Profitability Profiles: The Story of Bad Apple Bank

May 6, 2021 Dallas Wells

Welcome to the final installment of our series, “Profitability Profiles.” If you want to know more about the thinking behind these blog posts, you can check out the intro in part one, “The Story of Deadweight Bank.”   

If you just want a TDLR version, here goes:   

Most banks consider themselves “relationship banks,” but do they really know what that means? What constitutes a good relationship, or a bad one? And how can they attract more of the former?   

To find out what “relationship banking” looks like in reality, we did some deep dives into the data of a group of real (but anonymous) banks. We pulled out some of the most interesting stories – ones with themes we think will resonate with many of our readers – and are sharing them in this series.    

Reminder: We’ll be walking through some of the cases, and showing how PrecisionLender can help, in our May 11 webinar.    

And don’t forget to check out parts two and three in the series, the stories of “Surprise Bank” and “Consistency Bank” 

Okay, now on to our fourth and final installment:  the story of “Bad Apple Bank.” 

Bad Apple Bank  

Bad Apple Bank is a $25 billion institution located in a major metro area. They are known in the market for multi-family real estate lending, and among their investors as a solid operator focused on efficiency. Like the other banks we’ve reviewed in this series, their marketing pitch is all about relationship banking and a personalized, human-centric approach to service. 

A Strong Portfolio Except for …  

When we dig into relationship profitability data for a bank, we first look for obvious outliers in “the tails.” These outliers typically are what drive (or destroy) performance and are also most likely to yield actionable takeaways.   

Like most banks, Bad Apple is top heavy. Their top quartile of customers by net income account for 102% of total profits. The nearly 40,000 customers in quartiles 2 and 3  don’t make up for the losses in the bottom quartile. These ratios are all fairly normal, as is the fact that Bad Apple generates profitable deposits throughout the portfolio distribution but has a high risk/high reward outcome with loans.  

Income by Customer Profitability Quartile

What is unusual is that only 8% of Bad Apple’s relationships are unprofitable - the lowest ratio among the banks we researched for this series. It shows that Bad Aple has been disciplined with their commercial pricing strategies. 

In fact, pricing for credit risk stands out as a specific strength for Bad Apple. They typically report both past due and net charge off numbers that are above their peer groups. And those metrics are not just a function of loan mix. The trend holds true across nearly all loan types where the bank has meaningful balances. They are clearly booking loans on a consistent basis that are slightly riskier than their peers. But they get paid for that risk. Their yields on all loan types are also above their peer group, and again, they have very few relationships on which they lose money.  

But, oh boy, that small number of unprofitable relationships! Those few bad apples (thus the nickname) are really impacting the rest of the bunch.  

Bad Apple Bank’s bottom 25 commercial relationships lose a collective $35 million on an annualized basis. (These are all active loans, BTW. The truly ugly, non-accrual stuff has been removed from the analysis.) That wipes out the earnings on more than 55,000 relationships that rank ahead of them in profitability. To put that in context, that is 0.03% of their relationships undoing the work for 75% of their relationships. That is a whole lot of work from a whole lot of people that gets negated by a small handful of mistakes. 

What’s Making Those “Apples” Go Bad? 

The natural question, of course, is how do they avoid relationships with such big losses? This is an especially tricky problem for Bad Apple. Excessively tightening underwriting standards in order to avoid outlier credit problems would jeopardize their niche in the market as a higher-risk lender. Also, there is also no clear pattern: The least profitable customers come from a variety of different industries, and there are no clear concentrations among loan types, market, or bankers.  

The most common trait, it turns out, is that most of the least profitable relationships are loan-only customers. The reverse is true for the most profitable; across a wide variety of industries and markets and even sizes, the best predictor of high profitability is a deep relationship with multiple account types. 

         500 Most Profitable Relationships               500 Least Profitable Relationships                   

This difference in depth of relationship becomes most prevalent at the extreme ends of profitability. The top 100 customers generate more than $135 million in annual profits (about 1/3 of the total commercial bank profits). Just 2 of those 100 have only loans with the bank, and between the two of them they carry 10 separate loans. Bad Apple's most profitable relationships are deep and broad. 

Conversely, of their bottom 25 relationships that collectively lose over $35 million per year, 17 are loan only. The single biggest predictor of losing relationships is that they only have loans with the bank, and the biggest predictor of making the top quartile is that they have 10+ individual accounts with the bank. Especially for a bank focused on slightly higher risk CRE, cross selling is the single most important driver of future profitability. 

Examining the Rest of the Bunch  

Outside of those bottom relationships. Bad Apple has few glaring weaknesses. Their loan and deposit pricing are consistent and properly reflect the risk being taken. They are particularly proficient with their larger, more profitable customers. 

Returns by Customer Profitability Quartile

Bad Apple’s overall ROA on loans is 1.38%, which was the highest among the banks we researched for this series. That was driven not just by the top quartile (1.86%), but in particular by the top 25. Those customers generate a loan ROA of 2.54% (again the highest in our series) on average loan balances of around $50 million each. In our other banks, the most profitable borrowers were pristine credits that required minimal loss provisions. At Bad Apple, top customers included large developers that required loss provisions and higher capital holds, but that were priced at excellent credit spreads. Even in a hyper competitive large city, they have found a specialization where they can consistently generate healthy margins at scale. 

The other standout among the most profitable customers at Bad Apple were municipalities and other public entities. Municipal banking is profitable at many institutions, but Bad Apple has an outsized share of earnings coming from both loans and deposits with these customers. This is one of the benefits of operating in a large city; there is plenty of this kind of business to be had. Bad Apple tends to make about 4x from a typical school district customer what it makes from a typical multi-family real estate customer, and I doubt they spend nearly as much time and attention trying to source those accounts. 


In summary, Bad Apple Bank is another solidly profitable regional bank that performs better than their peer group both in terms of earnings and growth. While they are disciplined and make smart decisions in managing risk (only 8% of customers are unprofitable), they have a handful of very painful relationships that are hindering performance. Below are a few of our thoughts on strategies for improvement. 

Require Special Approval for Loan-Only Customers 

Bad Apple generally does a solid job of cross selling and has deep relationships. Over 97% of the commercial customers have non-loan accounts. However, those 3% without non-loan accounts have a high probability of ending up as big money losers. We would suggest that the bank institute a process for special sign-off from a line manager/executive for any new business that is loan only. This should not impact too many opportunities, since most already include deposits, and it will offer the most efficient means of “catching” those losing relationships before they make it on the books. 

Double Down on Municipal Business 

Bad Apple has a highly profitable set of municipal/public entity customers. Given their footprint, they can probably grow this business, and clearly have the infrastructure and products needed to serve those customers already in place. The bank should consider investing more in marketing and business development in this segment. They might also get slightly more aggressive in their pricing in order to gain some share (since these accounts are typically bid among financial institutions with a local presence). 

Aggressively Pursue C&I Loans 

Given Bad Apple’s solid deposit gathering performance across lots of industry types, they have a foothold with customers that are likely doing a lot of non-real estate borrowing. However, the bank has stuck with their real estate focus. While the strategy will differ, the bank is large enough that it should have the in-house expertise to be able to pursue C&I loans more aggressively. These will not necessarily have the same elevated risk profile and will require thinner pricing in order to improve current win rates. The bank should diverge their C&I strategy with lower, more aggressive hurdle rates for C&I while maintaining their current pricing strategy for real estate. 

See Anything Familiar?  

Chances are you may have seen a little of your bank in the story we just told. If not, you’ll likely come across an issue that hits close to home at some point in our series.  

If you’re looking to make some improvements on the relationship profitability front (and really what bank isn’t?) you should attend our upcoming webinar on the topic. We’ll walk through common problems we’ve seen and how PrecisionLender’s platform can help you solve them.  


About the Author

Dallas Wells

Dallas is a writer, speaker and former consultant who has held executive roles at two banks with experience in capital planning, liquidity forecasting, investments, budgeting, financial reporting and mergers and acquisitions.

Follow on Twitter Follow on Linkedin More Content by Dallas Wells
Previous Article
Scotiabank: No Pausing During the Pandemic
Scotiabank: No Pausing During the Pandemic

Despite a global pandemic, Scotiabank implements Q2’s PrecisionLender at record speed and sees great success.

Next Article
Commercial Loan Pricing Market Update (April 16-30)
Commercial Loan Pricing Market Update (April 16-30)

A look at what PrecisionLender data tells us about the commercial loan pricing market, for the second half ...