This article was released to members of the Cambridge Income Laboratory 30 days ago.
For the inaugural issue of The Chemist’s CEF Report (September 2016), describing the background and rationale of the Report, please click here.
The Chemist’s CEF Report is a monthly feature. This edition uses data taken from the close of December 30th.
Methodology
A database of CEFs was obtained from CEFAnalyzer. In instances where NAV data was delayed, premium/discount values were manually checked on CEFConnect. 533 actively trading CEFs were included in the analysis. Unless otherwise stated, all data from this article are from either CEFAnalyzer or CEFConnect.
All yields are quoted as the yield on price. All z-scores refer to the 1-year z-score, which I consider to be the most useful time duration for profiting from premium/discount reversion. For those new to CEFs, the 1-year z-score is calculated as the difference between the current premium/discount and the 1-year average premium/discount, all divided by the standard deviation of said premium/discount. Positive z-scores indicate that the CEF’s current premium/discount is higher than its historical average, while negative z-scores indicate that the current premium/discount is lower than the historical average. Incorporating the standard deviation into the z-score calculation enables comparison between CEFs that may have different magnitudes of absolute premia and discounts.
Changes in January 2017 Report
- Expense ratios are presented in the tables. In general, CEFs with low expense ratios are favored compared to CEFs with high expense ratios. Note that the expense ratios retrieved from CEFAnalyzer do not always correspond with data from CEFConnect, but it is impractical to confirm these manually from the fund sponsor for each CEF.
- Distribution coverage ratios (on NAV yield) are presented in the tables. I find this measure more useful than showing NII yield, excess yield, or UNII/Dist metrics because it is easiest to understand. A coverage ratio of 1 means that the fund is paying out exactly what it earnings in net investment income (NII). A coverage ratio of greater than 1 means that the CEF is earning more in NII than it is paying out, and vice versa. Note that the coverage ratio becomes tricky to interpret for option-income funds, which pay out a high degree of return-of-capital (ROC) distributions. This contributes to the tax-advantaged nature of the funds, and it is not always easy to determine whether the ROC is “constructive” or “destructive.” Therefore, I would use coverage ratios for option-income funds with caution. See this Douglas Albo article for further information about ROC for option-income funds.
- The top 10 yielding CEFs have been expanded to top 20 yielding CEFs due to investor desire to seek high income.
- An additional screening criterion has been included: absolute distance of current premium/discount from 1-year average. This is because a z-score, while telling you the relative distance of a fund’s premium/discount from its average, is affected by the volatility of said premium/discount. A fund could have a z-score of, say, -3, but if its discount is only 1% wider than its historical average it does not leave much room for arbitrage.
- I’ve also indicated in each section whether the metrics might be more interesting for buy-and-hold income investors, arbitrage investors, or both.
All of these improvements have been made thanks to feedback from subscribers and readers. If you have any suggestions, please feel free to share in the comments section below.
1. Top 10 highest premia and top 10 highest discounts
(May interest arbitrage investors)
CEFs with the highest discounts are potential buy candidates, while CEFs with the highest premia are potential sell/short candidates. The following data show the 10 CEFs with the highest premia and 10 CEFs with the highest discounts. Yields, z-scores and leverage are shown for comparison.
Top 10 highest premia:
Top 10 highest discounts:
2. Top 10 highest z-scores and top 10 lowest z-scores
(May interest arbitrage investors)
Similar to premia/discounts, CEFs with the lowest z-scores are potential buy candidates, while CEFs with the highest z-scores are potential sell/short candidates. The following data show the 10 CEFs with the highest z-scores and 10 CEFs with the lowest z-scores. Premium/discount, yields and leverage are shown for comparison.
Top 10 highest z-scores:
Top 10 lowest z-scores:
3. Top 10 highest yielding CEFs
(May interest buy-and-hold income investors)
Some readers are mostly interested in obtaining income from their CEFs, so the following data presents the top 20 highest yielding CEFs. I’ve also included the premium/discount and z-score data for reference. Before going out and buying all 10 funds from the list, some words of caution: [i] higher yields generally indicate higher risk, [ii] some of these funds trade at a premium, meaning you will be buying them at a price higher than the intrinsic value of the assets (which is why I’ve included the premium/discount and z-score data for consideration), and [iii] beware of funds paying out high yields from return of capital in a destructive manner.
*The premium/discount and z-score data may be unreliable as those funds report NAV values only infrequently.
4. Top 20 best combination of yield and discount
(May interest buy-and-hold income investors)
For possible buy candidates, it is probably a good idea to consider both yield and discount. Buying a CEF with both a high yield and discount not only gives you the opportunity to capitalize from discount contraction, but you also get “free” alpha every time the distribution is paid out. This is because paying out a distribution is effectively the same as liquidating the fund at NAV and returning the capital to the unitholders. I considered several ways to rank CEFs by a composite metric of both yield and discount. The simplest would be yield + discount; however, I disregarded this because yields and discounts may have different ranges of absolute values and a sum would be biased towards the larger set of values. I finally settled on the multiplicative product, yield x discount. This is because I consider a CEF with 7% yield and 7% discount to be more desirable than a fund with 2% yield and 12% discount, or 12% yield and 2% discount, even though each pair of quantities sum to 14%. Multiplying yield and discount together biases towards funds with both high yield and discount. Since discount is negative and yield is positive, the more negative the “D x Y” metric, the better.
5. Top 20 best combination of yield, discount and z-score
(May interest buy-and-hold income investors + arbitrage investors)
This is my favorite metric because it takes into account all three factors that I always consider when buying or selling CEFs: yield, discount and z-score. The composite metric simply multiplies the three quantities together. A screen is applied to only include CEFs with a negative 1-year z-score. As both discount and z-score are negative while yield is positive, the more positive the “D x Y x Z” metric, the better.
6. Top 20 best combination of yield, discount and distance
(May interest buy-and-hold income investors + arbitrage investors)
For those interested in potentially profiting from the highest degree of mean reversion, for this metric, I have substituted the z-score for distance (X). The more positive the “D x Y x X” metric, the better. Again, I have only included CEFs with a negative distance (i.e. they are trading at a lower premium/discount value than their average).
6. Summary statistics
The average premium/discount of all the CEFs in the database is -6.38%, a slight increase from -6.47% in the previous month. The following boxplot shows lower quartile, median and upper quartile data (the mean is indicated as the “x”).
The average distribution yield of all the CEFs in the database is 6.88%, a slight decrease from 7.06% in the month prior.
The average z-score of all the CEFs in the database is -0.12, a slight increase from -0.20 a month ago.
Commentary
In last month’s commentary (“The Chemist’s CEF Report – December 2016: A Recovery In CEFs?“), we noted that CEFs as a group had begun a healthy recovery with z-scores rebounding from -0.50 to -0.20. This month, CEFs have continued to recover, albeit only slightly, with the average 1-year z-score increasing to -0.12 and the average distribution yield decreasing from 7.06% to 6.88%.
Interestingly, the z-score boxplot shows that the spread in z-scores has actually expanded slightly from last month. This indicates that while the mean z-score value has remained mostly unchanged, there is a greater spread in the z-score distribution. This is good for an arbitrage trader like myself as it means that there are more opportunities to capitalize from premium/discount mean reversion.
Of note are the floating rate CEFs, which as a group have seen significant increases in premium/discount value and are mostly sitting in overvalued territory (see this Arbitrage Trader article). On the other hand, many munis have dropped to levels of substantial undervaluation. There are quite a few exciting CEF arbitrage opportunities that are being identified by the Cambridge Income Laboratory right now.
I intend to give a list of my top 3 CEF picks each month, which is based on my consideration of the data as well as my qualitative judgment. Note that me designating a fund as a top pick does not mean I am encouraging subscribers to buy the fund, nor am I necessarily going to include the fund in the Cambridge Income Laboratory Portfolio. Moreover, note that some of the picks may have a narrow mandate (e.g. utilities stocks or MLPs), and therefore, each investor should consider their own investment objective and risk tolerance before deciding to invest money into any of the picks.
For January 2017, my top 3 picks are:
- Goldman Sachs MLP Energy Renaissance Fund ((NYSE:GER)): 8.90% yield, -10.1% discount, -1.91 z-score, -6.7% distance, 0% leverage, 3.0% expense ratio.
- Macquarie/First Trust Global Infrastructure/Utilities Dividend Income Fund ((NYSE:MFD)): 10.9% yield, -12.6% discount, -1.39 z-score, -2.1% distance, 25.5% leverage, 2.3% expense ratio.
- Eaton Vance Tax-Managed Diversified Equity Income Fund ((NYSE:ETY)): 9.8% yield, -9.2% discount, -1.44 z-score, -2.0% distance, 0% leverage, 1.1% expense ratio.
The three picks have a decent combination of yield, premium/discount and z-score/distance values.
If you have enjoyed my article, please consider clicking the “Follow” button next to my name to be alerted to new content! Members of the Cambridge Income Laboratory (#2 on Marketplace for both ETF and CEF categories) are alerted of my best trade ideas in advance, can view my personal ETF/CEF income portfolio and have early access to my monthly CEF and ETN reports. For more details on what members receive visit here.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Article source: http://seekingalpha.com/article/4043151-chemists-cef-report-january-2017-cefs-continuing-improve-health