One of the metrics I track is the sector average spread of the dividend paying BDCs (or Business Development Companies) to the ten year U.S. treasury. The sector average spread has, since the beginning of November, been noticeably falling. This is a bad omen. This is a valuation shift that probably will not hold. And the news gets worse. This spread shift is most dramatic for the specific BDCs that one would want to own - the BDCs with the best dividend coverage ratios.
I will jump right in to the data. The first spreadsheet is current valuation data and the year to date returns for the sector. At the bottom of this spreadsheet is the current spread calculation.
Yield in the spreadsheet below is based on the Q4-13 dividend. Spreadsheet header abbreviations: Div = dividend; EPS = earnings per share; LTM = last twelve months; YTD = year to date. The dividend to EPS ratio is a measure of dividend safety. Due to calendar and fiscal years failing to overlap, I also include a dividend to the sum of the last four quarters of NII - in the Div/NIIltm column. The dividend to NAV ratio is a measure of safety and efficiency. The last four columns measure the percentage change in the 2013 EPS projection and the change in the price target since the beginning of the year; the change in the Q4-13 dividend from the Q4-12 dividend. Special dividends are not included in this data. ARCC will pay a special dividend of $0.05/share payable on 12-31-13 and again on 3-28-14. FDUS will pay a special dividend of $0.38/share payable 12-20-13. MAIN will pay a special dividend of $0.25/share on 12-24-13. FSC and FULL have cut their first quarter dividend - and that new rate is temporarily encoded as a Q4-12 div to show their forward yield
|Share Price||Div/||Div/||Div/||Div/||Q3-13||Price||YTD Percent Change||LTM||LTM|
|With the 10 Treasury at 2.73% and sector average yield (on Q4 dividends) at 8.7% - the spread is 597 bps.|
|The cap weighted ETF BDCS is up 8.00% year to date - with dividends its total return is 15.47%.|
|Sector yield, Dividend/NAV and Dividend/EPS ratio filter out the zero payout ACAS and SAR.|
|Weeding out ACAS and SAR, the average share price gain is 9.28%.|
Dates used to credit dividends:
|ACAS : none||AINV : 10-03-13||ARCC: 09-29-13||BKCC: 10-02-13||FDUS: 09-25-13||FSC : monthly||FULL: monthly|
|GAIN: monthly||GLAD;: monthly||GBDC;: 09-29-13||HTGC: 11-24-13||HRZN: 10-14-13||KCAP: 10-28-13||KED : 10-17-13|
|MAIN: monthly||MCC: 12-13-13||MCGC: 11-21-13||NGPC: 10-06-13||PFLT: monthly||PNNT : 10-01-13||PSEC: monthly|
|SAR: none||SLRC: 10-01-13||SUNS: monthly||TAXI : 08-22-13||TCAP: 12-24-13||TICC : 09-29-13||TCRD: 09-29-13|
Historical BDC Yield to 10 Year Treasury Spreads for the sector:
|Mar 10: 882||June: 649||Sept: 564||Dec: 509|
|Mar 11: 470||June: 605||Sept: 947||Dec: 873|
|Mar 12: 718||June: 760||Sept: 732||Dec: 761|
|Mar 13: 714||June: 683||Sept: 642||Dec:|
This second spreadsheet contains the late 2012 and 2013 data on historical spreads. From this data - one can see that the spread is at an 18-month low.
2013 Historical Spreads to the 10 year Treasury
This third set of information is output in paragraph form. It shows valuation and return data on BDCs broken into four different categories based on dividend coverage and the portfolio weighted average yield data. BDC EPS (earnings per share) projections are - in actuality - NII (or Net Investment Income) projections.
The Predictive Power of the EPS/Dividend Ratios on Valuations: Dividend to EPS ratios [for dividend paying BDCs] range from as low as the 60s to 110s [at 11-27-13] and may look all too random. But this ratio - during most time periods - helps explain the variety of yields, price-to-book ratios, and price to NII ratios. There is some correlation to YTD price changes as well. This first test is done for BDCs with weighted average portfolio yields of more than 10% in Q2-13. The stats are from Q4-13 and use 2014 EPS projections. Lower risk BDCs - or those with lower portfolio weighted average yields - have their metrics shown separately.
The following dividend paying BDCs had Q4-13 Dividend/EPS ratios of less than 91%: AINV, FDUS, GAIN, HTGC, MAIN and SLRC. Their YTD mean price gain = 16.97% and 2 of the 6 beat the sector mean yearly price gain [9.61%]. Their mean yield = 7.21% and they sold at an average price/book ratio = 1.26 and an average price/nii ratio = 12.25.
The following BDCs had Q4-13 Dividend/EPS ratios of more than 91% - but less than 100%: ARCC, FSC, GBDC, GLAD, HRZN, MCC, PNNT, TCAP and TCRD. Their YTD mean price gain = 7.06% and 4 of the 9 beat the sector mean yearly price gain [9.61%]. Their mean yield = 8.79% - and they sold at an average price/book ratio = 1.18 and an average price/nii ratio = 10.90.
The following companies had Q4-13 Dividend/EPS ratios of more than 100%: BKCC, FULL, KCAP, MCGC and PSEC. Their YTD mean price gain = 0.01% and 0 of the 5 beat the sector mean yearly price gain [9.61%]. Their mean yield = 11.42% - and they sold at an average price/book ratio of 1.02 and an average price/nii ratio of 9.43.
The BDCs that had weighted average portfolio yields of less than 10% [the second test] : KED, PFLT, SUNS and TAXI. Their YTD mean price gain = 17.98% and 2 of the 4 beat the sector mean yearly price gain [9.61%] Their mean yield = 6.80% - and they sold at an average price/book ratio = 1.18 and an average price/nii ratio = 12.79.
This fourth set of data shows the historical yields and spreads from the same four groupings used in the previous set of data.
Historical Yields & Spreads to the 10 year Treasury by type
|Date||High Coverage||Average Coverage||Low Coverage||Low Risk|
|Spreads||448 bps||606 bps||891 bps||407 bps|
|Spreads||630 bps||595 bps||780 bps||425 bps|
|Spreads||625 bps||635 bps||799 bps||457 bps|
|Spreads||613 bps||789 bps||841 bps||564 bps|
Conclusion: I do not include special dividends in my yield calculations. This is a potential reason for the collapse in spreads. Several BDCs with superior dividend coverage are paying special dividends. This would include Ares Capital Corporation, Fidus Investment Corporation and Main Street Capital. On the other hand, I have never included special dividends in my yield data - so it would be strange to have those events having an influence on current data while it was not there historically.
A professional analyst would be forced by convention to offer an explanation for all of this. I am not a professional. The best I can offer is my best guess. The market is always aware of the importance of a covered dividend. This fact shows up in valuation data all the time. But dividend safety is a nebulous valuation concept that is almost never fully priced into a stock. With the recent dividend cuts from Fifth Street Finance and Full Circle Capital Corporation - along with a cut in the dividend for Solar Capital earlier this year - the market has been hit on the head with the importance of the coverage metric. We are experiencing one of the almost never moments where dividend safety is correctly priced into the sector. This moment will pass.
I am confident this correct pricing will pass because I have tracked this kind of data in several sectors for many years. There has always been a revision to the mean on spread data. And the market has never had a long-term memory. If I had to buy something now - I would be overly tempted by the superior yields offered by BDCs with thinly covered dividends. It is rarely a good idea to give in to that kind of temptation. So I am currently doing something I rarely do - I am paying attention to market timing. And these stats are doing their version of yelling at me - this is not the right time to be buying the right BDCs.