Jim Cramer spent much of September arguing that Nvidia faced an unusual problem: extraordinary profits, but a share price that wasn’t responding strongly enough.
His solution was equally unusual. He wanted an Apple-style buyback of roughly $500 billion, enough to repurchase close to 10% of the company. On September 28, Nvidia moved in that direction — but stopped well short of what he wanted.
The company added $150 billion to its existing share repurchase program, raising the remaining authorization to $235 billion. Nvidia described it as the largest share repurchase authorization increase in history.
NVDA rose after the announcement, even as higher Treasury yields pushed the broader market lower. The immediate reaction was positive, but the longer-term effect depends on something less dramatic: how quickly Nvidia actually spends the money.
Nvidia’s Record Buyback Is Real, but It Has Not Happened Yet
According to Nvidia’s official announcement, the board approved an additional $150 billion in repurchases, bringing the total remaining authorization to $235 billion. The company expects to execute the program through the end of fiscal 2028, which falls on January 30, 2028.
Jensen Huang framed the decision as a reflection of Nvidia’s ability to invest in AI development while still returning capital to shareholders. That distinction matters. Nvidia is not presenting the buyback as an alternative to growth spending; it is arguing that the company can afford both.
Still, an authorization is not the same as a completed purchase. The board has given Nvidia permission to spend up to $235 billion, but the company is not contractually required to use the entire amount.
Its regulatory filings state that repurchases can be adjusted according to market conditions, operating requirements and other investment opportunities. The program can also be suspended at Nvidia’s discretion.
The announcement therefore creates potential demand for NVDA shares. It does not place $235 billion of immediate buying pressure into the market.
Jim Cramer Got the Direction Right, Not the Number
Cramer had called for Nvidia to follow the model used by Apple, which has spent years using excess cash to shrink its share count. His argument was that Nvidia’s valuation failed to reflect its earnings growth, AI leadership and cash-generation capacity.
He wanted a program worth about $500 billion—roughly five times the authorization Nvidia had available when he began making the case. Such a plan could have retired close to 10% of Nvidia’s outstanding shares at the prices he was discussing.
The approved plan is considerably smaller. With NVDA recently trading near $233 and approximately 24.1 billion shares outstanding, a $235 billion buyback could theoretically repurchase around one billion shares. That would represent roughly 4.2% of the company’s share count, assuming the average purchase price stayed near current levels.
The actual percentage could be lower if Nvidia’s stock rises before the purchases are completed. It could also be partially offset by shares issued through employee compensation programs.
Cramer nevertheless called the announcement “very significant.” His latest argument is that regular daily purchases could change the trajectory of NVDA stock by providing consistent demand when other investors are selling.
He may be right about the mechanism. Whether the effect is large enough to reshape the stock is a separate question.
The Execution Pace Matters More Than the Headline Number
Nvidia was already buying back stock before the latest announcement. During the second quarter of fiscal 2027, it spent $19.7 billion to repurchase 94 million shares. Across the first half of the fiscal year, it repurchased 203 million shares for $39.8 billion.
The company returned approximately $26 billion to shareholders through buybacks and dividends in the second quarter alone. That is already a substantial pace. It may not be enough to use the entire remaining authorization by January 2028.
A simple calculation illustrates the issue. With about 16 months remaining until the stated target date, Nvidia would need to spend an average of roughly $44 billion per quarter to complete $235 billion in repurchases. That would be more than twice the amount spent on buybacks in the latest reported quarter.
Nvidia could accelerate purchases when the stock weakens, use structured repurchase agreements or continue carrying part of the authorization beyond the expected period. The company’s actual quarterly activity will reveal more than the size of the program announced by the board.
If repurchases remain near $20 billion per quarter, the authorization will function mainly as a long-term source of support. If the pace moves closer to $40 billion or $50 billion per quarter, the market impact could be much more visible.
Nvidia Has the Earnings Power, but $235 Billion Is Still Enormous
The buyback is supported by a business producing numbers that would have looked improbable before the AI boom.
Nvidia reported second-quarter revenue of $96.2 billion, up 106% from a year earlier. Data center revenue reached $89 billion, while net income climbed to approximately $59.7 billion. The company expects third-quarter revenue of around $108 billion, even without assuming data center compute revenue from China.
Cash generation has also expanded rapidly. Nvidia produced $74.4 billion in operating cash flow during the first half of fiscal 2027. At the end of July, it held $56.6 billion in cash, cash equivalents and marketable debt securities, alongside $42.8 billion in marketable equity securities.
Those figures explain why Nvidia can discuss a $235 billion capital-return program without immediately alarming investors. They also show why the program cannot be funded from cash alone. Completing it will require Nvidia to continue producing enormous cash flows, sell investments, use additional financing or combine those sources.
That creates an important difference between Nvidia and a mature company conducting buybacks because it has few growth opportunities. Nvidia still faces heavy spending requirements across advanced chips, networking products, software, research infrastructure, strategic investments and long-term supply commitments.
What Could the Buyback Do to Nvidia’s Earnings Per Share?
When a company repurchases and retires shares, the same amount of profit is divided across fewer shares. That can increase earnings per share even if total net income does not grow.
If Nvidia reduced its share count by approximately 4.2%, unchanged profit would translate into an EPS increase of roughly 4% to 5%. Cramer’s preferred 10% reduction would have produced a mechanical EPS lift of about 11%, assuming no other changes.
The real outcome will be less precise. Nvidia issues equity to employees, the stock price will change during the program, and quarterly profits are unlikely to remain flat. A rising share price makes the buyback less effective because every dollar purchases fewer shares.
The most favorable scenario is one in which Nvidia’s earnings continue to grow while the share count declines. In that case, business expansion and financial engineering reinforce each other.
The less attractive scenario is Nvidia buying aggressively near a market peak just as AI spending begins to slow. Buybacks create value when shares are purchased below their long-term worth. They can destroy value when companies overpay.
A Buyback Cannot Protect Nvidia From an AI Slowdown
The new authorization strengthens the capital-return case for Nvidia, but it does not remove the risks surrounding the stock.
Nvidia still depends on hyperscalers, AI laboratories and cloud infrastructure companies maintaining aggressive capital expenditure. Its data center business also has meaningful customer concentration, while export controls continue to restrict access to China.
Competition is another long-term concern. Advanced Micro Devices, custom accelerators from large cloud companies and new AI chip designs do not need to displace Nvidia completely to affect pricing or margins. Even a gradual shift in market share could matter when investors are valuing the company on years of exceptional growth.
There is also the question of whether AI customers can generate sufficient returns from the infrastructure they are purchasing. Nvidia’s revenue can continue rising while its customers absorb the financial risk of building data centers. If those customers eventually reduce spending, a buyback will not replace lost chip demand.
The $235 billion program is therefore an amplifier, not the original investment thesis. It can increase the value delivered by strong earnings, but it cannot turn weak operating performance into sustainable growth.
The Number That Will Settle the Cramer Debate
Cramer can reasonably claim that Nvidia listened to the central part of his argument. The company adopted a much larger capital-return plan and explicitly connected it to confidence in the long-term AI opportunity.
But the debate will not be settled by the authorization itself.
The figure to watch is the number of shares actually repurchased each quarter. Investors should compare the cash spent with the reduction in diluted shares, rather than assuming every authorized dollar has already entered the market. They should also measure buybacks against operating cash flow, investment requirements and changes in Nvidia’s average purchase price.
If Nvidia starts spending close to $40 billion or more per quarter while AI revenue remains strong, Cramer’s prediction that the program could alter NVDA’s trajectory will become more credible. If the company maintains its previous pace, the announcement will still matter—but more as a signal of confidence than as an immediate transformation of the stock’s supply.
The Nvidia buyback is a major vote of confidence, but authorization should not be mistaken for execution. Readers can follow wider market developments through the Tapbit official website, access an existing account through Tapbit login, or register with Tapbit.
Frequently Asked Questions
How large is Nvidia’s new stock buyback?
Nvidia added $150 billion to its existing repurchase authorization. After including the unused portion of previous programs, the company now has $235 billion in remaining authorization.
Did Nvidia approve the $500 billion buyback proposed by Jim Cramer?
No. Cramer called for a program of approximately $500 billion that could repurchase close to 10% of Nvidia. The company’s remaining authorization is $235 billion, less than half of his proposed amount.
Has Nvidia already spent the $235 billion?
No. The figure represents the maximum amount currently authorized for future repurchases. Nvidia expects to execute the plan through fiscal 2028, but it can change or suspend purchases depending on market and business conditions.
