Gyrodyne - A Liquidation

Shares of Gyrodyne LLC (G-LLC) represent good value in the current market with shares available for purchase at $25.05 with a liquidation value of $31.24 with the potential for more. The company is in liquidation, thus providing its own ‘catalyst’ to realize value. The company intends to achieve liquidation by December 31, 2016, providing a 25% return with a margin of safety and a larger potential upside, especially from the Flowerfield property, if estimates turn out to be conservative.

Background:

Many value investors are familiar with the name Gyrodyne Company of America (GCA). In September 2013, GCA adopted a plan of liquidation. The company’s liquidation plan has a fairly complicated history and GCA as a whole has quite an interesting history as well. However, its history is largely irrelevant to the thesis. In order to not stray from the narrative, I will provide just a high level historical background.

Years ago, GCA owned a property which was taken by the state of NY under eminent domain. GCA was grossly underpaid for the property and chose to litigate. The courts came out in favor of GCA, which then received substantial payments from the state for additional consideration as well as for statutory interest. At the time of exercising its eminent domain rights, NY paid $26 million for the 245.5 acre property it “took”, whereas the court awarded an additional $98 million of consideration (this, total consideration of $125mm), $67 million in interest, and $1.5 million in costs and expenses.

GCA obtained a private letter ruling from the IRS allowing it to pay out amounts thus received to shareholders without incurring capital gains at the REIT level. Part of this dividend was paid in cash and part was paid in non-transferable interests in Gyrodyne Special Distribution, LLC (“GSD”), which held its properties. The transfer of properties by GCA to GSD resulted in the recognition of capital gain income (and thus REIT income) by GCA in 2013. In order to satisfy applicable REIT distribution requirements, Gyrodyne declared a second special dividend in December 2013. The second special dividend was paid in the form of nontransferable dividend notes (“DN”).

In order to finish the tax liquidation (which must be completed within the two year period from the adoption of the plan), the company undertook its latest transaction which combines all of GCA, GSD, DN into G-LLC. The merger to create G-LLC had to be delayed several times from an original date of August 14, 2014 to September 1, 2015, when it was finally consummated. The delay was caused by enough shareholders not responding to the vote proxy. G-LLC is now the only entity containing all of the properties to be liquidated.

Why is the opportunity available?

First, before the merger into G-LLC, the GSD and DN interests were non-transferable for almost one and a half years. Naturally, the merger into a liquid security gave holders of GSD and DN stakes an opportunity to sell. In fact, when the merger was completed, the trading level of G-LLC was pretty close to today’s announced liquidation value but soon dropped rapidly indicating significant selling. Second, all of this selling occurred in a fairly illiquid market. On an average day, anywhere from a few hundred to a few thousand shares of G-LLC may trade. The market cap is about $37 million and the free float is even smaller (due to large holders; Poplar Point, Tower View, and management together comprise about 21% of the outstanding shares without consideration of any other sub-5% holders), making it impossible for large buyers to take advantage of the situation.

Valuation:

Based on the latest 10-Q filed by G-LLC, the liquidation value per share is $31.24 per share, representing a 25% upside from today. Further, in the 10-Q, the company noted that the liquidation is “currently anticipated to be completed by December 31, 2016.”

The current properties of the company include (along with a description of each:

	Property	Description
1	Port Jefferson Professional Park in Port Jefferson Station, New York. (10 of 14 buildings owned by G-LLC)	On June 27, 2007, the Company acquired ten buildings in the Port Jefferson Professional Park in Port Jefferson Station, New York. The buildings were acquired for an aggregate purchase price of $8,850,000 or $225 per square foot. The buildings, located at 1-6, 8, 9 and 11 Medical Drive and 5380 Nesconset Highway in Port Jefferson Station, are situated on 5.16 acres with 39,329 square feet of rentable space. As of December 31, 2014, there were 15 tenants, comprising 14 leases; the difference reflects one long-term tenant under a month to month agreement. The annual base rent based on the rates in effect as of December 2014 is $743,000 which included month–to-month annualized base rent of $21,000 on approximately 800 square feet. The occupancy rate was 70% as of December 31, 2014.
2	Cortlandt Medical Center	On June 2, 2008, the Company acquired the Cortlandt Medical Center in Cortlandt Manor, New York. The property consists of five office buildings which are situated on 5.01 acres with 31,198 square feet of rentable space on the date of acquisition. The purchase price was $7 million or $231 per square foot. As of December 31, 2014, there were 15 tenants, comprising 15 leases. The annual base rent based on the rates in effect as of December 2014 is approximately $821,000. The property was 100% occupied as of December 31, 2014. Following certain capital improvements, the rentable square feet currently is 31,421 square feet. On August 29, 2008, the Company acquired a 1,600 square foot single-family residential dwelling located on 1.43 acres at 1987 Crompond Road, Cortlandt Manor, New York. The purchase price was $305,000. The Company was able to take advantage of a distressed sale by the seller. The property is located directly across the street from the Hudson Valley Hospital Center and adjoins the Cortlandt Medical Center. The property is zoned for medical office use by special permit and is potentially a future development site for expansion of the Cortlandt Medical Center. On May 20, 2010, the Company acquired the building located at 1989 Crompond Road, Cortlandt Manor, New York. The property consists of 2,450 square feet of rentable space on 1.6 acres. The purchase price for the property was approximately $720,000. This property is adjacent to the 1.43 acre property acquired by the Company in August 2008, and these two properties combined result in the Company owning approximately three acres directly across Crompond Road from the Hudson Valley Hospital Center in addition to the 5.01 acre Cortlandt Medical Center site. The property was 100% occupied as of December 31, 2014 by two tenants with a total annual base rent of $35,700.
3	Fairfax Medical Center	On March 31, 2009, the Company acquired the Fairfax Medical Center in Fairfax City, Virginia. The property consists of two office buildings which are situated on 3.5 acres with 57,621 square feet of rentable space at date of acquisition. The purchase price was $12,891,000 or $224 per square foot. As of December 31, 2014, there were 29 tenants, comprising 30 leases, renting space with an annual base rent of $1,463,000, based on the rates in effect as of December 2014. The occupancy rate as of December 31, 2014 was 93%.
4	Flowerfield	A 68 acre site, primarily zoned for light industry, and is located approximately 50 miles east of New York City on the north shore of Long Island. Flowerfield’s location places it in hydrological zone VII, one of the most liberal with respect to effluent discharge rates. The existing buildings are located in the hamlet of St. James, Township of Smithtown. The Flowerfield property is mostly undeveloped and is adjacent to the land that was taken by the state of NY under the 2005 eminent domain action. There is also about 130,426 square feet of rental space with an annual rent of $1,521,000.

On the whole, management has provided a gross sales estimate of $45,450,000 and a net estimate (net of selling costs of 6%) of $42,723,000 in their liquidation analysis. Using the 2014 impairment (Jefferson) and sale analysis (Cortlandt and Fairfax) and assigning the difference (by subtracting those three properties from $45,450,000, it seems Management has valued the properties as shown below. Also summarized below are the original purchase prices, annualized rents (from the 2014 10-K) and per square foot metrics.

For income producing portion of the properties, management used cap rates of 7.5% to 9.5%. Given the nature and location of the properties, the cap rate does not appear unreasonable. (Note that two properties have a combination of leased and redevelopment space and occupancy at the other two locations is partial so if you took the current property level NOI and applied the cap rate, you wouldn’t get to Management’s estimates as other components are involved). The CBRE office market report for Long Island shows cap rates for B & C properties in Long Island to be within the range used by management. Also, the properties seem to be garnering rents per square feet in line with the local market area rents according to the report ($24-26 psf).

Further, in looking at the transaction activity for REITs that own medical offices, cap rates of 7.5% to 9.5% for Class B & C properties seem to be within the ball park.

Cortlandt and Fairfax were acquired during the recession and the current valuation appears not to be much different from the initial purchase which gives some comfort that the properties are perhaps not being aggressively valued. Jefferson was acquired at the peak of the market but has subsequently been written down in 2013 and 2014 by about 30%.

As to the Flowerfield property, management has used a market value approach. It’s notoriously difficult to value undeveloped land. But one data point does exist from the state of NY condemnation mentioned above. The final consideration determined by the court in that case was $125 million for 245.5 acres of this same property, which would equate to $0.509 million per acre. Applying this against 68 acres yields $34.6 million. Of course a significant discount to undeveloped land value is probably warranted as a large transaction similar to NY probably wouldn’t occur for all of the acreage. Given the location next to SUNY, it’s difficult to imagine this land would be totally worthless. I believe a 50% discount from the judgment amount is adequate to account for the market correction and illiquidity, giving a value of $17.31 million for the 68 acres. The rentable area of 130,426 square feet has the lowest annual rent per leased square foot at $17, which is quite low even in C markets. In the past few years, this property has produced NOI at 50% of base rental income. IF it were to be 100% leased at $17 psf, this NOI would be $1.109 million (130,426 * 17 * 50%). At a 10% cap rate on $1.109 million, the property would be valued at about $11.1 million equating to $85 psf in gross sale value. However, it’s going to be very difficult to be leased 100% prior to sale, and perhaps a discount of 50% to $42.5 psf is warranted, giving a value of $5.54 million. Adding the land and the rental pieces together, Flowerfield could total up to $22.85 million. Appraisal estimates used in GCA’s litigation with NY were in excess of this amount. However, those appraisals were made using a residential zoning scenario for the acreage and the company, according to the latest 10-Q, is planning to rezone prior to sale.

I am not suggesting that Flowefield will necessarily fetch $22.85 million, but the point is to only show that management’s implied valuation of $16.4 million is likely achievable given that the value under fairly conservative assumptions produce a number above management’s valuation.

It is also worth noting that valuations for the properties similar to those used in liquidation accounting were also used to issue the GSD dividend to take advantage of the IRS private letter ruling and it would be highly unlikely that management would significantly overestimate the real estate value in the dividend.

Potential upside beyond current valuation:

Valuations in liquidations often tend to be conservatively estimated. Managements are loathe to set up high expectations in liquidations and not meet them. It’s much easier to set up a lower expectations and then to beat them handily. This is particularly true when it comes to distributions or dividends. It is also safer for management to make conservative estimates and to surpass them than to make aggressive estimates and risk being blamed/sued for breach of fiduciary duty if the final distributions come out less than estimated.

As noted earlier, Flowerfield’s land and rental values taken together may somewhat exceed the Management valuation. This property has perhaps the most potential for generating. It is possible a higher value will be realized, but difficult to say exactly how much.

In the case of G-LLC, Management has a further incentive for a conservative bias. The company has a bonus plan in place to reward management for liquidating assets at a premium to appraisal value. The bonus pool is to be funded upon the sale of each of the Company’s properties with an initial amount equal to 5% of the specified appraised value of such properties, so long as the gross selling price of the property is equal to or greater than 100% of its appraised value. Additional funding of the bonus pool would occur on a property-by-property basis when the gross sales price of a property exceeds its appraised value as follows: 10% on the first 10% of appreciation, 15% on the next 10% of appreciation and 20% on appreciation greater than 20%. Furthermore, if a specified property is sold on or before a designated date specified in the plan, an additional amount equal to 2% of the gross selling price of such property also would be funded into the bonus pool.

In order for management to earn the additional 2%, the properties must be sold by the following dates:

Cortlandt Manor – December 31, 2014
Fairfax – December 31, 2014
Port Jefferson – December 31, 2015
Flowerfield – December 31, 2016

The delay of the merger (noted above) caused the deadlines on Cortlandt and Fairfax to be missed. During October 2015, GLLC signed two non-binding PSAs to sell two of the ten buildings in the Port Jefferson Professional Park with closing expected in December 2015. The sales prices in the contract exceed the appraised value. The Cortlandt, Fairfax, and Jefferson properties are expected to be sold by June 2016.

The bonus pool is distributable in the following proportions: 15% for the Chairman, 50% for the directors other than the chairman (10% for each of the other five directors) and 35% for the Company’s executives and employees.

The property-level appraisals in the plan were withheld under confidential treatment but total appraised gross value stated in the plan is: $45,050,000 compared to gross real estate proceeds stated in liquidation accounting foot note of $45,450,000. As a result, management is certainly incentivized to seek the highest values for the properties.

Further, the Company owns a 10.12% LP interest in the “Grove” partnership. The Grove owned a 3,700+ acre citrus grove located in Palm Beach County, FL, which was the subject of a plan for mixed-use development. The Grove’s lender (Prudential) commenced a foreclosure action against the partnership. The property was sold by the partnership, the lawsuit was dismissed, and the debt repaid. GLLC did not receive any distribution in connection with the sale. Under the agreement with the purchaser, the Grove may receive additional payments if some development benchmarks are achieved by the purchaser. G-LLC is not providing a fair value and marking the investment at 0. It’s very likely that this the Grove is in fact a doughnut. It’s a long-shot but if they get something for it, it may increase the return, but it’s impossible to guess if and how much. Just something to be aware of.

Risks - there are two main risks:

1.     Timing: The entire saga of the Gyrodyne liquidation has been going on for some time and there’s no guarantee that management will stick to its December 31, 2016 guidance. This is not a capital risk as such, only a timing risk which may reduce annualized return. The only way this becomes a capital risk is if the liquidation takes so long that we go through the next major real estate correction. Currently, this possibility of the liquidation taking THAT long seems remote.

2.     Development: This risk relates to Flowerfield and Cortlandt undeveloped spaces. Management has indicated that perhaps pursuing zoning or developing projects at these properties will increase the ultimate liquidation value of the properties. A development project may require an investment of capital. If the returns on these projects are positive, the liquidation value and return could be enhanced but so would be the time required. It’s impossible to know which way management will take (although, indications are that zone-and-sell will be the preferred route) and how that would affect the return, given the time required.

Given the downside protection provided by the estimated liquidation value, the timing risk and potential additional upside (from 0% to 20%) is best seen in a sensitivity chart shown below. If distributions are made as properties are sold, the realized returns will be higher than those presented here as they assume one final distribution at the end of the period.                                 

	Final Distribution Value
	0%	+ 5%	+ 10%	+ 15%	+ 20%
Years	31.24	32.80	34.36	35.93	37.49
1.0	24.7%	30.9%	37.2%	43.4%	49.7%
1.5	15.9%	19.7%	23.5%	27.2%	30.8%
2.0	11.7%	14.4%	17.1%	19.8%	22.3%
2.5	9.2%	11.4%	13.5%	15.5%	17.5%
3.0	7.6%	9.4%	11.1%	12.8%	14.4%

As of this writing, the 10-year treasury trades at 2.25% and the S&P 500 index trades at 5% earnings yield on a trailing basis. Given the relative safety of this investment, I believe it provides a good risk-adjusted return on a relative and absolute basis compared to alternatives available in the market.

Please note:

Under NO circumstances should any content or communication here be construed as investment advice or a recommendation to buy or sell any security, whether expressed or implied. Factual statements are believed to be truthful and reliable, but are not warranted against errors or omissions. PLEASE do your own due diligence prior to investing.

Disclosure: Long GYRO

Nicholas Financial (NICK) - An Undervalued Auto Lender

I made an investment in Nicholas Financial (NICK) on 3/30/15, prior to the inception of this blog. The write-up below is as of that date.

I am increasing my holdings of Nicholas Financial at $13.70 per share because the recently completed leveraged recapitalization will substantially improve per share returns to continuing shareholders. I came across this specialty auto lender in November 2014 and had taken a small position, hoping to profit from the Dutch tender transaction that the Company was executing. However, as I followed the company more and the tender offer was successfully executed, I found it even more attractive as a longer-term holding.

Some background on Nicholas Financial, a short version of the November thesis, and the reasons for increasing my position today are given below.

Background on Nicholas Financial:

Nicholas Financial is a specialty auto lender, lending primarily to subprime borrowers. The Company generally acquires these loans from dealerships. To a lesser extent, the Company also originates direct loans and sells consumer-finance related products. Understandably, subprime lending has a very negative connotation among investors and deservedly so (recall the experience of the shareholders of New Century Financial Corporation, Countrywide, etc.) However, one must be careful not to paint with too broad a brush. While subprime home lending was one of the factors in the recent financial panic of 2008-09, subprime auto loans have a much better history, including during the recent crisis. In fact, according to Experian, about 30% of all auto-buyers in the US are in the subprime category. Lenders have long experience safely working with such loans. Collateral values are well-known, have not been observed to skyrocket beyond reason (unlike homes in the past cycle), and have proven to be an adequate basis for safe lending when combined with prudent assessment of borrower income and credit metrics.

Notice that in 2010, GM purchased AmeriCredit (a subprime lender), in order to rebuild its in-house financing arm following the separation of Ally financial (formerly GMAC) from GM. Ally itself, now free of government ownership and able to pursue any type of lending it pleases, is looking to expand into subprime lending as its CEO recently announced.

There are many subprime industry participants who have performed quite well over time by carefully making prudent loans at the right price. Examples of such operators in the sub-prime space are AmeriCredit (purchased by GM), Credit Acceptance Corporation, America’s Car Mart (which operates dealerships and a financing arm), Nicholas Financial, etc. Despite increased charge-offs experienced by the entire industry during the 2008-2010 period, Nicholas Financial did not have a single year with a net loss. Like the industry, Nicholas did experience higher losses but these were much more manageable than those experienced by competitors and by non-auto lenders. Consider this performance against the fact that Nicholas operated (and still does) with a significant exposure to Florida, a state hit harder than most during the crisis.

Quality of Nicholas Financial’s business:

Similar to analyzing banks, I prefer to judge non-bank lenders on the basis of the returns on equity they are able to safely generate. In this regard, Nicholas Financial’s track record has been a good one as shown in the table below. It falls short of some of its competitors results (particularly Credit Acceptance). However, it’s worth noting that Nicholas used far less leverage compared to competitors. For example, as of December 2014, NICK’s financing debt to loans receivable ratio stood at 47% compared to 70% for Credit Acceptance. Please note that in any given year, some adjustments to GAAP earnings, allowance, and the provision may be required in order to judge financial performance. However, over a credit cycle, these tend balance out. Thus, when looking at a 12 year period (below) ROEs calculated based on reported numbers can serve the purpose just as well.


Figure 1 – Nicholas Financial Results FY2002 – YTD15 (As of 12/31/14)



Note: YTD15 ROE is annualized

Recent periods include the following expenses that reduce ROE:

1. One-time professional fee expenses related to the failed sale of the Company to Prospect Capital (see background below) in the amount of $2.3 million (FY2014) and $0.4 million (YTD FY2015).

2. Dividend tax withholding charges of $1.2 million (FY2013) incurred under the Canada-United States Income Tax Convention, which imposes a 5% withholding tax on any dividends paid. During FY 2013, the Company paid a special dividend of $2 per share which is not expected to repeat.

3. Elevated loan-loss provision in 2015 due to the higher competitive activity and looser market lending standards at the current stage of the credit cycle, as observed by the Company.

There are certain nuances of Nicholas Financial’s business and accounting (dealer discounts, loss provisions, charge-off experience, interest rate swaps, etc.) which are important to analyze. I have left their detailed analysis out of the write-up in order to keep the length of this article manageable. Having considered these factors, though, they did not lead me to change the basic takeaway that Nicholas’ credit experience has been better than the industry. The high ROEs above are a reflection of prudent underwriting and not a reflection of leverage (often used in large quantities by other lenders) as Nicholas Financial was much more conservatively leveraged than competitors (see more comments on this below) during this period.

When analyzing allowance for loan losses and loss provisions, readers should take note of the error-correction made by the Company during FY13 (Form 10-K, Footnote 2). The impact on the balance sheet and the income statement was not material on a net basis. However, understanding this is important in analyzing the loan loss reserving. If, after doing your own analysis on these items, you would like to discuss any aspect, I would be happy to do so.

Rationale for the November 2014 purchase at $12.50:

In December 2013, Nicholas Financial reached an agreement for a tax-free stock-for-stock merger with Prospect Capital at $16 per share. Some shareholders decried this valuation as too low for the quality of Nicholas’ business. At the same time, the SEC raised some questions regarding Prospect’s historical practice of not consolidating certain entities. Due to this SEC development, Prospects’ registration statement could not become effective on time to close the deal and the deal was called off in June 2014. As a result of the failed merger, Nicholas’ stock price dropped below book value, eventually nearing $11 per share. In October, management announced that they had begun to look into various alternatives for returning cash to shareholders and/or selling the business. At this time, buying the stock around $12.50 looked appealing given several factors:

1. If a sale occurred, it would be at a price of at least $16. The growth of the business, between the December 2013 Prospect agreement and November 2014, had built up the value from $16 per share to at least $17 or $18. 

2. With the stock trading near book value, if a sale did not occur, purchasing at this price would give the investor respectable returns (recall that recent ROEs were around 15%). Through a leveraged recap or special dividend, management could safely increase leverage closer to the competitor average and boost ROEs. This would greatly enhance per-share value of continuing shareholders.

3. Given the small size of Nicholas’ footprint (Nicholas operates only in 16 states and has 66 locations) compared to the large size of the auto-loan subprime market and the growing sales experienced by the auto industry following the lean years of the recession, the business could continue to grow (i.e. reinvest earnings) without requiring many changes.

Rationale for March 30, 2014 purchase at $13.70:

In December 2014, Nicholas announced a tender offer (i.e., #2 above materialized). The size of the tender offer was large, $50-70 million compared to a market capitalization of $160 million. It was financed with 4%-interest debt (subject to change with LIBOR). But there was a catch: the offer was conditioned upon Nicholas obtaining all of the shares it sought, i.e. there would be no tender if all of the desired shares were not available to purchase (i.e. “all or none”).

In early March, the results were announced and the tender offer was successful, resulting in a reduction of approximately 38% of the stock outstanding at a price of $14.85 per share. This has greatly increased the economics of the current business for the continuing shareholders.

Below are some pro-forma calculations showing the effect of the leveraged recapitalization on Nicholas’ per share ROEs, book value, and EPS. A few things to note in those calculations:

1. The Company expects the costs of executing the recapitalization to be $800,000. These have been added to the $70 million tender amount below.

2. FDSO (Fully diluted shares outstanding) are estimated using the SC TO-I/A dated 3/19/15 filed with the SEC.

3. Interest expense of the debt issued for the leveraged recapitalization is calculated at 4% on $71 million of debt incurred.

4. Operating income excludes professional fees incurred in the failed sale of the Company to Prospect Capital and changes in FV of rate swap.

5. The first EPS calculation (2nd table) starts off with prior year (FY14) operating income adjusted for changes in #3 above and the second EPS calculation (3rd table) starts off with YTD’15 figures on an annualized basis, also adjusted for the changes in #3 above. The major reason for the reduction in the Operating Income between these two tables is a higher provision for loan losses recorded by the Company due to the higher competitive activity in auto loans at the current stage of the credit cycle and the resulting looser industry-standards observed by the Company.

Figure 2 – Selected Financial Results, Pro-Forma for Tender



Note: Tender reduction includes estimated fees of $800,000

Due to the tender above book value, the implied premium above book value for continuing shareholders has increased, i.e. price/book ratio has increased due to a reduction in book value. However, ROEs and EPS have been substantially improved on a go-forward basis making the shares attractive even at a higher price/book valuation. In the current environment it’s rare to find a company earning ROEs of 18-20% trading at 1.25x book value. Also Nicholas’ balance sheet does not contain any intangibles. Consider that American companies as a whole (whether measured by the Dow Jones, S&P 500, or Fortune 500), have generally earned ROEs around 12%-13% for many decades. For comparison purposes, the current price-to-book of the S&P 500 is about 2.7. Thus, it appears that ownership of Nicholas financial is attractive from a statistical standpoint.

Following the tender, the stock rose briefly and then began falling. I believe the principal reason for this is the oversubscription of the tender offer. Not all of the shares tendered were accepted for payment by the Company. Those of you involved in our earlier odd-lot tender arbitrages are aware of this “proration” factor. According to SEC filings, approximately 618,738 shares were not accepted. These were shareholders (or arbitragers) hoping to tender but unable to do so. The average volume of trading of Nicholas Financials’ shares was about 45,000 shares per day before the tender, which is quite low compared to the average large company stock. After the tender, this will presumably reduce even further because of the smaller float. Thus the unsold shares of those hoping to tender create downward pressure on the price as these investors or arbitrage funds begin to liquidate their shares in the open market. This phenomenon is present in many tender offers.

Another reason could be frequent coverage given to subprime auto-loan backed securitizations by several newspapers. Issuance in this market has risen substantially causing some to speculate whether a bubble exists in the securitization market for auto loans. However, Nicholas Financial does not participate in the securitization market as either an issuer or a purchaser and management is well aware of the currently loosening credit standards in auto lending.

Whatever the reasons for the lower price, the current price of $13.70, i.e. 1.25x post-tender book value and 7x estimated post-tender EPS is fairly attractive (1) compared to other options now available in the market, (2) given the historically disciplined and profitable credit underwriting of Nicholas’ management and, (3) a chance to continue growing the business gradually over time.

Downside protection analysis:

From a down-side protection standpoint, one can think of the company in three parts, each of which represents value the business could generate (similar to how some insurance companies may be evaluated

1. Value of net assets as of 12/31/14. For this we will use book value as a proxy since practically all of the significant items are liquid with an adjustment to be made for the deferred tax asset. Loans are stated as cost and could currently be liquidated at a premium to cost given the current market conditions and interest rates. However, for the sake of conservatism, these are considered at cost without any adjustment. Future interest to be earned in considered separately in ‘2’ below (if the loans were sold the present value of this future interest would show up at a premium to par in the price received and ‘1’ and ‘2’ could be considered together). Book value stands at $85 million as shown above less an adjustment for DTA of $1.6 million (due to the allowance being of a size smaller than that required for the entire DTA to be used). In total, this item ‘1’ is $83 million.

2. Future cash-flows in run-off (i.e. future revenues less expenses assuming the Company ceases extending new loans and lets the existing loans liquidate as they are paid off). As of December 31, 2014, the Company had approximately $135 million in unearned interest (i.e. interest expected to be earned from loans currently on the books) and dealer discounts of $17.5 million. I assume an average loan life of 3 years, loan charge-offs of 5% with a loss given default of 80% (this charge-off is in addition to what the Company has already reserved in the allowance), ongoing non-interest cash operating expenses of $16 million a year, interest expenses of $8 million a year (4% of $200 million LOC balance), and a 37% effective tax rate. Based on the above, aggregate income (3 years combined) in runoff would be approximately $39 million plus $4.5 million for the deferred tax asset, including the adjustment given above.

3. Value of future business to be conducted. This would include interest and discounts on all loans to be purchased from now until judgment day less all expenses. Value of customer relationships, existing contracts, assembled workforce, going-concern business value etc. are all items included in this item.

Since a downside protection analysis must be conservative by its nature, I assigned a value of ‘0’ to the third item above. (Readers might recall this is similar to the Symetra analysis from March 2012). Adding together the other two items gives total run-off value of $126.5 million. This compares to a market value of $105 million. Thus it appears that adequate downside protection exists. Even taking loan charge-offs to 10% would cause the aggregate run-off value to be $113.5 million, in line with current market capitalization.

Risks and other considerations:

As noted above and frequently seen in the news, at this time in the credit cycle, credit standards for auto lending are loose. It’s possible the loss experience in the coming years will be worse than has been the case in the last few years. However, Nicholas Financials’ management seems alert to the situation and has noted, in its SEC filings, that they are already taking actions in response. There is no doubt that if the underwriting cycle worsens Nicholas will have its share of losses. However, I believe these will be less than the industry and will not threaten Nicholas’ existence. Through one of the worst credit bubble bursts of our generation in 2008-09, management has shown their ability to profitably run the business through cyclical troughs. The performance was especially admirable given that a large portion of the loan book was in Florida which was one of the worst hit states during that crisis. A down-cycle may even provide an opportunity to increase the position at lower prices.

There is always a certain temptation to “wait” for such a credit problem to occur in the industry and then buy attractive securities (such as Nicholas Financial). However “waiting” is often simply wishful thinking in disguise. Guessing whether loosening credit standards will rise to the level of a “bubble” in the future is a gamble and not an investment approach. When a good business, with a good management, and a good price is available, the right thing to do is to buy it rather than trying to guess unknowable factors that may or may not occur. If adverse developments do occur and the stock trades down, one can take advantage to average down the purchase price. Over an entire cycle, if earnings are higher than over the previous cycle, the company will be worth more over time. Any ripples to disturb the tide during a given cycle could be used to increase the position if cash is available to the investor.

Regulation presents additional risk. Recently, much has been made of the CFPB’s direct supervision of non-bank auto finance companies. With more than 10,000 aggregate originations, Nicholas may qualify as a “larger participant” in the auto-lending market and thus be subject to CFPB regulation once a final rule is issued. Until a final rule is issued it’s difficult to estimate its impact on Nicholas and costs of regulatory compliance. I believe the discount from fair value (i.e. the margin of safety) in the purchase at $13.70 is sufficient to cover any eventualities of the final rule-making. If future costs of regulatory compliance turn out to be quite high, and impact earnings in a significant way, I will have to revisit the investment thesis. Regulatory compliance costs up to $1 million annually would not change my thesis.

For financial firms it’s also especially important to consider liquidity risk. To finance the recapitalization, the Company expanded its line of credit from $150 million to $225 million (contingent on successful completion of the recapitalization). The maturity was extended from January 2015 to January 2018, which reduces refinancing/liquidity risk. There is no other debt. The Company also does not finance itself in the auto loan securitization markets which further reduces liquidity risk.

Please note:

Under NO circumstances should any content or communication here be construed as investment advice or a recommendation to buy or sell any security, whether expressed or implied. Factual statements are believed to be truthful and reliable, but are not warranted against errors or omissions. PLEASE do your own due diligence prior to investing.

Nicholas Financial has fairly limited liquidity. If, after your own due diligence, you decide to transact in this security, I would suggest using limit orders rather than market orders to purchase shares.

Please note that any dividends paid by the Company in the future (currently none are paid) would be subject to Canadian withholding taxes.

Disclosure: Long NICK

MCG Capital merger announced

The investment in MCG Capital investment which has turned out to be a success. In November 2014 (prior to the inception of this blog), I had made three investments that were corporate action driven special situations. One of them was MCG Capital entered into in at $3.68. Given management's actions (tendering for over 50% of the then-outstanding stock below NAV, rapidly liquidating loans over several quarters, withdrawal of SBA licenses, etc. to name a few) and the presence of motivated shareholders, there were indications that the company would either be liquidated or sold, both of which seemed likely to happen at prices higher than where the stock traded.

On April 29, 2015, MCG Capital announced that it was being acquired by PennantPark Floating Rate Capital Ltd for consideration valued at $4.75. I have liquidated this position to look for other cheaper securities.

I had a high probability estimate of what would happen but the timing of the merger was pure luck and I had no indication it would be so soon. As a result, the total return on this investment has been 23% over the six months since November. This overstates the true economic result because short term capital gains taxes will be due.

If you are invested in this security, please take a look at the merger announcement, consideration, the attractiveness of PennantPark Floating Rate Capital (PFLT) stock to reach your own conclusions.

Under NO circumstances should any content or communication here be construed as investment advice or a recommendation to buy or sell any security, whether expressed or implied. Factual statements are believed to be truthful and reliable, but are not warranted against errors or omissions. PLEASE do your own due diligence prior to investing.