Fed Stress Tests, and One Interesting Bank

Every June, the U.S. Federal Reserve issues the Dodd Frank Act Stress Test results (found here), where they put critically important banks through recession simulators to ensure that they are resilient. Banks that pass the test can go on to boost their dividends and buybacks, while banks that come up short need to make changes first.

This year, no banks failed. One bank, Credit Suisse, has to submit a revised capital plan, which puts them in the gray area between passing and failing. A year ago, the Dodd-Frank Act was pared down a bit so that fewer banks are included in this 2019 report.

Their most severe simulation involves a hypothetical scenario where the official unemployment rate rises to 10%, real GDP contracts by 10%, average home prices fall by about 30%, the Dow loses about 2/3rds of its value in a crash while the VIX hits 70, 10-year treasury yields fall to under 1%, and BBB corporate bonds spike up to over 6%. Basically, it’s the same or slightly worse magnitude to the subprime mortgage crisis of 2007-2009, but applied to today’s economy where rates are already lower.

A shortcoming that some point out is that they strictly assume it’s a deflationary recession. That’s what we’re seeing throughout the developed world, where high debt and slowing growth are grinding inflation to low levels. It’s the most common type of recession in countries that control their own currencies (compared to emerging markets that often borrow in external currencies and thus encounter more severe inflationary recessions). The stress test does not consider the tail risk of stagflation, in a scenario where as the normal deflationary forces play out, fiscal or monetary policy to stimulate the economy and combat deflation could result in loss of confidence in currency or bonds, and result in deflation shifting towards inflation.

Loan Losses

Notable charts on expected losses from the stress test: 

There is a spectrum of volatility across loan types. Credit card and auto lending get slammed the hardest. Mortgages are safer, especially when there’s no housing bubble. Custodian banks have the least expected losses.

Since the number of tested banks was reduced this year to a narrow field of critical institutions, we don’t see some of the other credit card issuers. Last year’s report had results for Discover and American Express, for example, which had expected loss rates similar to Capital One.

The worst result I’ve ever seen was Synchrony Financial’s 2018 results, which weren’t included in the Fed’s 2018 report but were posted on Synchrony’s website. They had expected loan losses of over 20% during a recession as bad as the subprime crisis. After having done tests from 2015 through 2018, Synchrony doesn’t have to do one for 2019.

Even concentrated credit card issuers pass the test year after year with numbers like that, though. While their loan books are volatile during recessions, their return on equity is north of 20% during good times, which allows them to accumulate big reserves to withstand the bad times. So, even in scenarios where they lose 15-20% or more of their loan books, their capital ratios are sufficient. Of the concentrated credit card lenders, Capital One and Synchrony are the more aggressive ones with higher default rates while Discover and American Express are the more conservative ones (and held up reasonably well through the subprime crisis).

If we have a big sell-off at some point and credit card lender stocks drop significantly, Discover and American Express are ones I’ll be looking to buy. I already have a moderate position in Discover.

Basically, well-run credit card issuers are cases where the volatility is far worse than the risk of permanent capital loss, which can present buying opportunities in dark times. Opportunistic investors should pay attention to this space when times get tough and see how the fundamentals are doing compared to the stock price.

There are roughly three different types of bank loan risks rather than just one type of risk.

Credit Risk: The risk of experiencing asset losses. This is what most people focus on, and where well-capitalized custodian banks and mortgage banks hold up reasonably well, while credit card and investment banks get slammed.

Interest Rate Risk: Some banks are more focused on fee income (asset management, trading, insurance, etc) while others are more focused on interest income (the spread between longer-duration higher-rate asset yield assets and shorter-duration lower-rate liability yield liabilities). During a declining rate scenario, banks that are focused on the interest rate spread are more vulnerable. Mortgage banks tend to be a lot more sensitive here than credit card companies, because credit card companies have a very wide spread between yields of assets and liabilities.

Stagflation Risk: During a rare event in which interest yields unexpectedly spike, most banks would suffer, but banks that focus on fixed longer-duration assets, such as mortgages, can run into particularly severe trouble. Banks with variable rate lending activities (including credit card issuers), or that focus on fee income can hold up better. 

The Little Regional Bank that Could

I recently did a deep dive into BancFirst, a regional bank in Oklahoma with a market cap of under $2 billion, and found how much of a gem it has been.

Over the past 20 years, in addition to outperforming the S&P 500, they have outperformed the Big 4 banks, as well as the broad financial sector ETF:

Additionally, they have outperformed the SPDR Regional Bank ETF (KRE) since the ETF’s inception in 2006:

In fact, they have outperformed the regional bank ETF not only since inception, but also over the past 10-year, 5-year, 3-year, and 1-year periods. Well played.

Looking into management history, I’m not surprised that this has been the case. Insiders, especially executive chairman David Rainbolt, own a big percentage of the company worth hundreds of millions of dollars.

The company was founded by Gene Rainbolt. His son, David Rainbolt, was CEO from 1992 to 2017, at which point he retired and became executive chairman. Currently in his early/mid 60’s, he still has an active role in the company and has huge ownership of it. David Harlow has been the CEO since 2017, and he has been with the bank since the 1990’s.

Back in May, I wrote about how to find companies with good management. Executives that have big insider ownership, that serve long tenures, and that have a long history of strong capital allocation, generally continue doing well. While external pressures can weigh on BancFirst, shareholders can be assured that it’ll likely be managed as best as it can be. Executives have skin in the game, and clearly focus on long-term risk-adjusted returns rather than short-term bonuses at shareholder expense.

Regional banks are likely to face pressure, and many are already facing pressure, as the federal funds rate is likely to go down from here and we’re already seeing declining mortgage rates. Over the past few years as the Fed has been raising rates, the net interest margin of banks has been on the rise, which can easily reverse and start going down:

Most banks in Japan and Europe have lower net interest margins than United States banks due to their long-term zero rate policy. If the U.S. goes down that route during this next cycle, with a prolonged period of zero rates, the average net interest margin of banks in the U.S. could deteriorate.

As of the most recent quarter, BancFirst has a net interest margin of 3.85% (up from 3.66% last quarter), which is higher than the broader bank average of 3.36% (flat from 3.35% last quarter).

The bank has 25 years of consecutive annual dividend increases and weathered the subprime mortgage crisis well. They also avoided taking any major hit during the shale bust, where some banks that had lent to drillers faced loan losses.

The current balance sheet is diverse and well capitalized. Of their $7.6 billion in assets, $5 billion is invested in diverse loans yielding an average of 5.57%, with a focus on real estate and commercial. Real estate in Oklahoma is pretty cheap, and their risk controls are strong. The rest is held in liquid securities yielding between 2.35% and 3.00%. They have a good mix of fixed and variable assets. Three quarters of their assets have durations of under 5 years, while one quarter has durations of over 5 years. Overall, they have rather low credit risk, moderate interest rate risk, and moderate stagflation risk.

As of now, the stock’s P/E is under 15 and its dividend yield is about 2.2%. The stock hit a high of about $65 in Q3 of 2018, but has since sold off like many regional banks, and is now at $56, or about a 14% decline from its height.

If the U.S. economy continues to weaken, and if Fed cuts the federal funds rate, we could start to see broad declines in regional bank net interest margin, and subsequent stock price reductions. I’m not buying this stock at the moment, but it’s on my watch list as my regional bank of focus to see what the stock does as this rate cycle and economic cycle continue to unfold.