BankFinancial Corporation (NASDAQ:BFIN) has an opportunity to increase its earnings as interest rates rise. In general, the majority of its loans and investment securities are currently at relatively short-term maturities and/or adjustable rates, while its liabilities are concentrated in low-yielding savings/checking account types. If interest rates were to increase, BFIN could see an expansion of net interest margin.
In the analysis below, we attempt to estimate a “normal” net income figure by considering BFIN’s revenue relative to its expenses; then determine how these normalized earnings compare to the stock’s current market price. Please note that all figures in blue come from the company’s SEC filings, and all figures in peach come from author calculations. All figures are in thousands, except per-share amounts and share quantities.
BFIN earns revenues in two categories: interest income and noninterest income. Interest income makes up more than 80% of BFIN’s revenues, and is therefore the most important revenue category for BFIN’s financial performance:
We discuss each of these two categories below.
BFIN earns interest primarily from 2 types of assets: loans and investment securities. The recent September 2021 10-Q had the following balances and yields for BFIN’s interest-earning assets for the Q3 2021 period:
From the table above, we see that BFIN’s primary source of revenue is its loans. Further disaggregation of the loans results in the following mix:
As we can see from the table above, the primary loan types are currently multifamily residential and commercial loans/leases (which together make up 87% of the loan portfolio, per the September 2021 10-Q). This hasn’t always been the case though; before 2014, the primary loan types were real estate related (i.e., real estate loans represented 81% of the loan portfolio in 2011; now they represent only 54%, and are concentrated in multifamily residential).
I think their strategy shift from real estate to commercial loans is favorable for two reasons:
Let’s consider each of these in detail:
Per the recent September 2021 FOMC Meeting, the economic projections of the Federal Reserve Board members and Federal Reserve Bank presidents are as follows:
Given the projections for increases to interest rates in the future, it would behoove the bank to have either variable rate loans or those of shorter durations to allow for rolling over into higher rate loans as interest rates rise. With the company’s recent transition to more commercial loans, they are better able to achieve this result. Per their recent 10-Ks and 10-Qs, a summary of their typical terms for different loan types are as follows:
As such, I believe BFIN has positioned itself well for a rising rate environment; per their recent 2020 10-K, their liquidity management strategy goes like this:
We actively evaluate interest rate risk in connection with our lending, investing and deposit activities. In an effort to better manage interest rate risk, we have de-emphasized the origination of residential mortgage loans, and have increased our emphasis on the origination of nonresidential real estate loans, multi-family mortgage loans, and commercial loans and commercial leases. In addition, depending on market interest rates and our capital and liquidity position, we generally sell all or a portion of our longer-term, fixed-rate residential loans, and usually on a servicing-retained basis. Further, we primarily invest in shorter-duration securities, which generally have lower yields compared to longer-term investments. Shortening the average maturity of our interest-earning assets by increasing our investments in shorter-term loans and securities, as well as loans with variable rates of interest, helps to better match the maturities and interest rates of our assets and liabilities, thereby reducing the exposure of our net interest income to changes in market interest rates. Finally, we have classified all of our investment portfolio as available-for-sale so as to provide flexibility in liquidity management.
In more recent years, the company has had relatively muted loan provisions, but this has not always been the case. In the 2011 – 2012 years, the company recorded a significant provision for loan losses as follows:
Per the table above, more than half of the company’s interest income was absorbed by loan losses in 2012, resulting in a significant loss, as well as a greater loss in 2011. Let’s dive a bit deeper into what happened so we can understand whether such a misfortune could be recurring.
Looking back as far as 2008, we see that the loan losses are primarily a delayed result of the 2008 financial crisis. As we see in the following table, loan losses began in 2008 and slowly ramped up through 2012:
To get a better picture of the types of loans that had the worst problems, we pulled from SEC Filings the total charge-offs by loan type for the period 2008 – 2012; and taking the stance that the charge-offs are related to loans that were outstanding as of the end of 2008, we compared the total sum of charge-offs for the 2008 – 2012 period to the loan balances as of 12/31/2008:
Per the charge-offs as a percentage of loan balance in the table above, we note that the construction/land loans and the consumer loans had the highest charge-offs (at 19.20% and 15.07%, respectively), while the commercial loans/leases and the multifamily residential loans had the lowest charge-offs (at 3.12% and 4.44%, respectively). Looking at the company’s current loan mix (see earlier table), we note that practically 0% of its portfolio is construction/land or consumer loans, while 87% of its portfolio is composed of commercial loans/leases and multifamily residential loans (i.e., the types with lowest charge-offs in 2008-2012). As such, it appears the company has positioned itself to favor lower risk loan-types. Per a previous table, the average provision for loan losses over the past 5 years is approximately 0.44% of interest income, which is generally negligible; barring a significant economic catastrophe, I don’t see a material risk in loan losses going forward.
Having made the considerations above, let’s think about what this means for the company’s revenue going forward. To consider the interest income the company is expected to earn over various interest rate levels, we can compare the historical yields of its loan portfolio to the historical federal funds rates as follows:
We see in the chart above that the Securities and Other (which is cash) tend to follow the changes in interest rate levels, which makes sense based on our understanding of the shorter durations. The loans, however, have a delayed and muted effect as it takes a while for the changes in interest rates to work their way through the portfolio. Having said that, we would expect higher interest rates to be very good for the company’s revenue. To add to this, the company has also increased its proportion of loan balances that are at adjustable rates, and reduced the balances that are at fixed rates. Per the chart below, we see that in 2011, 62% of the loan balance was at fixed rates; but in 2020, this ratio is flipped, with approximately 59% of the loan balance now at adjustable rates:
This change is in line with the company’s pivot toward commercial loans, and away from residential loans. To reiterate, more adjustable rate loans will allow the company to better participate in higher interest rates going forward.
With that understanding, I believe the current interest income level is a floor for the company and will only go up as interest rates increase; at current yields and asset mix, the company might expect $46,125 in interest income as follows:
But under a rising rate environment, we might expect the next five years of interest rates to be very similar to the last five years (i.e., where the average federal funds rate increased from 0.39% in 2016 to 1.55% in 2020). So one method of calculating our expectation of a normalized interest income for the next five years might be to average the yields over the past five years, and multiply them by the current loan balances, as follows:
Per the calculation above, we might expect $60.8 million in interest income if interest rates were to move to levels seen in the past five years. In addition, if the company could put the $505,153 of cash (denoted as “other” in the table above) into higher yielding assets, additional interest income could be earned, putting them at the levels seen in 2018 and 2019 (which were very profitable years for the company). The company does have plans for doing that, as noted by CEO Morgan Gasior in the recent earnings call:
I think you could see us putting a little bit more into securities in the shorter duration world…So, in the securities area, we would still stay relatively short duration, but I do think there's some move -- there's some possibility of picking up some yield, and just holding it for a relatively short-term maturity and leaving it at that…I think the yield curve might've gotten to the point where a baseline investment might make sense.
Therefore, with this understanding, we might say that $46,125 interest income is a floor, while $60,848 is an achievable interest income within 5 years. And if the $505,153 in excess cash were put to work in loans or investment securities, we might see an 89 basis-point bump in yield (which is the difference between the current yield of the company’s investment securities and the current yield of its “other” assets), which would equal an additional $4,496 in interest income.
I didn’t want to spend too much time on non-interest income, since it represents a smaller proportion of total revenue; however, per an earlier table, it seems the company had unusually high non-interest income in the 2018 year. And since we are expecting the next 5 years to reflect those of the last 5 years, I decided to investigate what caused the higher non-interest income, so we can determine whether it is a recurring or non-recurring item. The following table disaggregates the non-interest income by type, and we can see that the 2018 year was quite unusual in that more than half of the non-interest income was due to gains on equity securities and a death benefit on bank-owned life insurance:
Per the 2018 10-K, the gains on equity securities are due to the sale of the company’s Class B Visa common shares that it owned, and the death benefit on a bank-owned life insurance policy as a result of the death of a retired bank executive. Since the company no longer owns any equity securities, and because the timing of such insurance payouts are unpredictable, I have removed those gains from the calculation of non-interest income above to normalize it. Doing so produces a smoother trend of non-interest income, with dampening deposit service charges and fees over time, and negligible earnings on bank-owned life insurance after the death of the retired bank executive.
With this analysis and the decreasing trend in deposit service charges and fees, I think a best estimate of normalized non-interest income is the annualized 2021 amount, which is $5,411.
BFIN’s expenses are composed of interest expense and non-interest expense:
Non-interest expense makes up the majority of expenses; but since it is mostly a fixed cost, we make less effort in estimating a normal amount. We see that the non-interest expense has been consistent over the past 5 years (with outsized administrative expenses around the time of the excessive loan losses in 2011-2012, previously discussed). Given that understanding, our best guess for “normal” non-interest expense is likely to be the Q1-Q3 2021 annualized amount, which is $40,792.
Next, we dive a little deeper into the more variable and risky of the two expense types: interest expense.
The liability side of the balance sheet gives us insight into the interest expense of the company. The recent September 2021 10-Q had the following balances and yields for BFIN’s interest-bearing liabilities for the Q3 2021 period:
As we can see from the table above, deposit accounts are certainly the cheapest sources of capital, while the subordinated notes are the most expensive. So the bank would obviously prefer to source its capital from depositors, which has proven to be the case: over the past 10 years, BFIN has continued to increase its deposits, while not needing to rely so much on other more expensive sources of capital; and in fact, the company currently has the most deposits it's had in 10 years:
Considering our previous discussion of interest-earning assets where we note that the company has $505,153 in cash (denoted as “Other” in the balance sheet), the company’s dilemma is not a need for more deposits but rather how to deploy the deposits it already has. In a recent earnings call transcript, the CEO Morgan Gasior notes, "We really -- obviously don't need any additional deposits" - which is a great problem for a bank.
If we were to try to estimate what interest expense to expect today, we would merely multiply the current yields by the current balances, resulting in $2,778:
This amount is less than half of the 2020 interest expense and significantly less than the average amount over the past 10 years. This is because the deposit mix in 2021 has leaned more heavily on checking/savings accounts than on CDs, so BFIN pays less in interest. In fact, the deposit mix is at an all-time low percentage for CDs currently:
Looking at the table above, we see that the deposit mix for the various categories is currently allocated to more liquidity instead of more yield (possibly due to savers’ uncertainty in the economy and their own expected personal/business cash flows). As such, I think we need to consider possible interest expense expectations based on various deposit mixes as well as interest rate levels; let’s address that now.
In staying consistent with our expectation (discussed above in the Revenue section) that interest rates over the next 5 years might look similar to the interest rates of the past 5 years, we calculate the 5 year historical average yield for each type of interest-bearing liability account:
In addition, we calculate the 5 year average deposit mix for each type of interest-bearing liability account:
Now, by multiplying the total deposit balance by the average deposit mix percentage and the average yields by deposit type, we can estimate an annual interest expense amount that we might expect in the next 5 years, which comes to $7,123:
Based on the analysis above, we can attempt to normalize the profitability of BFIN (i.e., determine what its profit might be in a “normal” year over the next 5 years). To do so, we will take into account our discussions above to come to appropriate amounts of revenues and expenses:
Since we have a range here, let’s put this into a table so that we can see the range of net income we might expect based on these assumptions:
Based on the table above, we might expect a net income of between $5,268 and $16,126, based on expected interest rates and deposit mixes. With that in mind, let’s consider how this looks in relation to the stock’s price.
As of 12/23/2021, BFIN’s stock was trading at $10.65 per share; and across 13,280,723 shares outstanding as of October 27, 2021, this results in a market capitalization of $141.440 million. Using our calculated net income range above, we come to an earnings yield between 3.72% (at current yields and asset mix) and 11.40% (at 5-year average asset mix and yields, while putting excess cash to work), as shown in the following chart:
Based on the analysis above, we might expect a 9.08% earnings yield (based on the stock’s current market price) at some point in the next 5 years if interest rates increase to be in line with the last five years (which were slightly higher than they are now). As pointed out earlier, the company now has more of its loan portfolio allocated to adjustable rate and shorter duration loans, which gives the company the ability to participate in an increasing rate environment, as well as have a reduced loan loss risk than it had 10 years ago. In addition, the company has $505,153 of dry powder it can use to invest in securities or originate additional loans as interest rates rise. Given these muted risks and potential for greater returns going forward, I think the company is a great component to hold in a conservative long-term investor’s equity portfolio. And to boot, the price of the stock is currently at an 11% discount to its tangible book value.
This article was written by
Disclosure: I/we have a beneficial long position in the shares of BFIN either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.