This is chapter seven of a guide on REITs. Begin here.
This is chapter two of a guide
to REIT stocks. Begin here.
Download the entire guide as PDF to read and reference at your own pace.
“We invest in a diversified portfolio of structure finance assets in the multifamily and commercial real estate markets, primarily consisting of bridge and mezzanine loans, including junior participating interest in first mortgages, preferred and director equity,” said Ivan Kaufman, boss of Arbor Realty Trust (ABR on Nyse).
“Through our agency business, we originate, sell and service a range of multifamily finance products through Fannie Mae, Freddie Mac, Ginnie Mae, FHA, HUD and CMBS programs. We retain the servicing rights and asset management responsibility on substantially all loans we originate and sell under the GSE and HUD programs.”
In simple language: Arbor is in the business of bridge financing. In bridge financing, the borrower typically seeks a short-term loan for the acquisition of mismanaged property. Once the borrower rehabs the property to market conditions - what is referred to in the industry as “stabilizing the property” - the bridge financier’s loan is paid with conventional lending.
Bridge financing is typically short term in nature, less than two years to maturity, and is priced higher than conventional financing due to the added risk of rehabbing a property.
Arbor also reports operations in niche markets, such as mezzanine finance, preferred equity investments, junior participating financing and structured transactions. Mezzanine loans are subordinate to conventional first mortgage loans but are senior to borrower's equity. Preferred equity investment is often used at times when the lender forbids the borrower to place additional debt on the property, and so, additional debt in the form of preferred equity is used instead. Junior participating financing is an agreement between two lien holders of the seniority of their position.
Arbor ended the year with 150 bridge loans, 12 preferred equity positions and 8 mezzanine loans. Of the 150 bridge loans, 120 loans were backed by multifamily properties. The total unpaid balance was $2.6 billion.
Bridge financing was about 70% of the total assets, equating to $1.8 billion. The weighted average interest rate was 6.28% for the loan portfolio, which excludes origination fees. The maturity of the portfolio was two years - 23.6 months, to be exact.
Over the past five years, Arbor substantially grew its loan portfolio. Loans and investments grew by 70%, from $1.5 billion to $2.6 billion and total assets grew by 94%, from $1.9 billion to $3.6 billion.
The increase in loan size was the result of increased leverage and equity raised in the secondary market. Total debt increased by 99%, from $1.2 billion to $2.5 billion and total equity grew by 98%, from $437 million to $864 million.
The weighted average last dollar LTV was 72% for the bridge loans; 90% for the preferred equity investments and 63% for the mezzanine loans.
The "Last Dollar LTV Ratio" is calculated by comparing the total of the carrying value of the loan and all senior lien positions within the capital stack to the fair value of the underlying collateral to determine the point at which Arbor will absorb a loss.
Arbor finances its lending activity using collateralized loan obligations (CLOs) and by issuance of preferred stock. As of 2017, collateralized loan obligations consisted of about 50% of Arbor's total liabilities and preferred stock issuance. The cost of the CLOs was 3.48% as of 2017, a slight increase from the 2016 cost of 3.21%.
Not unlike a bank, Arbor earns income by the difference between the money it lends and the money it borrows and by earnings fees from originating, selling and retaining mortgage servicing rights. In 2017, the banking-related side of the business contributed $157.2 million and the warehouse side of the business contributed about $178.9 million.
The cost of the banking-side of the business, in the form of interest expense, was $90 million and the cost of the warehouse-side of the business, in the form of employee compensation and selling and administrative cost, was $122.9 million. In short, the banking-side generates about twice the operating income: $67.2 million in gross income for banking and $34.3 million for the warehouse.
Before the common shareholder may determine how much cash is left to be distributed, there are additional stakeholders that stand in the way. To find the actual amount, the common shareholder needs to deduct some accounting-related expenses and some capital stack expenses.
Accounting-related costs include impairment of loss and provision for loan loss and loss sharing. Expenses related to the capital stack include preferred stock dividends and provision for income taxes.
In numbers: over the past three years, Arbor distributed $7.5 million to preferred stockholders and reserved for future tax payments a total of $14.1 million. It recognized a provision of $5.8 million.
Yet true expenses are often not found in the financial statement. Like many publicly traded companies, the cost of decisions made in a certain quarter creep back to the balance sheet only in several years after. If we go back to Arbor, the real cost of originating a loan at leverage greater than 70%, which is considered risky, can only be understood at times of duress.
During 2017, the stock traded as high as $8.99 and as low as $7.11. During 2016, the stock traded as high as $7.99 and as low as $6.92. Management paid 62 cents in 2016 and in 2017 it paid 72 cents, a similar dividend yield in both years. ABR started 2018 at $8.64 a share and now trades at $11.66 a share, an increase of 35%.
If you had bought $100 of ABR at the end of 2012, your investment at the end of 2017 would be worth $216, a compounded rate of return of 16%. By investing in the Nareit All REITs index your return would be $160, roughly 10% compounded return. If you had invested in the FTSE Nareit mortgage REITs index you would now have $155, a 9% compounded rate of return.
Arbor serves as prime example how the stock market carefully monitors trends in both FFO and AFFO measures. Over the past five years, Arbor's FFO grew by roughly three times, from $25 million in 2013 to $94 million in 2017, a compounded annual growth of 19%.
The AFFO grew by two times, from $27 million in 2013 to $84 million in 2017, a compounded growth of 25%. And in per-share numbers, FFO per share was $0.58 in 2013 and $1.17 in 2017; AFFO per share was $0.63 in 2013 and $1.04 in 2017.
Like mining for coal where if the value of coal declines, then miners stop their work. If the value of real estate drops, borrowers typically stop paying. Since Arbor is in the business of lending, the company is acutely dependent on real estate value and on its borrower's willingness to repay debt.
In addition, Arbor finances its operations using capital markets. And if Arbor’s lenders, whether stock holders or corporate debt providers, refuse to lend or require a much higher interest rate, then Arbor's business model - lending money at a higher cost than its original cost - greatly suffers.
If you invest in ABR, you must trust its boss. ABR’s chairman, chief executive officer and president own 70% of the outstanding membership interests and about 31% of the voting power of the outstanding stock.
Finally, the investor in a mortgage REIT is susceptible not only to the value of the collateral of loans, but also to market sentiment. If the stock market is fidgety about the future, then the value of your common stock will be penalized accordingly, irrespective of loan performance.
Mr. Ivan Kaufman, boss of Arbor, is a veteran of real estate finance business. He built the foundation of Arbor in 1983 when he founded Arbor National Holdings, which focused on the origination and servicing of residential mortgage loans. His chief financial officer, Mr. Paul Eleni, has been with the company since 1991.
Arbor's business is to provide customized financing and to execute transactions quickly, while guarding for credit quality. Customized financing is meeting the needs of borrowers that typically cannot access conventional bank financing.
And in an industry myriad with regulatory requirements and corporate bureaucracy, Arbor's ability to quickly finance loan transactions provide the borrower value. Credit quality means to assess track record of the borrower in repositioning investments while managing the overall loan portfolio.
Similar mortgage REIT companies, in terms of market capitalization, are KKR Real Estate Finance Trust, Hannon Armstrong Sustainable Infrastructure Capital and Granite Point Mortgage Trust. In the section below, I will describe how they differ in terms of valuation, profitability and leverage ratios.
Arbor faces competitors outside of the immediate peer group. If commercial companies and life insurance companies loosen their underwriting criteria and due diligence requirements, then borrowers are less likely to search for bridge financing. In addition, non-lender bank funds, such as Money360 or Lendingclub, are gaining market shares in the commercial real estate market. Since these funds are often financed solely with equity, they provide more flexible terms and pricing.
The stock market is discounting by 50% the earnings of Arbor Realty compared to its peers. One dollar of funds from operations from Arbor Realty is priced at $8, while one dollar of funds from operations from its peer groups is priced at $15. The difference in the valuation was not based on 2017 earnings but based on the average earnings over the past three years.
The reason for the difference in valuation is because of the loan assets owned by the varying mortgage REITs. For example, KKR Real Estate Finance holds a senior position loan with an average loan to value of 67% and floating based payment alongside a maturity date greater than 5 years.
Arbor, in comparison, is de facto a bridge lender. 150 of its 170 loans will expire over the next 2 years. What this means is an investor in Arbor must make plenty of assumptions in terms of where Arbor’s earnings will be in 3 years.
The stock market's fascination with companies can be seen in the price to book value ratio as well. Arbor Realty reports a book value per share of $14, while its stock price was $10.04, a 72% discount price to book value. KKR Real Estate Finance reported a book value of $19.73, while its stock was priced at $18.87, a 96% price to book value. And Granite Point Mortgage Trust reported a book value per share of $19.16 and a stock price of $18.10, a price of 95% to book value.
Hannon Armstrong is the only mortgage REIT priced at a premium. A quick glance at the reported book value of $12.75 and a stock price of $18.85 is a premium of about 50%. This premium can be understood as follows:
According to the stock market, if Hannon Armstrong sold its assets and paid off its lenders, what would remain for the shareholder is about 1.5 times the reported equity. In numbers: the sale will result in total proceeds of $1.16 billion, while the reported equity balance is $643 million.
There is a reason behind the apparent premium and it is related to the accounting profession. Specifically, the accounting behind the equity method investment results in a reduced amount of assets compared to what Hannon Armstrong truly owns. Because the stock market and Wall Street analysts are aware of the under reported assets, there is an "optic" premium to book value.
Reviewing the ratio of funds from operations to revenue generated by the firm, we see that most companies align with industry standards, which are 70 cents of funds from operations for every dollar of revenue. Yet with reported funds from operations of $94 million and revenue of $256 million, it appears that Arbor is not meeting industry standards. Its ratio is merely 26%.
The reason for the low ratio requires a careful read of Arbor's financial statement. There we will observe that this mortgage REIT has two lines of business. The first is the business of originating commercial bridge loans, and the second, which is not easily understood, is that Arbor originates and sells loans to government agencies, such as Fannie Mae and Ginned Mae. This line of business creates an additional source of income for the firm ($77 million to be exact), which is inflating the reported revenue. If we adjust for this business line, the ratio becomes reasonable, at funds 53%.
Arbor, KKR Real Estate Finance and Hannon Armstrong are paying a dividend yield of 7%. While dividend-seeking investors may be lured by the yield, I will argue that (1) to purchase stock because of the dividend is a mistake, and (2) it is utterly wrong in the case of mortgage REITS.
Regarding (1): What makes a business exceptional is not how management distributed the earnings of the company, but what is management doing with the retained earnings. Management that compounds retained earnings at 7% will outperform a management with a dividend policy of 7%. In addition, there is a tax disadvantage to dividends as both the giver and the receiver pay taxes.
Regarding (2): Mortgage REITs are highly susceptible to changes in the dividend policy. Consider Wheeler Real Estate Trust, for example. in 2017, the stock traded as high as $15 and as low as $8, and since management decided to cut the dividend, the stock market became wary of its on-going operations. WHLR now trades for 85 cents.
Current assets are cash held by the REIT and receivables net of loss allowance. These exclude the value of the real estate investments and any intangibles such as goodwill. Current liabilities include accounts that are payable in less than one year and any revolving lines of credit that renew every 12 months. SRG and PEI showed a higher positive net asset balance, while WPG and BFS showed a deficit account.
Financial theory states that if the working balance account (the difference between current assets and current liabilities) is positive, then the REIT can quickly pay all its outstanding debt and continue to operate, while a REIT that shows a negative account will surely cease it operations.
But financial reality is that these ratios, in today’s capital market, are practically meaningless. REITs’ business does not erupt in 12 months, and there are ample numbers of places to borrow short-term funds. But, nostalgic as I am, I still like to quickly glance over this ratio.
The difference between residential mortgage REITs in general and commercial real estate mortgage REITs is that the type of loan they originate and hold is very different. As mentioned earlier, Arbor is a bridge lender whose loans will expire in two years and is thus very dependent on the business cycle (the interest rate environment and real estate developer’s Animal Spirits.)
KREF's portfolio of loans is backed by office buildings and multifamily buildings. The two asset classes compose about 70% of the loan collateral. And since most of the portfolio is in New York, the investor must understand that New York, while a great city to live in, is not a "lender friendly" state.
What this means is that foreclosure proceeds take a long time and, the cost of a legal battle often outweighs the benefit. In short, the lender in New York, compared to Texas for example, faces additional risks that are unreported on financial statements.
Granite Point Mortgage has only 61 loans but has a market capitalization roughly the size of Arbor that owns over three times the amount of loans and has a whole origination business. This is because Granite Point loan size is much greater than the typical mortgage REIT’s loan size. With reported loans of $2.4 billion, the average loan is $39 million, a very high loan balance. The company reported that the initial LTV is 69.5%, the stabilized loan to value is 64% and loans are floating-based payments. But no matter the loan to value, such loan balances are surely keeping Granite Point's management awake at night.
I don't own nor am I considering owning any of the stock listed above. And unless commercial real estate mortgages completely freeze, which is when opportunities arise, I will not buy a single share in a mortgage REIT.
To me, these firms are highly leveraged, with little margin of safety, and often, the investor in these entities does not know the true health of the firm until it is too late.
Mortgage REITS are no different than your typical commercial bank mortgages. The bank's goal is to earn income based on the difference of the cost of funds and that cost of funds that you borrow the money. But while banks have a depository base that they use to finance their operations, mortgage REITs solely rely on capital market. That is much concentrated risk.
My purpose is this case study is twofold. First, to demonstrate that there are details that the reported financials miss; for example, that Arbor has a second business in originating and selling loans to government agencies. Or that because of Hannon Armstrong equity method accounting, its assets are under reported. And that these "hidden" details may lure an investor at a first glance, but may result in a different view of the company at a second look.
The second purpose is to provide a quick overview of mortgage REITs, because if you own passive investment vehicles, an exchange traded fund that tracks the performance of the Nareit index for example, then you are an explicit investor in these REITS.
Out of a total of 225 REIT companies, there are 38 mortgage-related REITS. There are 22 residential mortgage REITs and 16 commercial mortgage REITS. Their size ranges from as little as $35 million in market capitalization (Tremont Mortgage Trust) to as large as $12.6 billion (Annaly Capital Management.) The median market capitalization rate is $1 billion (included in the market civilization are firms such as Armour Residential REIT, PennyMac Mortgage Investment and New York Mortgage Trust.)
This concludes the guide. The following section includes a glossary of real estate terms, a reference section with links to the best books I have read about REITs and current research on REITs.