Do you want to buy CRCL?

Author: Zhao Haibei

To understand the business of stablecoins, we first need to discuss the three scenarios that make up stablecoins: storage, trading, and payment. For Circle, there are three assumptions:

1. The first two (storage + trading) determine performance, while payment determines valuation.

2. Since the demand for trading involves real-time pricing around the crypto market, it essentially expresses your expectations for BTC trends, which I believe are difficult to predict in the short term but can be roughly assessed in terms of bull and bear ranges in the medium term.

3. Due to the real-time transparency and verifiability of on-chain data, Circle's trading opportunities largely depend on the pricing changes around other businesses beyond USDC.

We can explain the above assumptions in sequence and think about investment opportunities around them.

If we understand the significance of stablecoins for payments, it essentially sacrifices the scale brought by the credit leverage of traditional banks (which typically have a leverage ratio of 10-20x) in exchange for velocity leverage (the turnover rate of on-chain DC payments is usually 5-12x that of traditional payments) as follows:

Bank Throughput = Capital × 10-20x Credit Leverage × Deposit Payment Velocity

Stablecoin Throughput = 1:1 Reserve Stock × On-chain Settlement Velocity

The existing model for stablecoin transfers is almost as shown in the diagram below (commonly referred to as the stablecoin sandwich):

(Payment service providers first convert fiat currency into stablecoins for cross-border transmission, and then the foreign receiving bank converts it back into fiat currency, using the intermediate stablecoin layer to replace the traditional correspondent banking network, thereby reducing the T+N time and liquidity costs of cross-border payments.)

In this model, the actual demand for holding stablecoins at scale is not much, mostly serving as an intermediary layer.

The profit model for stablecoins is very simple: scale * short-term interest rate. This means that for payment models, the gains brought to DC by scale are inherently capped. Compared to storage and trading scenarios (where holding demand leads to scale growth), payment itself is more like a narrative. We can check a few sets of data:

As shown in the figure above, the distribution of USDC transactions and volume across various chains shows that transactions are mainly contributed by Polygon, while volume is mainly contributed by Base; we can further break it down:

As shown in the figure above, starting from October 2025, the main driver of transfer numbers is driven by Polygon, growing from 33M/month (March 2025) to 502M/month (March 2026), an approximate 15-fold increase.

In the transfer counts on Polygon, 95% of the contribution comes from the well-known Polymarket. Since Polymarket is gas-free and due to various hot events and its aggressive marketing, the frequency here has been driven very high.

On the volume side, the Base ecosystem's volume is far ahead, accounting for 64.5% of USDC transfer volume, far exceeding the total of all other chains combined. If we continue to break it down:

Aerodrome (a DEX incubated by Coinbase) contributes nearly 95% of the USDC volume in the Base ecosystem (most of it is volume from arbitrage bots, not elaborated here), making it the absolute dominant factor.

Conclusion 1: Whether from the dimensions of transfer frequency or transfer volume, the main driver of USDC is not payment, but trading around DeFi scenarios. (Source: Dune SQL extraction)

Now, if we look at the distribution of issuance, based on the current total circulation of USDC at 73.81B:

• Coinbase approximately 19.00B, accounting for about 25.7% of the total USDC. (Coinbase products disclosed balance, extrapolated from last quarter's ratio)

• Circle's own platform approximately 13.09B, accounting for about 17.7%. (Extrapolated from the on-platform ratio disclosed in Q1)

• Binance approximately 8.33B, accounting for about 11.3%.

• Hyperliquid approximately 5.72B, accounting for about 7.8%.

• Sky/Maker approximately 4.22B, accounting for about 5.7%.

• AAVE approximately 1.9B, accounting for about 2.5%.

• OKX approximately 1.26B, accounting for about 1.7%.

• Bybit approximately 0.89B, accounting for about 1.2%.

(Source: On-chain Merkel tree extraction from various exchanges)

CEX + DeFi + Circle's own platform accounts for almost 75% of the USDC distribution, and this does not even consider many scattered large traders on-chain. Compared to the demand for trading and storage, the contribution of USDC around payment scenarios is actually negligible.

Counterexample 1: There is a conclusion in the market that states: due to the current BTC bear market (-60%), the issuance of USDC has not significantly decreased (-10%), so Circle is gradually breaking away from the influence of the BTC market. This conclusion is a factual error. If we look back at the historical correlation:

In fact, even during the last BTC bear market, the outflow of USDC was not high. It was the black swan event of SVB's collapse that led to massive panic selling, making the previous USDC circulation appear highly correlated with BTC trends. This is a fallacy of attributing correlation as causation.

Conclusion 2: USDC's performance is still highly dependent on the prosperity of the crypto market. A bear market in crypto will not significantly affect the stock of USDC, but it will certainly greatly stifle the mid-term growth of USDC's scale.

Having discussed scale, what about USDC's barriers?

Here I will cite a classic bearish viewpoint: a certain bank could directly convert hundreds of billions in assets into stablecoins, and USDC would be finished.

The market understands liquidity, which usually refers to the total scale of stablecoins, but my view is that the quality of liquidity is as important a dimension as scale. Over the past decade, USDC has distributed extremely decentralized liquidity across major exchanges, payment service providers, DeFi platforms, and even down to tens of thousands of individual merchants. The quality of this liquidity is a significant moat. For example, PayPal and USD1 can quickly ramp up volume through subsidies, but their actual application remains at the stage of giving out red envelopes to retail investors, fundamentally due to the insufficient quality of their liquidity. The long-term accumulation of quality further brings about the sedimentation of user mindset, forming a brand effect among American users, which is the value of time and management.

Stablecoins are not a medium but the network itself, with compliance/non-compliance as the boundary; both sides are destined to be an oligopolistic market.

Regarding the OUSD issue, let's not talk about the organization's bloat (from my understanding of these cooperative names, most people just sign a name to occupy a spot and may not even invest any resources). Its cooperation with various distributors distributes profits based on contributions. In fact, USDC has already been doing this (with CB, BN, HYPE's revenue sharing). Its product competitiveness is not particularly obvious, and free minting/burning is actually a completely irrational gimmick. For example:

In Tether's official fee structure, the subscription fee is 0.1%, and the redemption fee is the higher of $1,000 or 0.1%, with a minimum subscription/redemption amount of $100,000. So what would happen if USDC offered free minting/burning? Answer: Users would exchange USDT for USDC on exchanges and redeem dollars from Circle, thus making USDC the free exit for the entire stablecoin market, while simultaneously bearing the costs of withdrawals and deposits/KYC/AML/liquidity management, but could only accept arbitrage flows that contribute nothing to revenue. This is a business that is difficult to sustain long-term.

Moreover, from another dimension, based on the previous conclusion (payment does not constitute the basic performance of Circle; trading and DeFi scenarios around crypto are the basic performance), its impact on Circle's stock scale is minimal, more like a narrative diversion.

From scale and barriers, we have argued the performance composition of Circle. Now let's look at other businesses that have a significant impact on mid-to-short-term valuation: ARC, CPN, Ai agent;

First, to understand a company's new product line, we need to look at its past product history, as shown in the diagram below:

(Note: Red indicates failure, yellow indicates pending verification, green indicates success)

Among the 20 product routes in Circle's past, only two products ultimately achieved relatively good results. From this company's history, in February 2018, Circle acquired the exchange Poloniex for $400 million at the peak of the bull market, attempting to create a closed loop of payment + trading + investment + stablecoin. However, Poloniex was already under investigation by the SEC before the acquisition (suspected of listing securities tokens), and there were compliance loopholes regarding OFAC sanctions. Circle essentially bought a non-compliant platform riddled with issues. Subsequently, crypto entered a winter, and just 18 months later, Circle was forced to sell Poloniex to an associate of Justin Sun, incurring a net loss of $156 million, along with an additional SEC fine of $10.4 million and an OFAC investigation. That same year, Circle Pay was also shut down. After Poloniex, Jeremy cut all non-core businesses and went all-in on USDC.

Conclusion 3: USDC has made Circle successful; the company's product strength is relatively average. The chances of success for ARC and CPN are not that high.

If we look at the management style of this company, the distribution of the main speaking content in the last earnings call was as follows:

The emphasis on Agent was nearly 20%, while the trading volume of Agent is actually very sluggish:

Let's review the style of Circle's earnings reports:

• Q2 2025: Circle Payments Network has launched in May, with over 100 financial institutions in the pipeline; also launched Arc, an L1 designed for stablecoin finance; emphasized partnerships with Gateway, Binance, Corpay, FIS, Fiserv, OKX, etc. (The stock opened nearly +16%, closed at +1.3%, and turned to -5% the next day, down -16% after five trading days)

• Q3 2025: The Arc public testnet has over 100 companies participating; the company begins exploring an Arc native token; CPN has 29 financial institutions enrolled, 55 under eligibility review, and 500 in the pipeline; CPN's trailing 30-day annualized trading volume reaches $3.4B; USYC has also grown over 200% since the end of June, approaching $1B. (The stock dropped -12.2% on the same day, relative to the previous close -16.2%, and down -29.1% after five days)

• Q4/FY2025: Q4 total revenue and reserve income of $770M, Adjusted EBITDA of $167M, year-on-year +412%; RLDC margin reached 40%. Other Revenue began to grow, reaching $37M in Q4, including subscription, transaction revenue, blockchain rewards, etc. The market began to price Circle not just as USDC float but possibly with an additional layer of platform revenue. (The stock rose +35.5% on the same day, relative to the previous close +42.1%, and +71.5% after five days.)

• Q1 2026: ARC Token presale raised $222M, corresponding to a $3B FDV; released ARC Token whitepaper; launched Agent Stack, including Circle CLI, Agent Wallets, Agent Marketplace, attempting to bind USDC and AI agent payments; CPN's trailing 30-day annualized volume rose to $8.3B; Managed Payments launched, allowing financial institutions to make stablecoin payments without directly managing digital assets; USYC became the largest tokenized money market fund. (The stock rose +15.9% on the same day, but fell back to +8.8% relative to the previous close the next day, and basically returned to -2.0% after five days)

Combining our assumption 3/ due to the real-time transparency and verifiability of on-chain data, Circle's trading opportunities largely depend on the pricing changes around other businesses beyond USDC.

Conclusion 4: Market valuation is highly dependent on the imagination brought by the marginal changes of Circle's other businesses, and Circle's management is very clear about this, as they create momentum around this point in every earnings report.

Investment Strategy

As of the time of writing, CRCL has been hit down to around 61 due to OUSD rumors and the decline of the BTC market, with a current market capitalization of about 16.9B, corresponding to a TTM revenue of about 5.8-6.0x P/S, which is close to the low range since its listing. The next Q2 earnings report is expected around mid-August.

In fact, the current price is gradually pricing in a company without a dream premium (with an added bonus of a dream that may come from the earnings report).

Three short-term variables for Crcl:

1. Clarity is somewhat like a call option (losing time value, but if it goes public, it will surge): Currently, the probability for Poly is 40%, and the Senate needs 60 votes to pass. The Republican votes are insufficient, and they must secure at least 7 Democratic senators. Currently, the Democrats are demanding stronger ethics clauses to limit crypto conflicts of interest for the president, senior officials, and their families. National security factions are also concerned that Section 604 / BRCA may overly protect developers, potentially weakening anti-money laundering and sanctions enforcement. The bill needs to complete the Senate vote before the August recess, and it would be best to also complete the House's confirmation of the amended version. After the July recess, we enter the midterm election season, leaving very little agenda time in the fall. The Senate's official schedule shows that June 29 to July 10 is a state work period, and August 10 to September 11 is another long recess. This means that the truly available window is mainly concentrated in the weeks from July 14 to August 7. However, the Senate still has to deal with priorities such as appropriations, defense, border, intelligence authorizations, and nominations. If CLARITY is to complete the entire process, it cannot wait until the end of July to start; it would be best to see a clear floor path by mid-July, and time is actually very tight.

2. BTC Market: After MSTR cleared risks a few days ago, I believe it is a relatively good signal for BTC. The crypto market is at a relatively low point in its four-year cycle, providing Circle itself with a good margin of safety.

3. OUSD: Short-term overreaction, OUSD has almost no impact on the earnings baseline; USDC's average scale in Q2 should not be bad; management is likely to respond strongly to OUSD around the earnings report, emphasizing network liquidity and channel barriers.

I believe that below 60 is a good left-side entry point (BTC market is oscillating at a relatively low point, and the negative news of OUSD is oversold). I have already bought a position, but in the short term, due to a lack of catalysts and the theta decay attribute of Clarity, Circle's stock price will continue to be under pressure, making it unwise to go all-in; it is better to slowly build a position while waiting for Clarity.

I will establish all positions during the period of 1/ Clarity uncertainty landing, 2/ before the next earnings report, which means completing the target position establishment within a month and a half.

Returning to the previous article:

Most people's so-called value investing is merely speculation around fundamental factors.

------ Zhao Haibei, WeChat public account: Confessions of a Speculator by Zhao Haibei

I have been tracking Circle for a long time. Circle is actually a target with a lot of retail holdings and a strong sense of "value" (it looks very valuable and represents a certain advanced productivity of humanity). I do not doubt its future elasticity. Again, this is a speculation, not a long-term value investment. I continuously question the narrative that Circle conveys to the market around the payment scenario, and I believe its valuation's high volatility is a normalized event.