Page 1 of 3
Mea culpa. I was mistaken. . . There, I said it. And now, thanks to an enlightened editor, I get the opportunity to set matters right. Now I will respond to the reading public and various nettlesome bloggers.
For the sake of clarity, my confession of fallibility is in reference to an article titled "Passive Momentum Asset Allocation, Putting The Market To Work For You," which appeared in the Journal of Indexes First Quarter 2004 issue, on pages 36-39. It is page 39 that will be receiving most of our attention, and in particular, the comments and findings on that page regarding sectors.
So now that I have come to you all and bared my manager's soul and publicly admitted my error, please allow me to at least defend those portions of the offending material that I still stand foursquare behind, before I delve into the bothersome mistakes.
First, there is this comment on our passive momentum model: "While we were pleased that the model was successful at some level with sectors, we could not justify any practical utilization, given the negative factors outlined and the lack of return premium."
This statement is as true today as it was when the aforementioned article was originally published . Any momentum scheme that is based on a time period of less than a year is susceptible to a number of significant negatives. It is our observation that short-term variations of positive or negative momentum are in effect chaotic and resemble random noise. This will of course tend to produce random and chaotic returns, which are far from random in how they subject clients to volatility. Secondly, the tax consequences of any short-term momentum trading theory (which produces positive returns) must (in most cases) be negative. Moreover, there is the problemof the excessive expense generated by excessive trading.
Please allow me to stray just a bit here. My second point, that the tax consequences of the rare successful short-term momentum trading scheme can be a negative, brings me to a currently popular phrase in the money management business, "loss harvesting." Since it is my experience that most momentum plans produce losses, those who champion this whole idea of loss harvesting would applaud such a practice. In my opinion, the spin doctor who coined the loss harvesting phrase puts all those practicing the art (?) of spin in politics to shame. As a manager, I want at all costs to avoid losses in my clients' accounts, even if we attempt make them sound painless and fruitful by "harvesting" them. I like gains; losses - whether "harvested" or left in the field to rot - I do not.
Go to page 39.
The other issue I'd like to defend is the notion that "sector" and "asset class" are not interchangeable terms. This is a situation those of us in the trenches of real world portfolio management do face. I will still support the notion that sectors tend to muddy the waters when viewed through the lens of the Fama and French (JIM: IS THIS RIGHT? Not everyone will know - top of mind - what the Three factor Model is. You might ask Silver if we can insert this information, or how he would define it.) Three Factor Model, that famous asset pricing model that ascribes higher risks to small capitalization stocks and stocks with high book-to-market-value ratios . There can be no arguing that any sector will encompass stocks that cross the various boundaries of the Three Factor model -clearly, market cap and book-to-market ratios are in no way used to define any sector I am aware of (bloggers, come get me if you can). But since this is intended to be a full disclosure piece, when those Three Factor Model lenses are removed sectors are certainly a valid portfolio designation.
Three Factor or not, a sensible momentum discipline can be profitably applied to sector investing. And indexing is the key.
That seems to be about as much of my original observations on sectors that I can defend. Let's get on to what may be the most egregious oversight and get it out of the way first. And why none of you bloggers got this escapes me…
Error #1:"…we haven't found any sector product that is passively managed."
Technically, this is actually an accurate statement; we hadn't found such a product at the time. Of course, such a product exists now and it existed then, but we hadn't found it. Obviously, we didn't look very hard. As I will attempt to show in the remaining space available, the utilization of index sector ETF's (I will be using the iShare/Dow Jones family of products) allows for the successful implementation of an index-based approach to passive momentum investing.
Our original study, performed in 2001, as outlined on that now (in)famous page 39, was loaded with problems. From a passive manager or an investor's point of view, the single most offensive problem was the use of actively managed funds from the Fidelity family. I have no particular argument with Fidelity - we use them (sparingly) as a broker in our practice - but at the time, their sector funds seemed the only game in town. It is my belief that the inherent failings of most active managers lies at the root of this abandoned experiment.
Too much trading in the portfolios, which is likely impacted by the vagaries of short-term momentum (as well as other factors), surely led to too many incorrect choices within the fund portfolios. This resulted in unacceptably inconsistent annual returns, which I believe are the result of losing touch with long-term positive momentum, and forced the sector study to seek positive results from short-term trading horizons, with the expected sorry results.
All of which doesn't even touch on the issue that upsets so many fund investors, pass-through capital gains.
Error #2: "…..ETF's have far more appeal than do actively managed product, although you will run into the same issues in terms of trading the fund(s)….."
Again, I will make a modest defense that the above statement is at least partially true. ETFs certainly did and do have far more appeal than do actively traded funds. I don't think I need to rehash all the reasons that index products are more efficient and desirable investment vehicles than actively managed funds in any well thought out portfolio or investment policy. But as I will show, you really don't have to succumb to the inherently fatal flaws of an actively managed sector allocation just to pursue a momentum approach. Where I was as wrong as wrong can be is that with ETF's, you needn't suffer the same costly and inefficient fate befell the actively managed strategy in our original 2001 study.
So why, with the obvious shortcomings of an actively managed fund-based approach, didn't we seek out the ETF's back in 200? I could blame it on sponsors for not making enough noise so that we managers might have had a better understanding of the product. But that wouldn't be true for me. The truth is, as outlined again on page 39, we were so put off by the lack of an international Three Factor Model ETF product (a deficiency which, I have been told by some highly placed and reliable sources, is being looked into, with possible resolution actually coming to market in the reasonable future), and our own belief in the Three Factor Model for other portions of our practice, that we just overlooked the entire genre. As I told you at the very beginning, mea culpa.
We still hold that the Three Factor Model is a superior way to categorize the markets, and continue to manage our clients' funds with the model clearly in our crosshairs. But we are also believers that Adam Smith got it right when he focused attention on the power of supply and demand on markets. The extraordinary growth of the supply of ETF's has certainly drawn our attention, and this must be a factor of the demand for the product. Given that, and our original observation that ETF's are a superior product as compared to actively managed mutual funds, it is time to revisit the sector momentum issue in light of these funds..