Ok I'll ease up on the sarcasm now. A few days ago the WSJ had an article about ETF tracking errors for 2009 averaging 125 basis points and noted that 54 ETFs had tracking errors exceeding 300 basis points. This caused Felix Salmon to ask "Is this the beginning of the end of ETFs as an asset class?" I would note that ETFs aren't really an asset class but more like access to asset classes.
So ETFs, the plain vanilla ones, own stocks such that the combo within the fund tracks an index. The creation/redemption process and the arbitrage potential that exists are supposed to prevent the funds from straying meaningfully from their indicative values. In addition to the numbers on this from 2009 I know that in the fall of 2008 bond ETFs came unglued in the immediate aftermath of the Lehman weekend.
Obviously I have been a big believer in the product and I believe it is fair to say I was relatively early in writing about them and doing so a little differently than most folks. However I've also been consistent with several caveats and an important disclosure.
First thing to revisit is that ETFs are investment products. Like all investment products there are pros and cons to using them. Part of assessing the utility of an ETF is to weigh the pros against the cons and make a decision. If you are not aware of any cons for an investment product you are considering then you do not understand the product very well.Next is that ETFs simply offer access to various things (stocks, bonds, commodities, currencies and complex strategies) that is it. A particular ETF might be the best way to access something or not. Even if an ETF is the best way to access something that does not mean it is perfect.
If you build your portfolio at the sector level, for example, then you are figuring out the best way to build each sector while at the same time you might also be trying to work country decisions into your mix.
An example I have used before is with Brazil. The country hasn't sneaked up on anybody, it has been a popular investment destination for quite a few years. If you want to pure exposure to Brazil you are very likely going to choose a stock (there are plenty of them) or the iShares Brazil ETF (EWZ); you could also choose the Market Vectors Brazil Small Cap ETF or the soon to launch EG Shares Brazil Infrastructure ETF. If you buy an ETF you avoid single stock risk but, in the case of EWZ, you take on a lot of financial exposure. Buy a stock and you take on risk but avoid the financial sector (assuming that matters to you).
The above refer to composition issues. The risks, maybe cons is a better word, addressed in the two links above are more structural. ETFs can stray from their indicated value. Divergences can be short lived or last a long time. There is no way to predict when this will happen or how long it will last when it does. The possibility of such a divergence just goes with the territory just as a stock you buy today could have really bad, unforeseeable news tomorrow.
If the possibility of any tracking error is unacceptable to you then you should not use ETFs at all. From there it might get a little subtler. In 2009 EWZ was up about 100% versus 24 or 25% for the S&P 500. The person above wanting Brazil a year ago made the correct country decision. If they chose EWZ as their way in they added a lot of value to their portfolio. For the sake of discussion let's say that EWZ was up exactly 100%. If the underlying index was up 101% and you got 100% would you be ticked off? There is no wrong answer here, that 1% would either be acceptable or it wouldn't. If the 1% difference is ok then what about 2% or 4%? At what point would it be unacceptable? Again there is no wrong answer but how you answer goes a long way to whether or not ETFs are a good tool for you to use.
If you can't accept the possibility of a tracking error then, where Brazil is concerned, you probably need to pick a stock which is ok. I use a stock for Brazil as opposed to an ETF. Last year my stock lagged and other years it has beaten the ETF. That sometimes the stock does better while at other times the ETF does better is not an argument either way for ETFs.
One other point about tracking error is that the average was 125 basis points. The conversation is not about closed end funds trading 20-30% away from their net asset values. The potential tracking error of ETFs might still be unacceptable but this is not the same as closed end funds--at least not now.
The disclosure I referred to above is that I do not use many ETFs in larger portfolios. In accounts above a certain dollar size we can hold 40 names without giving away too large a percentage to commissions. In those accounts there are maybe half a dozen ETFs and the rest are individual stocks. If I think an ETF, warts and all, is the best way to access something then I use the ETF. I am agnostic about which product I use, I simply have to believe that going forward a thing is the best way to capture the space. Not that I can't be wrong of course but anyone can be wrong with anything they hold.
This is why I've never understood firms that say they only use ETFs. No product can be the single best way to access everything. The notion of limiting your portfolio to just one product makes no sense to me. Today an ETF might be the best way to capture Egypt (the only way really) but if a year from now Arab Cotton Ginning (a real company in the ETF) lists on the NYSE it might be worthwhile deciding if it could be a better way in; it might be (just an example, I know nothing of the company) better than the ETF, the 42% in financials could end up being a drag on the fund for all we know.
If you have been with this site for while then none of this should be new to you. ETFs are a great addition to the tool box but they have never been perfect and to the extent people are surprised tells me that they did not know the product very well. A less that perfect investment product can still be a great way to access a segment of the market.
12 comments:
ETFs are a small investors best friend IMO. They are the clearly the majority of my portfolio.
SEG
Roger, a very good post today full of wisdom. Thanks.
As a long time reader, I don't recall ever a discussion about assessing company management teams as in input in the decision making process. I am a strong believer in that concept. Corporate management teams deserve ratings from A+ to something significantly below F in my humble opinion. I think more attention needs paid to that in the discussions.
i have talked about management as a catalyst to sell; BAC after they bought MER.
very generally if management was truly and obviously stupid it would take much time to see this in the price and the annual report.
a little more subtle there are certain stocks that seem to go nowhere which can be a tell of something not quite right even if not overtly stupid.
clearly what i am saying is far from foolproof but the message of the market can be helpful with this.
A very timely commentary today. I do think that with the plethora of sector ETFs, while it nice to have variety, investors may be too smart by half and construct a portfolio that is much more speculative than they realise.
BTW -- I thought the You Tube "Hitler Corporate Taxes" was hilarious. Some of these 4 minute takeoffs from the made-for-tv pic are pretty raw, but this one, along with "Hitler Short Sale" are hilarious.
firms like iShares/Blackrock have invested tens of millions of dollars in their IT systems all in an effort to track indexes. They have portfolio managers with live, real-time feeds to the index providers and broker-dealers. I have personally been on the iShares trading floor in San Francisco.
They handle index changes and corporate actions and tax lots. They can lean on create-redeem arbitrage to keep the market price as close to NAV as possible with the click of a button. you are talking about world-class infrastructure here.
An index in not investable. if its not investable, you are comparing apples and oranges. for functional purposes, ETF's (and index mutual funds) ARE the index. If ishares/blackrock or vanguard have some tracking error -- then couldn't it be argued that an 'index return' is just a hypothetical number?
www.ETFreplay.com
I've wondered why someone like Ishares needs a trading floor. I guess they have to make slight adjustments daily since the index fluctuates, but I'd assume that's done by computer, rather than some process in which a person says "uh oh, GE has gone up 1%, I'd better buy some for the ETF."
lol Stephen,
funny you say that because I chatted with one of the PM's -- who had previously worked as an analyst at a wall street firm and said she said:
'when I heard 'index portfolio manager' I said 'wait, don't those things just run themselves?'
there is actually some work to do when you manage $353 billion. indexes change, there are adds and deletes, there are mergers and spinoffs, there are rights offerings and various corporate actions to respond to -- the tax-efficiency aspect to ETF's takes some work. there is interaction with broker-dealers for the create/redeem process.
the trading floor was not very big and each PM (portfolio manager) is part of a team that works on many different ETF's. It is quite leanly run and ishares is VERY big on hiring and investing time in people who are good at not only money management, but in ways to improve their IT systems.
Roger:
Have you been able to evaluate your stock picking skills by comparing your performance from all ETF portfolios with a mix of stock and ETF portfolios?
Roger,
did you ever think of using a hedge fund for your client portfolio? If your answer is positive what charateristics would you search for selection?
Tx,
Jeff from Milan, Italy
jeff, no
roger - ever think of starting your own hedge fund??
redsox2010, i like the moniker
no interest in starting a hedge fund. I have a very simple life with a job a love which pays a fine wage.
the way i prioritize things less simple is not ideal
Post a Comment