First National Bank Alaska

IThe business

A bank takes deposits, lends them out, and keeps the difference. First National Bank Alaska does that for individuals and businesses across Alaska from 28 locations in 19 communities, and it also runs a trust and wealth management department, a card business and a mortgage business.⁴³ It opened as the First National Bank of Anchorage on 30 January 1922 with $500,000 of capital.¹⁵ At 30 September 2026 it had $5.3B of assets, $2.8B of loans and equity of $581.3M, and it employed about 630 people.⁴⁵ It has no holding company: the bank itself is what you own, and it reports to the Office of the Comptroller of the Currency.³¹³

Two things set it apart from most banks. It holds a lot of cash and bonds, and a lot of capital. At the end of 2025 cash and securities were 44% of assets, and the bank has held between 44% and 60% for ten years C (calculated). Equity was 11.2% of assets, and the Tier 1 leverage ratio was 11.3%, against a 5% minimum for a well-capitalized bank.⁵³ The rest of this page keeps coming back to those two facts.

Where the money comes from

The bank reports one segment. Here is what the income statement looked like for 2025.³

Net interest income is 87% of revenue C (calculated). This is a spread bank, and the spread is what the rest of the page is about. Fee income of $29.1M is small and steady: it was $28.2M in 2024.³ The bank spends 50 cents to earn a dollar of revenue, which is a good ratio.⁵

What it owns

At 31 December 2025 the bank had $5,099.9M of assets: $258.7M of cash and balances due from banks, $1,989.0M of securities, all available for sale, and $2,712.0M of gross loans. Against them were $1,827.1M of noninterest-bearing deposits, $1,907.6M of interest-bearing deposits, $739.1M of overnight repurchase agreements, $56.9M of borrowings and other liabilities, and $569.2M of equity.³ By 30 June 2026 cash was $339.0M, securities $1,826.9M, loans $2,822.5M and equity $585.2M.²¹⁵ At 30 September 2026 the bank reports assets of about $5.3B, deposits and repurchase agreements of $4.6B and equity of $581.3M. It had not posted its detailed third-quarter balance sheet or call report when I wrote, so detail below is at 30 June 2026.⁴

The bank is lending more: loans were 46% of assets in 2016 and 53% in 2025, and gross loans grew 10% in 2025 alone.³ A large part of the bond book is also not idle. At 30 June 2026, $1.18B of the $1.83B was pledged as collateral, and at the end of 2025 $781M of the pledged bonds secured the repurchase agreements.¹³³ Those repurchase agreements are overnight secured borrowings. They cost interest at short-term rates, $21.2M in 2025, and they are not deposits. The bonds are also pledged to secure public deposits, and deposits of states and local governments were $346M at the end of 2025.³

Where it lends

The bank lends mostly against property. At the end of 2025 its gross loans were $2,725.7M before deferred fees.³

Two things stand out. First, this is a commercial real estate lender in a small economy: the number of nonfarm jobs was 354,000 in June 2026, down 900 from a year earlier.¹⁴ Second, the bank bought loans. The cash flow statement shows $208.8M of loan participations purchased in 2025 and $287.7M in 2024, while gross loans grew $242M in 2025 and about $194M in 2024 C (calculated). Without the purchases, other net lending was a gain of $41.8M in 2025 and a decline of $96.7M in 2024.³⁵ So most of the recent growth was bought. The notes say that $40.3M of the 2025 purchases and $68.9M of the 2024 purchases were government-guaranteed loans, and that is all they say. I cannot see where the rest of the loans are, who originated them, or how they are rated.

Where the customers are

The FDIC counts deposits at every branch. At 30 June 2025 Alaska had $16.04B of branch deposits. Wells Fargo had 42.2%, First National 22.4%, Northrim 17.5%, KeyBank 6.8% and First Bank 5.0%.⁶ In 2016 the shares were 50.5%, 20.1%, 10.5% and 10.3% for Wells Fargo, First National, Northrim and KeyBank.²² First National gained 2.3 points. The shares count branch deposits and say nothing about how many customers a bank has.

The state economy matters more here than it does for a national bank. Oil is about 25% of the state’s general-purpose revenue and the Permanent Fund transfer is about 60%, and Alaska has no state income or sales tax.²³ The state’s forecasters put the base case for North Slope oil at $75 a barrel for the year to June 2027, with a one-in-ten chance of $130 or more and a one-in-ten chance of $45 or less.²³ The bank does not give a breakdown of loans by industry, so I cannot say how much of its book depends on oil.

IIThe moat

A durable advantage is a cost or a habit that a rival with money and talent cannot copy in a reasonable time. A bank’s edge is often narrower than it looks, because loans are a commodity and a customer can usually go elsewhere. This bank’s case rests on deposits and on place. I can test each against numbers.

The deposits

Almost half of the bank’s deposits, $1,827M of $3,735M, are in accounts that pay no interest.⁵ That share was 53% in 2016, rose to 59% in 2021 when customers held cash, and has fallen back to 49% as rates rose C (calculated). In 2025 the bank paid $26.0M of interest on $3,707M of average deposits, 0.70% C (calculated).³ Adding the $21.2M it paid on repurchase agreements, the cost of all customer funds was 1.1% C (calculated). Those are low numbers in a year when the federal funds rate averaged 4.2%.³²⁴

Deposits like that are the main thing a bank can own. A business keeps its payroll account where it banks, and an Alaskan family keeps its checking account where its parents did. The bank has been in Anchorage for 104 years.¹⁵ I cannot measure loyalty directly. Deposits did drift away as rates rose: total deposits were $3.73B at the end of 2025 against $4.22B at the end of 2022, down 12%, and the noninterest-bearing balance fell 27% from its 2021 peak C (calculated).⁵ Deposits of the state and local governments were $346M at the end of 2025 against $412M a year earlier.³

The place

The second source of strength is that Alaska is a small market where size and branches matter. First National holds 22.4% of the state’s branch deposits, second to Wells Fargo at 42.2%.⁶ The bank calls itself the state’s largest locally owned and operated bank, with 28 locations in 19 communities.¹⁵ A branch in a small town is expensive to build and hard to justify for a rival with fewer customers. The FDIC data show deposit shares and nothing about branch counts, so I cannot test that.

What is not an advantage

Loans are a commodity: the loan yield is a market price for risk, and a competitor who funds at a similar cost can match it. Fee income is 13% of revenue, with no scale advantage I can see in cards or trust. Reader surveys that call the bank Alaska’s best place to work are the bank describing itself with praise from magazines, and I do not count them as a barrier.⁴ The advantage protects a margin of 3% to 4% and a fee line of $29M. It does not let the bank earn 20% on capital.

IIIThe competitive advantage

The last section was about why a rival cannot copy the bank. The advantage is what the bank earns because of it. For a bank, the best single test is the profit it makes on the tangible equity it holds. I set it beside the one other bank of its kind in Alaska, which is public and files with the SEC, and two large banks I have already valued.

Profit on assets is excellent, profit on equity is not

A bank’s return on equity is its return on assets times its assets per dollar of equity. First National earned 1.74% on its assets in the first nine months of 2026, which is high for any bank.⁴ Northrim earned about 1.65% in the third quarter C (calculated).⁹ But First National holds 10.9 cents of tangible equity per dollar of assets, and Northrim holds 8.5 cents C (calculated). So the bank that makes more on its assets makes less on its capital.

This is a choice. If First National ran at Northrim’s ratio of tangible equity to assets, it would carry about $452M of tangible equity and earn about 20% on it, on the same profit C (calculated). The difference is about $128M, or $40 a share C (calculated). I am not arguing for a return of it. A bank in a small, cyclical state that holds a large buffer is doing what the owners want, and the buffer is the reason its credit record is good and its dividend was held in 2020. But the price you pay for the buffer is a return that is lower than it would be.

There is a limit on how fast the money can leave. The bank’s statements say the OCC must approve dividends above the current year’s net profit plus the retained profit of the two years before. At the end of 2025 that retained amount was $30.6M, which matches the retained profit of 2024 ($16.4M) and 2025 ($14.2M) C (calculated).³ So a one-time return of $128M on top of the regular $63M of dividends would be about $191M, against a limit near $121M in a year in which the bank earns about $90M C (calculated). I have no sign that the bank plans anything of the kind.

What the numbers say about quality

The efficiency ratio is 50% for 2025 and 48.6% in 2026, net charge-offs averaged 0.04% of loans for ten years, and the risk-weighted capital ratio was 18.3% at the end of 2025, more than double the 6.5% a well-capitalized bank needs.⁵⁴³ That makes a bank that is hard to hurt. It does not make one that earns a lot on a dollar of capital: on a clean measure it earned 10.6% to 10.8% for six years, about what I ask of its equity.

IVManagement

Betsy Lawer joined the bank in 1974, started as a secretary, and worked in nearly every department before she became president on 1 May 2013, succeeding her father, Dan Cuddy, who had led the bank for more than sixty years.⁷¹¹ The bank says she has been Board Chair and CEO since 2015.¹⁰ On 7 October 2026 it announced that she will become Executive Chair on 2 January 2027 and that Darren Franz, now President of Lending and Administration and Chief Credit Officer, will become CEO and President. Franz brings more than three decades of Alaska banking and nearly seven years at the bank, and he sits on the board.¹⁰

Who owns it

I cannot give a holder list. The bank has no holding company and is not an SEC registrant: a search of EDGAR returns no filer by that name, and the SEC’s own ticker file has no FBAK.³⁰ It publishes audited statements, quarterly overviews and call reports, and I found no proxy statement, no beneficial-ownership table and no executive compensation table among them.¹² What I can say is that its press release calls it a Minority Depository Institution “as it is majority-owned by women,” that three of the former president’s daughters are on its board, and that the stock is worth $1.07B at the close.^1072931 I do not know how much of the stock the family holds, or how much of it ever trades.

How they are paid

I cannot follow the incentives as I would for a listed company, because they are not published. The statements show no stock-based pay at all: the equity statement moves only with income, dividends and the value of bonds, and the common stock and surplus have not changed since 2020.³¹⁷ Salaries and benefits were $72.8M in 2025, up 6% from $68.7M, for about 625 employees, an average of about $116,000 C (calculated).³⁵ The profit-sharing and 401(k) contribution was $2.8M.³ A non-qualified deferred compensation plan for some executives began in 2025, with a liability of $0.7M.³ A compensation committee exists; its decisions are not published.

What that implies is guesswork. If a family holds a large share of the stock, it is paid by dividends and by the value of the shares, and the incentive is to keep the bank safe and steady. That fits a thick buffer, low credit losses and a dividend held in 2020. It also fits a bank that keeps about $128M more capital than Northrim’s ratio would imply. I cannot tell whether the owners prefer that.

What they did with capital over a cycle

Over the ten years from 2016 to 2025 the bank earned $566M and paid $457M of dividends, 81% of earnings C (calculated).⁵ Equity rose from $493M to $569M, and with bond gains and losses taken out, from $499M to $594M, 19% in nine years C (calculated). That is a bank that pays out nearly everything and grows its capital more slowly than its balance sheet, which rose 41% over the same years C (calculated).

The dividend record has one cut. Dividends paid fell from $33.3M in 2009 to $16.6M in 2011, and stayed near $16M a year until 2016, when they rose to $27.0M, then to $31.7M in 2017 and $36.1M in 2018 R (reported).⁵ I did not find a reason for the 2011 cut in the documents I opened. The bank did not cut in 2020, and it paid a special dividend in December 2025 on top of the regular $4.00 a quarter, and has declared another for December 2026.³²⁵

The bank has also bought stock back, but not lately. Since November 2008 it has been authorized to repurchase up to $100M of its stock, and it bought 280,705 shares for $45.5M, an average of $162 a share C (calculated); it bought none in 2024 or 2025, and the authority was set to expire on 22 April 2026; I did not find whether it was renewed.³ The average of $162 is below today’s $338. In 2024 it sold its last 41,987 Visa Class B shares for a gain of $15.6M, about $11.6M after tax, which is why 2024 earnings were 17% above the bank’s run rate C (calculated).³ It has invested $33.3M in affordable-housing projects for tax credits, with $20.1M of further commitments.³

The bank has been careful with its credit and it pays out. Without a stated target return or a recent buyback, the owner’s return depends heavily on interest rates. I flag two things I cannot see: what changes when the founder’s heir hands the chief executive role to a banker, and whether the family would accept an offer.

VFinancials

The ten-year table has the figures the rest of this page leans on, with the nine months of 2026 in the last column. Four breaks in the series are worth flagging. The shares split ten for one in 2018, so earnings, dividends and book value per share are shown on the post-split count throughout. The tax rate fell from 41% in 2017 to 24% in 2018 C (calculated). Bond yields rose in 2022 and put an unrealized loss of $146M into equity, which I take out where I say so. And in 2024 a $15.6M gain on Visa shares is in profit.⁵¹⁹⁸³

The picture is a bank whose returns rose with interest rates and did little else. On a clean measure of tangible equity the bank earned 8.3% in 2016 and 7.3% in 2017 and then between 10.6% and 10.8% every year to 2023 C (calculated). That is about what I ask of its equity. The jump came in 2024 and 2025, to 11.8% and 13.2%, and it continues in 2026 at about 15.6%. The efficiency ratio fell from 57% to 50% and net charge-offs have been close to nothing, so the thing that moved is the margin.

The margin was 3.74% in 2019, 2.86% in 2021 and 2.77% in 2022 as the federal funds rate went from above 2% to almost nothing and then up. It was 3.88% in 2025 and 4.05% in the nine months of 2026.⁵⁴²⁴ The federal funds rate averaged 0.08% in 2021 and 4.21% in 2025, a gain of 4.1 points, while the margin rose 1.0 point, about a quarter of a point for each point of short rate C (calculated).²⁴⁵ That ratio is a crude measure of how the bank’s earnings respond to rates, and the sections below try to improve on it.

The latest five quarters

The third quarter was the best the bank has had: net income was $25.2M, 21% above the second quarter and 18% above the third quarter of 2025.⁴²¹ I can account for part of the jump from published numbers. The provision swung from a $1.2M charge to a $0.1M benefit, fee income rose $0.6M, expenses were about flat and the year-to-date margin rose.⁴²¹ The release gives no net interest income or balance sheet for the quarter, so I do not know whether it had a one-time item. I treat it as a possible high point and do not annualize it.

The bond book, and what it will do

The bank’s securities are $1.83B, all classified as available for sale and carried at market value, so the unrealized loss is already in equity. At 30 June 2026 the bonds were worth $35.2M less than their cost before tax, 1.9%, and the after-tax loss of $25.2M was 4.3% of equity C (calculated).¹³²¹ Treasuries were $758M, agency bonds $542M, mortgage securities $333M and municipal and corporate bonds $194M.¹³ Only 26% of the book matures or reprices in more than five years.

This is the part of the bank that most people miss. 45% of the bonds, $813M, mature or reprice within a year, and the whole book earns about 2.9%. If every dollar is replaced at the two-year Treasury yield of 4.61%, the bank earns about $14M a year more before tax, or $11M after tax, 12% of my 2026 profit estimate E (estimate). The bonds that mature in one to three years add about $8M more. These are upper bounds. I cannot see the yield by bucket, and some of the bonds that mature this year may already yield more than the average. The direction is still clear: if rates stay where they are, the margin has room to rise further.

What reprices in a year

At 30 June 2026, $819M of the loans, 29%, repriced within three months, $1,047M, 37%, within a year, and another $561M in one to three years.¹³ Put the two sides together.Within a year, $339M of cash, $813M of bonds and $1,047M of loans, $2.2B or 43% of assets, reprice or mature C (calculated).¹³ On the funding side, $785M of repurchase agreements are overnight and the $362M of time deposits outstanding at the end of 2025 were due in 2026, which together is $1.15B that costs what the market costs.²¹³ The $1.72B of noninterest-bearing deposits, 34% of assets, cost nothing at any rate.⁵ The rest of the interest-bearing deposits, about $1.5B, are savings, sweeps and money market accounts whose rates the bank sets.³ If those follow short rates by 20% to 60%, a one-point drop in all short rates costs the bank about $1M to $7M of pretax profit in the first year E (estimate). That is a crude static count, and the bank does not publish a sensitivity table that I could find. It says the bank is not strongly exposed to a modest fall in short rates near 4%. The exposure is to a return to the margin of 2024.

Read the table as a range. At the 2024 margin the bank earns about $61M, $19 a share and 11% on tangible equity, which is where it was before the last two years. At the 2026 margin it earns $90M. One point of margin is worth about $36M a year after tax, $11 a share, 40% of 2026 profit C (calculated). No other number on the page matters as much to the price.

Credit

The credit record is very good. Net charge-offs averaged 0.04% of loans over ten years and were never above 0.13% in a year.⁵ The recent signs are less good. Nonperforming loans rose from $4.3M at the end of 2024 to $10.6M at the end of 2025 and $13.2M in June 2026, then fell to $6.8M, 0.24% of loans, in September.⁵⁴ Loans the bank rates substandard or doubtful, a wider measure, were $45.5M at the end of 2025, 1.7% of loans, against $36.8M and $34.9M at the end of the two years before; special mention loans were $14.0M.³¹⁶⁸ Together that is $59.5M, 2.2% of loans and 10% of equity C (calculated). The allowance for credit losses was $21.5M, 0.79% of loans, down from 1.07% in 2021, and it was 0.76% in September.³⁴

The loan mix is what makes the thin allowance worth a second look. Commercial real estate, multifamily and construction loans were 261% of equity at 30 June 2026, and loans on property the borrower does not occupy were 180% of equity C (calculated).¹³ A 5% loss on that $1.5B would be $76M before tax, about 13% of the bank’s tangible equity C (calculated). The bank would absorb it, with earnings of about $90M a year and a leverage ratio near 12%, and that capital is what the buffer is for. But earnings would take two years to recover, and during them the multiple would not last.

Capital

The bank is strongly capitalized by any measure. At the end of 2025 its total risk-based capital ratio was 19.05%, its Tier 1 ratio 18.33% and its Tier 1 leverage ratio 11.27%, against 10%, 8% and 5% for a well-capitalized bank.³ At 30 September 2026 the leverage ratio was 11.81%.⁴ Regulatory capital leaves out the unrealized loss on the bonds. The OCC’s dividend limit is the current year’s profit plus the retained profit of the two years before, which was $30.6M at the end of 2025.³ One more risk is on the funding side: 42% of deposits were uninsured at the end of 2025, $1,572M, and the public deposits of states and local governments are $346M of the total.⁵³

VIValuation

Start with what would go wrong

Before any numbers, I ask what would most likely make this a poor investment from here, and work back from that. The bank is sound. These are the three ways it goes badly wrong, in the order I think most likely.

1. The margin falls back and the multiple follows. The bank earned 8% to 11% on a clean measure of tangible equity until 2023. Nearly all of the gain since is the margin, and one point of margin is worth $36M of profit a year C (calculated). If short rates fall and the bank has to pay more for deposits, the margin returns to the 3.24% of 2024, profit is about $61M, $19 a share, and the price of $338.00 is 18 times that C (calculated).⁵³ Nothing in the business has to break for that to happen.
2. One bad stretch in a concentrated loan book. 58% of loans are commercial real estate, loans on property the owner does not occupy are 180% of equity, the economy is one state’s, and a growing share of the loans were bought from others.³¹³ A 5% loss on the $1.5B of commercial property, multifamily and construction loans is $76M, 13% of tangible equity, and the bank would survive it C (calculated). The price, which assumes no such year, would not.
3. The buyer is not there when a holder needs to exit. The shares traded about 300 a day over 251 trading days, a median of 200 and a maximum of 2,800 C (calculated).¹ A $1M position is about 3,000 shares, ten days of the average day’s total volume. The price of $338.00 is the last trade, 46 shares on 4 November, and the average day’s turnover is $94K. If a holder with a large stake needed to exit, or two or three buyers stopped bidding, the quoted price could move a long way on small volume, and today’s close is within 1% of the highest since the 2018 split, $340.00 on 5 May 2026 C (calculated).¹

I can show what the first does, and I do below. The second I put in the bear case. The third is not in my values at all, except through the added return I demand for it, which I show as a separate item.

What the price implies

The shares are at 1.84 times tangible book and 12.3 times earnings. A buyer who sees today’s earnings as permanent can reasonably pay that. The question is whether they are permanent. Since the 2018 split the price rose from $252.00 to $338.00, 3.8% a year, and with dividends of $14.40 to $20.00 a year the total return was about 9.7% a year C (calculated).¹¹⁸³ Tangible book per share, leaving out the bond losses, rose 17% in that time. The multiple of earnings has not moved much: the year-end price was 15 times that year’s earnings in 2018, 13 in 2021, 12 in 2025 and 12 on my 2026 estimate C (calculated). What moved is earnings per share, up 62% from 2019 to my 2026 estimate against a price up 39%, so the market is paying the same multiple of a bigger profit, and tangible book is a smaller part of the price.

What an owner can take out

Reported earnings and owner cash are not the same. A shareholder’s share of the profit is free to leave the bank only if the bank keeps the equity it needs to stand still, which for a bank is equity growing with the balance sheet. The bank pays no stock-based compensation that I can find, so there is nothing to deduct for it.³

Owner cash is 77% of reported earnings, a gap of $20M that the bank keeps. That is close to the dividend: the bank pays out nearly all that it can. The owner’s return from today’s price is mostly the 6.5% of owner cash plus the growth the retained equity buys. At a 10.9% hurdle that leaves about 4 points of growth to find, from a bank growing its equity 3.5% to 4% a year C (calculated).

How I value a bank

For most businesses on this site I discount free cash flow. A bank cannot be valued that way, because its cash flow statement moves with the size of the balance sheet. I use the same method as on my Bank of America and JPMorgan Chase pages: an excess-return model, which values the shares as tangible book plus the present value of what the bank earns above its cost of equity.²⁷²⁸ A dollar of tangible equity that earns exactly the cost of equity is worth a dollar, and one that earns more is worth more. It gives the same number as discounting the cash an owner can take out each year, after keeping equity growing with the balance sheet. I do both, and they agree to the cent.

About 47% of the value sits beyond 2036, so it is sensitive to what the bank earns in ten years. Of $242 a share, $183 is the book itself and $8 is the fourth-quarter dividend. A bank earning one or two points over its cost of equity does not earn a big premium to its book, and for most of the last decade this one did not earn it at all.

Reverse the question: what return does the price need?

I take my base growth and capital path and ask the opposite question. What return on tangible equity, held every year for ten years and for ever after, makes the shares worth $338.00? I solve it.

Inputs: a 10.9% cost of equity, 2027 to 2036 and growth for ever after, tangible equity growing 4.0% a year to 2031 and 3.5% after, a payout of whatever profit is left after that growth (70% to 75%), no preferred stock, $578M of tangible equity at the end of 2026, 3.5% growth after 2036 with the return held at the solved level, and 3,166,885 shares E (estimate).³⁴

The answer is a return on tangible equity of 16.6% every year C (calculated). With the faster capital release and growth of my bull case it is 16.9%, and with the slower growth of my bear case, 17.4% C (calculated). A shorter route gives the same sense. If the bank earns a return R for ever and grows at g, the multiple of tangible book is (R minus g) over (cost of equity minus g). At today’s 1.84 times, R has to be 17.1% if growth is 3.5%, 17.6% if it is 3% and 16.7% if it is 4% C (calculated). That is a little above the 15% to 16% the bank earns in 2026 and a lot above the 11% it earned in the six years before 2024.

The price pays for today’s margin to last, and then some. It pays for a bank that stays at a peak. At a flat 10%, near the six-year average, the shares would be worth about $170; at 12%, $221; at 15%, $297; at 16%, $323; and at 17%, $348 E (estimate). The price sits between the last two.

The discount rate

I discount at 10.9%. It has three parts, and I list them because two are judgements and they matter. The first is the capital asset pricing model. The ten-year Treasury yield was 4.95% and Damodaran’s implied equity risk premium was 4.41% on 1 November, up from 4.20% a month earlier, which is the figure on my Bank of America and JPMorgan pages.³²³⁴ The bank’s own stock is too thinly traded to have a beta. As a proxy I used Northrim, the other Alaska bank, and measured its monthly price moves against the S&P 500 index fund: 1.39 over 36 months, 0.93 over 60 and 0.82 over 119 C (calculated).³⁵²⁶ The average is 1.05, which gives 9.6%. On my Bank of America page I took the highest of three betas, because a large bank’s beta reads low in calm years. Here I take the average, because the proxy is a different bank with its own noise, and the 1.39 reading alone would give 11.1% before the add-ons. The second is 1.0 point for the shares’ thin trading on an over-the-counter market and the thin disclosure. The third is 0.3 point for a business with one state’s economy behind it.

The last two are my judgement, and they change the answer. With the add-ons removed, at 9.6%, my base is $291, my bull is $373 and my bear is $184 E (estimate). The price is then 16% above the base and 9% below the bull. I keep them in because a holder who cannot exit without moving the price, or see executive pay, has reason to ask for more.

Three cases

My base case is the bank keeping most of what it has and giving some back. The margin eases from 4.05% to about 3.9% by 2028, because deposits get dearer and short rates drift lower, and the return falls from 15% to 12.5% by 2031 and stays there. That is above the six-year average before 2024 and below the 2026 pace. Equity grows with the balance sheet, so earnings per share are flat at about $27 to $28 for four years and then grow 3.5% a year E (estimate). The cash I model for the owner is $20.62 a share in 2027, easing to $18.68 in 2031, close to the $20.00 paid for 2025 and declared for 2026, so it counts the special dividend as part of the regular payout.

The bull case has the bond book reprice and the bank return some of its surplus capital, with a return above 15.5% for ten years. The bear case has the margin return to its 2024 level with higher credit costs: net income falls to about $55M, a return of 9%, below my cost of equity. Even the bear is a profitable bank that pays a dividend; its value of $158 is about 14% below today’s tangible book per share.

What I can’t judge

Some inputs I can’t know, and I would rather say so than pretend. The first is who owns the bank and what they want. The bank does not publish a holder list, a proxy or pay. Whether the family would accept a bid, issue stock, or keep paying out as it does is a question I can only guess at. The second is the loan book: I can see the classes, but not the property type, geography or loan-to-value, and I cannot see the participations bought from other lenders. The third is the bond yields by bucket and the bank’s own measure of its rate risk, which would let me test the $14M roll and the crude static count above. The fourth is the third quarter, which beat the second by $4.4M with no published income detail. The fifth is executive pay. The sixth is the trading market: with 300 shares a day I cannot tell what a large order would do. The seventh is why the dividend was cut in 2011.⁵ The range between my bear and bull is wide because these are wide questions.

At 1.25 times, near the multiple in my base value, the price returns about 6.1% a year to 2036, 1.1 points over a 4.95% Treasury. At today’s multiple it returns about 9.0%. Either is below my hurdle of 10.9%.

Multiply it out

The price needs several things together: a return near 16% for ten years (I would say 15%, since the bank has done it for less than two), credit costs near zero through a cycle (60%), and the payout, capital and control unchanged through a new chief executive (85%). Multiplied, that is 8% E (estimate). They overlap, so the true figure is higher, perhaps one in ten.

Quality and price are two questions

How good is the business? Good. It has a large local deposit base that costs almost nothing, a record of losses near zero, a very high capital ratio, and an efficiency ratio below 50%. How good is the price? High against my range. It is 40% above my base value and 9% above my bull value, at the highest multiple of book in the data, with the added problems of an OTC quote and thin disclosure. This looks like a good business at a price that assumes today’s margin lasts, and I say so without calling it either a bargain or a trap.

It helps to compare it with the next best use of the same dollar. A ten-year Treasury pays 4.95% with no risk to the promised dollars.³² This bank returns about 8.7% a year on my base case, nearly four points more, against my required 10.9%. The closest comparison on this site is Bank of America: its price was 8% above my base value there, it returned about 10.1% a year on my base against a required 10.6%, and it trades on the New York Stock Exchange.²⁷ The shortfall to the required return is 0.5 points there and 2.2 points here. Of the two, Bank of America is the better-priced on my numbers. A closer comparison, which I have not valued, is Northrim: about 19% on tangible equity at 1.79 times book, an SEC filer, trading about 40 times as much stock by value each day.⁹²⁶ On the crude formula with a 9.6% cost of equity its multiple needs about 14% held for ever. I do not know its credit, so I draw no conclusion, but it is the obvious alternative.

The margin of safety

My margin-of-safety price is $157, 35% below my base value. At $157 my bear case returns about 10.9% a year, close to the 10.9% I require, and the base returns about 15.3% C (calculated). I use the same 35% cushion as on the Bank of America and JPMorgan pages, for the same reason: I cannot see a bank’s credit from outside, and here I also cannot see who owns it or how the shares trade. $157 is 0.86 times tangible book. The price is 115% above it. The stock has closed below $157 once since the 2018 split, on 23 March 2020, at $125.00 on 1,900 shares.¹

Where this could be in 5 and 10 years

These are my estimates of what a share is worth at later dates, from the model above. The value at a date is the present value at that date of the owner cash that comes after it, on today’s share count and in that year’s dollars. It includes the value of the bank beyond 2036 and leaves out cash paid before the date, which the fourth column adds up from now to the end of 2036, including the fourth-quarter dividends of 2026.

For the base to stand at five years, the return has to ease from 15% to 12.5% by 2031 and earnings per share have to stay near $27. That pays out about $105 a share by then E (estimate). For it to stand at ten, the return has to settle at 12.5%, which is worth about 1.22 times tangible book; $321 is that multiple on $264 of tangible book per share E (estimate).

The most likely way the base is wrong is the fade. If the return stayed at 16% for ten years and after, instead of easing to 12.5%, a share would be worth about $445 at the end of 2036 and $323 today, a little below the price E (estimate). The odds in this section are my guesses and are not used in the values, except that the weighted value of $238 uses 25%, 50% and 25%. These are estimates of value, not forecasts of the share price. The market price can sit above or below value for years, and on a market this thin it can move on very little.

VIIThe case for the price

What most people miss

First, the bond book is also a source of future profit. The bank holds $813M of bonds that mature or reprice within a year, earning about 2.9% on the whole book, while the two-year Treasury pays 4.61%.¹³²¹³² Replaced at that yield, the gap is worth up to $14M of pretax profit a year, 12% of 2026 pretax profit E (estimate). Second, the bank earns more on its assets than Northrim and less on its capital, because it holds 10.9 cents of tangible equity per dollar of assets against 8.5. The difference is about $128M, $40 a share C (calculated).⁴⁹ Third, net charge-offs averaged 0.04% of loans over ten years, and in 2020 the bank held its dividend while its stock closed as low as $125.⁵¹

The bull case

The case is a century-old deposit franchise whose earnings are still catching up with rates, with a clean credit record and capital to spare. I found no named investor, analyst or price target on this stock in primary sources. What exists is the bank’s own account. On 4 November 2026 Betsy Lawer said in the third-quarter release that the quarter showed higher net income, an improved margin and strong asset quality.⁴ On 7 October she described the leadership change as the end of a long-term succession plan.¹⁰ The argument is that the margin holds above 4%, the bonds keep rolling up, the efficiency ratio stays under 50%, and the bank keeps paying out $20 a share.

Against reality

Much of that is true today: the margin is 4.05%, the efficiency ratio 48.6%, nonperforming loans 0.24% of loans and the dividend $20.00.⁴²⁵ If the margin holds above 4.1% for years and the bank returns some capital, my bull case gives a return averaging 16.0% and is worth $311 E (estimate). If the bank does that and hits its targets, $338.00 is 9% above that value at my 10.9% rate, and 9% below it at 9.6%, without the add-ons. The price needs 16.6% held for ten years, against 15.6% so far in 2026, 13.2% in 2025 and a clean 10.7% average for 2018 to 2023 C (calculated). The bank has done what the case needs for only two years, and not through a rate cycle.

What a 10x would take. Ten times $338.00 is $3,380 a share, about $10.7B of market value C (calculated). At today’s multiple of earnings that needs about $902M of net income. Revenue in 2025 was $217.9M, net interest income plus fees, and the net margin was 35.6%, so $902M of profit at that margin is about $2.5B of revenue, 11.6 times 2025’s C (calculated).³ Revenue grew 4.4% a year from 2016 to 2025, which would take about 57 years. At the 1.74% return on assets that is about $52B of assets, more than three times all the branch deposits in Alaska, $16.0B at June 2025.⁶⁴ The price has risen 3.8% a year since 2018 C (calculated).¹ A tenfold move is not plausible for a bank of this size in one state.

Where I come down

The bond roll, the clean credit record and the cost control moved me: they are why my base holds 15% in 2027 and not a straight fall to the 11% of the years before. The multiple and the family’s unknown plans did not move me. A margin above 4.1% into 2028 on flat or falling short rates, or a return of surplus capital, would raise my values; a margin heading toward 3.5% or a loss on commercial property would lower them.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• OTC market risk. The shares are quoted on OTCQX, a tier of the OTC Markets Group. It is not a national securities exchange: the SEC’s list of those names Nasdaq, the New York Stock Exchange, Cboe and others, and does not name OTCQX.²³⁶ The shares traded 300 a day on average, and the last two sessions were 200 and 46 shares at $338.00.¹³¹ A holder who needs to exit quickly may have to take a price well below the last trade.
• Limited disclosure. The bank is not an SEC registrant, so there is no 10-K, no proxy statement, no insider-trading reports and no required earnings call. It publishes audited annual statements, short quarterly releases and call reports. The detail I rely on for the latest quarter was not out on 4 November. I cannot see executive pay, director stock ownership or who the large holders are.³⁰¹²⁴
• Interest rates and the margin. 40% of 2026 profit rides on one point of margin. The ten-year Treasury yield is 4.95% and the three-month bill 4.21%, above the 3.88% fed funds rate, a sign that markets may expect higher short rates, and national inflation was 4.2% in May.³²¹⁴ A fall in the short rate with rising competition for deposits would shrink the margin. A rise would widen the $25M after-tax loss on the bonds.
• Credit and the loan mix. Commercial real estate is 58% of loans, and a share of loans was bought from other lenders without a disclosed location. Nonperforming loans rose 2.5 times in 2025 before they halved in the third quarter of 2026. The allowance is 0.76% of loans, lower than it was in 2021.³⁴⁵
• One state. Alaska had 354,000 nonfarm jobs in June 2026, 900 fewer than a year before.¹⁴ The state budget relies on oil for about a quarter of its revenue and the Permanent Fund for about 60%, and the oil forecast ranges from $45 to $130 a barrel in the same year.²³ I cannot tell whether the purchased loan participations are inside the state or outside it.
• Deposits. 42% of deposits were uninsured at the end of 2025, and 9% are public funds that need collateral.⁵³ Deposits fell 12% from 2022 to 2025.⁵
• Control and succession. Three of the thirteen directors are sisters, the chair becomes Executive Chair, and Franz becomes chief executive on 2 January 2027.¹⁰⁷²⁹ A change of chief executive can change how capital is used, for better or worse. I cannot see how many shares the family holds, so I cannot say how much say other holders have.

What would raise my base value: fourth-quarter results in February showing a return of 16% or more on normal credit costs, a margin above 4.1%, and a board that signals a return of capital. What would lower it: a margin drifting toward 3.5%, nonperforming loans climbing again, or a bank that grows by buying loans.⁴