Flagstar (FLG): Still Somewhere Between A Dumpster Fire & A Discarded Gem

I still like betting on the ugly ones. The names everyone swears are finished, the ones that make people laugh when you say them out loud. FLG is still living large in that bucket.

A year ago it was the poster child for bad credit, worse management, and New York rent laws that could crush a balance sheet. Now, quarter by quarter, the same “dumpster fire” is starting to flicker back to life. CET1 is 12.5 %, expenses are down 30%, NIM is rising for the third straight quarter to 1.91%, and the losses are shrinking fast. Credit isn’t clean yet at all, non-accruals are 5.17 %, reserves a thin 1.8 %, but the trend is no longer collapse, it’s repair. The market still hates it, which is exactly why I own it. The integrity check I ran in August holds: Otting said they’d cut fat, build capital, and stabilize credit, and they did.

But I also like beating my hypotheses to the gound so here’s a “bear take”.

Non-accrual loans climbed to 5.17% of total loans, up from 4.96% last quarter and 3.54% a year ago - still not great. The mix is predictable: about $2.44 billion of multi-family loans are non-accruing, another $551 million in commercial real estate, $154 million in C&I, and a smattering elsewhere. When you have $3.2 billion of non-performers on a $62.7 billion loan book, so despite my bullishness the math says there’s still a lot of skeletons in the closet.

For me, the allowance math is the tell that we’re not out of the woods. The total allowance for credit losses is $1.13 billion, 1.80% of total loans. That covers only 35% of non-performers, down from 53% last year. Management points to a 3.05% reserve on rent-regulated multi-family loans and 1.83% on the broader multi-family book. Those numbers look conservative in isolation, but they’re small when you consider how fast property-level cash flow has eroded and the fact that New Yorkers are about to elect a rent freezing communist to lead them. The provision dropped to $38 million from $64 million last quarter and $242 million a year ago, while the non-performer ratio kept rising. So why would you be trimming reserves as you have technically worsening credit? This kind of smells like goal seeking to an EPS number.

Charge-offs improved on paper, net charge-offs fell to $73 million, which is 0.46% of average loans on an annualized basis, down from 0.72% last quarter and 1.31% a year ago. Definitionally positive. But $46 million of that still came from multi-family loans, $18 million from CRE, $1 million from C&I. Year-to-date charge-offs sit at $305 million, nearly 0.63% of average loans. Delinquencies aged 30–89 days fell to $535 million from $604 million last quarter, but CRE past-dues actually ticked up 2%. I wrote how you have to trust Otting’s IndyMac experience here, and largely this side of it proves it out. They did mark most of the bad stuff appropriately.

Balance sheet wise, assets slipped to $91.7 billion, down 1% from June and 8% since December. Loans held for investment fell $1.5 billion (–2%) QoQ and $5.6 billion (–8%) YTD. Deposits dropped $600 million (–1%) in the quarter and $6.7 billion (–9%) for the year. They ran off $6.1 billion of brokered CDs with an average cost near 4.9%, which helped lower deposit costs by 13bps. That maneuver nudged net interest margin up 10 basis points to 1.91%, its third straight quarter of gains. Shrink to win is still intact.

Pre-provision net revenue (PPNR), the truest measure of earning power, came in at $15 million adjusted and the GAAP number was a $3 million loss. Efficiency sucks sitting at 100%, or 92% adjusted. It’s better than 115% a year ago, because operating expenses fell 30% year-over-year to $457 million adjusted. Again, Otting followed through on hacking costs and that cut was real, fewer employees, fewer branches, and no more mortgage servicing, yet it also marks the end of easy wins.

Capital ratios look strong on the surface. CET1 sits at 12.45%, Tier 1 at 13.25%, and TBVPS is $17.32. It’s trading at $11.85 and about 68% of TBV.

I listened to Otting’s remarks. He said they’re “on the path to profitability” like I expected him to. C&I balances did grow $448 million (+3%) this quarter after four straight declines, but that’s off a $14.4 billion base. Meanwhile, multi-family balances fell $1.5 billion (–5%) and CRE $473 million (–4%). The bullish case this quarter is simple: the direction of travel is still right. Losses are shrinking, NIM is climbing, and TBV is inching higher even as the bank sheds risk. NIM is almost above the 2% Mendoza line, credit remains mixed, but the core trend is constructive, criticized loans are down 19% year to date, and $1.3 billion of CRE loans paid off at par, with roughly 40% of those coming out of the substandard bucket.

Still bullish friends. Still convexity in this name and I think they sell this sooner than people think. Someone will want the NYC exposure, Mamdani or not.

The best is head (I hope),

Victaurs

Disclaimer:
This content is for informational and educational purposes only and does not constitute investment advice, an offer to sell, or a solicitation of an offer to buy any securities. The views expressed are solely those of the author and are based on publicly available information believed to be reliable at the time of writing, but no representation or warranty is made as to its accuracy, completeness, or timeliness. Any opinions or projections expressed herein are subject to change without notice. The author may have a financial interest in the securities mentioned. You should not rely on this content as the basis for any investment decision. All investments involve risk, including the risk of total loss. Please consult your own financial, legal, and tax advisors before making any investment decisions.