ownership of real assets and can increase to match inflation or declines in their native currency given time.
The only thing you gain is not having the regret of seeing someone in another country get bigger numbers than you – unless you pay for needless insurance and then miss out when the exchange rate moves in your favor!
If you think an exchange rate is wrong (for example, if you think the value of the US dollar is being undermined by excessive spending and it’s only temporarily high because of the demand created by bailouts) it might be cheaper to just hedge it separately and profit directly if you’re right or get insurance at a lower cost.
When I was setting up my TD eFunds for my RRSP a few years back, I think the currency neutral US index fund was the only one you could use for an RRSP if you wanted a US index, because you couldn’t hold an RRSP in US dollars. Is that right or did I misunderstand?
The US currency neutral index really seems like a bargain when you look at the price per share ($6.70 at the moment) so I’ve been tempted to put a more money into it; based on this though it sounds like a bad idea.
That’s a punishing fee, but what is the alternative? Especially if you already have a bunch of money in XSP. Converting it could be devastating if the USD goes down.
DM: I personally think that the debatable benefits of currency hedging for long-term investors are vastly outweighed by the high costs in the form of tracking errors.
SP: The currency-neutral funds currently in existence started out as “RRSP” funds that used derivatives to skirt the 30% foreign content rules in registered accounts. When the foreign content limits were eliminated, these funds changed their mandate into currency-hedged funds around 2005. Since then, the tracking errors shown by these funds are simply atrocious.
brad: You are right. See my previous comment. The unit price of the mutual fund shouldn’t matter… only the total amount you want to invest really matters.
chum: My opinion is that equity investors, by definition, have a long time horizon (at least 10 years and usually 20 years or more). Over that long time horizon, currency effects are likely to be a was. But say, that USD does depreciate. You only win if the cost of hedging is less than the depreciation of the USD. If the large tracking errors exhibited over the past 3 years continues, I’m not sure if the costs of hedging would ever be worth it.
CC
Very interesting post. If I may ask, how were the tracking errors for XSP calcualted in the three years?
CC
Sorry, I should explain the reason for the question. On the ishares.ca site, they have a tracking error chart for XSP, which shows an annual error of less than 1% for the past three years. So I was wondering what would be the explanation for the difference in calculations.
http://tools.ishares.com/tec2/view_chart.do
brad: is this relative to past prices? That could be an interesting comparison… with the value of US dollars declining until a few months ago it may have gotten cheaper pretty quickly. That doesn’t change the fact that it apparently has a very high cost to long-term investors though.
chum: if you want to get out of it, hanging on to a bad investment won’t bring back past losses. Like someone who is about to retire with too much money allocated to stocks and can’t afford to lose any more, if you realize that you made a mistake you want to switch to something else before it gets worse.
If the value of the US dollar continues to go down that affects all investors outside the US who buy american stocks – as they get cheaper to buy (regardless of their value in US dollars) demand may increase and raise the price. The mix of domestic and foreign shareholders probably has a large influence on the degree of this effect… anyone know of an easy way to find out what it is?
[...] Canadian Capitalist explains that the 15 bps (0.15%) extra fee on currency hedged versions of some iShares ETFs pale in comparison to their drag from high tracking errors. [...]
I was wondering why would you want to pay to be hedged, when that way you detract from the “foreign diversification performance” part of your portfolio.
Larry: I double checked my calculations and think they are correct. The reason why the tracking error chart shows wildly different values is that while their numbers agree with my computations for XSP, the returns for the benchmark index are very different. My suspicion is that the tracking error chart did not include dividends and hence their total return for the index is inaccurate. I’ll elaborate on my findings in a post next week.
CC – You mention calculating the return for XSP and IVV? The ishares websites post the yearly return on Dec 31 for 2008:
http://ca.ishares.com/product_info/fund_returns.do?ticker=XSP
XSP = -40.45
http://us.ishares.com/product_info/fund/returns/IVV.htm
IVV = -36.94
I guess the IVV return for a Canadian investor would be slightly lower (unless you had an account that let’s you keep USD), since the dividends will be converted to CND and then reconverted to USD when reinvested.
chum: Thanks for the additional data point. I checked ETFConnect and the annual returns reported for IVV agrees more closely with my calculations (shown in brackets):
2008: -36.94% (-36.64%)
2007: 5.44% (5.30%)
2006: 15.68% (15.78%)
http://www.etfconnect.com/select/fundpages/etf_funds.asp?MFID=45251
I’m convinced now that the tracking errors reported by the iShares tracking error tool wildly underestimates the actual tracking error.
[...] The Canadian Capitalist: Currency Neutral Funds Performed Poorly [...]
[...] 500 Index Fund (IVV). The MER for XSP is 0.15% higher, which doesn’t look too bad. But then, as Canadian Capitalist calculated, the XSP trailed IVV by -1.73%, -2.30% and -3.5% in 2006, 2007, and 2008, [...]
[...] Canadian Capitalist reports that currency neutral funds are no bargain. [...]
It’s actually not that bad; look at TD:TSE vs TD:NYSE… at the peak for 5 yrs the US traded was up 142% while the CAN edition was up only 65%… this is a common theme across many stocks I look into but at the end of the day by the I can’t say that buying in US dollars is a good idea as the forex fee eats a couple percent and you never know where the USD will be when you need money
I do not quite understand the discussions above about the relative merits of XSP vs IVV by focusing on tracking errors and MERs.
The reality over the kast year is that IVV performed about 20% better than XSP simply due to the depreciation of the C$ vis a vis the US$.
So…, if the CS$ might appreciate against the US$ over the next year or two (as occurred not too long ago), then XSP would do better in proportion to IVV (notwithstanding the MER and tracking error issues).