Wall Street is paying a surprisingly modest multiple for one of the fastest-growing companies on the planet. The real question is whether Nvidia's earnings can keep growing fast enough to justify even that discount.
NVDA | Investment Analysis | October 12, 2026
There's something almost absurd about the way Wall Street values Nvidia (NASDAQ: NVDA). Here we have a company selling the infrastructure powering much of the artificial intelligence revolution, generating extraordinary profit margins, and reporting revenue growth that would make most corporate executives consider selling their souls for a single quarter of similar performance. Yet investors are increasingly reluctant to assign Nvidia the kind of earnings multiple its growth might historically have commanded.
At first glance, that sounds like an opportunity.
After all, isn't finding exceptional businesses trading at reasonable valuations precisely what investors are supposed to be doing? Isn't that the entire attraction of buying quality companies when market enthusiasm temporarily cools?
Perhaps. But I've learned that whenever Wall Street appears to be offering something extraordinary at a discount, it's worth asking whether the market has overlooked an opportunity or recognized a risk that the earnings estimates haven't fully captured.
And Nvidia presents an especially interesting version of that dilemma.
The company's fundamentals remain remarkable. Revenue is expanding rapidly, earnings expectations continue climbing, and gross margins remain at levels that most hardware manufacturers could only dream about. Meanwhile, the world's largest technology companies are collectively committing enormous amounts of capital to AI infrastructure.
Yet the stock's valuation has compressed substantially.
I think that's where the investment debate gets interesting. Because Nvidia doesn't necessarily need to deliver disappointing earnings for its shareholders to experience disappointing returns. It merely needs to grow less spectacularly than investors currently expect.
And when you're dealing with a business whose customers are spending hundreds of billions of dollars building the computational infrastructure of the future, determining what constitutes sustainable growth becomes considerably more difficult than simply looking at a P/E ratio.
Nvidia's valuation is becoming difficult to ignore
October 9, 2026 closing price
$229.28
FY2027 P/E
24.6×
FY2028 P/E
14.4×
FY2028 EPS estimate
$15.91
At the close on October 9, Nvidia traded at $229.28. Consensus estimates compiled by Yahoo Finance placed normalized earnings at $9.31 per share for fiscal 2027 and $15.91 for fiscal 2028.
That produces a striking difference depending on which earnings period investors use.
At approximately 24.6 times fiscal 2027 earnings, Nvidia doesn't immediately scream deep value. But at roughly 14.4 times fiscal 2028 earnings, it begins looking unusually inexpensive for a company expected to deliver such substantial profit growth.
These are different fiscal-year valuation measures, rather than a conventional next-12-month P/E. Still, they demonstrate how dramatically Nvidia's valuation depends on the speed at which anticipated earnings become reality.
Interestingly, a recent Barron's report cited Nvidia trading at approximately 17.1 times forward earnings, down from 24 times under its chosen estimate methodology. That compression is worth paying attention to.
But comparing today's multiple with Nvidia's historical valuation requires caution.
Historically, the company deserved a premium partly because investors anticipated an extraordinary acceleration in its earnings. Today, those once-distant expectations are embedded in actual results and increasingly ambitious future estimates.
The market may simply be pricing the eventual maturation of the business.
In other words, Nvidia's lower P/E could represent an overlooked bargain, but it could also reflect investors recognizing that a company approaching several hundred billion dollars in annual revenue cannot compound at extraordinary rates indefinitely.
Both possibilities deserve consideration.
The earnings revision story is stronger than the share price suggests
This is where I find the bullish argument especially compelling.
When a stock becomes cheaper because analysts are aggressively reducing earnings estimates, that's one thing. It generally means the business outlook is deteriorating alongside market expectations.
But Nvidia is experiencing almost the opposite.
Its forward earnings forecasts have continued moving higher.
Consensus EPS estimates, 90 days ago versus current
■ 90 days ago
■ Current
USD per share; normalized earnings estimates from Yahoo Finance
The fiscal 2028 estimate has increased by approximately 25% over 90 days. That is a substantial upward revision for any company, particularly one of Nvidia's size.
And the direction of individual analyst revisions provides additional context. Over the preceding 30 days, Yahoo Finance reported 46 upward revisions and no downward revisions for fiscal 2028 earnings, while fiscal 2027 had 43 upward revisions against just one downward revision.
I consider that more meaningful than a collection of optimistic price targets.
Price targets can be influenced by sentiment, valuation preferences, and assumptions about what other investors might eventually pay. Earnings revisions, while certainly imperfect, reflect changing expectations for the company's underlying profitability.
Of course, there's a catch.
Analysts have a long history of becoming most optimistic just before the conditions supporting their forecasts begin changing. Consensus estimates aren't guarantees, and revision momentum can reverse when demand assumptions, margins, or customer spending plans shift.
Nevertheless, the combination of accelerating earnings expectations and compressed valuation multiples creates an unusual setup.
If those earnings materialize, Nvidia could become more attractively valued even without meaningful appreciation in the stock price.
The risk is that investors are placing increasing confidence in profits that remain several reporting periods away.
And that's where the balance of the argument depends on gross margins and AI infrastructure spending.
Gross margins: The number that could decide everything
There is one financial metric I believe investors should watch almost as closely as revenue growth: gross margin.
Nvidia's profitability is extraordinary not simply because the company sells large amounts of AI hardware, but because it captures a remarkable amount of the value generated by those sales.
In its fiscal 2027 second quarter, Nvidia reported revenue of $96.2 billion, up 106% year over year. Data center revenue reached $89 billion, an increase of 117%.
Its GAAP and non-GAAP gross margins were both 75%.
Consider what that means.
For every $100 in revenue, Nvidia retained approximately $75 in gross profit before accounting for operating expenses and other costs.
That's astonishing for a company operating in the semiconductor industry, where manufacturing complexity, product cycles, competitive pricing, and supply constraints can create significant pressure on profitability.
And the company's guidance suggested that these margins could remain remarkably resilient.
Management projected approximately $108 billion in third-quarter revenue, with gross margins around 74%, plus or minus half a percentage point.
There is no obvious profitability collapse in those expectations.
But I would argue that the margin outlook deserves more scrutiny than many bullish investors give it.
Nvidia is benefiting from a rare combination of strong demand, sophisticated hardware, a powerful software ecosystem, and customers willing to spend aggressively to secure computational capacity.
The question is how long that combination can support today's economics.
As AMD, custom AI accelerators, and internally designed chips become more capable, Nvidia may face pressure to compete more aggressively on price or deliver increasingly expensive technological improvements.
I don't necessarily expect margins to collapse. Nvidia's hardware and software ecosystem, particularly CUDA and its integrated networking capabilities, create substantial competitive advantages.
But the valuation implications of even modest margin compression shouldn't be underestimated.
What happens if gross margins fall?
Using Nvidia's $108 billion quarterly revenue guidance as a simple illustration:
Hypothetical gross margin | Quarterly gross profit |
|---|---|
75% | $81.0 billion |
74% | $79.9 billion |
70% | $75.6 billion |
65% | $70.2 billion |
These are mechanical illustrations, not forecasts. They assume unchanged revenue and do not account for operating expenses, taxes, or other changes.
Still, they make the sensitivity clear.
A decline from 75% to 70% would reduce quarterly gross profit by $5.4 billion at the same revenue level.
That doesn't make Nvidia an unattractive business. Far from it.
But investors buying Nvidia on the assumption of extraordinary earnings growth should recognize how sensitive those projections can become when exceptional profitability is assumed to persist.
I would feel considerably more comfortable underwriting Nvidia's long-term valuation using somewhat lower margins than today's record-level performance rather than assuming the company can indefinitely retain every advantage it currently enjoys.
If margins remain near 75%, that's additional upside.
If they normalize, I want the investment thesis to survive anyway.
AI spending forecasts: A trillion-dollar opportunity with a trillion-dollar question
If gross margins represent Nvidia's ability to capture value, capital expenditures represent the willingness of its customers to keep creating demand.
And right now, the spending projections are enormous.
Goldman Sachs Research estimated in August that global AI investment could exceed $1 trillion during 2026, including approximately $581 billion in the United States. In a separate September forecast, Goldman strategists projected that capital spending by the five largest U.S. hyperscalers could reach $1.2 trillion in 2027, compared with a Wall Street consensus estimate of $1.1 trillion.
These figures don't all measure the same thing. Global AI investment, total hyperscaler capital spending, and Nvidia's addressable chip market are distinct categories. Not every dollar spent on AI infrastructure flows to Nvidia.
Nevertheless, the magnitude of the projected investment is difficult to ignore.
The world's largest technology companies are treating AI infrastructure as a strategic necessity rather than an optional experiment.
Microsoft, Alphabet, Amazon, Meta, and other major participants are competing to establish the computing capacity necessary to support increasingly sophisticated AI systems.
That creates an enormous market for Nvidia.
But here's the uncomfortable question.
What happens when the companies buying all this infrastructure eventually need to prove that the spending is generating sufficient financial returns?
Building AI capacity is one thing. Monetizing it at attractive rates is another.
We've seen this distinction before in technology.
The internet transformed civilization, but not every company financing internet infrastructure produced acceptable shareholder returns. Telecommunications infrastructure proved essential, yet overinvestment still caused enormous financial losses.
AI can become a foundational technology while certain AI investments simultaneously prove uneconomic.
Those outcomes are not contradictory.
In fact, I consider that one of the most important distinctions investors need to understand.
The long-term success of artificial intelligence does not automatically guarantee that every dollar spent on today's infrastructure will produce an attractive return.
Nor does it guarantee that customers will continue increasing their capital budgets indefinitely.
A July Reuters analysis highlighted growing pressure on major technology companies' free cash flows as AI capital expenditure commitments expanded. The concern wasn't that AI had suddenly become irrelevant, but that the scale of investment might eventually outpace cash generation and monetization.
More recently, research discussed by Barron's suggested that AI infrastructure spending could exceed $1.4 trillion in 2027, with a potential adjustment in capital spending beginning in 2029 if industry revenue fails to justify the investments.
That's the risk I would focus on.
Nvidia's immediate customers might continue buying hardware aggressively even while the economics of their own AI services remain uncertain.
That dynamic can support extraordinary chip demand for several years.
But eventually, the economics have to work throughout the value chain.
And if they don't, Nvidia could experience a demand adjustment despite remaining technologically dominant.
Why I wouldn't confuse an earnings bargain with a risk-free investment
One thing I've learned from studying high-growth companies is that valuation compression can happen for perfectly rational reasons.
Investors aren't always being irrational when they refuse to pay the same multiples they paid during an earlier stage of a company's growth.
Sometimes the company's financial profile has changed.
Sometimes uncertainty about future earnings has increased.
And sometimes the market recognizes that the extraordinary period of growth supporting previous valuations cannot continue forever.
With Nvidia, I suspect all three considerations matter.
The company is no longer a relatively small semiconductor manufacturer participating in promising growth markets. It is an enormous infrastructure provider whose performance has become intertwined with the capital allocation decisions of some of the world's largest corporations.
Its customer base is financially powerful, but also concentrated.
Its technological position is formidable, but competition is increasing.
Its earnings growth is extraordinary, but the estimates themselves assume continued expansion of a capital-intensive investment cycle.
The apparent valuation discount therefore needs to be understood in context.
I wouldn't automatically assume that Nvidia deserves to trade at its historical average multiple.
I'd first ask whether its future growth, competitive advantages, and capital requirements justify that historical premium.
And then I'd ask a more important question.
What return can I reasonably expect if the stock never returns to that historical valuation?
Because that's where genuine value investing begins.
Not with the hope that someone else eventually pays more for the same earnings.
But with the expectation that the underlying business can create enough additional value to reward shareholders even under conservative assumptions.
What different earnings outcomes could mean for Nvidia's share price
To evaluate that question, I prefer running several scenarios rather than pretending I can predict the precise valuation Wall Street will assign two years from now.
Using the current fiscal 2028 normalized EPS estimate of $15.91 as a starting point, here is how different earnings and valuation assumptions translate into hypothetical share prices.
Illustrative FY2028 valuation scenarios
Bear case
$154.20
-32.7%
Base case
$270.47
+18.0%
Bull case
$412.50
+79.9%
| Scenario | EPS | P/E |
| Bear | $12.85 | 12× |
| Base | $15.91 | 17× |
| Bull | $18.75 | 22× |
I find this exercise particularly revealing because it demonstrates how dramatically the valuation multiple affects the investment outcome.
In a bearish scenario, Nvidia could generate $12.85 in annual earnings per share and still deliver a substantially lower stock price if investors assigned only a 12-times multiple.
In the base case, consensus earnings of $15.91 and a 17-times multiple would imply modest upside from the October 9 closing price.
In the bullish case, higher earnings combined with a stronger valuation multiple would create considerably more attractive potential returns.
These scenarios are not probability-weighted forecasts, and the implied share prices should not be mistaken for price targets.
Their purpose is to reveal the assumptions driving the investment.
Notice something important: the base scenario doesn't require a return to Nvidia's most enthusiastic historical valuations.
But it does require the company to deliver extraordinary earnings growth from current levels.
That's a critical distinction.
The investment case isn't necessarily that Nvidia deserves to become more expensive.
It's that the company may generate so much additional earnings power that even a relatively restrained multiple can support an attractive valuation.
Of course, the difference between an attractive stock and an attractive company often comes down to the price investors pay.
And Nvidia's share price already incorporates a meaningful expectation of future success.
The competitive moat is real, but it isn't invincible
One argument I hear frequently is that Nvidia's dominance makes its future earnings relatively secure.
I understand the reasoning.
The company isn't simply selling graphics processing units. It has built a broad ecosystem encompassing hardware, networking, software, development tools, and integrated AI computing systems.
That creates meaningful customer dependencies.
A company deploying complex AI workloads doesn't necessarily want to redesign its entire infrastructure simply because another manufacturer offers a less expensive chip.
Reliability, software compatibility, performance, developer familiarity, and deployment speed all matter.
This is one reason Nvidia has been able to maintain extraordinary margins even while expanding revenue at tremendous rates.
But I'd be cautious about extrapolating that competitive position indefinitely.
The largest cloud providers have compelling incentives to develop their own accelerators, particularly for workloads that can be standardized and deployed at scale.
Even if custom chips never completely replace Nvidia's products, they could reduce the growth of Nvidia's addressable market or weaken its pricing power in particular applications.
Competition doesn't have to destroy a dominant company's business to affect its valuation.
It only has to change expectations about future profitability.
And in Nvidia's case, those expectations are already substantial.
The next earnings report matters more than another price target
Given the company's current position, I would concentrate on four developments in upcoming results.
First, I want to see whether Nvidia delivers against its $108 billion fiscal third-quarter revenue guidance. Demand must remain strong enough to support the upward earnings revisions analysts have made.
Second, I want to see gross margins remain reasonably close to the company's 74% guidance. A modest fluctuation wouldn't automatically concern me, but a sustained downward trend would require revisiting long-term profitability assumptions.
Third, I want to see whether fiscal 2028 EPS estimates continue rising. If analysts begin reducing expectations while the stock's valuation remains compressed, the perceived bargain could quickly become less compelling.
Finally, I want to see evidence that hyperscaler spending continues translating into productive, economically sustainable AI workloads rather than an increasingly debt-dependent infrastructure race.
Recent reports about the financing requirements of AI infrastructure suggest this final issue deserves considerably more attention.
For me, the next phase of Nvidia's investment story is less about whether artificial intelligence will be important.
That question has largely been answered.
The more difficult issue is how much computing infrastructure businesses will need, how quickly they'll deploy it, and how profitably the entire ecosystem can absorb the resulting costs.
My investment conclusion: Attractive valuation, but earnings execution is everything
I see a compelling fundamental argument for Nvidia at current valuations.
The business continues delivering extraordinary revenue growth. Gross margins remain exceptionally high. Analyst earnings estimates have moved substantially higher, and global AI capital spending forecasts suggest continued infrastructure demand.
Those are legitimate strengths.
But I wouldn't buy the stock simply because its forward P/E is below a previous valuation range.
Historical multiples are reference points, not entitlements.
Nvidia's market position has become larger, its growth expectations more demanding, and its fortunes increasingly connected to one of the largest technology investment cycles in history.
A lower multiple may therefore reflect a rational adjustment in how investors assess the durability of future earnings.
My view is that Nvidia looks more attractive on an earnings-growth-adjusted basis than its recent valuation compression would suggest, but its margin of safety depends heavily on the credibility of fiscal 2028 forecasts.
If Nvidia approaches consensus earnings while maintaining strong gross margins, I believe the current price could prove attractive for patient investors.
If the AI spending cycle weakens materially or competition begins pressuring margins, the apparent discount could evaporate without the stock ever returning to its previous valuation multiples.
That's why I would characterize Nvidia as a potentially attractive growth-at-a-reasonable-price opportunity rather than a straightforward deep-value investment.
There is a significant difference between those classifications.
And for investors who want to avoid overpaying for growth, that distinction matters.
The greatest opportunity might not come from waiting for Nvidia's P/E to return to historical highs. It could come from owning a company whose earnings compound so rapidly that the share price eventually catches up, even at a restrained multiple.
But that's the wager.
We aren't simply betting on artificial intelligence changing the world.
We're betting that Nvidia can continue capturing an extraordinary share of the economic value created by that transformation.
Those are not the same thing.
Nvidia may be cheaper than its history suggests. Whether it is genuinely cheap depends on how much of its extraordinary future is already reflected in the earnings estimates investors are using today.
And that, rather than the historical P/E discount alone, is the question I would want answered before putting more capital to work.
Investment research based on information available October 12, 2026. Scenario valuations are illustrative and not investment advice. Earnings forecasts and capital spending estimates are subject to change.
Comments
Post a Comment