Specialty Finance
September 20, 2010
BDC Best Practices Our view of the optimal BDC dividend policy
Greg Mason, CFA Troy Ward Jonathan Bock, CFA (314) 342-2194 (314) 342-2714 (314) 342-2918 masong@stifel.com wardt@stifel.com bockj@stifel.com
The main point Here we outline what we believe is the optimal BDC dividend policy. Based on our 10-year coverage of the BDC sector, during good times and bad, we have come to the conclusion that the optimal dividend policy should be centered on recurring income, should not habitually include a return of capital and, while it should strive to be steady and stable, it should have room for flexibility to changing economic scenarios. Figure 1 provides our current coverage universe and, as highlighted by the table, we believe there is strong 2011 dividend coverage in the BDC sector. Figure 1 BDC Dividend Coverage (as of 9/17/10)
2011E Consenus Earnings $1.32 $1.23 $0.68 $0.89 $1.20 $1.37 $0.97 $1.51 $0.89 $1.71 $1.12 $2.40 $1.26 $1.09 $1.20 $0.60 $0.66 2011E Consenus Dividend Coverage 660% 154% 142% 131% 115% 110% 110% 108% 106% 104% 100% 100% 98% 91% 94% 88% NA 110% Current Dividend Yield 1.7% 8.1% 8.2% 7.7% 10.2% 8.5% 9.0% 9.3% 7.4% 10.1% 11.0% 11.8% 11.5% 12.3% 12.1% 11.7% 0.0% 8.9%
Company THL Credit Hercules Technology MCG Capital Natural Gas Partners Pennant Park Golub BDC TICC Capital Ares Capital Gladstone Capital Triangle Capital Apollo Investment Solar Capital Blackrock Kelso Prospect Capital Fifth Street Capital Kohlberg Capital American Capital Average*
*Excludes TCRD and ACAS
Ticker TCRD HTGC MCGC NGPC PNNT GBDC TICC ARCC GLAD TCAP AINV SLRC BKCC PSEC FSC KCAP ACAS
Stifel Rating Buy Buy Buy Hold Hold Buy Hold Buy Sell NR Hold NR Hold Hold Hold Hold Hold
Current Dividend $0.20 $0.80 $0.48 $0.68 $1.04 $1.24 $0.88 $1.40 $0.84 $1.64 $1.12 $2.40 $1.28 $1.20 $1.28 $0.68 $0.00
Source: Company Reports, Stifel Estimates
Introductionwhat is the best policy for a BDC? You wouldnt think this would be a very complicated question. According to the 1940 act, a BDC must distribute 90% of its earnings to remain in compliance with BDC rules. So at first blush, it seems relatively easy. All a manager must do is pay out 90% of your earnings and voila there is your dividend policy. Unfortunately, it is not that easy. In order to have a constructive conversation on the optimal dividend policy we must first have a discussion of terms. These terms are: Net Operating Income, Taxable Earnings, and GAAP Earnings. Net Operating Earnings (NOI) This is the most common earnings metric used in the BDC sector and it is often referred to as company earnings. NOI is best described as the income generated from the portfolio (both cash and non-cash interest & fees minus expenses) each quarter. NOI does not include realized and unrealized Stifel Nicolaus does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.
All relevant disclosures and certifications appear on pages 6 - 7 of this report.
Specialty Finance
September 20, 2010
movements from portfolio asset value changes. NOI is the most appropriate measure of earnings in the BDC sector because it gives investors the on-going income potential the investment portfolio is able to deliver. Taxable Earnings This is the accounting earnings that determines how much a BDC must pay out to meet the BDC rules (90% of Taxable Earnings). The easiest definition of taxable earnings is the interest income from the portfolio (both cash and non-cash) plus all fees earned in the period minus realized losses. This means that differences between NOI and taxable earnings will largely result from (1) how the BDC accounts for fees and (2) certain realized losses. In our view, fees are the biggest difference between NOI and taxable earnings. When a BDC originates a new loan, they usually get an upfront fee anywhere from 1% - 3% of the total amount of the loan. On a taxable earnings basis, all fee income is booked in the quarter it is received. However on an NOI basis, this fee is often amortized into income over the life of the loan (a few take it all upfront into NOI). As a result, taxable income for most BDCs will exceed NOI in that quarter because the upfront fees are included while NOI only counts a small portion of the fee because it is amortized over the loan life. This means that in periods of high originations (i.e. putting lots of capital to work) taxable earnings may exceed NOI due to the upfront fees. The other major difference between taxable earnings and NOI results from certain realized losses. If a manager takes a realized loss on a lead arranged transaction (meaning they were the biggest owner of the debt tranche and a lead arranger of the credit), the BDC may be able to use that loss to lower taxable income (thus lower the required distribution). However, NOI wouldnt be impacted by the realized loss because it excludes realized and unrealized losses. ACAS, FSC, MCGC and NGPC are examples of BDCs who have taken realized losses and offset taxable income. This resulted in a lower required dividend payout for the managers and allowed them to retain capital. GAAP Earnings This number is Earnings Per Share (EPS). It is exactly what you would see reported for a c-corp; however, it is less meaningful in the BDC sector. GAAP earnings are essentially NOI plus all the movements in asset values (i.e. realized & unrealized gains and losses). For example if a debt investment in the portfolio becomes impaired it may drop in value by 50%, which may cause that quarter GAAP income to be a large negative earnings per share number for the quarter. While movements in asset value is very important, using an earnings metric that includes that volatility each quarter doesnt really provide much value to the investor because it blurs the true earning power of the portfolio. Thus, we believe the changes in asset values are a much better tool for evaluating credit quality metrics and book value changes than dividend paying ability. With these terms in mind, we define our optimal BDC dividend policy. Understand that this is not designed to put into question any one BDCs dividend policy; rather, it is designed to familiarize investors with the risks and benefits associated with various dividend policies. We believe more education on this subject will lead to a better discussion on BDC dividends. As a result, our optimal dividend policy is defined according to four key themes: (1) it should be in-line with Net Operating Income (NOI), (2) it should not include a habitual a return of capital, (3) it should strive to be stable but have the flexibility to change if economic scenarios change and (4) it can be paid quarterly or monthly. Theme # 1 BDC dividends should be in line with Net Operating Income (NOI) We know that BDCs must pay out 90% of taxable earnings, but NOI, in our view, is the best operating metric to evaluate the on-going BDC earnings ability, thus we believe it should be the primary determinant of a BDCs dividend. We prefer BDCs use NOI instead of taxable income to determine the dividend level because of the treatment of origination fees under taxable income. Taxable income includes upfront origination fees and this may artificially inflate earnings during periods of high origination activity which would lead to an elevated dividend that may not be sustainable. This high dividend could incentivize management teams to continue originating deals regardless of the market opportunity because if they dont it could lead to a dividend reduction. In addition, we believe investors looks more at the BDCs ability to generate NOI than just the level of dividend yield. In Figure 2, we show the current dividend coverage for the BDCs as well as the current dividend yields. We think it is very telling that some of the lowest dividend yields have the greatest coverage and the highest yields have some of the lowest dividend coverage. On an NOI yield basis the BDCs trade much closer to each other (i.e. the standard deviation for earnings yield is 1.0% while the standard deviation for dividend yield is 1.7%).
Page 2
Specialty Finance
September 20, 2010
Figure 2 Dividend Coverage (as of 9/17/10)
2011E Consenus Earnings $1.32 $1.23 $0.68 $0.89 $1.20 $1.37 $0.97 $1.51 $0.89 $1.71 $1.12 $2.40 $1.26 $1.09 $1.20 $0.60 $0.66 2011E Consenus Dividend Coverage 660% 154% 142% 131% 115% 110% 110% 108% 106% 104% 100% 100% 98% 91% 94% 88% NA 110% Current Dividend Yield 1.7% 8.1% 8.2% 7.7% 10.2% 8.5% 9.0% 9.3% 7.4% 10.1% 11.0% 11.8% 11.5% 12.3% 12.1% 11.7% 0.0% 8.9%
Company THL Credit Hercules Technology MCG Capital Natural Gas Partners Pennant Park Golub BDC TICC Capital Ares Capital Gladstone Capital Triangle Capital Apollo Investment Solar Capital Blackrock Kelso Prospect Capital Fifth Street Capital Kohlberg Capital American Capital Average*
*Excludes TCRD and ACAS
Ticker TCRD HTGC MCGC NGPC PNNT GBDC TICC ARCC GLAD TCAP AINV SLRC BKCC PSEC FSC KCAP ACAS
Stifel Rating Buy Buy Buy Hold Hold Buy Hold Buy Sell NR Hold NR Hold Hold Hold Hold Hold
Current Dividend $0.20 $0.80 $0.48 $0.68 $1.04 $1.24 $0.88 $1.40 $0.84 $1.64 $1.12 $2.40 $1.28 $1.20 $1.28 $0.68 $0.00
Source: Company Reports, Stifel Estimates
Our argument for a dividend policy that matches NOI usually prompts the question, what happens when NOI and taxable income differ? Youre forced to payout taxable income, so what do you do if taxable income exceeds NOI? In this case, we suggest managers build a dividend spillover. See the Appendix section for a detailed discussion of the dividend spillover. Theme # 2 An optimal BDC dividend policy should not include a habitual return of capital. Ideally no BDC dividend would ever include a return of capital. A return of capital is when the dividend exceeds the taxable income. In this case management is simply returning shareholders a piece of the company instead of investing it and making money. In our opinion, this is especially ridiculous when the BDC is also raising new capital at the same time when a portion of the dividend is a return of capital. There are few instances, for a very short time, when we believe a return of capital may be appropriate. One is when a realized loss is eligible to be deducted from taxable income. This loss will temporarily lower taxable income (and the dividend requirement), however depending on the magnitude it may be best to keep the dividend steady and have a temporary return of capital. In these cases we have seen management teams cut the dividend, which allows them to retain the lost capital and we have had management teams keep the same dividend and it included a return of capital. The former who retained capital will have higher future earnings (reinvested that low cost capital) and the latter will likely have a more stable share price because didnt have a dividend reduction. We can see pros and cons to both decisions and do not have a significant conviction either way; as long as the time frame for the return of capital is limited. Where we draw the line is when a dividend has an on-going return of capital component and there is not a clear path to full coverage from taxable income. In our opinion, this is burning the furniture to heat the house. The BDC is essentially monetizing current or future earning assets to support the dividend which, in turn, causes the dividend shortfall to grow and ultimately makes the problem worse. In addition there have been instances where BDC managers are consistently paying capital out the back door in the form of a dividend and bringing new capital in the front door (share issuance) with significant expenses. Theme # 3 An optimal BDC dividend policy should strive to be stable but flexible enough to change if the economic environment changes. It should come as no surprise that BDC investors like stable dividends. A stable dividend not only provides shareholders with an attractive income stream but it also signifies an operating model that is working well which should limit volatility in share price. It is very important to note that our optimal dividend policy doesnt indicate that the dividend can never be changed; it signifies that management should strive to have a stable dividend. We know that seems obvious but during the recent credit market turmoil we saw too many BDC managers try to stubbornly hang-on to a dividend rate that was just too high. Our belief is that managers should adopt a dividend policy that has a level of conservatism that allows stability during stressed situations but in the event of a significant unforeseen event the
Page 3
Specialty Finance
September 20, 2010
manager needs to be willing to make the tough decision and reset the dividend rate. We believe a dividend policy that is based on NOI (refer to Theme #1 above) provides the highest likelihood for a stable dividend. Theme # 4 The dividend can be paid either quarterly or monthly but monthly seems to be gaining momentum. We are starting to see some of the BDCs shift to monthly dividends and after some thought, the monthly dividend seems a net positive to us. Initially we frowned on the monthly dividend because we thought institutional clients disliked it due to administrative issues. However based on our discussions with a number of institutional clients, none seemed to really care if the dividend is paid monthly or quarterly. So in our opinion if institutions dont care, we know retail investors like the monthly dividend then we view it as a net positive. One other aspect of a monthly dividend is it makes it a bit more difficult to short the stock without incurring the additional dividend cost (if an investor is short when a stock goes ex-dividend they incur that dividend as a cost). If there is an ex-dividend date every month it may discourage some investors to short the stock looking to capitalize (or create) a short-term trend.
Page 4
Specialty Finance
September 20, 2010
APPENDIX: Meeting the Distribution Requirements; Dividend Spillover Despite the rule of paying 90% of taxable income, BDCs actually have significant flexibility when it comes to setting the dividend policy. The rule states that 90% of taxable income must be declared by the time taxes are due. This means BDCs have up to 9 months after the year end to meet the payout requirement. Any undistributed income after the year end is subject to a 4% excise tax but the distribution rules are not violated as long as 90% is paid out by tax day. This means a BDC could retain (spillover) up to three quarters worth of dividends. While we view spillover earnings as a good thing, we would warn investors that it still doesnt protect dividends against all downside scenarios. Before the credit crisis, many investors erroneously believed the spillover was this magic bucket of cash that guaranteed the ability to pay future dividends. The reality was/is that spillover income is reinvested in the portfolio to generate additional returns. The problem was that some investors viewed the spillover as protection against future dividend cuts and that just wasnt the case So, should a BDC use the Spillover and pay 4% excise tax? We believe the answer is Yes. After years internal debate we believe any taxable income above NOI should be spilled over into the next year and pay the 4% excise tax. A spillover allows the BDC to retain capital with a 4% cost and reinvest that into higher yielding assets (earn a spread). This cost of capital is significantly lower than issuing new equity which would have a 5-10% upfront cost plus the annual dividend yield (~10%). Additionally, that spillover could serve as a reserve account absorb realized losses. This concept is a bit confusing so we will provide an example. Assume Z-BDC had a $100 of taxable income in 2010, which requires them to pay out $90 before they file their tax return in Sept 2011 (assume their year end is December 2010). They pay $80 ($20 per quarter) in dividends in 2010 and spillover the additional $20 into 2011 and pay a 4% excise tax. When they pay their first $20 quarterly dividend in 2011 they have essentially fulfilled their distribution requirements for 2010 (remember they had to pay out an additional $10 before they filed their tax return in Sept). So in 2011 if they pay another $80 in dividends $20 will be from 2010 earnings and $60 will be from $2011 earnings so if they earned $100 again in 2011 they would now spillover $40 which will be paid out through the first 2 dividends in 2012. In this example ultimately the spillover bucket will be maximized and the dividends will have to rise to $100. However, if the BDC has a realized loss (which at some point they all will make a lending mistake) they can lower their taxable earnings in the year the loss was taken. The loss will effectively eliminate the spillover, but the dividend can remain at a stable level because it was already running lower than the taxable income. . Figure 3 Dividend Spillover Example
2010 Income/Distributions 2010 Taxable Income XYZ Dividend Policy 100 20/qtr 1Q 20 2Q 20 3Q 20 4Q 20
Dividend Spillover 20
In 2010, the BDC earned $100 in taxable income, but only paid out $80 in 2010 dividends. This results in a $20 spillover to be used in 2011
2011 Income/Distributions 2011 Taxable Income XYZ Dividend Policy 100 20/qtr 1Q 20 2Q 20 3Q 20 4Q 20
Dividend Spillover 40
Paid from 2010 $20 dividend spillover
In 2011, the BDC earned $100 in taxable income and paid $80 in dividends. HOWEVER, the 1Q dividend was paid with the 2010 spillover. As a result, the BDC only paid out $60 of 2011 dividends which were paid in 2Q, 3Q, and 4Q and this will result in a $40 dividend spillover ($100 in 2011 taxable income less $60 of 2011 dividend paid)
2012 Income/Distributions 2011 Taxable Income XYZ Dividend Policy 60 20/qtr 1Q 20 2Q 20
Paid from 2011 $40 dividend spillover
3Q 20
4Q 20
Dividend Spillover 20
In 2012, the BDC earned $60 in taxable income for 2012, but only paid out $40 in 2012. Remember 1Q and 2Q where funded by the 2011 dividend spillover. As a result, the spillover actually absorbed the loss and prevented the BDC from cutting its dividend.
In 2012, let's say the BDC earns $100 million of ordinary income, but takes a $40 million loss. This would lower taxable income to $60 million
Source: Stifel Nicolaus
Page 5
Specialty Finance Important Disclosures and Certifications
September 20, 2010
We, Greg Mason and Troy Ward, certify that the views expressed in this research report accurately reflect our personal views about the subject securities or issuers; and We, Greg Mason and Troy Ward, certify that no part of our compensation was, is, or will be directly or indirectly related to the specific recommendation or views contained in this research report. For our European Conflicts Management Policy go to the research page at www.stifel.com. Stifel, Nicolaus & Company, Inc.'s research analysts receive compensation that is based upon (among other factors) Stifel Nicolaus' overall investment banking revenues. Our investment rating system is three tiered, defined as follows: BUY -For U.S. securities we expect the stock to outperform the S&P 500 by more than 10% over the next 12 months. For Canadian securities we expect the stock to outperform the S&P/TSX Composite Index by more than 10% over the next 12 months. For yield-sensitive securities, we expect a total return in excess of 12% over the next 12 months for U.S. securities as compared to the S&P 500 and Canadian securities as compared to the S&P/TSX Composite Index. HOLD -For U.S. securities we expect the stock to perform within 10% (plus or minus) of the S&P 500 over the next 12 months. For Canadian securities we expect the stock to perform within 10% (plus or minus) of the S&P/TSX Composite Index. A Hold rating is also used for yield-sensitive securities where we are comfortable with the safety of the dividend, but believe that upside in the share price is limited. SELL -For U.S. securities we expect the stock to underperform the S&P 500 by more than 10% over the next 12 months and believe the stock could decline in value. For Canadian securities we expect the stock to underperform the S&P/TSX Composite Index by more than 10% over the next 12 months and believe the stock could decline in value. Of the securities we rate, 51% are rated Buy, 47% are rated Hold, and 2% are rated Sell. Within the last 12 months, Stifel, Nicolaus & Company, Inc. or an affiliate has provided investment banking services for 35%, 24% and 10% of the companies whose shares are rated Buy, Hold and Sell, respectively. Additional Disclosures Please visit the Research Page at www.stifel.com for the current research disclosures applicable to the companies mentioned in this publication that are within Stifel Nicolaus' coverage universe. For a discussion of risks to target price please see our stand-alone company reports and notes for all Buy-rated stocks. The information contained herein has been prepared from sources believed to be reliable but is not guaranteed by us and is not a complete summary or statement of all available data, nor is it considered an offer to buy or sell any securities referred to herein. Opinions expressed are subject to change without notice and do not take into account the particular investment objectives, financial situation or needs of individual investors. Employees of Stifel, Nicolaus & Company, Inc. or its affiliates may, at times, release written or oral commentary, technical analysis or trading strategies that differ from the opinions expressed within. Past performance should not and cannot be viewed as an indicator of future performance. Stifel, Nicolaus & Company, Inc. is a multi-disciplined financial services firm that regularly seeks investment banking assignments and compensation from issuers for services including, but not limited to, acting as an underwriter in an offering or financial advisor in a merger or acquisition, or serving as a placement agent in private transactions. Moreover, Stifel Nicolaus and its affiliates and their respective shareholders, directors, officers and/or employees, may from time to time have long or short positions in such securities or in options or other derivative instruments based thereon. These materials have been approved by Stifel Nicolaus Limited and/or Thomas Weisel Partners International Ltd., authorized and regulated by the Financial Services Authority (UK), in connection with its distribution to professional clients and eligible counterparties in the European Economic Area. (Stifel Nicolaus Limited home office: London +44 20 7557 6030.) No investments or services mentioned are available in the European Economic Area to retail clients or to anyone in Canada other than a Designated Institution. This investment research report is classified as objective for the purposes of the FSA rules. Please contact a Stifel Nicolaus entity in your jurisdiction if you require additional information.
Page 6
Specialty Finance
September 20, 2010
The use of information or data in this research report provided by or derived from Standard & Poors Financial Services, LLC is 2010, Standard & Poors Financial Services, LLC (S&P). Reproduction of Compustat data and/or information in any form is prohibited except with the prior written permission of S&P Because of the possibility of . human or mechanical error by S&Ps sources, S&P or others, S&P does not guarantee the accuracy, adequacy, completeness or availability of any information and is not responsible for any errors or omissions or for the results obtained from the use of such information. S&P GIVES NO EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE. In no event shall S&P be liable for any indirect, special or consequential damages in connection with subscribers or others use of Compustat data and/or information. For recipients internal use only.
Additional Information Is Available Upon Request 2010 Stifel, Nicolaus & Company, Inc. One South Street Baltimore, MD 21202. All rights reserved.
Page 7
