BOK FINANCIAL CORP (BOKF) stock intrinsic value fair value estimate & owner earnings analysis

BOK Financial is a Tulsa-based bank holding company (about $53 billion of assets) operating BOK Financial, Bank of Texas, Bank of Albuquerque, Bank of Arkansas, Bank of Kansas City and CoBiz Bank across Oklahoma, Texas, Colorado, Arizona, New Mexico, Arkansas, Kansas and Missouri. It is controlled by its chairman, George Kaiser, who owns about 63% of the shares. Two features set it apart from a typical regional bank. First, fees supply about 38% of revenue, led by a trust and asset-management business overseeing about $127 billion, plus mortgage servicing and securities trading. Second, its balance sheet is conservative: loans were about 65% of deposits at the end of 2025 and its core capital ratio was 12.9%. The main concentrations are healthcare lending (mostly senior housing, $4.0 billion or 16% of loans) and energy ($2.9 billion or 11%, down $372 million in 2025, three-quarters to oil and gas producers), a specialty the bank has run since its founding. What the owners earn. For a bank, deposits are raw material and interest is an operating cost, so we treat earnings available to common shareholders as the owners' cash, with no separate debt deduction. Per share, these were $3.53 in 2016 (the energy-bust trough), $5.11, $6.63, $7.03, $6.19 in 2020 (a large pandemic reserve build), $8.95 in 2021 (flattered by releasing that reserve), $7.68, $8.02, $8.14 and $9.17 in 2025. The 2018 purchase of CoBiz reshaped the bank, so we judge it from 2019. The median of 2021-2025 is $8.14. Three straight quarters since then ran well above that: first-half 2026 earnings were $5.49 a share against $4.05 a year earlier, as net interest income rose 8% and fees 14%. Those quarters set aside nothing for loan losses, though, and 2025 set aside only $2 million. After charging a normal $40 million a year or so, the recent pace is about $10.40. We start the base case at $9.00. That is above the five-year median because three quarters back it up, and below the recent pace because today's near-zero loss provisions will not last. Earnings rose 12.7% in 2025 and the base case assumes a fall of about 2% in its first year. Share count has fallen from 70.8 million in 2019 to about 60 million. Base case: about $160 a share. Earnings start at $9.00 and grow 2.5% a year, roughly the pace of the bank's deposit and fee base over the past decade (about 3-4% a year excluding the acquisition). The bank keeps back enough each year to fund that growth at its present capital ratio and pays out the rest. Beyond year ten, the deposit franchise and the trust business, which have earned through every cycle in the bank's history, keep earning at the year-ten level with no further growth credited. Wider margins, faster loan growth and Texas share gains are not credited because none of them is secured. Bear case: about $105 a share. This case includes one full credit cycle like 2008-2010 for regional commercial lenders, along with an energy price collapse like 2015-2016, which pushed this bank's earnings to $3.53. We charge $1.2 billion of extra pre-tax credit losses over years two to four, about 4.5% of the $26.8 billion loan book, which matches the industry's commercial loss experience in that cycle and sits above this bank's own record. Earnings fall to roughly $1-$3 a share for three years. After that, the bank settles at $7.00, about an 8% return on today's book, which reflects interest margins back at the 2024 low of 2.65% and a smaller energy book. After tax, the losses cost about $950 million against $6.1 billion of equity, so the bank would not need to raise capital or issue shares, and the bear case contains no dilution. Its value is above tangible book of about $83 a share, and here is why: the fee businesses and low-cost deposit base kept the bank profitable through 2008-2010, the 2016 energy bust and 2020. Even in its worst recent year it earned $3.53 a share, so a downturn eats into earnings rather than capital. Bull case: about $200 a share. Earnings start at $10.00, close to the recent pace after a normal loss charge, and grow 4.5% a year. That is the growth rate the trust, servicing and commercial books actually achieved from 2019 to 2025. The case also adds about $8 a share as roughly $250 million of unrealized bond losses (marked against equity at mid-2026) pull back to par as the bonds mature. Accounting notes. Stock compensation is already charged in reported earnings and is not deducted twice. Intangible amortisation is small (intangibles are $30 million against $1.04 billion of goodwill) and does not affect the result. The bank's reported operating cash flow swings by billions from year to year with its trading inventory and mortgages held for sale, so it says nothing about earning power and was not used. Previous estimate. Our prior figures valued the bank on a return-on-book approach and are not comparable to these. The business news since then is mixed. On the positive side, earnings, interest income and fees all rose in the first half of 2026, and share count fell after a $413 million buyback in 2025. On the negative side, loss provisions stayed at zero, which cannot last; unrealized bond losses widened from $166 million to $250 million; and energy loans kept shrinking. Margin-of-safety price: $80. We are fairly confident in the earnings level. The bank has earned between $6 and $9 a share in every normal year since the CoBiz deal, and the dividend has risen every year for more than a decade, to $2.52 a share at the current rate. The worst plausible outcome is a severe energy and commercial real estate downturn hitting at the same time as falling interest rates. Losses at the 2008-2010 pace would wipe out about two years of earnings but leave capital well above regulatory minimums. Because there is little holding-company debt and a controlling owner who has backed the bank through earlier cycles, the business would survive that hit. The likeliest mistakes are three: the earnings level could be too high if margins return to 2024 levels; senior-housing loans could prove worse than energy did in 2016; and the stock rarely trades much above book because a single owner holds most of it. If all three go wrong, the shares are worth roughly tangible book, about $83, and they have spent long stretches near that level in the past. The $80 price is about tangible book for a bank that earns more than 10% on that capital, with a dividend of roughly 3%. If the base case unfolds over three to four years, the base case implies the 20% annual target from that price. Above that, a credit cycle arriving early could delay a 20% annual return by years.