Nvidia added $150 billion on September 28, 2026. The announced $235 billion remaining authorization includes that increase.
Does the authorization mean Nvidia has already bought $235 billion of stock?
No. The $235 billion is remaining authority for future purchases, which Nvidia expects to execute through fiscal 2028.
Will Nvidia’s buyback raise its stock price?
A buyback can reduce shares outstanding, but the result depends on actual purchases, their price, new share issuance, operating earnings and market valuation. A higher price is not guaranteed.
This article is for informational purposes only and does not constitute financial, investment, tax, or legal advice. Market data, tax rules, and prices can change after the article date. TECHi and its authors may hold positions in securities or digital assets mentioned. Always conduct your own research and consult a licensed financial, tax, or legal professional before making decisions.
Jazib Zaman founded TECHi in 2010 and is chief executive of its publisher, TechAbout LLC. He writes about AI infrastructure, semiconductors and the companies financing them, from Palantir's growth expectations to the toolmakers behind AI memory demand. He is a former member of the Forbes Technology Council.
A semiconductor wafer and processor illustrate Nvidia’s expanded share repurchase authorization. TECHi illustration.
Nvidia’s new $150 billion buyback authorization sounds enormous because it is. But against a company valued at about $5.52 trillion, the additional spending capacity equals roughly 2.7% of its equity value. That is meaningful support for each remaining share, not a replacement for the earnings growth investors already expect.
Nvidia announced the increase on September 28, taking its remaining repurchase authorization to $235 billion. Management expects to execute that remaining program through fiscal 2028. Those are authorized future purchases, not $235 billion already spent.
For investors assessing NVDA after Monday’s session, the sharper question is how much stock that money can retire—and how much of the benefit survives new share issuance and the prices Nvidia pays.
Article Brief
Key takeaways
4 Points24s Read
01Scale-The new $150 billion authorization is roughly 2.7% of Nvidia’s September 28 market value.
02Total remaining-The announced $235 billion balance already includes the increase; the two figures must not be added.
03Illustration-At a fixed $228.86 share price and unchanged income, spending the full balance could lift EPS roughly 4.4% before dilution and cash effects.
04What to watch-Actual repurchase spending, average purchase price and diluted shares will show the realized benefit.
Nvidia’s $150 billion buyback adds about 2.7% of market value
The TECHi Nvidia quote page recorded a September 28 regular-session close of $228.86, up $3.79, or 1.68%, with market capitalization of approximately $5.52 trillion. The closing price comes from Yahoo Finance through TECHi’s integrated quote stack, timestamped 4 p.m. EDT. It is a completed-session observation, not a live after-hours quote.
Using that price and the rounded market value gives the authorization a more useful scale:
The additional $150 billion: about 655 million shares at $228.86, equivalent to approximately 2.7% of current equity value.
The full $235 billion remaining: about 1.03 billion shares at the same price, equivalent to approximately 4.3% of current equity value.
The pre-increase remainder: $85 billion, calculated by subtracting the new authorization from the announced total.
These are TECHi calculations using a fixed price. They are not Nvidia’s forecast of the number of shares it will purchase. Transaction costs, taxes, price changes and future issuance are excluded.
The distinction between $150 billion and $235 billion also matters. Adding both figures would double-count the new authorization. The larger number already includes the increase.
Authorization is not a purchase schedule
A board authorization gives management room to buy shares; it does not require Nvidia to spend the entire amount or buy on any particular trading day. The September 28 release says the remaining program is expected to be executed through fiscal 2028. It does not disclose a daily buying plan, a guaranteed average price or a floor under NVDA stock. Investors should therefore resist describing the announcement as $150 billion of immediate market demand.
There is a useful precedent inside Nvidia’s own reporting. In its first-quarter fiscal 2027 results, released May 20, the company said its board had approved an additional $80 billion on May 18. It also reported approximately $20 billion returned to shareholders that quarter through repurchases and dividends combined. Four months later, it has announced another increase. The sequence shows a company expanding its authority while continuing to return cash, but it does not let us infer a stable quarterly repurchase rate: the return figures include dividends, and authorizations can be spent at different speeds.
The $235 billion remainder is a stock of permission at the announcement date. The $26 billion returned in the July quarter is a flow of completed spending and dividends during an earlier period. Comparing those amounts can describe scale; subtracting the earlier flow from the newly announced balance would mix dates and misstate what remains. Nvidia’s future disclosures, rather than the board ceiling, will establish how much of the program is actually used.
A 4.3% share reduction would mean roughly 4.4% more EPS
Hold total net income constant, assume the entire remaining authorization is spent at the reference price, and ignore new issuance. Reducing the share base by approximately 4.3% would raise earnings per share by about 4.4% on that simplified, fully completed basis. The additional $150 billion alone produces a roughly 2.8% lift under the same assumptions.
The calculation is earnings divided by a smaller share count: 1 ÷ (1 − the repurchased fraction), minus 1. Reported EPS uses weighted-average diluted shares, so purchases spread across several quarters would feed into reported results gradually. Cash spent on repurchases can also reduce interest income; financing purchases with debt could add expense. Neither effect is included here.
That is the limit of the mechanical argument. A buyback can increase each remaining share’s claim on earnings. It does not create an equivalent percentage increase in revenue, and a 4.4% illustrative EPS benefit does not guarantee a 4.4% stock-price gain.
The denominator deserves care. Nvidia’s August results show 24.285 billion weighted-average diluted shares for the July quarter. Dividing the illustrated 1.03 billion shares by that historical diluted count gives about 4.2%, close to the 4.3% market-value shortcut above. They are not identical measures: the share count is a quarterly accounting average, while the market value and price are a September 28 snapshot. Neither captures the actual future mix of repurchases, employee awards and other potential dilution. The point of putting both measures alongside each other is to show the likely order of magnitude, not to manufacture precision from mismatched dates.
For a cleaner test after each earnings report, compare diluted weighted-average shares with the prior quarter and the same quarter a year earlier, then read the financing cash-flow line for actual shares repurchased. A falling count would show that purchases exceeded dilution over the measurement period. An unchanged count despite large cash outlays would narrow the per-share benefit. Even that comparison needs care if acquisitions, stock splits or changes in convertible securities alter the denominator. The board authorization by itself answers none of those questions.
Nvidia’s purchase price will decide how far $235 billion goes
A rising share price makes the same authorization retire fewer shares. At an illustrative average purchase price of $300, $235 billion would buy about 783 million shares. At $200, it would buy about 1.18 billion. Those prices are sensitivity assumptions, not targets or predictions.
The difference is substantial: roughly 392 million shares for the same cash budget. Investors therefore need the eventual purchase prices and the change in diluted shares, not just the headline dollars.
This also explains why a falling share count is a better test than a large repurchase total alone. New stock issued to employees or for other purposes can offset some of the shares bought back. Gross spending and net dilution answer different questions.
The stock-price sensitivity creates another asymmetry. If management buys aggressively when the shares are expensive, it retires fewer shares and leaves less cash available for other uses. If the average price is lower, the same budget has more per-share effect. The authorization is flexible enough to let Nvidia adjust the pace, but the announcement does not tell investors how it will make that trade-off. A buyback can be sensible capital allocation and still be a poor reason to pay any price for the stock.
Capital returns still depend on the operating business
Nvidia’s August 26 earnings release reported approximately $26 billion returned through repurchases and dividends during its second fiscal quarter. That is an earlier, completed-period figure combining two forms of capital return; it should not be treated as quarterly buybacks alone or deducted again from Monday’s remaining authorization.
The same release reported $96.2 billion of quarterly revenue and projected $108 billion, plus or minus 2%, for the following quarter. That guidance remains a management expectation. The September 28 buyback announcement did not replace it with a new revenue outlook. Readers can track the next reported results and guidance on TECHi’s Nvidia earnings page.
Cash generation puts a limit on the headline. Nvidia reported $24.1 billion in operating cash flow for the July quarter and $21.3 billion in free cash flow under its stated definition, after $2.7 billion of purchases related to property, equipment and intangible assets and about $0.1 billion of related principal payments. The remaining $235 billion authority is about eleven times that single quarter’s free cash flow. That ratio measures the size of the authorization against one completed quarter; it is not a forecast that eleven similar quarters will occur, nor a claim that the whole program must be funded from that one cash-flow line.
Quarterly cash conversion may move with customer payment timing, inventory, taxes and investment needs. Nvidia could choose to retain cash, spend on capacity and product development, or return it to shareholders. The board has created room for the last choice without announcing a binding allocation among them. The July-quarter return of $26 billion, which includes dividends, exceeded the quarter’s stated free cash flow. One period can be supported by cash accumulated in prior periods; it does not establish an indefinitely repeatable pace.
Investors should watch the operating cash-flow and free-cash-flow lines alongside repurchase spending, then ask whether the diluted share count actually falls. That combined check is more informative than either the authorization headline or one quarter’s EPS in isolation. It also protects against a common reading error: treating higher EPS caused by fewer shares as proof that demand for Nvidia’s products has improved.
The size of those operations is why the buyback deserves attention without becoming the entire investment case. Investors still need to judge the cash the business generates, the capital it commits elsewhere and the valuation paid for the earnings left over. TECHi’s Nvidia financial statements provide the reported-period cash-flow and balance-sheet context for that comparison.
Monday’s authorization gives Nvidia a larger pool of potential demand for its own shares. The next useful evidence will be cash actually spent, the average purchase price and the diluted share count in subsequent filings. Until those arrive, the defensible conclusion is narrower than the headline: a very large dollar commitment can deliver a useful, single-digit mechanical earnings benefit at Nvidia’s current scale, while operating performance remains the larger driver.
This article is for information and analysis, not personalized investment advice.
Jazib Zaman founded TECHi in 2010 and is chief executive of its publisher, TechAbout LLC. He writes about AI infrastructure, semiconductors and the companies financing them, from Palantir's growth expectations to the toolmakers behind AI memory demand. He is a former member of the Forbes Technology Council.
Comments