Accounting Form vs. Economic Substance in Banking

As I observed in an earlier post, large banks tended to respond to rising interest rates not by hedging or rebalancing their portfolios but by classifying more securities at held-to-maturity.  This is one of many instances over the years where banks (and often their regulators) chose accounting form over economic substance.  Supervisory goodwill, trust preferred securities, and redemptions in kind are three examples that come to mind.

Goodwill

In 1982, the Federal Home Loan Bank Board authorized some regulatory accounting practices that deviated from GAAP.   The FHLBB looked to avoid a wave of failures by propping up reported capital levels.  One practice was appraised equity capital, which allowed banks to write up the value of their office buildings to their appraised value rather than merely carrying them at depreciated historical cost.  Mark to market was applied selectively, however, in that you could write up your office building but did not need to write down your underwater mortgages or investments.

At least appraised equity capital indicated tangible value for some of a thrift’s assets.  The same cannot be said of another practice – deferred loan losses.  When rates shot up in the early 1980s, many thrifts were stuck with long-term, below market mortgages and MBS.  To encourage portfolio restructuring, thrifts could sell mortgage-related assets at a loss but amortize that loss over the remaining contractual life of the asset.  The unamortized amount, called deferred loan losses, was treated as an asset, even though the money was already out the door.

GAAP recognizes the concept of goodwill, but goodwill also represents a huge disparity between accounting form and economic substance.  Goodwill is essentially an accounting plug that equals the difference between what you paid for an acquisition and the fair value of the underlying assets and liabilities.  It has no tangible value and even its implied value presumes that the acquisition brings long-term benefits to the acquiring bank.  The Bank Board further sweetened the pot by allowing acquirers to accrete the purchase accounting discount faster than they needed to amortize the goodwill.  That resulted in a net gain on day one.  Including goodwill in capital made little sense since it provided no real protection against losses. 

The Competitive Equality Banking Act of 1987 got rid of appraised equity capital and deferred loan losses.  Goodwill remained and proved more difficult and costlier to remove from regulatory capital.  The Financial Institutions Reform, Recovery, and Enforcement Act of 1989 excluded from capital goodwill arising from new acquisitions and phased out existing supervisory goodwill over a five-year period.  That was hardly the end of the story, however.

In United States v. Winstar, the Supreme Court ruled that the goodwill provisions in FIRREA represented a breach of contract by the government and the affected institutions could sue for damages.  The court rejected the government’s “unmistakability defense.”  In other words, the acquisitions of the failed bank need not include contract language that protected the acquirer from future legislative changes.  In fact, the Bank Board was not authorized to make such guarantees.  Winstar set off a wave of more than 100 lawsuits.  The US Court of Federal Claims was sympathetic to the plaintiffs and awarded some eye-popping amounts in damages.  The awards were typically reduced on appeal.

Many of my fellow regulators sympathized with the Winstar plaintiffs, but I don’t see it.  I could see some sympathy for those that the government shut down, but in many of the plaintiff banks remained open.  They just couldn’t grow as fast as they wished because … they didn’t have enough capital.  Also consider what this says about the acquirers.  They apparently lacked such a fundamental understanding of finance to believe that goodwill is a real asset just because regulators said so.  Or they were attracted to the moral hazard that allowed them to leverage like crazy knowing the taxpayers would end up holding the bag.

Trust Preferred Securities

Then we have trust preferred securities.  The regulatory treatment of trust preferred securities (TruPS) allowed banks to have it both ways.  TruPS have both debt and equity features.  Like debt, TruPS were payable at maturity (usually 30 years) and dividend payments were tax-deductible.  However, starting in 1996, the Federal Reserve decided to treat TruPs as Tier 1 holding company capital, if it had certain features, such as allowing the issuer to defer interest payments for up to five years.  The Federal Reserve implemented the policy with a one-page press release.  I kid you not.  At the time Alan Greenspan still enjoyed his reputation as a financial maestro, so the Fed could get away with a lot of dubious moves.

TruPS look a lot like subordinated debt, which only counts as Tier 2 capital.  In fact, bank capital rules always treated TruPS as Tier 2.  This was a distinction without much of a difference, however.  The holding company could downstream the TruPS proceeds to the bank as paid in capital, which qualified as Tier 1.

Bank and thrift regulations generally limit investments to debt securities.  But banks could also purchase investment grade TruPS if they met certain other requirements.  Guidance issued in 1998 limited qualifying purchases to publicly traded securities due to liquidity concerns.  Subsequent guidance opened the door for some private placements under SEC Rule 144A.  (Spoiler alert:  the securities still weren’t very liquid.)  Regulators also prohibited reciprocal purchases of TruPS.  For example, Bank A and Bank B couldn’t work out a deal to purchase each other’s TruPS.  That’s like farmers agreeing to steal each other’s sheep.

While TruPS issuers initially consisted of large financial institutions, pooled TruPS opened the issuance market to a wider range of banks.  The miracle of diversification (more of a thing prior to 2008), meant that most pooled TruPS could receive investment grade ratings.  They still weren’t very liquid but qualified under 144A.  Pooled TruPS were notoriously opaque.  Basic information, like a list of the underlying bank issuers, was as closely guarded as the Colonel’s Secret Recipe.  After the pooled TruPS market fell apart, banks and placement agents became more willing to share with regulators.  In at least one instance, one of the pooled TruPS a bank had purchased included its own issuance.  In that case, the bank was stealing its own sheep.

Section 171 of the Dodd-Frank Act excludes TruPS from BHC Tier 1 capital.  Regulators issued a Final Rule, effective January 1, 2014, implementing Section 171.  However, the Final Rule includes generous grandfathering provisions.  BHCs under $15 billion can continue to include indefinitely TRUPS issued prior to 2010 in Tier 1.  The language in the preamble to the Final Rule is revealing:

Although the agencies continue to believe that TruPS do not absorb losses sufficiently to be included in tier 1 capital as a general matter, the agencies are also sensitive to the difficulties community banking organizations often face when issuing new capital instruments and are aware of the importance their capacity to lend can play in local economies.

In other words, we know that including TruPS in Tier 1 capital doesn’t make a lot of sense.  But the smaller banks complained so we’ll keep it in anyway.  Regulators may have also had Winstar in mind.

Redemptions-in-Kind

A little known but fascinating instance of form triumphing over substance involved mortgage related mutual funds.  Going back to the 1990s, some smaller thrift institutions invested in mortgage-related mutual funds.  These funds invested in short duration securities, so helped institutions satisfy Qualified Thrift Lender requirements while also controlling their interest rate risk.  Many of these thrifts lacked internal expertise and essentially delegated investment decision making to the fund manager.  The funds offered lackluster but consistent returns.  Starting in the 2000s, the fund managers juiced up yields by investing in private label CMOs, including mezzanine tranches.  Those tranches initially carried investment grade ratings.

When the mortgage market collapsed, the fund’s performance collapsed as well.  Since these were equity investments, thrifts needed to reflect the decline in net asset value in their regulatory capital.  Some investors decided to cut their losses and redemption requests soared.  Facing a liquidity crunch, the fund invoked a redemption in kind (RIK) provision.  The fund would limit cash redemptions to a de minimis amount per quarter, but investors could still redeem in kind by exchanging their investment in the fund with a pro rata share of the underlying securities.

Investors didn’t have a lot of good options, but choosing RIK was the worst possible decision.  Some thought RIK would stop the bleeding, but it made things worse.  They owned the same underlying securities, but in small, hard-to-sell odd lots.  They still faced onerous capital treatment (in some cases a 1250% risk weight) and impairment charges.  Patience would have made more sense.  Some banks steadily decreased their exposure by taking the maximum allowable cash redemption each quarter.  When the market finally began to recover in 2010, the fund resumed cash redemptions.

I’m aware of at least one bank failure resulting from an RIK decision.  A tiny (<$20 million) thrift invested a huge portion of its assets in mortgage mutual funds.  Bank management made a bad decision worse by opting to redeem in kind.  Impairments and 1250% risk weights caused the bank to bleed into insolvency, with regulators shutting it down in early 2012.  The Failed Bank Review by the Treasury’s Office of Inspector General gave high level summary of what went wrong.  However, the OIG decided against a more in-depth review due to the bank’s small size.  That’s unfortunate.  That bank’s failure vividly illustrates the dangers of concentration risk, asset illiquidity, and the pitfalls of prioritizing form over substance.