I Got Your Factor

Are factor ETFs the real deal or just marketing hype?

Last week or the week before @ldrogen, the founder of Estimize Tweeted something along the lines of buy a quality/momentum ETF and call it day (he Tweets so much that there was no way to easily find the exact Tweet). Arguably the first factor ETF was the FTSE RAFI US 1000 Portfolio (PRF). It screens for book value, dividends, earnings and revenue. The fund has been successful, outperforming the S&P 500 since its inception.

When I first wrote about PRF almost 12 years ago I said it looks like it will correlate to small cap. Rob Arnott called my house to tell me why I was wrong. Actually, it has correlated to small cap but it has correlated even more to large cap value. In the last ten years, ETFReplay reports the correlation ranging between 0.96 and 1.00 the whole time. And while you could argue that PRF is the first factor fund, really it is growth and value funds that are the first factor funds.

Growth and value have been around for ages. Between the two one must be the better performer in any given period. In 2016 value outperformed by a noticeable amount and this year growth has trounced value by a shocking amount.

As ETFs have proliferated so have many other factors like volatility, momentum and quality to name a few of the more popular ones. Dividend-centric funds have been around for a while; factor. The PowerShares S&P 500 Low Volatility ETF (SPLV) launched in 2011 and came out of the starting blocks blowing away the S&P 500.

Remember the Rydex (now Guggenhaim) S&P 500 Equal Weight ETF (RSP)? Most of the time it blows away the S&P 500 and has a track record of long term outperformance. In a similar vein how about the ALPS Equal Sector Weight ETF (EQL)? Where RSP equal weights all the names in the index, EQL equally weights the originally ten, now 11 sectors.

Now circle back to the newer ones I mentioned above. There are more factor funds than can be easily rounded up to be looked at (if anyone knows where please let me know, ETF.com’s screener doesn’t seem to have smart beta/factors, or I can’t find it). The iShares Edge MSCI USA Momentum Factor ETF (MTUM) has been trading for a few years. YTD MTUM is up 33% versus 21% for IVW which is a growth fund and 16% for IVV, the iShares S&P 500, a good year thus far. In 2016 MTUM was up 3.8%, IVW up 7.58% and 12.4% for IVV. Is MTUM just growth on steroids? Yeah, maybe.

The secret for SPLV’s debut? It was heavy in utilities and utilities had a great run when SPLV first launched. RSP has compared so well to the S&P 500 because it is more of mid cap fund but it has lagged MDY, the first mid cap ETF. EQL is another one who came in with a crazy strong back test because of how well utilities did for an extended period as interest rates fell.

The big macro on this is that most of the factor strategies are valid but none can outperform all of the time. Nothing can. If you want low volatility, the expectation before you even look under the hood should be that you will give away some upside in years like 2017 so you probably need to be very confident that it will spare the full brunt of any large declines, if not, why bother? If I were going to buy a low volatility fund now, I would want it to not be that sensitive to rising interest rates.

Momentum funds don’t necessarily use fast indicators which means they can lag at certain times like earlier in the cycle per one momentum manager I spoke to recently unrelated to this post and as we saw above, momentum might be expected to lag when growth lags. I doubt it is that difficult to figure out what sorts of environments a certain factor might be vulnerable to but there is no way to know when it will lag.

The behavioral thing to look out for is buying into a factor when it is doing well, have it underperform right after it gets bought, giving up and rotating into another factor that has been doing well only to have that one then struggle. Going down this road in your portfolio requires patience but then I can’t think of any strategy that doesn’t require patience.