Betting Against Bank of Internet

Disclosure: I am short against Bank of Internet as of July 28, 2014 and at the time this post is published. You can download a zip folder containing some of BOFI’s public filings, two Kerrisdale PDF reports, and Citron’s report. Compared to my two previous public reports (Gamestop and Castlight Health), Bank of Internet is a more complex business. I’ve tried to keep the content interesting and education for financially literate and non-financially literate readers alike. Also, given the essay format, I’m only able to tell part of the story. The zip folder contains the material to answer any further questions one might have.

What Is Bank of Internet (BOFI) Holdings?

Bank of Internet is exactly as its name suggests: It’s an online-only bank. Individuals can make deposits online and the company uses those deposits to make investments. Started in 1999, BOFI has grown to over 300 employees and a market value of over $1 billion. It’s stock has rocketed upwards over 900% in the past five years. This tremendous run is about to end.

The BOFI Short Reports

Two short sellers have come out with public research reports against BOFI: Kerrisdale Capital (with a 36 page PDF and 37 slide Powerpoint presentation) and Citron Research (10 pages, shorter but more scathing).

Particularly scary for BOFI is that Kerrisdale Capital was once one of its biggest fans, bought the stock in 2009, held it as their largest investment for the firm, and published a positive report back when the stock was around $8 per share in 2009. Now that it’s at $79, Kerrisdale is calling it quits. They have a target price of $50 for BOFI stock and have flipped their position, betting it drops down to reasonable levels after having made huge sums of money buying the stock the past five years.

Citron is even more negative, with a $40 price target for reasons you’ll soon learn.

As a fan of Citron and Kerrisdale, I encourage you to read their reports. I’ll explain their key points against BOFI here, which should reveal the understandable problems with BOFI’s stock price. If you are worried about being bored by numbers, jump to section “Reason Five” which might intrigue you to read the rest (although I listed it last as its probably the least directly relevant to the company’s value).

Reason One: Earnings Driven by Mortgage-Backed Securities Are Ending

Page 16 of Kerrisdale’s report highlights the primary reason for BOFI’s huge growth and eventual decline: It’s assets are residential mortgage-backed securities (RMBS). For my less financially-inclined readers, RMBS are financial assets that represent pieces of residential mortgages.

Kerrisdale writes, “During the financial crisis, BOFI recognized that RMBS were undervalued and aggressively used depositors’ money to build up what was, by the end of FY2009, a $458 million portfolio, amounting to a whopping 35% of total assets.”

This investment by BOFI in cheap, post-financial-crisis mortgage assets was smart and has driven much of their profits in the past five years. However, these were opportunistic, non-recurring purchases. Once these assets, which are structured to last for a fixed amount of time, end or are rolled over into new assets with new terms, BOFI’s advantage versus its competition will be gone, as Kerrisdale demonstrates in this slide:

BOFI Advantage is Securities, Soon Ending

Kerrisdale continues, “Roughly 45% of the company’s securities book, equivalent to 15% of earning assets, is in non- agency RMBS with a high yield of ~5%. These securities are expected to decline by 15%-20% annually and could lead to a ~175 basis point decline in the overall securities yield. This would translate to a 25bps decline in the net interest margin over the next few years.”

And here is a demonstration of how the company’s overall earnings will be hurt by vanishing derivatives profits:

BOFI Profit Falls

Kerrisdale elaborates: “BOFI management, to its credit, recognized during the financial crisis that non-agency mortgage-backed securities were attractive investments. By putting almost a third of its balance sheet into these securities at low prices, it built up a store of future earnings that it has been gradually recognizing over time. As these assets continue to pay down, BOFI will have to reinvest at much lower yields, further depressing its NIM and reducing profitability. Moreover, in a post-Dodd-Frank regulatory environment, we question whether a bank of BOFI’s current size would ever again be allowed to make such an extreme gamble with its depositors’ money.“

Bank of Internet did make money from buying these mortgage-backed securities. Yet, as those investments expire, their money will have to be invested in new assets with different rates of return. Except….

Reason Two: Interest Rate Sensitivity Could Cause Cash Crunch

“BOFI has one of the largest negative interest-rate gaps among publicly traded banks. In other words, its assets reprice much more slowly than its liabilities. As rates increase, its funding will become dramatically more expensive, but its asset yields will stagnate. At a time when almost every high-profile bank has sacrificed short-term earnings to make its balance sheet “asset sensitive,” with assets repricing faster than liabilities and thus positively levered to higher rates, BOFI has made the opposite bet, pumping up its earnings today at the cost of returns tomorrow,” writes Kerrisdale on page four of their report.

An interest-rate gap is the difference in interest-rate sensitivity between assets and liabilities. In the case of a negative number, that means if interest rates go up, liabilities would increase faster than assets, so profits are likely to decline.

Bank of Internet’s interest-rate gap, by its own admission on page 57 of its March 2014 third quarter 10Q filing with the SEC, is -157%.

The not-so-subtle Citron explains, “Banks are not supposed to take on interest rate risk any more than a Vegas Sports Book is supposed to bet on one horse in a race. They are supposed to be structured so that most of their rate risk is hedged…. Most banks are
within 10% or 20% of zero. Any steep interest rate rise would obliterate earnings.“

BOFI’s liabilities are the deposits individuals and businesses keep at the bank (remember from personal finance 101 that you are a liability to the bank), and its assets are the loans it makes and the mortgage-backed securities it buys. The reason BOFI has such a dramatic negative interest-rate gap is that mortgage securities are long dated while deposits like savings and checking accounts can be moved around by the account holders (you and me) any time. I am admittedly not an expert on interest-rate risk management and how banks manage assets to maintain a zero interest-rate gap, but this table comparing BOFI’s rate gap to other online banks makes BOFI’s problem pretty clear.

BOFI Rate Gap

Given interest rates have been low for years, there’s not much room for rates to go lower and mostly talk of rates going higher. When the Federal Reserve chairwoman is debating whether rates will rise slowly or quickly and doesn’t even mention them declining further, that should scare the crap out of BOFI.

“Even a modest increase in interest rates, which is largely expected at this point, will result in a relatively significant reduction in BOFI’s earnings power due to the rapid increase in funding costs for its deposit base. Larger rate increases would put BOFI in a position where it could have capital shortfalls,” adds Kerrisdale.

In plain English, Kerrisdale is saying when interest rates rise, the short-term depositors will cost the company more money because the rates for their savings accounts will be higher. But because so much of BOFI’s money is tied up in mortgage loans and securities that it can’t get cash out of quickly, the company could be in a position where it’s tight on money.

Company management probably sees this coming since they announced two weeks ago they will be raising $50 million by selling more stock. Even if this works in protecting them from a cash shortfall, selling additional stock diminishes the value of current shareholders stocks and will probably push the stock price lower as the new shares hit the market.

One would think that if a company is systemically at risk of running out of short-term cash, even with a low probability, and is selling its own stock ownership stake for more money, it’d start getting conservative with the cash it does have. But if you’re Bank of Internet, conservatism is for pansies who don’t pay themselves enough. Why ensure the future stability of your business when you can double down on a new risky business venture….

Reason Three: Regulatory Risk in New Stored Value Card Business and H&R Block Bank Acquisition

In April, BOFI and H&R Block announced a deal for BOFI to buy H&R Block Bank (a subdivision of H&R Block) for $200-$250 million. In exchange, BOFI receives $450 to $550 million in customer deposits and prepaid card balances. Prepaid cards are generally things like gift cards or metropolitan subway cards where there is no account connected to the card. The money is tied directly to the card. However, there are businesses that issue these kinds of cards for other general purposes.

There is notable increased regulatory risk with the new prepaid card business BOFI is in the process of acquiring.

In June, The Bancorp’s stock was cut by 20% in a single day by a “Consent Order” issued to them by the Federal Deposit Insurance Corporation (FDIC) to improve their regulatory compliance (read the list of required company changes in Bancorp’s SEC filing here). The Bancorp is the USA’s largest issuer of prepaid or “stored value” cards.

Poor Bancorp

An article by Computer Services Incorporated explains why the government cares so much about the prepaid card business. It’s ripe for terrorist activity. Since there is no account linked to a prepaid card, these cards can be used as an alternative to cash, and used for illegal purposes with no name or bank attached.

This became apparent in 2010 when prepaid cards issued by Meta Financial (the second largest issuer of stored value cards in the country) were used by the suspects of the assassination of HAMAS commander Mahmoud al-Mabhouh. Meta Financial was quickly dealt regulatory fines which also cut their stock in half and required many years to recover.

Meta Financial 2010

The government has a slew of new rules (such as the Prepaid Access Rule, the Bank Secrecy Act, and Anti-Money Laundering Rules) banks must comply with to avoid fines and penalties.

This regulatory risk is Citron’s primary short thesis:

“Citron notes that we are not saying that BOFI either supports illicit activities or even that stored value cards do not offer useful service. What we are observing is this area of the business will soon get disrupted and stay disrupted for a long time.

BOFI added to their risk of regulation with their recent acquisition of H&R Block bank. Citron notes that The Street’s initial euphoria over the H&R Block deal has completely worn off – the stock rallied from 75 to 100, but has now retraced all the way back to where it was before the deal was announced. The reason? The Street wised up to the reality that nobody else wanted H&R Block’s bank because of the regulatory risk – principally, becoming a major player in the stored value card space.

Here’s the issue, according to the CFO of a huge credit union who didn’t want to be named in the article: “It’s very difficult for a bank in the Stored Value Card business to comply with the Bank Secrecy Act (BSA – the banking regulation on money laundering) because the bank needs to verify every person who buys a Stored Value card. For example, if they give that card to someone else there are possible compliance problems. Without an over-the-counter deposit and withdrawal system, tracking the stored value cards is difficult. The government is highly concerned with terrorism activity right now and Stored Value represents a high-growth potential problem that needs tight regulation.“

I’m just speculating, but it seems to me that Bank of Internet is most likely buying into the prepaid card business in an effort to boost top line revenues and its total depositor and asset numbers. It’s an attempt at keeping up its high growth while exposing its internal operations to painful compliance cost increases and/or regulatory fines.

Reason Four: Fundamentally Overvalued

Disregarding the potential increased regulatory costs, Bank of Internet is overvalued right now at its market price.

Kerrisdale has done a better valuation job than I can. They’ve taken two approaches to deciding what BOFI is really worth: First, they calculate what they think the company is reasonably worth. Second, they take the company’s $73 per share, over $1 billion valuation and figure out what the company would have to do to earn this valuation.

Kerrisdale finds, using its model, that BOFI is worth $48.34 per share.

Kerrisdale BOFI Valuation

So what would BOFI’s business have to look like for it to earn enough money to be worth $73 per share?

Market BOFI Valuation

Here are the remarkable conclusions Kerrisdale draws from this model:

  • BOFI will have to triple its deposits by 2021 from $2.4 billion to $7.5 billion. Kerrisdale notes that US GDP growth is only a couple percent annually, so most of these deposits would have to be taken from competition, not economic growth. This would have to be accomplished while competition is increasing.
  • BOFI must maintain Net Interest Margins at 3.5%. Due to BOFI’s negative interest rate gap, a rise in interest rates will destroy its margins.

For an easier side-by-side comparison based on simpler metrics like price-to-book (for the non-finance readers, book value basically being the value of a company’s assets minus its liabilities), Citron put together these handy tables:

BOFI Price to Book Overvalued

Reason Five: The CEO and CFO Previously Ran Two Financial Firms….Which Went Bankrupt

My personal favorite reason to be skeptical of Bank of Internet is that it’s top management has a pretty poor record of running financial institutions.

Chief Executive Officer Gregory Garrabants was previous job was Senior Vice President in charge of business development at IndyMac until October 2007. IndyMac was the seventh largest mortgage originator in the United States. That is, until the housing crisis began in 2007 when the stock collapsed, the company’s mortgages and mortgage-backed securities lost money, and IndyMac was put into government “conservatorship” in 2008, becoming the fifth largest bank failure in US history.

Chief Financial Officer Andrew Micheleti was Controller and Vice President of Finance of Imperial Savings. Imperial used depositor money in the 1980s to buy junk bonds from the now-defunct investment bank Drexel Burnham Lambert. Both Drexel and Imperial would not survive the 90s (Drexel shut down after its leader Michael Milken went to jail and Imperial was taken over by the Federal government before it lost all its depositors’ money).

Given management’s background and the mortgage-related investments previously explained, what Greg and Andrew have done with their careers is this: Learn as much as possible about the mortgage and mortgage securitization industry as possible, find conservative financial companies that are moseying along, come in as upper management aiming to juice profits with risky mortgage loans and derivatives bets, cash out (as evidenced by Garrabant’s $2 million in compensation just last year), and move on when the company crashes when the penalty for short term risks over long term sustainability comes calling.

It’s ingenious really: pump up a shaky market, pay yourself good money while it goes up, quit as it’s about to collapse, and come back after everything has crashed and buy back all the old crappy assets you sold to people at higher prices just a couple years prior on the cheap. Wash, rinse, repeat.

It’s possible for Greg and Andrew to have learned from their previous debacles how to run and grow a sustainable bank. However, given the derivatives on their balance sheet I’ve shown earlier, the damning narrative seems to fit.

How I Could Be Wrong

With any idea, it’s important to try and figure out how it could be wrong, and an easy way to do that is support the inverse of your arguments. Any of the following items could help Bank of Internet grow profitably and prove me wrong. For their sake, I hope so.

  • The OCC or FDIC Do Not Force New Compliance Costs: If for some reason the government does not take action against BOFI for its prepaid card business, then this new division could add to the profitability of the company.
  • Interest Rates Remain Low For Years: If interest rates remain low, then BOFI’s net interest margins will take longer to decline.
  • BOFI Improves Its Organic Growth: Of course, if BOFI improves its competitive position either through improved loan selection and underwriting, improved marketing/branding to consumers, or some new investments I’m unaware of that could replicate the success of its mortgage-backed securities portfolio, then it could outgrow its competitors. There is little sign of any of these things happening, but they’re not impossible either.

The Bottom Line

Given all the reasons why BOFI’s stock may take a hit, it seems unlikely they can avoid them all. Even if earnings and their deposit base grow, interest rates stay low, and their H&R Block Bank acquisition is purely beneficial, the company is still valued at twice the earnings multiple of its competitors, leaving buyers less room for potential upside. If any of these positive pillars holding the stock up collapse, investors will be inclined to reevaluate their position. If we learned anything from the financial crisis, it’s that our banks should not be treated as high-flying growth businesses. This is particularly true when the growth is from investing in the types of derivatives that caused the last financial crisis from executives who crashed companies in that crisis.