Stock Market Rally 2025: What's Really Driving It

The Rally Nobody Wanted to Believe Is Now Impossible to Ignore

The 2025 stock market rally has done exactly what the best rallies always do: it punished the cautious and rewarded the decisive. Since early April, US equities have posted their best winning streak since 1985, executing a near-vertical V-shaped recovery that left most retail investors — and more than a few institutional ones — holding cash on the sidelines while the market climbed without them.

Now, with Goldman Sachs raising its S&P 500 year-end target to 8,000, semiconductor options premiums running at 5x their historical average, and margin debt growing at an annualised rate of 81.6%, the question is no longer whether the rally is real. The question is what phase comes next — and whether the signals underneath the surface are flashing opportunity or danger.

Here is a clear-eyed breakdown of what the data actually says.

The Numbers Behind the 2025 Stock Market Rally

Let's start with the hard figures, because they tell a more nuanced story than the headlines do.

• Margin debt (FINRA data, April): up 6.8% month-over-month, representing an annualised increase of 81.6% — nearly a doubling of margin debt year-on-year
• Semiconductor options premiums: currently running at 5x their historical average, driven by a 2.7x surge in semiconductor-related options volume
• Free credit balances in cash accounts: down for the first time since January, meaning retail investors are now actively deploying cash into equities
• Goldman Sachs S&P 500 EPS forecast for 2026: up 24%, almost double the 13% growth recorded last year
• VIX (Dow Jones Volatility Index): approaching its long-term support level around 15–16, currently sitting near 15.64

Taken together, these numbers describe a market that is enthusiastic to the point of aggression. Margin nearly doubling on an annualised basis is not a green flag. It is a yellow one — not a reason to exit, but a reason to be precise about where you are putting new capital.

Phase One Is Hardware. Phase Two Is Software — And It's Starting Now

The structural story of this rally has two distinct chapters, and most investors are still reading chapter one.

Phase one — roughly April through mid-June — has been almost entirely driven by hardware. Semiconductors, AI infrastructure, and related equities carried 70–80% of the index gains in both the S&P 500 and the NASDAQ 100. That is extraordinary concentration. It also means the foundation of the rally, impressive as it looks, is narrower than the headline numbers suggest.

Phase two is software. And there are early signs it has begun.

Consider what happened on a single trading day: Palantir climbed over 7.8% intraday. Axon gained 13%. Snowflake surged following earnings that showed a narrowed loss and a raised forecast, driven largely by aggressive cuts to general and administrative expenses. Service Now also posted strong results.

Here is why this matters structurally: software represents 21% of the NASDAQ 100. If software companies begin participating in the rally in earnest, the index has a genuine runway to new all-time highs — not because sentiment improved, but because the earnings base broadened. Hardware got the market here. Software is what takes it further.

The IGV software ETF is already attracting elevated volume. That is worth watching.

What Citadel's Data Reveals — and What It Misses

Citadel's Scott Rubner has published data identifying the current "pain trade" as US equities continuing to rise. His argument: improving geopolitical sentiment (including progress on an Iran deal), subdued volatility, and relentless mega-cap leadership are forcing cautious investors to abandon their defensive positioning and chase the market higher.

He is right about the mechanism. Institutional and retail investors who sat out the April-to-May surge are now facing a difficult choice: buy in at elevated levels or remain underweight equities and risk further underperformance. That repositioning pressure is real, and it creates a self-reinforcing momentum dynamic.

But Rubner's framework appears to be missing something significant: the software rotation. His analysis focuses heavily on semiconductor momentum and mega-cap tech, yet the most actionable move in the near term may be the catch-up trade in software — a sector that has lagged hardware by a wide margin and is now showing the early technical signs of a short squeeze and institutional re-entry.

When market makers and hedge funds are focused on one sector, the alpha often lives in the adjacent one they are ignoring.

The Bear Case Is Real — Just Misdirected

It would be intellectually dishonest to write about this rally without acknowledging the legitimate risks. Economist Mark Zandi's bearish consumer thesis deserves serious consideration, even if it does not invalidate the equity bull case.

His key points:

• Q1 real GDP was revised down to 1.6%, and that figure includes a bounce-back from the government shutdown at the end of 2024
• Core capital goods orders declined in April, following strong Q4 readings — a signal of softening business investment
• New home sales remain weak, a predictable outcome when the 10-year Treasury is pressing against its ceiling near 4.57%
• Consumer financial buffers are thinning — excess savings built up during the pandemic era are largely exhausted for a significant portion of households

Add to this the accelerating displacement of service-sector workers by AI — call centres, customer support, back-office roles — and you have a genuine bifurcation in the economy. The stock market is not the economy. Businesses deploying AI are winning. Workers being replaced by AI are losing. Both things are true simultaneously.

The bear case is not that markets will crash imminently. The bear case is that this is a K-shaped recovery playing out in slow motion: strong corporate earnings, compressed wages, and a consumer who is increasingly relying on credit to maintain spending.

Rate Expectations Are Mispriced — and That Is a Tailwind

One of the least-discussed drivers of the current market environment is the mispricing of rate expectations.

The market is currently pricing in approximately a 60% probability of rate hikes in the next year. That expectation is almost certainly wrong. With PCE inflation continuing to print below consensus, GDP growth softening, and the labour market showing signs of stabilisation rather than overheating, the Federal Reserve has neither the mandate nor the political room to hike further.

As that 60% rate-hike probability compresses toward zero — and it will — the impact on equities will be meaningful. Lower rate expectations reduce the discount rate applied to future earnings, which mechanically increases the present value of growth stocks. Software companies, which are valued heavily on future cash flows, stand to benefit disproportionately from this repricing.

This is not a prediction. It is arithmetic. Watch the Fed Funds futures curve. When it shifts, equity positioning should shift with it.

How to Think About Positioning Right Now

This is not a market that rewards either blind aggression or excessive caution. Here is a practical framework based on the current data:

What the data supports:

• Continued exposure to the NASDAQ 100 and S&P 500, with recognition that further gains depend on the software rotation materialising
• Selective software exposure — individual names with strong earnings catalysts or short-squeeze setups, or a diversified approach via software ETFs like IGV
• Monitoring the 10-year Treasury yield as a key variable for both growth equity valuations and housing-related equities
• Treating the NASDAQ 100's 725 level as a meaningful technical floor that has been tested and held multiple times

What the data flags as risks:

• Margin debt at near-doubling annualised rates is a late-cycle signal. It does not predict a crash, but it does predict that any correction will be sharp when it comes
• Semiconductor options premiums at 5x historical averages mean the risk/reward on new options positions in semis is unfavourable right now — buyers are overpaying
• Hardware concentration: if the software rotation fails to materialise, the indices are more vulnerable than the headline levels suggest

The bottom line: Phase two of this rally — the broadening into software — is the most important variable to watch. If it holds and spreads, new all-time highs are not just possible, they are the base case. If it stalls, the hardware-only foundation starts to look fragile.

Conclusion: Precision Over Panic, Discipline Over FOMO

The 2025 stock market rally is real, it has fundamental support, and it likely has further to run. But it is entering a phase that requires more precision than the one that preceded it.

The easy money in hardware has been made. The next leg depends on software earnings and a rotation that is just beginning. Margin debt at 81.6% annualised growth is a warning sign, not a crash signal — but it narrows your margin for error. Rate expectations are mispriced in a direction that favours equities, and that correction will take time to play out.

The investors who will do best from here are not the ones chasing semiconductor options at 5x premiums. They are the ones who identified the software rotation early, sized their positions with discipline, and have a clear framework for when the thesis has played out.

Watch the software sector. Watch the 10-year yield. Watch margin debt. The data is there. Use it.

Frequently Asked Questions

What is driving the 2025 stock market rally? The rally has been driven primarily by three factors: a sharp improvement in geopolitical sentiment (including progress on Iran-related trade uncertainty), strong Q1 corporate earnings — particularly in the semiconductor and AI hardware sectors — and a compression of volatility (the VIX) that has encouraged institutional and retail investors to increase equity exposure. Goldman Sachs now projects S&P 500 earnings growth of 24% in 2026, nearly double last year's pace, which provides fundamental support for elevated index levels.

Is the stock market rally sustainable, or is a crash coming? The rally has real fundamental underpinnings, but several indicators warrant caution. Margin debt is growing at an annualised rate of 81.6% — nearly a doubling — which historically signals elevated risk in the event of a sentiment shift. However, margin growth alone does not predict a crash; it signals fragility. A broadening of the rally into software stocks would provide a more durable foundation. A crash becomes more likely if the software rotation fails and the market remains dependent on hardware alone.

Why are semiconductor options premiums so high right now? Semiconductor-related options volume has surged 2.7x, pushing options premiums to approximately 5x their historical average. This reflects intense retail and institutional interest in semiconductor stocks as the primary beneficiary of AI infrastructure spending. The practical consequence is that buying options on semiconductor stocks is currently expensive — buyers are paying a significant premium for leverage. Sellers of those options (including market makers) are benefiting from elevated premiums in a low-volatility environment.

What is the software rotation and why does it matter for the NASDAQ 100? The software rotation refers to capital moving from hardware and semiconductor stocks — which have led the 2025 rally — into software companies that have lagged but are now beginning to show earnings momentum. Software represents approximately 21% of the NASDAQ 100. If software stocks participate meaningfully in the rally, the index has a structural path to new all-time highs because it would represent a genuine broadening of earnings growth rather than concentration in a single sector. Early indicators — including strong earnings from Snowflake and Service Now, and a surge in Palantir's share price — suggest this rotation may be beginning.