SpaceX’s First Results After IPO: Why SPCX Stock Crashed Despite Strong Q2 Earnings

SpaceX beat Wall Street’s revenue, EBITDA and earnings estimates in its first quarterly result after going public. SPCX stock still fell 7.46% in after-hours trading. The reason is simple: investors were impressed by what SpaceX earned, but alarmed by what it had to spend. Capital expenditure reached $18.37 billion in one quarter, nearly as much as SpaceX spent during the whole of 2025, and management indicated that the next two quarters could look similar.

Let's break down SpaceX’s Q2 2026 earnings, why the stock fell despite a sizable beat, what Elon Musk and management said, and the numbers investors should track before deciding whether the post-IPO fall has created value or merely reduced an expensive valuation.

SpaceX Q2 2026 Earnings Results

On the surface, this was a strong quarter. Revenue reached $7.81 billion, rising 92% from a year earlier. The net loss narrowed to $541 million from approximately $1 billion, while adjusted EBITDA almost tripled to $3.54 billion.

More importantly, SpaceX outperformed almost every operating hurdle discussed in our SpaceX Q2 earnings preview.

Sources: SpaceX’s official Q2 earnings release

The company did not simply beat because of one accounting adjustment. All three business segments produced more revenue than expected, adjusted EBITDA was significantly ahead of forecasts, and the consolidated operating loss narrowed to just $143 million from $970 million a year earlier.

The one major miss was capital expenditure. Unfortunately for SPCX stock, it was also the number the market cared about most.

Why Did SpaceX (SPCX) Stock Fall After Q2 Earnings?

SPCX closed the regular trading session on August 4 at $125.33, up 9.43%. It then fell to $115.98 after hours, a decline of 7.46% from the closing price, according to after-hours market data.

There is an important nuance here. Even after the after-hours decline, the stock remained about 1.3% above its August 3 closing price of $114.53. The earnings reaction therefore erased most of the pre-result rally rather than creating an entirely new one-day collapse.

Even so, the after-hours reversal tells us that investors found several parts of the result uncomfortable.

The result was not weak. It was capital-hungry.

That difference matters. A weak quarter would suggest that demand is missing. SpaceX’s problem is almost the opposite: demand and revenue are growing, but the company is spending an extraordinary amount to build the capacity needed to serve them.

Why SpaceX’s $18.37 Billion Q2 Capex Worried Investors

SpaceX spent $18.37 billion on capital expenditure during Q2, including $15.83 billion on AI infrastructure. AI alone accounted for approximately 86% of the company’s quarterly capex.

To put that spending in perspective:

The simplified free cash flow figure subtracts capital expenditure from operating cash flow. It excludes acquisitions and some other investing items, so it should be treated as a diagnostic measure rather than SpaceX’s official free cash flow.

Still, the message is hard to miss. SpaceX generated $3.47 billion from operations during the first six months of 2026 but spent $28.48 billion on capital assets.

Think of adjusted EBITDA as the income produced by a new factory before counting the cost of building the factory. It can show that demand and day-to-day operations are improving, but it does not pay for the concrete, machinery and power infrastructure. SpaceX’s factories happen to include rockets, satellites and some of the largest AI data centres in the world.

The cash bill is very real.

How High Could SpaceX’s 2026 Capital Spending Go?

CFO Bret Johnsen said capital expenditure in each of the next two quarters should be similar to Q2, according to the company’s earnings-call summary.

If Q3 and Q4 each match Q2’s $18.37 billion, the rough calculation would look like this:

This is not company guidance for an exact full-year figure. It is a simple extrapolation from the CFO’s statement.

A $65.2 billion outcome would be more than three times SpaceX’s approximately $20.74 billion of capital expenditure in 2025. It would also consume roughly three-quarters of the $85.7 billion raised through the IPO, although SpaceX has other cash, operating inflows and debt financing.

Here is another way to see the acceleration. From Q1 to Q2:

• Revenue increased by approximately $3.12 billion.
• Capital expenditure increased by approximately $8.26 billion.
• Capex therefore increased by about $2.65 for every $1 of sequential revenue added.

That does not mean SpaceX permanently spends $2.65 to earn $1. Capital expenditure creates assets that can generate revenue for several years. It does show how heavily the current growth is front-loaded with cash spending.

How Starlink Profits Are Funding SpaceX’s AI and Starship Growth

The most revealing table in SpaceX’s earnings release was not the consolidated income statement. It was the segment breakdown.

Source: SpaceX Q2 earnings release.

Connectivity, which primarily includes Starlink, generated $1.66 billion of operating profit. Space and AI together produced approximately $1.80 billion of operating losses.

Subtract those two loss-making businesses from Starlink’s profit and the result is almost exactly the company’s $143 million consolidated operating loss.

That gives investors a simple way to understand the current financial structure: Starlink is not merely SpaceX’s largest business. It is the financial shock absorber for Starship and AI.

This is encouraging because Starlink’s economics improved materially. Connectivity revenue rose 66% year over year, operating income increased 79%, and the segment delivered an operating margin of approximately 38.6%.

Starlink finished the quarter with 12 million subscribers, double the year-earlier level. Average revenue per user remained at $66 per month compared with Q1, although it was down from $85 a year earlier as the company expanded into lower-priced international markets.

Enterprise and government connectivity revenue was particularly strong, rising 108% to $1.81 billion. SpaceX also reported over $6 billion in new multi-year U.S. government contracts for Starshield.

The risk is that Starlink’s profits are being asked to support several expensive projects simultaneously. If Connectivity margins weaken while AI and Starship continue consuming capital, the financial model becomes much more dependent on the IPO cash pile.

AI Had a Strong Quarter, but EBITDA Does Not Tell the Whole Story

SpaceX’s AI segment delivered one of the quarter’s biggest improvements. Revenue reached $2.56 billion, up 213% sequentially and 247% year over year. Adjusted EBITDA improved from a $609 million loss in Q1 to a positive $1.15 billion in Q2. Yet the segment still reported a $1.26 billion operating loss.

How can both numbers be correct?

Adjusted EBITDA excludes depreciation, amortisation and share-based compensation. The AI segment recorded approximately $1.89 billion of depreciation and amortisation during the quarter, reflecting the cost of its enormous computing infrastructure.

Imagine a taxi fleet claiming a profit before counting the ageing of its cars. The cash may have been spent when the vehicles were purchased, but the vehicles still lose economic value as they are used. AI servers work in a similar way, only they can become outdated much faster.

For a software business, EBITDA may provide a useful view of underlying operations. For a company spending $15.83 billion a quarter on GPUs, power and data centres, ignoring depreciation can leave out a large part of the economic story.

This does not make the EBITDA improvement meaningless. It means investors should track both numbers:

• Adjusted EBITDA shows whether the installed compute is producing operating contribution.
• Operating profit shows whether that contribution covers depreciation and other recurring costs.
• Free cash flow shows whether the company is funding the expansion internally.

SpaceX passed the first test this quarter. It has not yet passed the other two.

Are SpaceX’s AI Contracts as Valuable as They Look?

SpaceX announced $14.1 billion of contracted cloud-services sales, which generated $1.6 billion of incremental AI infrastructure revenue during Q2. Management also said another $6.7 billion of cloud-services revenue had been contracted early in Q3 for a six-month period beginning to ramp in October.

CFO Bret Johnsen said new compute deployments currently have a payback period of less than one year. If sustained and confirmed through cash flow, that would be an unusually attractive return on data-centre investment.

But the SpaceX Q2 10-Q adds two important qualifications:

1. A significant portion of AI infrastructure revenue comes from a small number of customers.
2. After an initial ramp period, the cloud agreements can generally be terminated by either party with 90 days’ notice.

This makes the company’s $100 billion annualized revenue ambition harder to value.

Management expects to reach an annualized revenue run-rate of approximately $100 billion based on December revenue. That would mean December monthly revenue of roughly $8.3 billion.

However, an annualized run-rate is not the same as $100 billion of revenue earned during 2026. Nor is it automatically equivalent to highly predictable subscription revenue. It takes one strong month, multiplies it by 12 and assumes the activity can continue.

That assumption may prove correct. Investors will need renewal rates, customer diversification, utilization, cash collections and contract margins to verify it.

SpaceX Earnings Call: What Elon Musk and Management Said

Management used the first public earnings call to set some extremely ambitious operating targets.

Musk also said SpaceX would build new AI systems exclusively on Nvidia’s Vera Rubin architecture and expects to begin launching Starmind AI satellites next year. The company is targeting space-based computing as a future extension of its terrestrial data-centre business.

The ambition is enormous. So is the mathematical burden. Moving from a $100 billion annualized run-rate at the end of 2026 to $1 trillion of revenue in 2030 requires revenue to grow at approximately 78% annually for four consecutive years.

That is possible mathematically. Operationally, it would require SpaceX to keep adding the equivalent of very large public companies to its revenue base every year. Investors should treat the $1 trillion figure as a management aspiration, not a conventional earnings forecast.

What Actually Went Wrong for SpaceX?

This was not a conventional earnings miss. In fact, describing it as a “hollow beat” would be slightly unfair.

SpaceX beat expectations across every reported business. Starlink subscribers reached the expected 12 million, Connectivity margins improved, AI revenue was stronger than forecast, AI adjusted EBITDA turned positive, and the operating loss narrowed sharply.

The more accurate description is a capital-heavy beat. Our preview argued that SpaceX needed to prove three things together:

1. Starlink could continue producing high-quality growth.
2. AI revenue could scale without uncontrolled spending.
3. Starship progress could eventually reduce deployment costs.

The result offered good evidence for the first point and partial evidence for the third. It failed to settle the second.

AI revenue increased by $1.74 billion sequentially. AI capital expenditure increased by $8.11 billion. That equals approximately $4.65 of additional AI capex for every $1 of sequential AI revenue added.

Again, this is not a permanent unit-cost calculation. The new servers should support revenue beyond one quarter. But it explains why the market wants more than a revenue beat. It wants to know how many dollars of durable cash flow each dollar of investment will eventually create.

SpaceX Balance Sheet: Cash, Debt and IPO Funding Explained

SpaceX ended Q2 with approximately $100 billion in cash, cash equivalents and marketable securities. It also reported a $47.5 billion backlog and another $5 billion of available borrowing capacity.

That gives the company substantial room to invest. It is not facing an immediate liquidity problem. However, the source of that cash matters.

Source: SpaceX Q2 10-Q.

The cash balance increased primarily because SpaceX raised $85.68 billion through its IPO and issued $25 billion of senior notes. Operating cash generation contributed only a small portion of the increase.

SpaceX therefore has a large financial fuel tank, but much of that tank was filled by shareholders and lenders. The next phase of the story is whether the company’s operating businesses can refill it faster than AI infrastructure and Starship consumes it.

Is SPCX Stock Cheap After the Earnings Fall?

SPCX traded at $115.98 after hours, approximately 48.6% below its $225.64 post-IPO intraday high and about 14% below the $135 IPO price.

Using the 13.18 billion Class A and Class B shares outstanding at the end of Q2, the after-hours price implies an equity value of approximately $1.53 trillion.

These are simplified calculations, not fair-value estimates or price forecasts.

The $123 billion erased after hours was approximately 6.7 times SpaceX’s entire Q2 capital expenditure. That tells us the stock was not reacting to the $18.37 billion already spent. It was repricing the possibility that similarly large bills could continue for years.

The valuation also illustrates why a 50% share-price decline does not automatically create a value stock. At the current after-hours price, SpaceX still trades at nearly 49 times annualized Q2 sales.

Even if management reaches its $100 billion December run-rate objective, the stock would trade at approximately 15 times that run-rate. That may be easier to justify if the revenue is high-margin, diversified and recurring. It is harder to justify if much of the growth comes from capital-intensive compute contracts that customers can terminate with limited notice.

SPCX is therefore not conventionally cheap. It remains a public stock carrying venture-style assumptions.

What Would SpaceX Need to Show for the Stock to Find a Bottom?

One earnings result cannot confirm a lasting bottom, especially when the first major post-IPO share unlock follows on August 6. Up to 911.5 million shares held by employees and early investors may become eligible for trading, according to MarketWatch’s lock-up analysis.

Eligible does not mean all those shares will be sold. It does mean the stock will no longer benefit from the same scarcity that helped drive its early post-IPO surge.

A more credible fundamental bottom would require evidence across four gates:

The critical measure is no longer whether SpaceX can grow. Q2 answered that question convincingly. The critical measure is whether revenue can eventually grow faster than the capital base required to produce it.

How Investors Can Read the Next SpaceX Results

This framework avoids reducing the investment debate to a single EPS beat or a one-night stock move.

A Starlink-focused investor should watch Connectivity operating income and ARPU. An AI-focused investor should watch contract renewals, customer concentration, utilization and cash returns on new compute. A Starship-focused investor should watch flight cadence, reuse and the Space segment’s path away from operating losses.

Different theses require different evidence.

Final View: A Strong Quarter With an Expensive Message

SpaceX’s first public earnings result proved that the company can generate exceptional revenue growth. Starlink is highly profitable, AI monetisation has started far faster than expected, and the overall operating loss is approaching breakeven.

But the result also revealed the price of that growth. SpaceX is currently committing more than twice its quarterly revenue to capital expenditure, and management expects the spending pace to remain elevated. Much of the AI revenue comes from a small number of contracts that may be terminated with 90 days’ notice, while Starlink remains responsible for offsetting losses in both Space and AI.

The stock’s 7.46% after-hours decline was therefore not the market saying that SpaceX failed. It was the market demanding a higher standard of proof.

The next stage of the SPCX story will not be decided by revenue growth alone. It will be decided by whether SpaceX can convert an $18 billion quarterly investment bill into durable, diversified and internally funded cash flow.

Until that conversion becomes visible, the lower share price may represent a cheaper entry into an extraordinary growth story, but it does not yet make SPCX a conventional value stock.