Lower ROE’s in Banking: “Inefficient” has become “Resilient”

The world is in the midst of a global crisis.  So far, the COVID public health and employment crisis has not turned into a widespread banking crisis.  Why is this?  One reason is that banks are, in general, far better capitalized than they were in the last crisis.  While banks return on assets recovered to pre-financial crisis levels, return on equity never fully recovered due to increased capital requirements.  What was long seen as a downside in bank performance may finally be seen by investors and us all as a virtue.  Want to know more?  We go through all the math below.

It's worth pointing out that this analysis is based on Call Report data, which is reported quarterly by all FDIC-insured banks. This month, the first quarter of 2020 will be reported with limited visible impact from COVID. The second quarter will be reported in July, so we’re still a few months away from seeing the impact in these numbers.

Figure 1

Banking Margins: A Decade in Review

Breaking down banking margins isn’t difficult. (If you’re well versed in these metrics, skip ahead a bit.) To start, all banks make revenues from interest they charge on loans. Banks ‘pay for’ this with interest expense to deposit account holders, the Fed, and other banks. The difference between these interest rates is ‘Net Interest Margin,’ or ‘NIM’. Every bank will experience a different NIM depending on product mix, where in the credit spectrum they operate (especially consumer-heavy banks), the size of their deposit base, etc. At a high level, banks with more than $100B in total assets have markedly lower average NIM’s than other banking subsegments (Figure 1), largely because these banks operate at a scale where lower NIM can still cover fixed costs for the bank.

Of course, NIM is only part of the story. Banks also experience charge offs. The net charge off rate (charge off dollars less recoveries divided by total assets) is often subtracted from NIM to result in ‘Risk Adjusted Return,’ or ‘RAR.’ Unsurprisingly, charge off rates (and, subsequently, RAR’s) are much more cyclical than NIM’s. Again, notice how the larger banks (in this case, those banks with >=$50B in Total Assets) operate on tighter margins than smaller banks. In most periods, these banks have the highest net charge off rates. How can the largest subsegment of banks afford to have the lowest NIM’s and the highest net charge off rates?

Figure 2

Scale. Banking requires large fixed costs in the form of compliance teams, finance departments, tech investments, marketing, and more. To evaluate the impact of non-interest expenses, we turn to the efficiency ratio and the operating ratio. The efficiency ratio is the ratio of non-interest expenses to operating income, whereas the operating ratio is the ratio of non-interest expense to total assets. Each measure has its own place in analyzing banks, but here they tell a similar story: the largest banks spend less, relative to assets and to income, than the smaller banks (Figure 2).

Now we get into the chicken and the egg question. Are smaller banks forced to underwrite lower-risk, lower-yield loans because they operate less efficiently? Or do they operate less efficiently because they’ve traditionally operated in a low-risk, low-yield, low-cost of funds environment? Either case is scary considering these subsegments aren’t experiencing low-cost deposit growth. Thus the cycle begins of writing lower and lower margin loans, preventing investment in consumer-preferred technologies like mobile banking, e-Check deposits, peer-to-peer transfers, and other expensive infrastructure; which in turn limits deposit growth, and makes it difficult to find high-margin loans at a fair cost to acquire, accelerating the vicious cycle.

Figure 3

Interestingly, this all stacks up to somewhat similar ROA’s across the board. There is certainly noise, including periods where some subsegments are clear winners or losers (we’ll have to analyze the $10-50B banks some other time), but in 2019 banks with more than $50B in assets had average ROA’s within a few bps of one another, and the smallest banks were only around 40 bps behind (Figure 3). There is no denying that there are periods with more differentiated results (post-crisis recovery varied significantly among subsegments). Finally, many of these subsegments are back to ROA levels not seen since before the financial crisis.

Yay! ROA’s have recovered, we can pack our bags and call it a day, right? Of course, this isn’t the end of the story of banking margins. It’s important for banks to think about their returns per assets, but investors really care about return on equity. And while ROA’s are at pre-financial crisis levels, more stringent capital requirements have kept ROE from returning to pre-financial crisis levels (Figure 4). In fact, the smallest banks haven’t been able to consistently break the 10% ROE threshold that investors are looking for.

Figure 4

Lower ROE’s, Maybe a Good Thing?

When I first wrote this post this section wasn’t going to exist, and the post would end on the note that the industry may need to get comfortable with lower ROE’s. And then the outbreak of COVID-19 changed everyone’s plans and kicked off the next recession.

This isn’t the place to make specific forecasts as to the impact of the pandemic. But I don’t need specific forecasts to know that, during a recession, the game changes for banking. Suddenly, those lower ROE’s are looking more ‘resilient’ than ‘inefficient.’ Maybe the banks will be in a position to help in this one. Sure, the rollout of the Paycheck Protection Program has struggled, but that’s due to bureaucratic bottlenecks – no one is talking about systemic risks (as they shouldn’t).

ROE’s won’t creep up toward 20% again anytime soon, even if ROA’s jump back from coronavirus. Instead, let’s focus on the benchmark set by the 2010s and return to a profitable, resilient banking industry that is ready for the next crisis instead of causing the next crisis.

There is a lot to unpack in these views. Do you see something I didn’t cover that you would like me to investigate? Shoot me an email and we can dive deeper into the call reports together!