The Write-Down

When Concentrix Corporation closed its $4.8 billion acquisition of Webhelp in September 2023, the deal was supposed to create something formidable: a 400,000-employee customer-experience operation running call centers and digital support from Lyon to Bogotá to Manila, serving some of the world's largest brands. The combination left Concentrix with nearly $5 billion of goodwill on its balance sheet — up from about $1.6 billion before the acquisition — the accounting premium that reflects a buyer's conviction that the future will justify the price.

On January 13, 2026, Concentrix filed a six-page 8-K with the Securities and Exchange Commission. On page two, in a table of preliminary fourth-quarter results, the company disclosed a $1,523.3 million non-cash goodwill impairment charge. The explanation was a single clause: the write-down "primarily result[ed] from the recent trading range for the Company's stock price and market capitalization." In plain English, the market had marked down the combined company, and Concentrix's own impairment test had caught up to that verdict. About 30.5 percent of the goodwill on its books — roughly 31 cents of every dollar — was gone.

The Webhelp deal had been closed for twenty-eight months.

Concentrix was not alone. In one earnings cycle, four companies in the business of selling human labor and expertise disclosed $1.76 billion in impairments. Kelly Services, the 79-year-old Detroit staffing firm, wrote down $102 million in goodwill — its second consecutive year of charges. Forrester Research erased $110.7 million, nearly all of its goodwill, across two quarters. GEE Group, a smaller professional staffing company, took $22 million in charges against goodwill and acquired intangible assets, also for the second year running. All four filings landed within ten weeks. No major outlet has aggregated these charges or connected them to the data that links them: the longest sustained decline in U.S. temporary help employment since the Bureau of Labor Statistics began tracking it in 1990 — thirty-nine consecutive months and counting.

---

To understand what these numbers mean, you have to understand what goodwill is and when it breaks.

When one company acquires another, it almost always pays more than the target's net assets — the buildings, receivables, and contracts on the books. The excess gets recorded as goodwill, an intangible asset that sits on the buyer's balance sheet as a placeholder for the deal's future promise: customer relationships, revenue synergies, the strategic logic that justified the premium. Unlike a truck or a server, goodwill doesn't depreciate on a schedule. It just sits there until something forces a reckoning.

That reckoning comes through impairment testing under ASC 350, the accounting standard that governs goodwill. Companies test annually or whenever events suggest value has eroded — a sustained stock decline, weakening revenue, rising discount rates. When the test shows that carrying value exceeds fair value, the goodwill gets written down. No cash leaves the building. But the entry is a formal admission that an acquisition didn't deliver what was promised.

Concentrix's stock had fallen far enough, for long enough, that the gap between its market capitalization and book value became impossible to explain away. The resulting $1.52 billion impairment was among the largest goodwill write-downs in the staffing and BPO sector in recent years. And the company still carried $4.64 billion in long-term debt as of November 30, 2025 — much of it from the borrowings that financed the very deal whose goodwill it had just impaired. Goodwill impairments are typically excluded from the EBITDA calculations that govern loan covenants, but they shrink shareholders' equity and worsen leverage ratios. For a company carrying that much debt, with tranches that will eventually need to be refinanced, the math tightens.

The 10-K filed two weeks later confirmed the charge but added little about what had changed in the underlying business. Concentrix still employed hundreds of thousands of people, still ran operations on multiple continents. The accounting, though, now reflected a different judgment on what all of that was worth.

---

If Concentrix were an isolated case — one big deal gone sideways — this would be a corporate story, not a sector story. What makes it a sector story is what happened next in Detroit, and what had been happening for three years in the federal employment data.

Kelly Services is not a flashy acquirer. Founded in 1946, it is a fixture of American staffing — the kind of firm that places office temps, warehouse workers, and nurses across dozens of markets. According to its most recent 10-K, Kelly connects approximately 375,000 people with work annually. The company had been expanding through acquisitions in recent years, building up the goodwill on its balance sheet. And Kelly, too, was now reckoning with the distance between what it paid and what those businesses were producing.

In its February 12 8-K, Kelly disclosed a $102 million goodwill impairment for fiscal 2025 — the year ended December 28. That followed an $86.3 million impairment the prior year. The filing was blunt: "Operating loss of $69.8 million compared to a loss of $15.1 million reported in 2024; both years reflect non-cash impairment charges of $102.0 million and $86.3 million, respectively." Over two fiscal years, Kelly had written down $188.3 million — well over half the goodwill it carried at the start. Before the 2025 charge, Kelly's goodwill stood at $304.2 million. After: $202.1 million.

The accounting effects compounded. Kelly cited cumulative losses in recent years — including goodwill impairment charges — as the reason it established a $197.6 million valuation allowance against its tax credit carryforwards, the work opportunity credits and foreign tax credits the company had been counting on to offset future tax bills. Kelly was telling investors that not only were its acquisitions worth less than it paid, but the resulting losses made it uncertain the company would earn enough in coming years to use its own tax benefits. The total balance-sheet hit — impairments plus the tax-asset write-down — approached $300 million in a single year.

---

By February 12, temporary help employment in the United States had declined year-over-year for 39 straight months.

The Bureau of Labor Statistics tracks temp employment as a subset of Professional and Business Services. The series — TEMPHELPS in the Federal Reserve's data system — is a favorite of labor economists because temps are the first workers hired when demand rises and the first let go when it softens. In March 2022, the same month the Federal Reserve raised interest rates for the first time in the current cycle, temporary help employment peaked at roughly 3.08 million. By November 2022, the year-over-year comparisons turned negative. They have stayed negative every month since.

Thirty-nine consecutive months of year-over-year decline through January 2026. The longest such streak in data going back to 1990. At its worst, in October 2024, temp employment was falling at 8.1 percent year-over-year. By January 2026, the pace had moderated to negative 2.3 percent — still declining, just more slowly. In absolute terms, temp employment stood at 2.41 million, down about 662,000 from the peak. Roughly one in five temporary jobs that existed in early 2022 had disappeared.

Yet the broader Professional and Business Services category — which includes temps but also consultants, accountants, and engineers on permanent payrolls — barely budged: down only 0.18 percent year-over-year in January 2026. At the October 2024 trough, it fell just 1.0 percent, versus temp help's 8.1 percent at the same moment. The pain was specific, concentrated in the market for commoditized, flexible labor hours — exactly the market that staffing firms had spent billions acquiring their way into.

ManpowerGroup CEO Jonas Prising, speaking at Davos in late January 2026, acknowledged the pressure directly. He told Reuters that 2025 had been "difficult" for the staffing industry. He expressed "hope" — his word — that 2026 would bring growth opportunities, citing improving economic outlooks in Europe and the United States. Not a plan. Not guidance. Hope.

---

The pattern extended beyond traditional staffing into the adjacent business of selling packaged expertise.

Forrester Research, the advisory firm whose reports land on the desks of CIOs and chief digital officers at Fortune 500 companies, disclosed $110.7 million in goodwill impairments across fiscal 2025 — $83.9 million in the first quarter, another $26.8 million in the fourth, per its February 12 8-K. The charges amounted to a near-total write-off of the firm's accumulated goodwill, the premium from years of acquiring smaller research and consulting practices. During fiscal 2025, Forrester also cut guidance and announced restructuring measures — signs that the demand weakness behind the write-downs was showing up in operations, not just on the balance sheet. Forrester's model is different from Kelly's or Concentrix's — it sells subscriptions and advisory engagements, not temp hours — but it shares a fundamental input: human knowledge and labor, billed to corporate clients who have been tightening budgets.

GEE Group, a professional contract staffing firm that places IT professionals, accountants, and engineers, added $22 million in impairments of goodwill and acquired intangible assets in its fiscal year ended September 2025, per its December 18 8-K. Like Kelly, it was the second consecutive year of charges — $19.4 million the year before. GEE's filing offered something none of the other three did: a specific diagnosis. The "proliferation of various artificial intelligence (AI) applications and tools," the company wrote, "has had a dampening effect on many organizations' hiring plans and in many cases led to job terminations." The other three companies were silent on AI. GEE said out loud what the rest of the sector may be reckoning with privately.

Across four companies, impairments of goodwill — and, at GEE, acquired intangible assets — totaled $1.758 billion. GEE's filing hit in December 2025, Concentrix's in January 2026, Kelly's and Forrester's on the same day in February. The clustering suggests common forces at work: falling stock prices and deteriorating demand pushing acquisition-heavy balance sheets past their impairment thresholds almost simultaneously.

---

One major staffing firm conspicuously avoided the charges. Robert Half, which also operates the advisory practice Protiviti, filed its 10-K on February 13 and reported zero goodwill impairments. The reason is visible on its balance sheet: Robert Half carried just $56.6 million in goodwill — 0.7 percent of its $8 billion in total assets. The company had grown primarily through organic expansion, not acquisitions. It had little goodwill to impair because it had never paid large premiums to buy its way to scale.

The contrast clarifies what these write-downs are really about. All staffing companies face the same 39-month demand decline. But the acquisition-heavy firms are carrying balance sheets loaded with goodwill from deals premised on a different future — one where companies kept buying temporary labor at the volumes of 2021 and 2022. When that future didn't materialize, the accounting eventually caught up. The impairment charge is the moment of reckoning.

---

These are not abstract balance-sheet entries. Their consequences are real, even if the full scope remains uncertain.

Kelly's February 12 filing disclosed an operating loss for fiscal 2025 of $69.8 million — nearly five times the prior year's loss. A company absorbing that kind of hit has less room to invest in the sales teams, branch offices, and recruiter headcount that drive a staffing firm's revenue. According to its 10-K, Kelly connects approximately 375,000 people with work annually. Whether a weakened intermediary can compete as effectively for assignments in 2026 is an open question — one Kelly's full 10-K, with its segment breakdowns, may begin to answer.

Concentrix faces a different version of the same pressure. Its $4.64 billion in long-term debt dwarfs Kelly's roughly $298 million, and while the impairment itself doesn't violate loan covenants, the reduced equity base against which leverage is measured matters when debt eventually needs to be refinanced.

For enterprise buyers — the procurement offices that hire staffing and BPO firms — the write-downs carry a signal about vendor stability. Procurement teams often consolidate supplier lists during periods of uncertainty, and a staffing provider writing down its acquisitions is one whose strategic bet didn't pay off. The impairments don't cause those decisions, but they can accelerate the calculus.

Several caveats belong here. The $1.76 billion captures four companies that filed during a specific window. It is not a census. ManpowerGroup, ASGN, TrueBlue, and TTEC Holdings had not yet reported full-year results at the time of this analysis; additional impairments may surface. The comparison across firms is imperfect: Concentrix is a BPO company, Forrester is an advisory business, Kelly and GEE are traditional staffing firms. Their clients, margins, and competitive dynamics differ substantially. What unites them is a common economic input — human labor and expertise — and a common accounting result.

The 39-month TEMPHELPS decline, while unprecedented in duration, demands careful interpretation. Temp employment is falling from what may have been an artificially inflated peak, pumped up by the reopening scramble of 2021–2022. Some of this is reversion to a more normal baseline. The streak is historic, but so was the surge that preceded it.

Still, four firms, three subsectors, nearly $1.8 billion, all within ten weeks, all during the longest sustained temp-employment decline since the data began. Either this is coincidence, or it is the labor market pricing something that the deal makers didn't see — or didn't want to.

---

The next chapter arrives soon. Concentrix reports first-quarter fiscal 2026 results in the spring. Kelly's full 10-K, with detailed segment breakdowns, will show where the remaining $202 million in goodwill sits and how close those reporting units are to their own tipping points. Several industry peers have yet to file their annual reports. And TEMPHELPS updates monthly — each release a data point on whether the 39-month streak is finally bending upward or extending into a fourth year.

Sometime in the weeks ahead, a CEO at one of these companies will face the earnings-call question that always follows a write-down: Is the worst behind you? The answer depends on something no impairment test can model — whether companies start buying temporary labor again, or whether the three-year contraction reflects a permanent shift in how businesses staff. A world where they have decided they need fewer humans, on less flexible terms, than the staffing industry spent billions assuming they would.