TFS Financial Corp. (TFSL) - $9.50
I do not know when regulators will allow TFSL to resume buying back shares. But I do know that TFSL has met or exceeded all the stipulations of the regulators, and that this has been the case for three quarters now. Meanwhile, the Company’s shares trade at 44% of current, and 79% of my estimate of fully-converted (“second-step”) book value. The buyback math under book value is obviously quite compelling, as is the share price impact pro forma for the return of a steady purchaser of shares, as is the potential of resumed dividends.
TFSL trades at approximately $9.50, while its current book value is $21.50 (SE/shares issued). Fully converted (this is a Mutual Holding Company - MHC), I estimate book value per share at ~$12.00. By the complicated but interesting math of MHCs, management has incentivize to buy back as much existing and issued stock as possible. They started doing so after going public. Then, in June 2010, the OTS issued an MOU citing, among several items, that the level of home equity line of credit loans was too high. TFSL was (and is still) prohibited from buying back stock and/or paying dividends until A) the MOU’s conditions are met and B) the OCC (OTS was merged into the OCC) actually lifts the MOU. TFSL management asserts that A) has occurred; now it is a waiting game for regulatory approvals B).
TFS Financial (TFSL) is an Ohio-based (Cleveland) thrift which went partially public in 2007 in a “first step conversion.” Like many sleepy MHCs, TFSL didn’t loan aggressively in the halcyon years of stupid lending and, as a result, didn’t implode subsequently. In fact, management bought back shares aggressively during this period ($186 million in 2008 and $103 million in 2009). Loan loss provisions peaked in fiscal 2009 (September) at $115 million against $9.2 billion in loans. Today total past due and nonaccrual loans stand at $307 million out of $10 billion in total loans (3%). Loan loss reserves are $100 million and shareholder equity is $1.8 billion (16% of total assets). Overall asset quality has been improving steadily, while pre-tax pre-provision earnings are on the order of $100 million annually.
Today there are 308.9 million total shares outstanding, of which 81.8 million trade publicly as what I call “minority shares.” The remaining 227.1 million shares stand in reserve as yet to be issued to depositors upon a second step conversion. These shares would not be issued for free, however. Depending on the conversion valuation (historically ranging from 60% to 100% of fully converted book value), a second step conversion would yield $1.3 to $4.4 billion in new proceeds. This is where financial data warehouses get BV/share wrong (stated by FactSet, for example, at $5.74 per share) because the additional capital received for issuance of shares in reserve is not calculated, though the full share count is calculated.
By any measure, TFSL is overcapitalized today. Here are some statistics from their December 2011 10Q:
-Total Capital to Risk Weighted Assets is 22%, implying $885 million in excess of the required 10%;
-Core Capital to Adjusted Tangible Assets is 14%, implying $967 million in excess of the required 5%;
-Tier 1 Capital to Risk-Weighted Assets is 21%, implying $1.08 billion in excess of the required 6%.
-TFSL’s market cap is $779 million on current shares actually issued. At this valuation and given this egregiously overcapitalized state, the company could theoretically buy back all but one share and divided, say, $35 million in anticipated 2012 earnings to it. Though an impractical scenario, the company was an aggressive buyer of its shares pre-MOU, and will be so again when the MOU is lifted and when they are granted authority to resume buybacks.
Why buy back shares:
1) Asymmetric dividend payments to minority holders: Some MHCs, including TFSL, can elect to pay dividends solely to the minority shares outstanding. This policy was allowed by the now-merged-away OTS and was since grandfathered (see TFSL 9/30/11 10K page 49 under “Waivers of Dividends by Third Federal, MHC”). However, there is pending language in Dodd-Frank whereby there may be a requirement “that a majority of the mutual holding company’s members eligible to vote must approve a dividend wavier…” This means there is a threat that an asymmetric dividend cannot be practically paid. We leave this to the regulators. This may have teeth or it may be a sideshow. Management states in the above-referenced 10K (same paragraph): “…While Third Federal Savings, MHC believes that the FRB does not have the authority to require a member vote to approve mutual holding company dividend waivers, there can be no assurance that the final rule will not require such a member vote…” This is a nice option looking ahead, but not imperative to this thesis.
2) Buybacks increase current and fully converted BV: Even if an asymmetric dividend is regulated away by a required vote of the “members,” buybacks still inure to the benefit of minority shareholders as there is a fixed number of shares in reserve for issuance to depositors. Therefore, every minority share purchased increases the percentage owned by minority shareholders. If the company were to buy back, say, 25% of the minority shares outstanding at $11, pro forma book value pre-second-step goes from $21.50 to $25.25 and, post-second-step from $12.00 to $12.40 (using my calculations at a second step done at 80% of fully converted book value).
The second step math is beyond the scope of this writeup, as management has no stated plans to fully convert. The existence of the second-step option, however, would make TFSL more over-capitalized and would bring more firepower (as if they need some) for subsequent buybacks. Suffice to say that TFSL currently trades at a valuation that implies a second step conversion at 55% of fully diluted pro forma book value, while 80% (or above) would be a more realistic target, based primarily on comparable completed second step conversions. Whether you believe my 80% figure or not, the current valuation implying the 55% is plain silly.
State of the MOU
Management states in the Dec ’11 10Q, that all the conditions of the OTS’/OCC’s MOU have been addressed and met or exceeded: “…As indicated above, we believe that to date, we have complied with all of the stipulations of the MOU and New MOU…” (page 28). They have had this language in their docs for several quarters now. I have no idea when the MOU will be lifted or what any kind of timing is involved, or whether or not there will be additional stipulations or delays due to the merger between the OTS and OCC. I am satisfied to know that TFSL has played ball with the regulators and met or exceeded their requests, and that the ball is now in the regulators’ court. TFSL management has asserted that this delay is due to the OTS/OCC merger.
There is no clear timeline currently. At some point the following happens: The regulators meet with the company formally to assess TFSL’s compliance with the MOU, then it will take a month for sign off, then there will be a waiting period (maybe 45 days) for TFSL to apply for and be granted the ability to resume buybacks.
Once this timeline is known, the wayward share price pattern, and perhaps the current opportunity, ceases.
In conclusion it is possible TFSL could pay out hefty dividends and/or run its shares to mid-teen levels through the resumed buyback process, or both. The current share price is at a very attractive level given the upcoming catalysts pertaining to the lifting of the MOU and given the current and pro forma BV/share figures. Where should TFSL trade on a normalized basis? Pre-MOU, for example, shares traded (mind you, this was in a more fearful bank environment) between $12 and $14 per share, levels at which the company was repurchasing shares. Second, similar fully-converted thrifts trade near book value, which is consistent with TFSL’s trading level pre-MOU.