Circle doubles in a month—what is the market betting on?

Title: Circle Doubles in a Month—What Is the Market Betting On?

Author: 0x2333

Source:

Repost: Mars Finance

Last June, Circle went public at $31 per share. Two weeks later, the stock hit $299. Then, it dropped back to $50, a decline of over 80%. This February, after the earnings report, the stock doubled in two weeks, reaching $111 today.

During the same period, Bitcoin fell 40%. The correlation between Circle’s stock price and the crypto market has broken.

“You’re seeing a decoupling.”

The drop to $50 is understandable within the industry.

In September and October 2025, the Federal Reserve cut the benchmark interest rate twice by 25 basis points, bringing it down to 3.75%. Reserve yields declined accordingly. Q3 data showed that the USDC reserve yield decreased by 96 basis points year-over-year.

What does this 96 basis points mean in dollars? Circle’s prospectus estimates that for every 100bp rate cut, the annual interest income loss is about $618 million. Half of this loss is offset by reduced distribution costs, resulting in a net loss of around $300 million. This is a static estimate assuming USDC’s scale remains unchanged. In 2025, the Fed cut rates by a total of 75bp, which alone would have taken nearly $200 million annually from Circle’s revenue.

It’s not over. Circle and Coinbase have a revenue-sharing agreement. All reserve income generated from USDC on Coinbase’s platform goes to Coinbase, while outside platform reserves are split 50/50. In Q4, total reserve income was $733 million, with $461 million in distribution costs, leaving Circle with a net reserve income of $273 million. Non-interest income was $37 million, less than 5% of total revenue.

This structure is the fundamental reason why Circle’s stock fell from $299 to $50. Rate cuts compress reserve yields, Coinbase’s fixed revenue share structure remains unchanged, and Circle is caught in the middle—its revenue ceiling is clear, and its profit floor is also clear.

But after the February 25 earnings report, Circle’s stock surged 35% in a single day.

EPS was $0.43, versus an analyst expectation of $0.16. It’s not just a slight beat; it’s a significant outperformance. But what’s driving the revaluation isn’t just the quarterly numbers; it’s a larger structural reality that this report has finally made clear to the market.

By 2025, the total crypto market cap has fallen over 40% from its peak. During this period, USDC’s circulating supply increased by 72%, reaching $75.3 billion, a new all-time high. The total stablecoin market cap also surpassed $314 billion, also a record high.

This isn’t a bullish rally in a bull market; it’s a countertrend in a bear market.

The significance of this is fundamental to Circle’s valuation logic. Previously, the market valued CRCL as a beneficiary of the crypto cycle—more USDC trading volume during bull markets, less during bear markets. Coinbase’s logic is similar: trading volume equals revenue, and in a bear market, fee income collapses.

But 2025 data refutes this framework. USDC’s growth didn’t stop in the bear market; it accelerated.

Circle CEO Allaire said during the earnings call, “You’re seeing a decoupling.” He was referring to the decoupling between Bitcoin and stablecoins. But the half-sentence he didn’t finish is that stablecoins are shifting from being a crypto trading unit of account to becoming a global payment settlement infrastructure.

The driving force is no longer speculative trading demand but a new wave of participants—those who have never appeared in the crypto space before. Visa announced expanding USDC settlement, allowing US-based Visa card issuers to settle outside normal banking hours using USDC. Mastercard followed suit. JPMorgan launched several USDC-related products last year. Intuit announced a partnership with Circle to bring low-cost programmable payments to millions of businesses and consumers. Polymarket migrated its core settlement assets to USDC on a large scale.

This isn’t about crypto-native users saving or withdrawing money. It’s traditional financial institutions embedding USDC into their payment and settlement pipelines. These two use cases have entirely different valuation logic. The former follows the crypto cycle; the latter follows the global payment volume. The global cross-border payments market is about $150 trillion annually, and USDC’s on-chain trading volume in a quarter is $11.9 trillion, up 247% year-over-year. These numbers can’t be directly compared, but the market is starting to price Circle based on the second framework.

GENIUS Act, and a pure entry ticket

There’s a widely circulated saying in the industry: “If your thesis is that stablecoins will eat the global payments market, CRCL is the most direct bet. COIN is a comprehensive platform that also earns USDC revenue, but they are different tools for different narratives.”

This explains why $CRCL could break out independently while Coinbase’s stock remains flat. Coinbase operates an exchange, a wallet, the Base chain, and institutional custody; USDC is just one of many business lines. Circle only does one thing: issue and circulate USDC. If you’re betting on the stablecoin sector itself, there’s only one pure play available.

This logic became clearer after the GENIUS Act passed.

In July 2025, the law was enacted, establishing the first federal regulatory framework for stablecoins, requiring compliant issuers to hold 100% cash or short-term government debt reserves and undergo regular audits. On the day the GENIUS Act passed, $CRCL surged 34% in a single day. The market understood this signal. It’s not just a compliance boost; it’s a regulatory moat—an insurmountable line that Tether cannot cross in the short term.

JPMorgan’s data confirms this view: after the law’s passage, the overall stablecoin market grew 19%, with USDC’s market share increasing from 24% to 25.5%, while Tether’s share declined from 67.5% to 60.4%. On-chain trading volume figures are even more direct: in Q4, USDC surpassed USDT, capturing about 50% of stablecoin on-chain trading volume—Tether’s first decline in years in this dimension.

But Tether didn’t give up. After the GENIUS Act, Tether launched USAT, partnering with Anchorage Digital and Cantor Fitzgerald, designing reserves according to GENIUS standards. USAT’s CEO, Bo Hines, is a former White House crypto advisor. Currently, USAT’s circulation is about $20 million, negligible compared to USDC’s $75.3 billion, but Tether has the world’s largest stablecoin user network, and Cantor Fitzgerald brings Wall Street connections—not a trivial advantage.

Meanwhile, a wave of new entrants that never appeared in the stablecoin space before is emerging. Fidelity issued FIDD on Ethereum, fully compliant with GENIUS standards, targeting institutions and retail. Robinhood and Revolut are reportedly developing their own stablecoins. JPMorgan and US Bancorp are expanding stablecoin plans. After Stripe acquired Bridge, it integrated stablecoin settlement into its $1.9 trillion annual payment flow. Treasury Secretary Bessent said the US stablecoin market could reach $3.7 trillion by the end of this decade. USDC’s current $75.3 billion is less than 3% of that.

The GENIUS Act doesn’t just open the door for Circle; it opens the entire traditional financial system. The first-mover advantages are real: support on 30 blockchains, deep integration with Visa and JPMorgan, years of enterprise API infrastructure—things that new entrants can’t replicate in a year or two. Bernstein’s target price of $190 is based on the regulatory moat and technological barriers.

But the depth of this moat remains uncertain until a real stress test occurs. Another rarely discussed question in the industry: Circle and Coinbase’s revenue-sharing contracts have fixed terms. When renegotiated, Coinbase’s bargaining power won’t be less than now. USDC holdings on their platform increased from 5% in 2022 to 22% today. The outcome of negotiations will directly impact how much of USDC’s growth Circle can actually retain.

$23 billion—betting on a story that hasn’t yet happened

More than half of Circle’s current valuation is based on a story that’s still unfolding.

Allaire spent considerable time discussing AI Agent payment needs during the earnings call. AI Agents executing autonomous tasks require small, high-frequency, cross-time-zone payments—calling APIs, purchasing computing power, settling cross-border. Allaire calls this the “machine economy,” arguing that once AI Agents outnumber human users, the main users of payment infrastructure will be machines.

In traditional payment systems, each step involves friction. Credit cards have business hours, manual authorization, and minimum fee structures that make sub-$0.01 payments uneconomical. Stripe’s minimum fee is $0.30 plus 2.9%. Cross-border fees for Visa and Mastercard average 1.5–3%. Bank wire transfers don’t operate on weekends.

USDC, at the technical level, has none of these limitations. It operates 24/7, settles on-chain instantly, with transaction costs on high-speed chains like Solana under $0.001, aiming for $0.00001. Circle has developed dedicated infrastructure for AI Agent payments, with the Arc testnet already live. This isn’t just a small improvement; it’s a fundamental shift in cost structure by orders of magnitude.

Benchmark’s analyst Mark Palmer said, “AI Agents need programmable money that can be embedded directly into software workflows, without long settlement windows.” Traditional card networks are designed for human checkout processes, not for machines.

The reality of this demand can be seen in protocol-level action speeds.

In May 2025, Coinbase launched the x402 protocol, using the rarely used HTTP 402 status code to enable AI Agents to make automatic payments—requiring USDC payment before responding to requests, without manual approval. Over five months, x402 processed over 100 million payments. Google introduced AP2 (Agent Payment Protocol), and OpenAI tested “Instant Checkout” within ChatGPT, with settlement layers combining Stripe and stablecoins. These are not white papers—they are operational infrastructure already in production.

Visa’s data provides a scale reference. By November 2025, Visa’s monthly stablecoin settlement volume was about $3.5 billion annualized, rising to $4.5 billion by January 2026. Compared to Visa’s total annual transaction volume of about $16 trillion, this is negligible. But the trend shift is more important than the absolute numbers: Visa is expanding this pipeline. Coinbase CEO Brian Armstrong also said in early March, “Soon, the number of AI Agents initiating transactions will surpass humans.”

The gap between narrative and reality is clear in the data.

In the past 30 days, x402 processed $24 million in total transactions, involving 94,000 buyers and 22,000 sellers. Meanwhile, the global e-commerce market is estimated at $6.88 trillion. McKinsey estimates that the real annual use of stablecoins for payments is about $390 billion, with B2B accounting for roughly $226 billion, retail less. ECB data shows organic retail transfers account for about 0.5% of total stablecoin volume.

Circle expects a net loss of $70 million in 2025. The Arc mainnet is planned for 2026, currently still in testing. AI and non-interest income combined account for less than 5% of total revenue.

Gartner predicts the AI Agent economy will reach $30 trillion by 2030, and Bessent forecasts the stablecoin market will hit $3.7 trillion by the end of this decade. If these numbers are true, Circle’s current $75.3 billion USDC circulation is just the beginning. But the jump from $24 million in x402 monthly volume to a $30 trillion AI economy is a road no one has traveled.

The $23 billion market cap bets on this journey being completed.

Circle went public at $31 in June last year, rose to $299 in two weeks, then fell back to $50, and has now doubled again. A question that’s never been truly answered in this curve: what kind of company will Circle ultimately become?

Is it a rate arbitrage business, a compliant stablecoin infrastructure, or a settlement layer for the AI economy? Allaire says, “You’re seeing a decoupling,” but where that decoupling leads remains another question.