The top 20 things learned from Amitabh Bhargava when writing the Bank Brief over the past 20 years have resulted in some our most favorite sayings.
1. Management teams really matter. I contend one could have gone to our 2007 Global Financial Services Conference, ahead of the financial crisis, and seen the top 20 U.S. banks present. If I had gone long the 10 best presentations (based on company positioning, strategy, poise) and short the worst 10, I believe I would have gone roughly 7 out of 10 on each and could have made a great deal of money.
2. The cockroach theory is real. When a company reveals bad news to the public, additional and related negative events may be revealed in the future. With respect to earnings, we have observed when a company misses lowered expectations, a string of misses sometimes ensues.
3. The first loss is sometimes the best loss. In May 2012, JPM disclosed a roughly $1bn mark-to-market losses in its synthetic credit portfolio (aka The London Whale). Its realized loss ultimately exceeded $6bn. More recently, in 2Q23, COF sold a full $0.9bn of its office portfolio (albeit while providing seller financing) with hefty loss. Jury is still out, but this could be a good call.
4. Just because a stock goes up on its earnings print or a conference presentation, doesn’t mean it was good. Sometimes I’ll look at an earnings release or listen to a conference presentation and opine that it wasn’t good. Still, the stock gaps up. Assuming I’m right, it’s usually because the market is missing something (which does sometimes happen) or ‘positioning’. Still, I’ve observed over the years, in the ensuing weeks, those that had good quarters/conference presentations tend to outperform those with bad ones.
5. Banks are not extinct. Shortly before I started covering bank stocks on the sell-side, former MSFT CEO Bill Gates said in a 1994 speech at a BAI conference that “Banks are dinosaurs, they can be bypassed.” Thirty years later, banks are bigger than ever.
6. There is only one commodity no matter who you are that you want more of, and that’s money. I adapted this from former WFC/NOB CEO Dick Kovacevich. Fairly self-explanatory, but it is a nice secular tailwind and should aid banks over time. It also helps explain why almost all banks seem to be pushing on wealth management. When asked why he robbed banks, Willie Sutton replied “Because that’s where the money is.”1
7. Interest rate risk can in fact kill banks. This is a relatively newer learning. I was brought up on the notion that credit issues cause banks to fail, not interest rate risk. Last year proved this wrong.
8. Borrowers don’t always pay back their mortgage. Another falsity I was taught is consumers will always pay their mortgage bill first. Well that’s not necessarily the case when it’s underwater on a 3rd home and the borrower has no job and didn’t need a down payment. Learned that one the hard way.
9. Regulators are undefeated and consent orders tend to last longer and cost more than initially projected. The first point we borrow from former PNC CEO Jim Rohr. Banks don’t have a great history of battling the regulators, though Basel III endgame will be interesting to watch. With respect to consent orders, WFC followed by FITB, MTB, USB and BBT jump to mind, but there are others. Looking out, all eyes on C.
10. There are some positives to being regulated. There are some advantages to being regulated and its does create barriers to entry for certain products. It also allows us to compile long-term data sets, including several back to 1934, when FDIC insurance first went into effect (see Bank Chartbook 1934-2023, 3/11/24, for example).
11. There is a great deal of value in the reading the 10K/Qs. I can’t stress this enough. When the 10K/Q comes out, read it. I’m always amazed how few people do. If you can’t find the time, we do provide handy summaries in the Bank Brief after each filing. At least look at those.
12. Very detailed disclosures don’t mean everything is ok. In fact, sometimes the opposite is true. In 2007, I’d say the two companies with the most comprehensive financial disclosures were Wachovia and National City. Enough said. Also, whenever a company decides to give you monthly financial figures, that could also be a bad sign. NCC and WB both did it. WFC did it in the aftermath of its sales practices issues. SVB did it in the late 1990s.
13. Not all bank mergers are bad. While bank investors typically like to hammer the buyer’s stock, done right, mergers can create value. Still, we have admittedly seen our fair share of bad mergers.
14. Still, the first quarter out of the box for a large bank merger can be very messy. Acquisitions are complicated. The accounting can be noisy. Fair value marks rarely match expectations. Intra-quarter deal completions make comps difficult for a couple of quarters. It typically takes time for the financials to work out and synergies to materialize.
15. Hockey sticks are tough to make happen. Many times over the years, companies have given upbeat full-year guidance after a weak 4Q, while not providing a great guide for 1Q. We’ve learned a sharp ramp-up in earnings is typically tough to come by. In many such cases, the full-year forecast was not achieved. This year, for several line items, banks have pointed to a better 2H than 1H. We’ll see.
16. Widget companies make widgets, banks make loans. Banks are always looking to make (good) loans. While there can sometimes be certain ‘supply constraints’ (RWA mitigation, liquidity management, increased funding costs, tighter lending standards, focus on existing full relationships), slowdowns tend to be much more demand-driven (softening economy, businesses continue to reduce capex amid slowing demand/uncertainty as well as expectations for lower rates, elevated interest rates/higher debt service costs, lagging effect of higher costs). Still, there can be certain expectations (like office loans at the moment).
17. Net interest margins typically go down. I know this is hard for those newer to the space to grapple with, given industry-wide NIMs increased in 2016, 2017 and 2018 and then again in 2022 and 2023. Still, in my first 20 years on the sell side, they expanded only two times (2002 and 2009; we don’t count 2010 b/c that was solely due to an accounting change).
18. Modest increase in loan losses, no problem. Rapid rise, get out of the way. Looking at the past 86 years, NCOs were stable to lower in 43 years, up 20bps or less in 35 years, and up by more than 20bps in 8 years. In both the ‘stable to lower’ and the ‘up no more than 20bps’ scenarios, bank stocks outperformed the market slightly more than half of time. Still, in the ‘up more than 20bps’ scenario, bank stocks underperformed 7 out 8 times (by 20 percentage points, on average).
19. Analysts typically under-forecast provisions when credit conditions are getting worse, and overestimate them when things are improving/stable. For better or worse, provisions are often mis-modeled. When credit is deteriorating, analysts typically don’t forecast high enough loan loss provisions and when provisions are declining, analysts tend to underestimate how quickly they can come down. We’ll see how much CECL, or lessons learned, changes this.
20. Banks are levered, so their mistakes are magnified. Still, PPNR power is often underappreciated. While capital ratios have trended higher since the GFC, tangible common equity ratios are still only 8% (so 12.5x levered). That said, we think investors sometimes under-appreciate the power of PPNR. For example, PPNR for our coverage this year should exceed the typical annual stress test loan losses.