$28.26
$60.10
bear $28.26
base $49.27
bull $60.10
margin of safety $23.00
market price* $43.58
National Bankshares is the holding company for The National Bank of Blacksburg, chartered in 1891, with offices across southwest, western and central Virginia. It had $1.83 billion of assets and $1.63 billion of deposits at June 30, 2026. Loans of about $1.0 billion produced 65% of 2025 operating revenue and a $654 million securities portfolio another 19.5%. Borrowings are negligible (interest on borrowed money was $7 thousand in the first half of 2026). The bank bought Frontier Community Bank on June 1, 2024 and finished a core-systems conversion in the second quarter of 2025. For a bank, interest is an operating cost, so we value earnings after interest and subtract no debt. Earnings history. Owner earnings per share (net income per diluted share; stock compensation is already expensed and is small, $193 thousand in 2025) for 2016-2025 were $2.15, 2.03, 2.32, 2.65, 2.48, 3.28, 4.33, 2.66, 1.24 and 2.49. Ten years of swings with little net growth: the decade median is $2.49 and the 2019-2025 median is $2.65. Two years are distorted. 2024 carried $2.9 million of merger costs; 2025 carried $2.1 million of one-time conversion costs (about $0.27 per share after tax), so 2025 on a clean basis was nearer $2.76. Since then results have stepped up: first-half 2026 net income was $10.0 million ($1.57 per share) against $5.5 million ($0.87) a year earlier, and net interest income was $13.05 million in the second quarter of 2026 against $10.99 million in the second quarter of 2025. The last four quarters add up to about $3.19 per share, but they include almost no loan-loss provisions (2025 was a net recovery of $16 thousand; the first half of 2026 about $0.2 million), whereas provisions were $1.65 million in 2016 and $2.0 million in 2020. Underlying costs are also rising: noninterest expense was $18.6 million in the first half of 2026 against $17.2 million in the second half of 2025. We therefore start the cases from about $3.00, the recent run rate less roughly $1 million a year of ordinary provisions, rather than from the $2.65 long-run median, because four consecutive quarters corroborate the higher level. Credit and capital. Net charge-offs were 0.03% of average loans in 2025, nonaccrual loans were $0.3 million at June 30, 2026 and the allowance was $10.0 million (about 1.0% of loans). The main structural risk is commercial real estate: $468 million, 46.8% of loans at year-end 2025, mainly multi-family, non-owner-occupied and other commercial property. Tangible book value was about $28.3 per share at June 30, 2026 (equity $192.2 million less $12.0 million of goodwill and intangibles); equity was 10.5% of assets even after a $37.7 million accumulated other comprehensive loss. That loss reflects the securities portfolio: at year-end 2025 its $706 million of amortized cost was carried at $654 million, of which $24 million matures within a year and $230 million in one to five years, and bonds that mature return their face value. Base case, about $49 per share. Earnings of $3.00 a year for four years, a mild credit episode in years five and six (roughly $4 million a year of extra provisions) that cuts earnings to $2.50, then $3.00 a year with no growth. No meaningful debt is owed, so the business is worth the same as the equity. Growth reconciliation: owner earnings per share grew 101% in 2025 (from a merger-depressed $1.24 to $2.49); year one of this case is $3.00, 20% above 2025 and well below the trailing rate. No growth beyond that is credited: ten years of data show earnings of $2.15 in 2016 and $2.49 in 2025, there is no contracted growth, and Frontier was a single acquisition rather than a repeated process. The franchise is assumed to keep earning that flat amount because 135 years of operation, $1.63 billion of deposits and a decade of owner earnings with no downward trend are the evidence for it; we do not assume it rises. Bear case, about $28 per share. Three years of squeezed margins: earnings fall to $2.00, $1.30 and $1.55 as the margin slides back toward the 2.19% of 2024 on today's cost base. Then comes a full credit cycle modeled on 2008-2012. Federal Reserve data show commercial real estate charge-off rates for U.S. banks averaging about 1.2%, 2.4%, 2.3%, 1.4% and 0.7% in those years. We apply roughly 7% cumulatively to the $468 million of commercial real estate and about 1.5% to the other $531 million, giving provisions of $12 million, $15 million, $10 million and $4 million (about $41 million, roughly three years of the pre-provision earnings in that case). Earnings are $0.10, a loss of $0.28, $0.36 and $1.15 across those years; they recover to $1.70 and then settle at $2.10, the level of 2016-2017, with no growth. Pre-provision earnings over those four years are larger than the cumulative provisions, so equity is not depleted and no new shares are assumed. The bear value is about equal to tangible book value, so we do not argue the franchise is worth more than its capital in a downturn. Bull case, about $60 per share. Margin widens about a quarter of a percentage point as maturing securities and loans reprice (the securities maturing at par noted above), and credit costs stay at ordinary levels with no cycle: earnings of $3.20, $3.40 and $3.55, then $3.65 flat. No new acquisitions, markets or products are credited. Our earlier analysis (September 2026) used a different valuation basis, so its figures are not comparable. Challenge to the case. We tested whether the recent jump was cost-driven: the drop in expenses versus a year ago is only the conversion charge falling away, and underlying costs have risen, so the improvement comes from net interest income, not cost cuts. That is why the base level rests on income already reported rather than on any further cost saving. What would make this wrong: another margin squeeze like 2024's, or a commercial real estate downturn larger than the 2008-2012 pattern we used. How we set the margin-of-safety price. We are reasonably, not highly, sure of the earnings level: four quarters support about $3.2, but the same bank earned $1.24 in 2024 and $4.33 in 2022, and the recent figures carry nearly no provisions. The worst plausible event is a margin squeeze followed by a commercial real estate cycle; the bank has almost no borrowings, an equity cushion of about $190 million, and a deposit base that has not shrunk meaningfully (down 1.2% in 2025, as the bank deliberately cut time-deposit rates), so it would survive the bear path, though earnings would be close to zero for two years and the dividend would be under pressure. If we are wrong on the margin and on credit costs together, the shares are worth about tangible book value, roughly $28. The price assumes a three-to-five-year holding period; the dividend was $1.51 per share in 2025 and retained earnings add roughly the same again each year at current profits. At $23 the shares are priced at about 7 times the current run-rate earnings and sit about 19% below tangible book value, so dividends plus retained earnings alone would imply over 13% a year, and an exit near 1.2 times tangible book value after four years of retained earnings would carry the implied return to 20%. The margin-of-safety price is $23.