Know Your Narrow-Based ETFs

Narrow-based ETFs often draw a lot of criticism for being too risky, but I think a blanket label like that misses the point entirely. Some narrow-based ETFs end up being great tools.

Of course, some end up being useless, too. A narrow-based ETF's fate can hinge on its composition method, or it may fizzle because some subsectors or themes don't lend themselves to indexing. For example, I think that among small-cap drug/biotech companies, some stocks win at the expense of other stocks, and putting some of each in an index may mean they just cancel each other out.

Recently, we have seen two ETFs in the vice-group list that may allow for a good dissection of how narrow-based products work and what they can and cannot do.

In December the (PUF Quote - Cramer on PUF - Stock Picks)FocusShares ISE SINdex Fund (PUF) listed. Just last week, the (BJK Quote - Cramer on BJK - Stock Picks)Market Vectors Gaming ETF (BJK) listed. The general theme seems to be the same but the construction is very different.

BJK is just gaming. Its largest holding is International Game Technology(IGT Quote - Cramer on IGT - Stock Picks) at 9.9%. The holdings are a mix of casino/hotels and gaming-equipment stocks -- and that's it. PUF is a little broader, with a little over half of its assets in gaming-related stocks, with the rest split between tobacco companies and alcohol stocks.

Gaming is considered to be in the discretionary sector, so BJK is a narrow theme within discretionary. PUF is an almost even split between discretionary (gaming) and staples (booze and smokes). This will not make one better than the other per se, but at different times, if the methodologies are good, one should do better than the other.

When discretionary lags staples, as was the case in 2007, it makes sense to expect BJK to lag PUF. Conversely, when discretionary outperforms staples, as it probably will when the next cycle starts, BJK should lead.

Does this hold water? The index underlying PUF was up, on a closing basis, 12.71% vs. 5.96% for the index underlying BJK in 2007. That BJK was positive at all is commendable because the discretionary sector, as measured by the (XLY Quote - Cramer on XLY - Stock Picks)Discretionary SPDR (XLY), was down 15% in 2007. BJK lagging PUF in 2007 makes perfect sense to me based on where we are in the stock market cycle.

If you are someone who would consider using subsector ETFs, you need to be aware of the cyclical issues that will likely affect the fund you are considering. Most of the coverage of ETFs I see fails to realize the importance of this.

A themed fund with a tilt to industrial stocks, for example, is not very likely to do well during recessions. Expecting otherwise will likely lead to misunderstanding and disappointment.

As ETFs track indices, some more established and legitimate than others, a top-down view that assesses whether this is the "right" point in the cycle for a given fund makes much more sense than a bottom-up study of the price-to-earnings ratio or book value. Indices can appear to be cheap or expensive for very long periods of time without, I believe, much in the way of predictive utility.

Back to PUF and BJK: The back test result for 2007 is impressive for both funds. BJK, being discretionary, did much better than XLY (which should be thought of as a benchmark for BJK) and PUF outperformed the (XLP Quote - Cramer on XLP - Stock Picks)Staples Sector SPDR (XLP) 12.71% to 10.26% despite having the potential drag of being 50% in discretionary.

While the funds do offer promise, they are not immune to risk. PUF has roughly 20% in casino stocks, while BJK is slightly higher. Every so often these stocks get crushed (after 9/11 as an example) and any event that takes down all of the casino stocks in a meaningful way will hurt these two funds.

There have been a lot of very narrow funds listed in the last couple of years, and there are sure to be many more coming out in the future. Learning how to evaluate them will become increasingly important for investors who would rather not select individual stocks.