Description

Disclaimer:  This is intended for information purposes only (not investment advice) and should not be relied upon as a basis for investment.  The author holds a position in the issuer and undertakes no obligation to update any future changes in the position or in the investment opinions expressed herein.

This writeup is for people who look for baby with bathwater type of banks investments.

Axos was written once on VIC by huqiu and 3 times under its previous name BOFI.  They provide a great background for the company.  I prefer to get this out early and fast than later and longer, but happy to answer anything in the comments.

Summary:

Axos been tied with the regional CA banks and is down 30% even though it’s been having a record year, growing a diverse deposits base, no HTM risks and a tiny AFS securities portfolio.  In the last decade, there were probably 4-5 months the company could have had at the current valuation (and the 4-5 months were all concentrated around the COVID scare), even though it’s a high quality bank with low charge offs and high ROE.

Overview:

The company is an online only bank, highly efficient (mid / high 40s efficiency rate) with a high ROE (15-20% range).  On both the asset side as well as the liabilities side, Axos curved niches in verticals it can compete on service rather than on price.

Let’s start with the asset side.

The company is focused on CRE loans and jumbo mortgages.  Axos lends at low LTVs (58%) and as a result has a long-term track record of low single-digit NCO.  Also, their loans won’t be the cheapest, but they can underwrite fast and can do “complicated” situations.  Thanks to this they get a relatively high yield on the loans (6.6% at the time of the last report).

In addition, most of the loans are either variable or short-duration:

As can be see above, the company had only $4B of loans that are fixed rate and that mature in 3+ years (this represents only 13% of the loans). 

Next thing to look at is the securities portfolio.  On a 03/12/2023 report the company announced that it had only $248M of securities, all of which are AFS with unrealized losses of $7M.  Based on the recent Q, the company had 5.6% yield on these securities, which is again consistent with a recently acquired level of return on securities (as opposed to say a 2% yield from a 2 years old securities portfolio).

Next let’s look at the liability side.  88% of the deposits are covered by the $250,000 FDIC insurance.  Additionally, the company has an Insured Cash Sweep program that covers up to $150M in deposits per depositor.

The deposits tend to be relatively sticky, low-costs (ranging 2.2%-2.6%) and spread across consumer, commercial and broker.  The broker deposits mostly came from the E*TRADE Advisor Services acquisition, which provided ~$2B of deposits.  It is expected that these will normalize to lower % of AUM:

Even at a normalized level, these balances provide a low-cost funding source for the bank.  Currently, the bank has about $700M of these brokered deposits deposited with other banks, but they can be used for growth if / when needed.

If the economy deteriorates Axos could have a unique source of deposits.  In 2018 the company acquired Epiq, the largest software company for handling ch. 7 bankruptcies.  The program requires an escrow account with Axos and will provide counter cyclical (and cheap) deposit base if the economy takes a nosedive.

Finally, the company has only $260M of FHLB advances (less than 2% of loans).  This could be dramatically increased if needed to stem a run on the bank.

Valuation:

In the most recent quarter, the company earned $1.35.  There was nothing one-time in the numbers, so annualizing this number gets us to an EPS of $5.4 and a PE x6.  This is less than half of the mean PE of 15x of the last decade.

Based on the company’s guidance for mid-teens loan growth, EPS run rate will be close to $5.6-$5.8 by FQ4 (calendar Q2). 

From here, there’s a lot of upside in a distress environment, as the company will have cheap deposits and capital unencumbered by securities losses.

Finally, the company likely has a March end BVPS of ~$31 and a market price of $34, so the PB is 1.1x.

This would be one of the cheapest it had been in the past decade and exactly half of the last decade's median PB of 2.2x.

Assuming a fiscal 2025 BVPS of $45-$50 and 1.5x-2x PB (the high end of the range would still be 10% below last decade’s median) would get us a $68-$100, or 2-3x from today’s close price.  These prices will likely represent 10-15x earnings for 2025, which isn’t demanding given the quality of the business and the fact that the mean PE of the past decade was 15x.

Risks:

1.       It’s a bank.  Anytime you lever 10x there are risks.

2.       Jumbo mortgages are 24% of loans and are concentrated in CA.  A flight of rich people from CA would hurt the company.  This is offset by the low LTV (57% across the single-family portfolio).

3.       Bank run.  I think that this is a lower risk than most banks, because of the diverse source of deposits, 88% under $250K, additional coverage of up to $150M and FHLB advances of ~1-2% of loans (which provides a lot of firepower to stem a run).
Finally, on the report linked above, the company reported a GROWTH in deposits year to March 12.
Anyway, this is not a risk one can ignore today. 

4.       Increase in regulation to the financial industry following March’s events.

Disclaimer:  This is intended for information purposes only (not investment advice) and should not be relied upon as a basis for investment.  The author holds a position in the issuer and undertakes no obligation to update any future changes in the position or in the investment opinions expressed herein.

Catalyst

A stop to the non-discriminate decline in bank stocks regardless of their deposit sources / securities holdings