I am recommending a long position in Arbor Realty Trust (the “Company”, NYSE:  ABR).  ABR is an externally managed real estate finance company primarily making first lien bridge loans into the multi-family (64% of assets and ~80% of loans made over the past twelve months) and commercial real estate markets.  Loans are collateralized by assets that are not able to get permanent financing because of large CapEx needs, mismanagement, high vacancies, etc.  Arbor is providing financing to financial sponsors who are acquiring these types of properties with the goal of turning them around. Typical loan terms are 6 – 36 months, are interest only, and have interest reserves.  The loans are small, averaging $7mm - $8mm of principal.  71% of the Company’s loans are bridge loans with typical rates of L + 500 – 650, and also include LIBOR floors.  The Company’s top three markets are New York (36% of loans), Texas (10%), and Florida (8%).  Approximately 70% of the Company’s assets are floating rate.  LTV for these loans is currently 77% based on management’s estimates compared to 88% in 2009.  Arbor qualities as a REIT for federal income tax purposes.  

The reason I am buying the stock is that GAAP accounting is currently obscuring the true book value of the Company resulting in the Company trading in-line with peers based on reported book value (0.9x book), but at 0.7x book when adjusting to reflect the Company’s true economic book value.  Further, of the $2.31 difference between GAAP and true book value, $1.17 will be recognized on the books during 2014, providing a catalyst to earning a ~30% return over the next year, assuming that the Company trades at 1x book.  The following are the primary adjustments to book value:

1.  Real Estate Owned:  The Company’s owned real estate is underwater, but the loans to that real estate are non-recourse, so in no event will the value to the Company from the owned real estate be less than $0.

2.  450 West 33rd Transaction:  In 2007, the Company had warrants in a property that was sold that resulted in a $58MM gain net of incentive fees.  The property was sold for ~$700MM with debt financing of ~$500MM.  In order to defer taxation and to avoid paying out a distribution to shareholders when liquidity was quickly becoming scarce, the Company entered into a “bottom-up guarantee” of $77MM and retained a 2% interest in the property.  This essentially means the Company has guaranteed the buildings’ value from $0 to $77MM through the life of the loan.  This allowed the Company to defer that gain until the guarantee expires in May 2014.  The Company now has ~$180MM of NOLs that will be useful in fully offsetting the taxable gain that should be recognized during 2014.  This guarantee created a net $58MM liability on the books, which should reverse itself this year generating a $1.17 gain in book value per share in 2014.

3.  Unrealized loss on interest rate swaps:  The Company has hedged certain debt instruments, and those interest rate swaps are currently showing a loss.  That will reverse as those debt instruments mature.

4.  VIE Debt Adjustment:  The CLO and CDO liabilities are marked above par.  In no event will the Company have to pay above par to extinguish those liabilities.

Reported Tangible Book	$7.53
Owned Real Estate Adjustment (1)	0.24
450 West 33rd Transaction (2)	1.17
Unrealized loss on interest rate swaps (3)	0.50
VIE Debt Adjustment (4)	0.40
Warrant / Option Dilution Impact	(0.09)
Adjusted Tangible Book Value	$9.75

I should also note that I feel confident in the Company’s book reserves, and believe they may even be over stated.  Since the financial crisis, the Company has incurred provisions equal to $549mm, and has only charged off $387mm over that same time period.  Management stated on their last conference call that they do not expect to take any material additional loss reserves against its legacy assets, and that any they do take will likely be offset by gains on recoveries.

Further, I believe the Company is significantly under-earning, as ~$0.9bn ($1.3Bn at the beginning of 2013) of the loan book relates to legacy assets underwritten prior to the crisis, which are earning ~4% interest rates.  During the 2013, the Company earned $0.33 adjusted for one-time items or a 5% return on the current stock price.  As management continues to work down these legacy assets, ROEs and returns on the stock price should improve.  For instance, the Company reduced its legacy book by ~$400MM during 2013, which resulted in interest income on earning assets rising from 5.2% last year to 6.3% during 2013.  Further, management noted that that the expense side decline as a % of assets going forward, as they put the equity capital they just raised to work, and as the legacy assets stop requiring so much expense to workout.  Offsetting these benefits is the fact that provisions are probably artificially low during the LTM period (after being very high over the past 5 years).  Since 1991, the charge-off rate on commercial real estate loans has been 0.68% of principle according to the Federal Reserve.  Now, credit quality in the Company’s portfolio is probably worse than average.  If I assume double the loan loss provisions that would result in $11MM of provisions per year compared to $6.5mm during 2013.  I made the following assumptions in my analysis of potential ROEs after the legacy assets run-off:

  •  Interest Income to Assets:  Determined the implied interest rate on new loans by assuming that the loans in the old portfolio were are 4%, and used that implied 8.0% new rate for the high ROE scenario.  I assumed in the low ROE scenario that it could come down by 50 bps based on guidance from management that they are seeing some pressure on rates.

 •Interest Expense / Assets:  Left it consistent with the LTM period.

 • Provision for loan losses / Assets:  Used the assumptions outlined in the above paragraph for the low case, and reduced that by 50 bps in the high case, based on management guidance that they expect loan loss provisions to be very low.
 • Costs / Assets:  Have this coming down 30 bps from the LTM period to account for increased benefits of scale and the fact that there are expenses flowing through the income statement related to working out the legacy assets in the high case.  Left costs / assets the same as in the LTM period for the low case.
 • Equity to Assets:  Consistent with the LTM period.

		Future
	LTM	Low	High
Interest income / assets	5.4%	7.5%	8.0%
Interest expense / assets	2.3%	2.3%	2.3%
NII	3.1%	5.2%	5.7%
Provision for loan losses / assets	0.2%	1.5%	1.0%
NII After Credit Provisions	2.9%	3.7%	4.7%
Costs (includes pfd dividend) / assets	1.9%	1.9%	1.6%
Return on Assets	0.9%	1.7%	3.1%
Equity to Assets	25.7%	25.7%	25.7%
Return on Adjusted Equity	3.6%	6.8%	12.0%
Return on Stock Price	5.2%	9.9%	17.5%

These assumptions result in ROEs ranging from 7% - 12%, and yields on the stock price of 10% - 18%.  This analysis makes me comfortable that the stock doesn’t have much downside from here given that even on the low end of the earnings power estimate, the stock shouldn’t trade much below its current price, as the investor could “create” an 8.5% ROE (average and median of comps’ ROE) if the stock traded materially above where it currently trades.  

So, if the Company can continue to reduce legacy assets at the current rate, it will be fully out of its legacy assets in two to three years.  At that point, if the Company is earning the mid-point between the high and the low earnings power, then it will likely trade around book.  If the Company continues its dividend over a three year period, then book will grow to $10.20 by the end of year 3, resulting in a 22% IRR (including dividends) over the 3 year period.  

Management:

Ivan Kaufman has served as ABR’s Chairman, Chief Executive Officer and President since June 2003.  Mr. Kaufman has also been the Chief Executive Officer and President of ACM since 1993.  In 1983, he co-founded a predecessor of Arbor National Holdings Inc. and its residential lending subsidiary, Arbor National Mortgage Inc., which became a public company in 1992 and was sold to BankAmerica in 1995.  Mr. Kaufman has also served on Fannie Mae's regional advisory and technology boards, as well as the Board of Directors of the Empire State Mortgage Bankers Association. 

ACM is ABR’s 2nd largest shareholder with 11% of the shares outstanding.  In addition to his indirect ownership through ACM, Ivan Kaufman owns 350,000 shares (worth ~$2.3mm).  

Risk and Mitigants:

There are three primary risks to my thesis (i) the Company issues equity below intrinsic value over the next year to fund the dividend (I don’t believe they need to), (ii) management merges the Company with its other privately held company in an effort to extract value from the public shareholders (I think this is possible, but would still likely result in a good risk adjusted return), and (iii) the multi-family / commercial real estate market deteriorates materially prior to the catalyst occurring, resulting in write-offs in the Company’s loan portfolio.    

1.  The Company Issues Shares below Intrinsic Value:  Management clearly believes the Company is worth ~$9 per share.  However, they have issued ~$175mm worth of stock below that value over the last two years.  This is likely because they did not want to cut the dividend in, because while the Company can cover the dividend with earnings (including some one-time items) on a consolidated basis, the Company’s CDOs are past their investment period, and thus cash is being trapped in those structures to pay debt amortization and cannot be sent to the equity holder (ABR HoldCo).  Management expects to be able to refinance the CDOs during 2014, as LTVs come down a bit.  

The Company is also currently launching new CLOs with the cash raised.  While these CLOs probably do not offer as good of a return as buying back stock, management believes the Company will earn 13% - 15% returns on that equity (I think those returns will be lower due to the effect of defaults, which management doesn’t appear to be baking into their return analysis).  The Company currently has ~$80MM of cash on HoldCo’s books (including recent preferred stock issuance), which would cover ~3 years of dividends.  The Company did not issue equity after the crisis until this year.  Management stated on its last call that it would rather raise capital in the preferred equity markets over issuing common at the current valuation.   

2.  Management is Currently Contemplating Merging its Privately Held GSE based Business with ABR:  The Company has formed a committee of independent directors to evaluate the merger.  No terms have been announced – just the formation of the special committee.  That special committee appears to have disbanded in Q4, and management has indicated they are having trouble reaching an agreement on valuation, as the independent directors want to make sure that it is accretive to ABR.  I like the fact that the independents are mindful of valuation.  Management thinks that by combining the companies, ABR (i) can become internally managed, (ii) will be larger and have better analyst coverage, etc., and (iii) can more efficiently incent their loan sales force to generate the best opportunities for the Company as a whole.  They believe all of these factors will result in a higher stock price.  I am obviously skeptical that they are just doing this to grab more shares at an attractive price (after all, they think the shares are worth at least $9 compared to $6.70 today).  Based on some scuttlebutt on the relative sizes of the business, I am assuming a 75mm share issuance or a $500MM increase in market capitalization, I believe is a conservative number.  Assuming those shares are issued for consideration worth fair value, then ABR would still have attractive upside from the current share price (~18%), but it obviously would significantly erode the upside and margin of safety.  One would start to lose money if management issued shares worth $500mm to purchase their privately held business worth $350MM of less.  In other words, ABR would have to overvalue the privately held business by ~40% in order to erase one’s upside.  

Below is a table showing each independent directors stock holdings.  This compares to $140k of annual BoD compensation.  

Name	Ownership
William Green   $200k	$0.2mm
Michael Kojaian	$8.0MM
William Helmreich	$1.2MM
Karen Edwards	$0.4mm
Archie Dykes	$0.4mm
Melvin Lazar	$0.8mm
Stanley Kreitman	$0.1mm

Management does seem like straight shooters – for instance, the way they structured their base and incentive fees seems much more fair than others I have seen (incentive fee has appropriate hurdle rates, high water marks, etc.), so that also gives me some comfort.  

Further, Leon Cooperman owns over 5% of the stock, and that should help keep management honest.  

3.  Material Decline in Multi-Family and Commercial Real Estate Markets (especially in NY):  A deterioration in the underlying collateral of the company’s loans due to a real estate correction, would result in increased write-offs and a deterioration in equity value.  

4.  Not Covering Its Dividend:  The Company is not currently covering its dividend when excluding one-time gains.  While I believe asset value supports my valuation/thesis, and that the Company can earn the dividend after it reduces its legacy asset book, it is a risk that if (i) the equity markets seize up or (ii) the Company is unable to continue to reduce its legacy asset book, then the Company may need to cut its dividend, causing the stock to fall further.  I would likely be a buyer on that news, but it is a risk to the stock price.

GAAP accounting is currently obscuring the true book value of the Company resulting in the Company trading in-line based on reported book value (0.9x book), but at 0.7x book when adjusting to reflect the Company’s true economic book value. Further, of the $2.29 difference between GAAP and true book value, $1.17 will be recognized on the books during 2014, providing a catalyst to earning a ~30% return over the next year (including dividends), assuming that the Company trades at 1x book.