What Drives Bank Valuations?

I've been researching what is driving bank stock valuations. It has been a subject that has lived rent free in my head for years. What I would like to drive valuation: long-term EPS growth, financial performance and financial condition. There is correlating factors to these drivers. But I also can't ignore how much tangible book value per share (TBVPS) growth is also a key driver.

Banking is one of the only industries that anchor valuations so heavily to equity. Not just my opinion. A CEO just lamented about this. What makes it troublesome is that retail bank stock investors like a clean balance sheet, good profits, and a strong dividend. Pay a higher dividend and you stunt TBVPS growth. And since this is a valuation driver, you might be slowing share price appreciation. 

You can see this play out with banks that have poor earnings. They trade at relatively high P/EPS multiples versus very profitable banks. It is common to see a very profitable bank trade at less than 10x P/ EPS while a terrible earner trading at 15x-20x. The poor earner might be trading at book, while the good earner at 1.25x book. Telling you all that you need to know about the peculiarities of bank stock valuation. 

To do a deeper dive, I ran a lot of data. I downloaded the following data points for years 2021 - 2026Q1:

TBVPS Growth, EPS Growth, Leverage Ratio, NPAs/Assets, ROAA, ROAE, Net Interest Margin, Cost of Funds, Yield on Earning Assets, Efficiency Ratio, P/EPS, P/TBVPS

I asked Copilot to find correlations between the financial performance and financial condition factors and compare it to investment banks and consulting firms' observations on bank stock valuations. Here are the most salient findings with accompanying tables and charts.

Interesting that Copilot made no correlation to Cost of Funds, even though those publicly traded bank with the lowest cost of funds traded at the highest P/TBVPS, and second highest P/EPS (see table below). Low COF banks trading at higher multiples than higher Yield on Earning Assets banks has been consistent over the 20 years I have made the comparison. 

Here is what Copilot had to say after consuming all of the data and referencing industry experts.

 ✅ A. ROE / ROTCE ↔ P/TBV (strongest positive correlation)

Conclusion:
👉 This is the #1 driver of bank valuation multiples

• Banks with higher ROE consistently show higher TBV growth and would command higher P/TBV

Supporting evidence:

• Price-to-book rises with expected ROE and capital strength [bis.org]
• Bank analysts explicitly link P/TBV to ROE vs cost of equity [breakingin...street.com]

The dataset implication:

• ROAE + EPS growth columns will show highest explanatory power for valuation dispersion

---

✅ B. ROA ↔ Multiples (very strong, often underappreciated)

Conclusion:
👉 ROA drives quality of earnings, and therefore valuation durability

• Clean ROA → high NIM, low NPAs, strong efficiency
• High ROA banks don’t rely on leverage → more sustainable

Observed pattern in your data:

• Banks with:
  • ROAA > ~1.2%
  • Low NPAs
  • Efficiency <55%

→ also show:

• Strong EPS growth
• Strong TBVPS compounding

Industry confirmation:

• ROA reflects true operating performance and asset quality [linkedin.com]
• Markets reward “clean ROE built on ROA,” not leverage-driven returns [linkedin.com]

---

✅ C. EPS Growth ↔ P/E (very strong)

Conclusion:
👉 Growth is the primary driver of P/E dispersion

From your dataset:

• EPS growth volatility is very high across banks
• High-growth names (especially >15–20%) stand out sharply

Relationship:

• Higher forward EPS growth → higher justified P/E

Analyst view:

• Bank valuation embeds expected earnings growth + cost of equity [breakingin...street.com]

---

✅ D. TBVPS Growth ↔ P/TBV (structural correlation)

Conclusion: 👉 TBV growth is the long-term anchor of returns

In the dataset:

• TBVPS growth aligns tightly with:
  • ROE
  • retained earnings
  • efficiency

Banks with steady TBV compounding:

• Show stable ROE
• Have better efficiency ratios

---

✅ E. Asset quality (NPAs) ↔ Multiples (negative correlation)

Conclusion: 👉 One of the strongest negative drivers

From the data:

• NPAs vary dramatically
• Higher NPAs → lower ROA → lower ROE → compressed multiples

Industry evidence:

• Credit quality directly affects risk and valuation via expected losses [intrinsic-...vestor.com]

---

✅ F. Efficiency ratio ↔ Valuation (negative correlation)

Conclusion: 👉 Efficiency drives operating leverage and ROA

From the data:

• Banks with efficiency <50–55% consistently show:
  • higher ROA
  • better EPS growth

Interpretation:

• Markets reward cost discipline + operating leverage

---

✅ G. NIM ↔ ROA ↔ Multiples (moderate–strong chain)

Conclusion: 👉 NIM is an indirect but important driver

From your dataset:

• Higher NIM → higher ROA → higher ROE → higher valuation

Industry context:

• NIM compression in 2025 is already affecting profitability expectations [mckinsey.com]

---

✅ H. Leverage ↔ Multiples (non-linear / weak-to-negative)

• High leverage can boost ROE
• But:
  • increases risk
  • lowers valuation quality

The dataset:

• Shows mixed relationship
• Supports known reality: markets discount “leveraged ROE”

---

3) Synthesis: What actually explains bank multiples (ranked)

Tier 1 (dominant drivers)

1. ROE / ROTCE
2. EPS growth
3. TBVPS growth

Tier 2 (core operating drivers)

1. ROA (asset-level profitability)
2. Efficiency ratio
3. NIM

Tier 3 (risk modifiers)

1. Credit quality (NPAs)
2. Capital / leverage

---

4) How this aligns with recent investment-bank / industry commentary

🔹 1. Profitability (ROE) is still the anchor

• Valuation gaps across banks persist largely due to ROE dispersion
• Investors reward sustainable ROE above cost of equity [bis.org]

---

🔹 2. Growth skepticism is suppressing multiples

• Despite strong 2024–2025 earnings:
  • bank P/B and P/E remain structurally low
• Investors are questioning long-term growth durability [mckinsey.com]

---

🔹 3. Markets increasingly distinguish “quality of earnings”

• Banks are not valued just on ROE:
  • but how ROE is generated
• Clean ROA + low credit cost outperform leverage-driven models [linkedin.com]

---

🔹 4. Strategy + credibility now matter more

• Bain: valuation dispersion driven by:
  • execution consistency
  • growth narrative
  • capital allocation credibility [bain.com]

---

🔹 5. Structural valuation framework

Investment banks (FIG coverage) consistently anchor:

• P/TBV ↔ ROE / ROTCE
• P/E ↔ EPS growth + cyclicality
• Adjusted for:
  • credit risk
  • capital strength [acetheround.com]

---

5) Bottom line (what the dataset is telling us)

🔴 The single most important takeaway:

👉 ROE (and ROTCE) is the strongest direct driver of bank trading multiples

---

🔵 But the true causal chain in your data is:

NIM + Efficiency + Credit quality → ROA → ROE → TBV growth → Multiples

---

🟢 And the market overlay is:

• Growth (EPS, TBV) determines P/E vs P/TBV premium
• Risk (NPAs, leverage) determines discount vs premium

Banks On Sale

My bank stock portfolio was comfortably in the black, with a solid 2.5% dividend yield at year-end. My have times changed.

Before I begin, I feel compelled to disclose that I am not a registered broker or financial advisor. I am not giving you investment advice.

At December 31, 2019, the SNL Bank & Thrift Index stood at 193% price/tangible book, 13.6x EPS, and a 2.63% dividend yield. Then Covid-19.

At April 14, 2020, I measured all banks and thrifts with total assets between $1 billion and $10 billion. Still community banks, and have decent trading volume for more efficient pricing. At the median, these banks had a market metric slash line (P/E, P/TB, Dividend Yield): 8.8x / 95% / 3.39%. But the differences among banks varied greatly. The greatest drop in market price was Marlin Business Services Corp., at 64%. At the other end of the spectrum, Community Bancshares in McArthur, Ohio GAINED 8%! Do they finance respirator production?

I should point out that Marlin Business Services has many subsidiaries, most of them finance companies, but one is a bank.

Statistics

On average there was a significant drop in valuations. But "on average" is a pesky phrase. What is the standard deviation? Recall from statistics class, which gave me math stress by the way, the closer the standard deviation is to zero, the lower the data variability and the more reliable the average is. The higher the standard deviation, the more variation there is in the data and the less accurate the average is. The standard deviation of stock price decline between December 31st and April 14th for the banks I measured was 11.8. Yikes!

That means, within the data, there is opportunity. And I hear about this opportunity from bankers that really, really want to buy back their stock at these valuation levels. But an abundance of caution because of the unknown, plus optics, is preventing them from doing so. 

Fear of the unknown and the gravitational pull that whispers in our ear to buy high, sell low may be holding back the rest of us. It is tempting to be as liquid as possible during this period of uncertainty. But it is the uncertainty that has otherwise healthy and profitable banks trading below book value. Some comfortably below book value.

First Quarter

Early earnings releases, primarily by the big banks may be fanning flames of fear. Citi announced a $7 billion provision for loan loss in 1Q, up from $2.1 billion the quarter prior. JPMorgan announced a 1Q provision of $8.3 billion from $1.4 billion. But $1.3 billion in assets MainStreet Bancshares in Fairfax, Virginia announced a provision of $350 thousand, down from $358 thousand the quarter prior. 

I think it likely though that most community banks will announce increases in provision, and some significant increases, due to early forbearance and payment deferment requests. And, as I said to a bank publication reporter yesterday, big banks take more of a macro-economic approach when assessing their loan portfolio. Community banks are more credit-by-credit. And that may not come to bare until the second quarter. 

Spreadsheets

And this is likely contributing to the wide variation in valuations we are seeing. The below two tables represent the top 10 price declines in banks with $1 billion to $10 billion in total assets from December 31st to April 14th.

Some of the above banks have stories. For example, at first glance, First Defiance looks compelling. A 1.50% ROA, 12.15% ROE, only 63 basis points non-performing assets/assets and a 9.58% tangible capital ratio. Even if NPAs spiked, they have a loan loss allowance as first defense and their capital was very good. Why in the heck did their stock drop 53% and now trades at a 6x earnings and a nearly 6% dividend yield?

Probably because they closed on a previously announced all-stock acquisition of a $2.9 billion in assets bank on January 31st. After their December 31st earnings, but before quarter end. And before the precipitous Covid induced bank stock decline. The deal value at January 31st was actually greater than at announcement! 

So there is uncertainty there, at least until FDEF announces 1Q earnings, which is not anticipated until April 28th. Even then it may not be clear because it would be challenging for FDEF to get their arms around the Covid impact to their own loan portfolio, let alone another, almost equally sized bank. Uncertainty equals discount.

And so it goes with many of the banks in the $1 billion to $10 billion in assets cohort. They have stories that are not easily analyzed by summary spreadsheet. Investors must look at loan types, non-performing loan trends, capital levels and trends, and percent of allowance to total loans. And, perhaps most importantly, management. Because good management rarely falls victim to bad circumstances over the long haul.

There are deals out there. You have to put in the work to find them.

~ Jeff

The Real Reason for Bank Scale: Trading Multiples

"Get big or get out." "You must be twice the size that you are to succeed." These are bromides that some industry talking heads might be telling you. I hear it and read it frequently. And in today's social media, non fact-based opinion society, if you say it enough, people may start to believe it.

I moderated a strategic planning retreat with a bank that achieved top quartile financial performance. Their growth was solid too. Their asset size was less than $500 million. A director challenged me: Does our size matter so long as we continue to perform the way we have performed? My answer: Not really, with one exception.

Trading multiples. I referenced this phenomenon in a 2013 blog post, Too Small to Succeed in Banking. In that post I opined, "As we migrate towards greater institutional ownership, stock liquidity is becoming increasingly important." What I said then likely remains true today. Institutional owners (funds, etc.) now own two-thirds of shares outstanding in publicly traded US banks. 

But why does this matter to my sub $500 million in assets bank client?

So I ran some charts for you (courtesy of S&P Global Market Intelligence).

The bottom table was a bonus so readers can see that at the end of 2017 bank p/e's relative to the S&P 500 p/e surpassed their 10-year median in every asset category. Significantly so for banks $500 million to $5 billion in assets. So, on a relative basis, valuations are higher. 2008 was an anomaly because the S&P 500 companies traded at stratospheric p/e's because they had no "e".

Back to my main point. Over the past 10 years, banks that have less than $500 million in total assets have traded at lower price-to-tangible book multiples. In every year. And the differences in multiples match up nicely by asset size. The price-to-earnings chart is a little murkier based on the choppiness of earnings. However, investors tend to value smaller financial institutions more on book value than earnings. A good earner, like my client, tend to trade at relatively low p/e's because they have great earnings. 

Does that sound right? Earn better, and get rewarded with a lower p/e?

Fair is in the eye of the beholder. If you remove nearly 2/3 of the potential shareholder base because you have little daily trading volume, then you have less buyers seeking your shares. Supply and demand.

And that is where the economies of scale argument has merit. If you intend to remain independent, and continue to perform well and grow sufficiently, then you are likely delivering total returns acceptable to shareholders. Even without trading multiple expansion.

But if you would like to acquire a nearby financial institution, and you are trading at lower trading multiples than other would-be acquirers, you would be at a disadvantage. Your "currency" isn't worth as much as your larger competitors. Which may also make you vulnerable to an aggressive buyer's offer to buy you, if the buyer is large and has much better trading multiples than your bank.  

Fortunately, unsolicited offers are not common in our cordial industry. But we shouldn't rely on it.

~ Jeff

The State of Banking

Where are we and where have we been? Trends are telling. In 2013, there were 6,812 FDIC-insured financial institutions. At September 30, 2017 there were 5,737, a 16% decline. There were 166 mergers, and four new charters in the first three quarters of 2017. As an industry, the trend is down. Ski slope down.

What about the financial performance and condition of our industry? The Presidential State of the Union address was supposed to report to Congress the Administration's view of the condition and performance of our country. It has turned into a sea of words amounting to nothing more than a wish list and priorities. Because the nation's balance sheet and income statement is not improving.

But what of banking? 

I broke down banking's financial condition, performance, and trading multiples into thirteen charts, seen below. Charts 1-6 are financial condition trends, 7-10 are financial performance, and 11-13 are trading multiples.

The numbers are medians from all publicly traded financial institutions between $1 billion and $10 billion in total assets. That yielded 291 total institutions, broken down by region. 

Financial condition ratios are promising. So I will say to you that the condition of the industry is strong. Assets, Loans, and Deposits continue to grow, although at a more moderate pace. And capital ratios have held steady and strong. In fact, if you listen to some institutional investors, the industry is over-capitalized. Many view an 8% leverage ratio as the "right" number. Although this should depend on an institution's risk profile and growth trajectory. And I have never heard a regulator say the phrase "over capitalized".

Non-performing asset ratios are in a long term downward trend. They have leveled off in the 60-80 basis point range. Some regions are experiencing slightly elevated non-performers from the previous year. A trend to watch.

The challenge with industry balance sheets is that loans have grown faster than deposits over the past few years. And loan/deposit ratios are steadily increasing as liquidity positions steadily decrease. Many bankers are less concerned about this citing their access to wholesale funding to bridge any shortfalls, or that they have been mopping up excess liquidity.

But rates have been rising slowly, and I believe Fed rate increases will accelerate this year, perhaps crossing the rate threshold where depositors now care what you pay them, and dooming those Betas in your ALCO assumptions to irrelevance. My opinion is that one or two more rate increases will trigger more skirmishes on the deposit battlefield. Those that have not positioned their bank to have strong liquidity will have to compete, giving back deposit mix gains they have worked so hard to achieve.

Net interest margins have leveled off from long-term industry declines. Good news! In 2017, NIMs ranged from a high of 3.8% in the West and Southwest, to a low of 3.1% in the Northeast. Are these anomalies due to region, competition, or business models? I would argue a mix of all three. But if I were a Northeast bank, I would ask why other regions achieved between 3.5%-3.8% NIMs and we're at 3.1%. That's a tidy sum to leave on the table.

Efficiency ratio trends look fantastic! And since NIMs are holding steady, it leaves me to think that operating expense control or increased profitability in fee-based businesses are at work. Based on my firm's experience with the profitability of fee lines of business, I am guessing the former. As balance sheets grow, operating expenses grow less, creating greater efficiency. Positive operating leverage!

Both ROAA and ROAE declined 2016-17, although efficiency is better. So what doesn't the Efficiency Ratio measure? Provision, and income taxes. Most of the institutions, if not all of them, probably took a Deferred Tax Asset (DTA) writedown in the income tax line item, impacting these bottom line ratios. But this probably doesn't account for all of the decline. Are assets growing faster than profits, therefore reducing ROAA? Is equity accumulating faster than an institution's ability to deploy it, therefore reducing ROAE? Or, is provision expense up throughout the industry. I believe a combination of the three. But if provision expenses are rising, credits could be moving from pass, to watch, to substandard, and onward. Take note.

Trading mutliple trends are jolting. Banking is not a long-term growth industry. In a past blog post I wrote about the PEG ratio (P/E divided by EPS growth), and to keep an eye out for anything that moves too far from 1. I further deconstructed a bank's p/e ratio because the industry is more capital dependent than most, if not all industries. I estimate that a bank's p/e can be reduced by 5-7 times to calculate PEG due to high capitalization. If I took 7x, and applied it to the Mid Atlantic's 22.1x p/e (the lowest of the six regions), then those banks would have to achieve earnings growth rates of 15% to earn that valuation. Note that p/e measurements for the below charts were done on 12/31, after the new tax law passed, and after most bank's announced their DTA writedowns. But prior to their earnings releases, so the reduced earnings were not yet baked in the cake.

Banks might earn that valuation depending on how they take advantage of the new tax law. Do they take the one-time earnings injection, or make strategic investments for longer term earnings growth. If the former, I do not believe the p/e's will last long term. And therefore I believe, as an industry, bank stocks are likely at peak valuations.

Or as industry stock analysts put it: Neutral.


What are your thoughts on the state of banking?


~ Jeff


Note: I make no investment recommendations in my blog. Please do not claim to invest in any security based on what you read here. You should make your own decisions in that regard. FINRA makes people take a test to ensure they know what they are doing before recommending securities. I'm sure that strategy works well.


Source for all charts: S&P Global Market Intelligence

Are Banks Overvalued?

The S&P 500 Bank Index is up 41% in one year. US Regional Banks' price-earnings multiple was 16.6x and price to tangible book value was over 2x (see chart). So are banks over-valued?

It depends. One way to compare is to look at the p/e ratio compared to the market. The S&P 500 p/e currently stands at 24.6x. So it looks like bank stocks are not overvalued.

But hold on. One ratio that can help us out is the PEG ratio. Remember that in Finance class? It's the p/e ratio divided by the earnings growth rate. According to Peter Lynch's iconic book One Up on Wall Street, a stock is fairly priced if its PEG ratio was 1. Meaning if it's p/e is 16.6x, like the US Regional Banks mentioned above, then the earnings growth rate should be 16.6%. I know I'm comparing a multiple to a growth percent. But, hey, I didn't invent the PEG ratio.

The challenge with banks' PEG ratio, as the chart shows, is that it is way over 1, by a factor of over 5 (5.7). I checked it against other industries in the Financial Services sector. Insurance brokerage has a PEG of 3.4. Specialty Finance: 0.4. The regional banks' PEG ratio, if I do the reverse math, implies that earnings are growing around 3% for the banks in that index. Which is very close to the 3-year annual net income growth for all FDIC insured banks.

So by the PEG ratio, banks would appear to be over-valued. Which may be true. But I want to bring up two mitigating points about banks:

1.  Banks are capital intensive. We must contemplate that implicit in their p/e ratio is some level of their tangible book value.

2.  Banks are not, in general, growth stocks long term. Risk management and the legions of regulators work in tandem to limit growth. 

Relating to 1, I did a data run of all banks and thrifts between $1 billion and $10 billion in total assets that were profitable. I checked their median p/e ratio. I then took their market cap and deducted their tangible common equity to deconstruct their p/e between tangible book and their market cap over tangible book (see chart). 

It is true that other industries can deconstruct p/e in this fashion. But would such an analysis of other industries equate 56% of an industry's p/e to it's tangible book value?

Even if we deducted tangible book from p/e, the industry PEG would still be 2.53 (7.6x / 3% growth), more than double Peter Lynch's prediction of fairly valued.

Which brings me to 2. How long can a company, a sector, and an industry grow faster than its markets? Certainly not forever. And for many, earning their p/e's means stoking growth either through acquisition, or greater risk taking. One is risky, and the other can be deadly. 

For these reasons, bankers may want to consider more moderate growth objectives, maximize earnings, and pay a larger portion of shareholder returns in dividends. 

What is your opinion on bank valuations?

~ Jeff

Note: I make no investment recommendations in my blog. I have a difficult time with my own portfolio. 

Too Small to Succeed in Banking

Conventional wisdom: The onslaught of new banking laws, regulations, and regulatory activism requires scale to absorb costs. Or, the rapid pace of technological change and the sophistication of hackers requires resources not found in small community banks.

There is truth to conventional wisdom. But with all generalities, there are exceptions. And in banking, lots of exceptions. How do I tell the $250 million in asset client that I just visited that they are too small to make it, even though they sport a 1.47% ROA? The financial institution landscape is littered with banks and credit unions like my client.

Almost two years ago I wrote that bank shareholders were changing. (see The Coming Bank Consolidation) Community bank investors used to be the local insurance agent, mortician, and family that sits next to us in church. Today, many of traditional bank investors put their money in mutual funds, and leave the investing up to fund managers. No longer does the local barber show up at our annual meetings complaining about the pastries. Instead, professional money managers call to tell us how to run the bank.

As we migrate towards greater institutional ownership, stock liquidity is becoming increasingly important. Occasionally money managers call me with their criteria for their fund. One criteria is typically float and volume. Under 10,000 shares trading volume you say.... fuggedaboutit! 

My firm is occasionally called upon  to value banks that don't trade. Part of the valuation includes a discount for the lack of liquidity. Not a term foreign to other industries, by the way. In determining the discount, we look to trading markets to see the discounts applied, if any, to thinly traded bank stocks compared to their high volume brethren.

I recently ran an analysis of banks that trade over 10,000 shares per day, to those that trade 500-10,000 shares per day. The results are in the table below.

I controlled for financial performance (greater than 1% ROA), and asset quality (less than 2% NPAs/Assets). Of course, more performance and market data factor into trading multiples, but I couldn't control for everything. The result: Low Trading Volume financial institutions trade at a price/tangible book ratio of 110.5%, and a price/earnings ratio of 12.1x, compared to High Trading Volume FIs at 160.6% and 13.2x respectively.

What do I think community banks can do about the disparity?

1. Get a real investor relations program. Bankers think we should actively court investors when we need capital. Not so. Trading multiples are a direct result of supply and demand. If you want greater multiples, build demand... always.

2. Focus on retail investors. Community bankers think they can tap the institutional market. Investment banking firms tell them so. The truth is, institutions willing to invest in a bank that trades 2,000 shares per day are typically those that expect to exit the stock by selling the bank. How else can they exit your stock at such low trading volume? One benefit of having local, retail investors is they tend to have greater patience when implementing strategic change. To institutional investors, you are a number on a spreadsheet. If you take institutional money, plan their exit before they invest anything.

3. Perform. There is a positive correlation between financial performance and condition, and trading multiples, period. 

Banks that trade at low trading volumes trade at lower multiples and may be shut out of the institutional market for capital. Those that don't build retail investor demand for their shares may be required to sell if they need capital. You may be too small to succeed, but not because you can't deliver superior performance. But because nobody will invest in your bank.

~ Jeff