Does timing high yield ETFs work?

In past posts I’ve discussed how a timing system, based on ETFs, works to lower risk, preserve capital, and at worst match long term market returns. This system use commonly available ETFs such as VTI, SPY, IEF, etc… Also, in the past I’ve loosed into whether timing high dividend stocks works in the same manner. I’ve found that the same timing methodology does not work for high dividend stocks. But what about high yield ETFs in general? Maybe this a potential sweet spot for income investors; yield and capital preservation. Lets take a quick look.

I picked a handful of high yield ETFs from various sectors and assets types and did a historical analysis using a monthly timing system based on the 200 day simple moving average. The periods for the analysis varied quite a bit due to the fact that many of these ETFs have not existed for a very long time. In one case, that of MLPs, I used the index itself since the ETF covering the sector, AMJ, is one of the newest of the high yield ETFs with a very brief history. The results of my analysis are shown below. All of the results include re-invested dividends.

The table shows the starting period of the analysis, the ending period was end of Dec 2011 for all ETFs, the buy and hold total return during the period and the timing total return during the period. As you can see in the table, the timing system outperformed buy and hold for 6 out of 8 ETFs. Those 6 ETFs varied quite a bit; US dividend stocks (DVY), foreign dividend stocks (IDV), US real estate (IYR), energy stocks (XLE), high yield bonds (HYG), and mortgage stocks (REM). The two ETFs that timing returns were lower than buy and hold returns were long term gov’t bonds (TLT), and the MLP index (AMZ).

A few observations on these results. Obviously the results are not cut and dry but I think AMZ and TLT results have a couple of unique circumstances. First, AMZ is not an investable product. There was no ETF for the MLP index until relatively recently. Thus one of the momentum factors, program trading of ETFs driving individual stock prices, that I think drives the success of timing systems was not in play at all. It will be interesting to see how AMJ does over time. The other factor that I think influences the TLT and AMZ results is the under performance of timing systems during strong bull markets. Both AMZ and TLT have been in strong bull markets for a while.

Now, does it makes sense for the income investor to time high yield ETFs? It depends. The downside of timing these products is that there will be periods where the income investor is not invested and thus not receiving income. Thus if you’re trying to live solely off an income stream such a strategy would not work. On the other hand if you focus on total return and don’t mind tapping into principal during certain periods to supplement your income, as in the 4% SWR model, then such a strategy may make sense. Personally, this is not my style of investing. I’d rather own individual great dividend paying stocks with a steady and growing income stream. But I understand the attraction of high yield ETFs especially given the paltry yield of common ETFs such as SPY, IEF, etc…

In short, it looks like timing high yield ETFs works as well as timing other ETF products and could have a role in an income investors investment strategy.

P.S. It’s pretty easy to run your own analysis. Historical prices for almost any ETF product is available on Yahooo Finance. Download monthly historical prices, calculate a 10 month SMA, simulate a buy when the adjusted price is above the 10 mo SMA, simulate a sell when its below, and calculate the total returns based on the adjusted prices. That’s it. Also, you can run these kind of analyses on ETFreplay.com but you have to subscribe which is $30/month. Not for me, thanks.

About these ads

2 Responses to Does timing high yield ETFs work?

  1. J Carroll says:

    Paul, would your timing analysis differ significantly from actively monitoring the 200-day SMA and acting on this indicator; buy when an ETF crosses from below to above, and sell when it crosses from above to below. Thanks.

    • libertatemamo says:

      J, this analysis is based on the 200 day SMA. It just waits till the end of the month for the signal. A monthly signal based on the 200 day SMA outperforms more frequent signals such as weekly or daily. So, yes the return results would be less if a more frequent signal had been used. Waiting for monthly signals avoids a lot of false entries and exits.

      Paul

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 1,699 other followers

%d bloggers like this: