Friday, January 20, 2017

Banking Economies of Scale Revisited

In 2011, on these pages, I wrote my most read blog post ever, titled: Does your bank achieve positive operating leverage? Even today, nearly six years later, it receives a material amount of views. Particularly from larger financial institutions.

Economies of scale has eluded our industry in its purist form. For some time, banks between $1B and $10B in total assets tend to wring out the best expense ratios (operating expense/average assets) and efficiency ratios. So economies of scale hucksters walk with this chink in their armor as to why their story-line falters at a certain size.

I also noted in my most recent and in all of my past Top 5 total return posts that community banks deliver superior returns to their larger brethren. So, although there are plenty of consultants and investment bankers with pitch books telling you to get bigger, there are also contrarians such as myself that believe that bigger is not always better. And my pitch book is simply a bunch of spreadsheets. No fancy bubble charts, green light/red light tables, or tombstones. 

In this post I would like to revisit a couple of tables. First, I broke down all commercial banks by asset size to show expense and efficiency ratios as banks became larger. The results are below.

As the table suggests, financial institutions of all sizes reduced their relative operating expenses since 2011, with the only blip being a slight expense ratio increase in the $500MM-$1B commercial bank category. I should note that the efficiency ratio from that sized bank actually went down between 2011-16, suggesting a slightly better net interest margin.

The economies of scale argument clearly has merit, as you can see from the table. As asset sizes increase, expense and efficiency ratios tend to decrease. With that pesky exception of financial institutions between $5B-$10B in assets. These are averages. So there are exceptions. And I have often spoken about there being a significant number of exceptions to the economies of scale bromide. 

One example is German American Bank, highlighted in American Banker's Community Banker of the Year issue, and on this blog. It is a $3B bank with a 55% efficiency ratio. Open Bank in Los Angeles is a $722 million bank with a 58% efficiency ratio. I didn't have to research small efficient banks. They rolled off my tongue. Actually, my fingertips.

The below table was taken from my firm's profitability outsourcing database. We do the cost accounting for dozens of financial institutions that includes calculating operating cost per product account. Did costs go down at this granular level as assets grew?

Obviously, no. But why? If assets grew, and the bankwide expense and efficiency ratio has generally declined as banks grew, how did these costs go up? It is a fully absorbed cost system, so all costs within the bank are allocated to products and services.

My theory is this. Average balances per account have been growing since the low end of the yield curve has hovered near zero, and today is below 1%. The cost to originate and maintain a $100,000 money market account is nearly identical to originating and maintaining a $50,000 money market account. The growing average balance per account phenomenon has been occurring in most bank products. So, bankwide, costs would appear to go down because denominators, average assets in the expense ratio and total revenue in the efficiency ratio, are going up with the average balance per account.

Number of accounts, however, have not been increasing at nearly the same pace as the balance sheet, if at all. So all of those resources at your financial institution designed to grow new account relationships have not been efficiently utilized. 

In other words, in account originations, financial institutions are generally, and on average, over capacity. 

Financial institutions have tried to reduce this capacity in branches by consolidation and staff reduction. That is why you don't see material increases in cost per account in deposit categories. 

But either through expense reduction or new account acquisition, there is more left to do.


  1. Hi Jeff. I stumbled upon your blog as I was curious about bank's cost structure and how do banks think about cost because when we teach management accounting, most of the cost structure applications is based on manufacturing sector.

    The question that I have is given that more than 50% of the cost is human resources (as you mentioned in one of your other blogs), how do banks respond to situations when they have sudden increase in inflow of deposits, say during the shale boom. One possibility is they might wait to see whether the effects are persistent and use the current capacity because from what you say looks like there is excess capacity and this windfall liquidity inflow can be taken care off or the banks can expand by hiring employees, opening branches, etc. once they observe that the shale boom effect is persistent. I am curious to know your views and also on banks cost responses to uncertainty.

  2. Banks have step-variable cost structures, where they buy large chunks of capacity at once. There are two types of activity, origination and maintenance. Technology and productivity improvements increase capacity and therefore banks don't often perform at peak capacity, in my experience. When they do, they purchase chunks (either tech or additional staff). Most times, spikes of volume can mop up excess capacity... and also overtime, etc. The key challenge to banks is to shift a very high percentage of fixed costs to be more variable, therefore allowing them to reduce costs in times of low utilization.

    That was a pretty wonky answer. Best I can do tho'! Thanks for the question.