Chapter 4

Capital Allocation

Gartner returns capital in one channel only — share repurchases — and pays no dividend [1]. In 2025 it bought back $2.0 billion of stock, about 1.5 times operating cash flow and 2.7 times net income, funded partly by $800 million of new notes; shareholders' equity fell from $1.36 billion to $0.32 billion in a single year [2]. The multi-year record is genuinely mixed: accretive when the stock was cheap, roughly half underwater on the peak-price purchases of 2024–2025.

One channel: all buyback, no dividend

Gartner has never paid a cash dividend and states it does not currently intend to; its 2024 revolving credit agreement carries a covenant that would in any case limit dividends [3]. Excess capital goes to repurchases. The board first authorized a $1.2 billion program in 2015, added $5.8 billion in tranches between February 2021 and September 2025, and authorized a further $500 million in January 2026; $0.7 billion of authorization remained at year-end 2025 [4]. For an asset-light business — capital expenditure runs near $115 million against $6.5 billion of revenue — a buyback-only return policy is a defensible default. The question is not whether to buy back stock, but at what price and with whose money.

2025: the largest year, funded beyond cash flow

The repurchase pace stepped up sharply in 2025. Gartner bought 7.0 million shares for approximately $2.0 billion — recorded at $2,004 million including the 1% excise tax [5] — against operating cash flow of $1,290 million and net income of $729 million [6]. The gap was bridged with balance-sheet capacity: $800 million of new senior notes issued in November and a $210 million draw-down of cash [7].

Source: FY2025 Annual Report (Form 10-K), Consolidated Statements of Cash Flows [8]; MD&A [9].

In only one prior year of this window — 2021 — did buybacks exceed operating cash flow, and that year net income and cash flow were both stronger relative to the spend. The 2025 combination is distinct: the largest buyback in the company's history landed in the year operating cash flow fell (to $1.29 billion from $1.48 billion) and reported net income more than halved on the $150 million Digital Markets goodwill impairment and the absence of 2024's insurance gain [10].

Sources: Statement of Stockholders' Equity [11]; Note 6 Debt [12]; net debt / EBITDA derived from reported financials.

The most visible consequence sits in the equity account. Repurchases of $2,004 million against $729 million of net income drove shareholders' equity down from $1,359 million to $320 million, and cumulative treasury stock up to $9.0 billion [13]. Thin book equity is a feature of a company that has bought back far more stock than it has retained in earnings, not a solvency signal on its own — but it does mean the buyback is now consuming capacity rather than distributing surplus.

The value question: right idea, wrong prices in 2024–2025

A buyback creates value only if the stock is repurchased below its worth. Gartner's timing has been the opposite of the textbook. The company spent the least when the stock was cheapest and the most when it was dearest.

Source: average annual closing price from market data.

Sources: average annual closing price from market data; buybacks from FY2025 Annual Report (Form 10-K), Consolidated Statements of Cash Flows [14].

The 2025 purchases averaged about $284 per share — 7.0 million shares for $2.0 billion [15]. Against the roughly $142 the stock trades at in mid-2026, that capital is about half underwater. The 2024 buyback of $735 million, executed near the $476 average of that year, is down closer to 70%. The purchases from 2019–2021, made below $250, look accretive by the same measure. The record is not a blanket indictment — it is a specific one: Gartner leaned hardest into repurchases at the top of its own price range.

Two facts cut the other way, and both belong in the read. First, the 2025 average of $284 was well below that year's $358 average close, so the company did concentrate buying into the second-half decline rather than at the January peak — the Q4 program alone averaged $239 per share [16]. Second, over the full span the buyback has done its structural job: roughly $6.7 billion of repurchases since 2018 lowered the weighted-average share count about 17%, from 90.8 million to 75.4 million, concentrating the recurring franchise's cash flows on fewer shares [17]. Stock-based compensation of $156 million a year continually refills the share pool, so gross repurchases buy down more shares than the net count reduction shows [18].

The balance sheet after the spend

The debt used to fund the 2025 buyback is worth reading precisely, because it corrects a common shorthand. The November 2025 notes were not Gartner's first bond issue: the company has carried public senior notes since 2020. It ended 2025 with five tranches totaling $3,005 million of principal [19].

Source: FY2025 Annual Report (Form 10-K), Note 6 Debt [20].

The November deal — $350 million of 4.95% notes due 2031 and $450 million of 5.60% notes due 2035 — did two things: it repaid the $274.4 million then drawn on the revolver, terming out short-term borrowing, and it funded repurchases [21]. The pricing marks a step up in the cost of capital: the new money carries 4.95% and 5.60% coupons against the 3.625%–4.50% legacy stack [22]. Gartner is now retiring equity at a mid-single-digit pre-tax cost of debt while the earnings yield the buyback captures has risen as the stock fell — a more sensible trade at $142 than the same trade was at $476.

Leverage rose but remains modest. Net debt — $3.0 billion of borrowings against $1.7 billion of cash — sits near $1.28 billion, roughly 1.05 times EBITDA of about $1.23 billion, up from roughly 0.4 times a year earlier [23]. Cash interest paid was $100 million against $1.03 billion of operating income, so coverage is not in question [24]. Gartner also kept about $1.0 billion of undrawn revolver capacity [25]. The doubling of net leverage is real, but from a base low enough that it constrains nothing.

Digital Markets: a small pruning

The one portfolio move of the period fits the same picture. Gartner recognized a $150 million goodwill impairment on Digital Markets in the third quarter of 2025, reclassified the unit as held for sale, and completed the disposal on February 5, 2026 for approximately $110 million [26]. It is a minor transaction — the proceeds are roughly one-eighteenth of a single year's buyback — but a directionally consistent one: recycling a sub-scale, slower-growth asset rather than adding acquisitions, in a period when management's capital instinct is to concentrate on the core and return cash.

Reading the record

The capital-allocation record cuts both ways. The policy is coherent for the business, the balance sheet is not stretched, and the long-run share-count reduction is real. The weakness is timing: management spent its largest sums — roughly $2.7 billion across 2024 and 2025 — at the highest prices in the company's history, and on today's quote those purchases are well underwater. The aggressiveness itself is a signal worth naming: a management team funding a record buyback with new debt is voting, with cash, that the 2025 stall is cyclical rather than structural. That is a view, not a proof.

What would change the read in either direction is observable. Continued repurchases at a pace that outruns operating cash flow, while the stock stays depressed, would compound the timing problem and keep leverage climbing. Slowing the buyback to let free cash flow rebuild equity — and buying more heavily only at prices like today's — would convert the same policy from value-destructive to value-accretive. The instrument is sound; the discipline question is when management chooses to use it.