$25.05
$56.92
bear $25.05
base $43.17
bull $56.92
margin of safety $24.00
market price* $71.85
Exponent is an engineering and scientific consulting firm best known for failure analysis: when a product catches fire, a vehicle crashes, a structure fails or a chemical is challenged by a regulator, companies and their lawyers hire Exponent's roughly 970 technical staff to work out what happened and, often, to testify about it. It has two reporting segments: engineering and other scientific services (about 1.19 million billable hours in 2025) and environmental and health (about 280,000). No client accounted for a tenth of revenue in any of the last three years, it has no borrowings, and it needs very little capital: equipment spending has averaged under 3 percent of revenue for a decade. It grows almost entirely by hiring and has made essentially no acquisitions. What the business earns. We measure owner earnings as operating cash flow with working-capital timing removed, less stock compensation (treated as a real cost) and the capital spending needed to stay in business. On that measure Exponent produced $96.6 million in 2021, $91.3 million in 2022, $93.6 million in 2023, $109.0 million in 2024 and $97.4 million in 2025, or $1.81, $1.75, $1.81, $2.11 and $1.90 per diluted share. 2024 is flattered by a 53rd week. The 2022 figure is no longer depressed, as a plain cash-flow measure shows, because the $33 million build in unpaid client bills that year is a timing effect we remove. The past five years are the right window: the business has not changed shape, and they include a soft patch, a strong year and a flat year. The median is about $96.6 million. One adjustment matters. That figure includes roughly $7 million a year of after-tax interest on a cash pile that no longer exists: the company spent $97 million on buybacks in 2025 and a further $146 million in the first half of 2026, taking cash from $222 million to $67 million. We therefore value the operating business at about $90 million a year before any interest, and add the remaining cash separately. Earlier years (2016 to 2019) were lower, at $0.68 to $1.20 per share, but they predate both the cut in the corporate tax rate and a decade of bill-rate increases, so they describe a smaller business. The recent trend is mixed. In 2025 owner earnings per share fell about 10 percent, partly because 2024 had an extra week, and billable hours fell 2 percent while technical headcount rose only 1 percent, so all of 2025's growth came from price. The first half of 2026 then turned sharply: revenue was up 17.5 percent and net income up 11 percent on the same period of 2025, according to the 10-Q filed in August 2026. That is the dated reason our base case starts the first year at about $93 million, modestly above the roughly $90 million operating level, rather than extending 2025's decline. Operating cash flow in that half was weaker, at $30 million against $43 million, because client receivables rose by $39 million. We treat that as the same billing-timing swing seen in 2022 rather than a change in earning power, but it is worth watching. The cases. There is no debt and no interest expense, so the business value and the owners' value differ only by the cash. We add about $37 million of net cash, keeping roughly $30 million of the $67 million for day-to-day operations, and count the shares at about 48.5 million diluted after the buybacks. The deferred-compensation assets of $123 million are matched by a larger liability to employees and are not counted. Office leases are paid out of operating cash flow and are already charged. Base case, about $43 per share. Operating owner earnings of $93 million in year one, growing 2.5 percent a year over ten years to about $116 million. That growth is credited only because it is the continuation of something Exponent has done every year for a decade: raising bill rates faster than wages through the 2020 shutdown and the 2022 to 2025 plateau, while slowly adding staff from its own cash. It is well below the roughly 6 percent a year revenue has grown since 2016, because volume growth from hiring has stalled since 2021 and we do not assume it resumes. After year ten we let earnings grow 2 percent a year for a further forty years. That is justified by the structural feature of the business: product-liability litigation, product recalls and regulatory submissions are permanent features of the American economy, and Exponent's fifty-plus-year record of courtroom credibility has kept it the go-to firm through every cycle in its public history. The business is worth about $2.06 billion, owes nothing, and with $37 million of net cash the shares are worth about $43. Bear case, about $25 per share. Earnings fall back to $88 million with no growth for ten years: billable hours keep slipping as they did in 2025 and price only offsets wage inflation. In years three and four a downturn cuts earnings by 13 percent and then 7 percent. The 2020 shutdown is the worst stretch in our ten years of data, and revenue fell only 4 percent; the bear charges a deeper hit because staff are a largely fixed cost, so a fall in hours lands on profit at several times its rate. After year ten earnings shrink 2 percent a year and the story is ended after forty years. That reflects the company's own statement that barriers to entry in consulting are low, and the risk that software, including artificial intelligence, takes over some of the routine analysis it bills by the hour. The business is worth about $1.18 billion, owes nothing, and with the cash the shares are worth about $25. Bull case, about $57 per share. Starting at $98 million, close to the 2025 level including the strength of early 2026, and growing 5 percent a year for ten years: the first-half 2026 pace persists and the firm resumes adding technical staff at the 3 to 4 percent a year it managed before 2021, on top of its bill-rate increases. After year ten, growth of 2.5 percent a year for forty years. We exclude any windfall from new practice areas such as battery-fire investigation or autonomous-vehicle testing beyond what is already in the run rate, because none is contracted. The business is worth about $2.72 billion and the shares about $57. What changed since our previous analysis. Our earlier estimate was built on a different basis and its figures are not directly comparable. The business itself has moved in both directions since then. In its favor, the first half of 2026 confirmed real revenue reacceleration. Against it, heavy buybacks have spent most of the cash that was earning interest, and receivables have swollen. The margin-of-safety price. How sure are we of the earnings level? Fairly sure. Owner earnings have stayed between $91 million and $109 million for five straight years, capital needs are trivial, and there is no debt to service. The uncertainty lies in growth, not in the level. The worst plausible outcome would be a lasting fall in demand for expert-witness and failure-analysis work, for example tort reform that sharply reduces product-liability litigation, or clients replacing a large share of billable analysis with software. Neither case models a sudden collapse of that kind. The structure of the business survives it, though: with no debt, no customer concentration and a variable bonus pool equal to a third of pre-bonus profit, costs fall when profits fall, and the company could shrink rather than fail. If we are wrong about the two things most likely to go wrong, namely that headcount never grows again and that the bill-rate premium erodes as competitors and software close the gap, the shares look worth something in the low-to-mid $20s, which is the bear case. The price assumes a three-to-five-year holding period. During that period, the business pays a dividend of about $1.30 a share and keeps buying back stock. At $24 a share, the bear case implies roughly breaking even. If the business delivers its base path and the shares come to reflect it within about three and a half years, the base case implies a move from $24 to the mid-$40s plus a dividend yielding over 5 percent on that price, which exceeds 20 percent a year. If the recovery takes the full five years, the base case still implies close to 20 percent. A higher price would make that target depend on the bull case. That is why the margin-of-safety price is $24.