Preferred stock investors are frequently advised to keep a sharp eye on the interest rates of various government instruments as a sort of indicator of what they can expect for the dividend rates to be offered by new preferred stocks. But the actual data show little or no relationship between the direction of U.S. government money rates and the direction of preferred stock dividend rates.
While a prior article looked at the relationship between increases in the Federal Funds rate and preferred stock market prices, this article examines the relationship between three government money rates and the dividends being offered by newly issued high quality preferred stocks (defined as those that meet the ten risk-lowering selection criteria from chapter 7 of my book, Preferred Stock Investing. For example, high quality preferred stocks offer "cumulative" dividends, are rated as investment grade and are issued by a company that has a perfect track record of never having suspended a preferred stock dividend).
Source for all preferred stock data in this article: CDx3 Notification Service database and Preferred Stock Investing, Fourth Edition (PreferredStockInvesting.com). Disclosure: The CDx3 Notification Service is my preferred stock email alert and research newsletter service and includes the database of all preferred stocks and exchange-traded debt securities traded on U.S. stock exchanges used for this article.
The following chart includes four lines. The top (blue) line shows the average monthly dividend rate being offered by new high quality preferred stocks over the last twelve years (2001 through 2012).
The bottom three lines (in various shades of grey) are average monthly U.S. government money rates that are frequently referred to as being predictors of preferred stock dividend rates (to one degree or another).
The bottom line (darkest grey) is the Federal Funds rate; the next line up (lighter) is the average monthly yield of the five year Treasury note and the lightest grey line is for the ten year Treasury.
The macro direction of preferred stock dividend rates over the last twelve years is indicated across the top of the chart just above the preferred stock dividend rate line (blue). The dividend rates being offered by newly issued high quality preferred stocks, at a macro level, were decreasing between January 2001 and March 2004 (terrorist attacks followed by war and recession). This period started with D-A from Dominion Resources (NYSE:D) at 8.4% issued in January 2001 and ended with HE-U from Hawaiian Electric (NYSE:HE) at 6.5% in March 2004.
Investors then enjoyed an overall rate increase between April 2004 and March 2009 as a housing boom was followed by the Global Credit Crisis during which cash-starved banks bid up preferred stock dividend rates to record levels. During this period high quality preferred stock dividend rates increased continually for five years, all the way up to 9.6% from BB&T Corporation's (NYSE:BBT) BBT-B.
Since April 2009, high quality preferred stock dividend rates have been decreasing once again. The first high quality preferred stock for 2010, for example, came in March from JP Morgan's (NYSE:JPM) JPM-C at 6.7%, quite a drop from BBT-B's 9.6% just a few months earlier. Similar high quality preferreds are now being introduced at dividend rates closer to 6%.
While the relationship between the three government money rates is pretty easy to spot, the notion that high quality preferred stock dividend rates follow along is less obvious. There are periods where preferred stock rates and government money rates seem to be somewhat related, moving in the same direction, but there are just as many occasions where they seem to move in entirely opposite directions.
Measuring the Relationship
The most common method for determining the extent to which two sets of numbers are related is to calculate the "correlation coefficient," the result of which is a number between zero and one.
A correlation value of 0.0 indicates no correlation between the two sets of numbers (such as the number of craters on Mars and the average weight of newborn elephants) while a correlation value of 1.0 indicates a perfect correlation (such as the number of Hollywood marriages and the following year's divorce rate).
For government money rates to be a good predictor of what upcoming preferred stock dividend rates are likely to be doing (as we are often told), we are looking for correlation coefficient values that are close to 1.0. After all, we're talking about our investing strategy here; we need a solid and consistent indicator.
Also, we are looking for a strong positive correlation (when one rate goes up so does the other). A negative correlation value indicates that when one rate goes up the other tends to go down, which would be even further from the advice that many preferred stock investors are routinely given.
Macro Period Relationship
The first table directly under the chart shows the correlation coefficient values that quantify the strength of the relationship between preferred stock dividend rates and government money rates for our three macro periods. Looking over the results, it would be hard to argue that preferred stock dividend rates were being driven by any of the three government money rates included here. The correlation values indicate a weak correlation (if any), certainly not enough to base an investment strategy on.
The strongest relationship is a negative correlation during the April 2004 through March 2009 time frame (the five year long second macro period). Investors following the traditional advice and seeing government rates start to fall in late 2007 would have concluded that preferred stock dividend rates were likely to take a corresponding nose dive. As we now know, and as the chart illustrates, exactly the opposite happened; when government money rates went down, preferred stock dividend rates increased at a breakneck pace (producing the negative correlation values seen in the first table under the chart).
Annual Period Relationship
Hoping that a more granular analysis would reveal the often-cited relationship between government money rates and preferred stock dividend rates, the second table under the chart presents the correlation values for each of the past twelve years.
2007 shows the strongest single-year correlation but, as mentioned earlier, the relationship is going the wrong way; as government rates were falling, more often than not preferred stock dividend rates were on the rise during 2007.
The only instance where government money rates have had a strong relationship to preferred stock dividend rates over the last twelve years was with the 10 year Treasury in 2004 (+0.77).
Even if you slide the data back and forth in time and watch the numbers the story is pretty much the same.
The advice often given to preferred stock investors regarding the relationship of government money rates to preferred stock dividend returns is not supported by these data, certainly not to the extent needed to provide the basis for a preferred stock investing strategy.
Source for all preferred stock data in this article: CDx3 Notification Service database and Preferred Stock Investing, Fourth Edition (PreferredStockInvesting.com). The CDx3 Notification Service is my preferred stock email alert and research newsletter service and includes the database of all preferred stocks and exchange-traded debt securities traded on U.S. stock exchanges used for this article.
Additional disclosure: Securities identified within this article are for illustration purposes only and are not to be taken as recommendations.