We have been working on additional analysis as a follow-up to our initial article on Medallion Financial (NASDAQ:TAXI) published last week. We are a generalist value investor, not a banking analyst. Our goal is to remove layers of complexity associated with financial reporting of banking/financial services entities and to focus on underlying asset value and earnings power potentially obfuscated by GAAP accounting.
Why is it difficult to see TAXI's underlying dividend power? Mainly due to complexities related to parent GAAP accounting - 1) GAAP EPS and 2) unconsolidated Medallion Bank
One of key pushbacks to our short thesis is TAXI's high dividend yield of ~12% today. At first glance, TAXI's dividend is being supported by relatively robust "net increase in net assets resulting from operations" (equivalent to GAAP EPS). As you can see in the parent's (or holdco's) income statement below, GAAP EPS has comfortably exceeded dividend per share ("DPS") every year in 2012-15 with a reasonable payout ratio in the 70-84% range.
Parent (Holdco) income statement:
However, GAAP EPS is not a good reflection of TAXI's earnings power as it includes sizeable non-cash items such as net realized/unrealized gains/losses. For example, GAAP EPS in 2015 included ~$16m non-cash benefit from "write-up" of the book value of Medallion Bank ("MB"). In our view, a better metric for a lender's earnings power is pre-tax income (essentially net interest income from loans less operating expenses). TAXI as a RIC (Regulated Investment Company) is exempt from corporate taxation as long as it distributes to its shareholders at least 90% of its investment company taxable income. Thus, currently pre-tax income and net income are the same (obviously, TAXI's ability to maintain its status as a RIC is critical).
As shown in the holdco income statement above, pre-tax income per share has been significantly below DPS in each of the past 4 years. DPS/pre-tax income ratio has been alarmingly high at 145-197% during this period. For example, in 2015, pre-tax income per share was only 69c vs. DPS of $1.00 and GAAP EPS of $1.20.
Another complexity associated with the parent financial reporting is the accounting methodology of MB. While MB is a wholly-owned subsidiary of TAXI, MB is an off-balance sheet entity and is NOT consolidated with the holdco. Holdco financial statements would show dividend received from MB in the income statement and fair market equity value in the balance sheet (currently 11% premium to book equity value).
While dividend income from MB was only $18m of the holdco's $33m net interest income, MB accounts for almost of all of the holdco's net earnings because dividend income from MB is equivalent to net income (net of operating expenses and taxes) and should be compared to holdco pre-tax income (i.e. $18m dividend vs. $17m holdco pre-tax income, which implies holdco ex-MB contributed $2m of net loss in 2015).
It is also worth noting MB is a taxpaying entity, and as a non-RIC subsidiary of TAXI, is allowed to retain all earnings in the business to fund future growth. Thus, it is necessary to analyze MB's standalone financials (thankfully disclosed separately by parent) separate from the holdco stub (defined as holdco ex-MB).
Our framework: TAXI earnings power = holdco stub + MB standalone
Let's try to figure out TAXI's true earnings power based on 2015 reported financials. Holdco stub earning is simply determined by subtracting dividend income from controlled subsidiaries (which is almost entirely MB) from consolidated holdco pre-tax income.
Holdco stub earnings:
Here is the income statement for standalone MB from TAXI's 10-K.
MB income statement:
Now underlying dividend power of TAXI can be calculated as follows. Since MB is NOT required to distribute all of its net earnings, we are showing dividend from MB at various payout ratios to give you a range of possible dividends for TAXI shareholders. Even at a 100% payout ratio at MB, the maximum DPS for TAXI shareholders is only 88c. However, 100% payout ratio would not be prudent because 1) MB will not have any additional capital to grow and make new loans, 2) MB's tier 1 leverage capital ratio of 15.1% is barely above the regulatory minimum of 15% ordered by FDIC.
TAXI underlying dividend range:
What does this mean for the fair value of TAXI shares? We see substantial downside even under the best-case scenario
Let's consider the scenario of 100% payout ratio at MB. Under the scenario, unless TAXI stock trades at premium to book value (extremely unlikely given overstated book value), TAXI will not have access to additional capital to grow its loan book at both holdco and MB. What is the appropriate equity yield for such a static portfolio? We have believed for some time that at least double-digit equity yield is warranted to compensate for all the risks associated with the current earnings.
Last week, TAXI provided us with a latest valuation benchmark by issuing a new 5-year bond with a 9% coupon. Obviously, equity yield will need to be meaningfully above 9%. By the way, no bank with a strong capital position and a well-performing loan book would issue at such a high rate! If you are long TAXI and have to maintain long exposure, it makes no sense to own the stock instead of owning a bond, which is senior to equity with ~$200m of loss cushion (= current TAXI market cap).
Moreover, as mentioned in the first article, we believe the current level of allowance is highly inadequate given our view on the medallion value and recent delinquency trends. TAXI's earnings (holdco + MB) will almost certainly be significantly lower than 2015 for the foreseeable future if TAXI's balance sheet is properly marked and provisioned. At 80-100% payout ratio for MB and 13-15% dividend yield, TAXI shares will be worth only $4.5-6.5/share, or 20-45% below the current price of $8.4 as shown in the table below.
Unfortunately, this represents the best possible outcome for current TAXI shareholders. When we think about a real downside case for the stock, the dividend analysis below becomes irrelevant. As the medallion loan portfolio deteriorates further, investors will likely begin to focus on book value and capital adequacy, and will be eventually concerned about rising risk of dividend cut and/or dilutive capital raise. TAXI stock appears significantly overvalued due to misunderstanding about the underlying book value and sustainable dividend power. Our dividend analysis further strengthens our conviction in the short thesis as the stock is dramatically overvalued even under the best-case scenario.
Disclosure: I am/we are short TAXI.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.