When first we heard that Charter Financial Corp (NASDAQ:CHFN), the 80% owned subsidiary of mutual thrift holding company Charter Financial, MHC ("MHC"), was initiating a share-buyback and de-listing strategy that would essentially undo its 2001 IPO, we were curious. Under the buyback, shareholders will have the opportunity to tender at a price not less than $43.00 per share, nor more than $52.00, for up to 1 million of the 3.9 million shares currently outstanding in public hands.

But the more we look, the less the CHFN buyback�transaction makes sense.� Since reporting a special cash dividend of $1.75 per share back in August, CHFN has run from the low $30s to the mid-$50s, culminating in the subject share repurchase announcement. At Monday's closing price of $51.75, that is 4x book, 70x trailing earnings, and 5x the 2001 IPO valuation of $10 per share. So what's wrong with this picture?

Despite the apparent premium equity market valuation, the 100% owned bank sub of CHFN, Charter Bank of West Point, GA, is a poor financial performer -- and has been for some while.� In Q2 2006, Charter Bank's ROA was a less that stellar 0.71% and that is the best level achieved since the 2001 IPO.� The bank also hit its best ever ROE in Q2 2006,�a whopping 4.6%, according to the IRA Bank Monitor, using data from the FDIC.� The UBPR peer group for Charter Bank reported ROA and ROE of 1.52% and 13.7%, respectively, in Q2 2006.

Part of the reason for Charter Bank's pedestrian financial performance may be size and asset mix. A billion dollars in assets is too small to even partly leverage the $245 million in tangible equity on CHFN's balance sheet.� Of interest, Charter Bank's asset allocation mix is so conservative that The IRA Bank Monitor Basel II model generates a ratio of Economic Capital to Tier One Capital of just 0.038:1.� Who says that Basel II is bad for small banks?

Since the IPO, Charter Bank's loan default experience has fluctuated between a "AA" and "BBB" bond equivalent rating based upon industry breakpoints and calculations by the IRA Bank Monitor.� Charter Bank's Loss Given Default rate has been volatile and, most recently, above peer at 85.6% or an average of 85.6 cents of loss per $1 of default in Q2 2006.�� More than 94% of Charter Bank's loan book is in real estate, which sported a 90% LGD in Q2 2006, a level that has been consistent several years back.�

With only 30% of its assets funded off deposits and the same proportion supported by FHLB advances, Charter Bank is unable to finance existing assets via its nine branches. Our question: With loans equal to just 39% of total assets at June 30, 2006, why doesn't Charter Bank simply sell some of the MBS, repay the FHLB advances and focus on growing the banking business organically -- instead of running a fixed-income hedge fund?� But given the gruesome 90%�LGD for the real estate book, maybe holding MBS is a better choice for Charter Bank!

By opening�additional branches�in attractive markets like Georgia and Alabama, in theory there should be great opportunities for�CHFN to generate sustainable growth and profit.� However, both have so far eluded CHFN management, which instead of using its capital pile to fund growth�seems content to shovel money into the furnace via special dividends and value-destroying share repurchase schemes.�

Another interesting aspect of the CHFN balance sheet is the fact that more than half of its $12 per share in book value of equity is accounted for by one investment: �4.4 million shares of Freddie Mac (NYSE:FRE). Indeed, CHFN's management brags about the fact that it has parked a large chunk of its capital in shares of FRE.

Charter Bank's performance has been so poor that the parent of CHFN, MHC, waived taking dividends, allowing CHFN to pay out 62% of earnings to public shareholders in the past year. This selfless policy may make CHFN a more attractive investment for public shareholders, but certainly is no deal for the parent holding company MHC or its mutual stakeholders.

Here's the question raised by all of this data:� Why�is CHFN�management�willing to pay between 4x and 5x the 2001 IPO valuation to escape the benefits of public ownership?� Buying back shares at this time seems to make little sense and does little�to solve basic performance and�business model issues.��Let's not forget that� in August CHFN declared a $6.8 million special dividend�applicable to�public shareholders before announcing the share repurchase program.

More to the point, paying $42 per share or more for paper�sold at $10 five years ago seems like a bad way to start the process of improving financial performance. Worst of all, CHFN's assumption that a million share repurchase will get them below the magic 300 shareholder level necessary to de-list is not a given.�

In view of�the bank's weak historical financial performance, holders of CHFN who can participate in the tender�at $42 or more should do so, in our opinion.� Tis the season for giving and the management of CHFN seems to be in a very generous mood.

Questions? Comments? [email protected]