If BDC Price Appreciation Is Mostly About NAV Growth, Why Are There So Many Exceptions?

|
Includes: ABDC, ACSF, AINV, ARCC, BKCC, CMFN, CPTA, FDUS, FSIC, GAIN, GARS, GBDC, GECC, GLAD, GSBD, HCAP, HRZN, HTGC, KCAP, MAIN, MCC, MFIN, MRCC, NMFC, OCSI, OCSL, OFS, OHAI, OXSQ, PFLT, PNNT, PSEC, SAR, SCM, SLRC, SUNS, TCAP-OLD, TCPC, TCRD, TPVG, TSLX, WHF
by: Factoids

Summary

My first rule of BDC investing is "never buy a falling NAV".

That is a rule that works, in general, year after year - and by a large degree.

But if this suggestion is so good, why are there so many exceptions?

This is an article by a metrically guided investor and for metrically guided investors. I will show a lot of numbers and attempt to explain the correlations. In this specific article, I will examine why a good rule of thumb sometimes fail to perform. Let's start with the year to date stats, where I have made red squares around the year to date returns and NAV changes of BDCs with negative total returns. Negative NAV changes and negative price changes are strongly correlated. But for 2016, there have been several exceptions to that strong correlation - and those exceptions have the green squares around them. The data:

Overly cheap valuations can trump the favorable attribute of NAV change. I will provide two examples.

AINV began 2016 with a non-accrual problem that overlapped a heavy weighting in oil and gas loan problem. It also began 2016 with a 15.33% yield and a 67% Price/NAV ratio.

Let's start with the assumption that PNNT's oil investments are totally worthless. What would the valuation should be?

AINV's Net Asset Value at the end of the first quarter was $1,645.581 million and ending shares were 226.156 million. One way to partially verify the upcoming math is to take 1,645.581 and divide by 226.156 shares to arrive at a $7.2763 compared to the NAV given in the earnings release of $7.28.

We know from the 10-K that 'oil and gas' energy investments were 11.9% of the portfolio and the total portfolio was $2,916.829. That comes to $347.103 million in oil and gas. That number needs to be subtracted from the NAV dollars. $1,645.581 million minus $347.103 million = $1,298.478 million. That number needs to be divided by ending shares of 226.156 million to arrive at $5.74 'adjusted NAV' to zero out the value of all oil and gas investments. With sector average Price/NAV being 85% - I would with some bullishness project an energy free AINV to sell at 90% of NAV. And 90% of $5.74 results in a price target of $5.1674.

AINV closed on 5-20-16 at an irrationally low $5.21. $5.21 would be 90.76% of the adjusted NAV of $5.74. Thus, as of 5-20-16, AINV is selling as if that $347.103 million (which is both 11.9% of investments and 21.1% of NAV) of oil and gas investments are worth zero.

Let's start with the assumption that AINV's oil investments are near worthless and never recover and then what do you think the valuation should be?"

PNNT began the year with a moderately heavy weighting in the problem category of oil and gas loans. It also began the year with an 18.12% yield and a 63% Price/NAV ratio.

PNNT's Net Asset Value at the end of the quarter was $627.619 million and ending shares were 71.061 million. Let's dividend the $627.619 NAV by the 71.061 share to arrive at a NAV of $8.8322 compared to the NAV given in the earnings release of $8.83. We know from the 10-Q that 'oil and gas' energy investments were 6% of the portfolio and the total portfolio was $1,231.498 million. That comes to $73.890 million in oil and gas.

That number needs to be subtracted from the NAV. $627.619 - $73.890 = 553.729 million. That number needs to be divided by ending shares of 71.061 to arrive at $7.7923 'adjusted NAV'.

With sector average Price/NAV being 85% - I would with some bullishness project an energy free PNNT to sell at 90% of NAV. And 90% of $7.7923 = $7.01.

PNNT closed on 5-20-16 at $6.18. $6.18 would be 88.15% of the adjusted NAV of $7.01. Thus, as of 5-21-16, PNNT is selling as if that $73.890 million of oil and gas investments are worth less than zero.

I believe the numbers show that both AINV and PNNT began the year at irrationally low valuations. Put in different words, too much bad was priced into these 'problem' BDCs. And that justifies why - even with falling NAVs - these two stocks have not underperformed the sector.

Now that I have provided examples of how NAV can fail to explain price change - let me show the stats to prove the power of NAV change to influence price change:

The Importance of NAV Growth in 2016

The following companies had NAVs that rose since Q3-15 - the NAV we had at the start of 2016: FDUS, GAIN, GBDC, MRCC, OFS and MFIN. Their mean price change since the beginning of 2016 is 4.90%. Their mean total return for the year is 8.36% - and 5 of the 6 beat the sector median yearly price change. Their mean Price/NAV ratio is 88.73 - their mean yield is 10.43% - and their mean Price/NII ratio is 9.37. Their mean Dividend/NII ratio is 94.71%. Their mean Q1-16 NII/TII ratio is 51.32%.

The following companies had shrinking NAVs of less than 4%: ABDC, ARCC, HCAP, HRZN, HTGC, MAIN, PFLT, SAR, SLRC, SUNS, TCAP, TCPC and TSLX. Their mean price change since the beginning of 2016 is 4.06%. Their mean total return for the year is 7.38% - and 12 of the 13 beat the sector median yearly price change. Their mean Price/NAV ratio is 99.49 - their mean yield is 9.83% - and their mean Price/NII ratio is 10.17. Their mean Dividend/NII ratio is 98.29%. Their mean Q1-16 NII/TII ratio is 52.27%.

The following companies had shrinking NAVs of more than 4%: ACSF, AINV, BKCC, CMFN, CPTA, FSC, FSFR, FSIC, FULL, GARS, GLAD, GSBD, KCAP, MCC, NMFC, OHAI, PSEC, PNNT, SCM, TCRD, TICC, TPVG and WHF. Their mean price change since the beginning of 2016 is -7.87%. Their mean total return for the year is -3.16% - and 9 of the 23 beat the sector median yearly price change. Their mean Price/NAV ratio is 77.48 - their mean yield is 14.26% - and their mean Price/NII ratio is 7.48. Their mean Dividend/NII ratio is 104.20%. Their mean Q1-16 NII/TII ratio is 49.82%.

Here are the historical stats of the same kind of parsing of the data. In most years I have divided the data into three groupings. I began dividing the data in four grouping in 2015. The high growth grouping in 2015 was for BDCs with more than a positive 4% change. Thus far in 2016, there are zero BDCs with that high of a change. As a result, the number of groups fell back to three.

Historical Price Changes By Degree of LTM NAV Growth:

Year Best Growth Small Growth Small Fall Large Fall
2012 +26.31% +15.25% +13.61%
2013 +23.79% + 2.97% - 2.68%
2014 -11.08% -10.08% -20.16%
2015 +12.10% - 7.76% - 8.24% -19.14%
YTD 2016 +4.90% + 4.06% - 7.87%

Knowing that NAV growth matters is a rear view mirror observation. How does one know in advance that NAV is going to grow? The answer is - one does not know. But there are clues or omens. NAV growth is correlated with (1) BDCs with stronger dividend coverage; (2) BDCs with lower fees or higher NII/TII ratios; (3) BDCs with better historical "NII projection accuracy"; and (4) BDCs with historical NAV growth. Putting point four in different words - there is some NAV change inertia. There is a light correlation of NAV change to portfolio company weighted average yield. (5) The lower risk BDCs - or those with lower portfolio weighted average yields - have moderately out performed.

Let's wrap this up by answering the question in the title line - why are there so many exceptions in 2016 to the rule that falling NAVs result in falling total returns? I can not answer that question with confidence - but I can provide a good theory. I believe the answer can be found in the following two sets of data:

Historical BDC Yield to 10 Year Treasury Spreads for the sector:

Year End of Q1 End of Q2 End of Q3 End of Q4

2007 March: 347 June: 312 Sept: 453 Dec: 663
2008 March: 881 June: 1044 Sept: 1096 Dec: 2910
2009 March: 1562 June: 624 Sept: Dec: 1179
2010 March: 882 June: 649 Sept: 564 Dec: 509
2011 March: 470 June: 605 Sept: 947 Dec: 873
2012 March: 718 June: 760 Sept: 732 Dec: 761
2013 March: 714 June: 683 Sept: 642 Dec: 593
2014 March: 656 June: 618 Sept: 689 Dec: 853
2015 March: 832 June: 798 Sept: 1118 Dec: 1087
2016 March: 1040 June: Sept: Dec:

Historical BDC sector average Price/NAV Ratios:

Year End of Q1 End of Q2 End of Q3 End of Q4

2007 March: 1.37 June: 1.31 Sept: 1.18 Dec: 0.96
2008 March: 0.90 June: 0.82 Sept: 0.77 Dec: 0.46
2009 March: 0.40 June: 0.58 Sept: Dec: 0.76
2010 March: 0.91 June: 0.84 Sept: 0.95 Dec: 1.02
2011 March: 1.08 June: 0.97 Sept: 0.80 Dec: 0.90
2012 March: 0.98 June: 0.99 Sept: 1.06 Dec: 1.04
2013 March: 1.11 June: 1.06 Sept: 1.09 Dec: 1.08
2014 March: 1.04 June: 1.07 Sept: 0.99 Dec: 0.91
2015 March: 0.95 June: 0.92 Sept: 0.81 Dec: 0.83
2016 March: 0.86 June: Sept: Dec:

The data shows that we entered 2016 at relatively low points when it comes to NAV valuations and high points when it comes to yield spreads. That is an indication that there was a lot of fear priced into the sector average valuations at the start of the year.

There are two good reasons for the low valuations.

BDCs will tend to sell at logical "yield + dividend CAGRs" (CAGRs being the acronym for projected dividend Compound Annual Growth Rates). Once upon a time, there were regular occurrences of dividend increases. And BDCs sold at more than NAV when there were expectations of dividend growth. At this point in time, I have several dividend CAGR projections that are negative; a whole lot at zero; and the few with positive CAGRs with positive CAGRs only have projections of 1% average annual growth. BDCs will sell at higher Price/NAV ratios when the CAGR projections increase.

BDCs will tend to sell a higher Price/NAV valuations when NAV are rising and lower valuations when NAVs are falling. At the end of 2012, in my coverage universe of 27 BDCs, 17 had LTM NAV increases. At the end of 2013, in my coverage universe of 30 BDCs, 19 had LTM NAV increases. As if right now, in my coverage universe of 42 BDCs, 6 had LTM NAV increases.

There were strong and metrically justified reasons for BDCs to enter 2016 at low valuations as a group. But as the two examples of AINV and PNNT provided, there was a whole lot of fear priced into BDCs that were perceived to be bad. The reasons we are having some outperformance in the NAV falling BDCs is due beginning the year with several instances of the mis-pricing of fear.

Disclosure: I am/we are long AINV, ARCC, FDUS, MRCC, PFLT, PNNT.

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.