A fee is supposed to be a cost. It is the toll you pay for access to another party's liquidity, and its function is to make markets sustainable. When an exchange announces that it will rebate 110% of that toll, the system's grammar has broken.
Over the past several weeks, HTX โ the exchange formerly known as Huobi โ ran a "Trade to Earn" campaign built on precisely that inversion. Users received negative trading fees on perpetual contracts tracking QQQ, NVDA, MSFT, and gold, alongside a daily 6,000 USDT prize pool, all wrapped in the language of $HTX buyback and burn. The first phase has concluded, ostensibly successfully. A second phase has been promised, with details withheld.
I have spent enough years auditing incentive structures to recognize when a flywheel is actually a treadmill. This campaign deserves forensic attention, not because its mechanics are complex, but because the narrative architecture around them is doing the heavy lifting.
HTX is one of the oldest names in digital assets. Founded in 2013 as Huobi, it survived multiple market cycles and a succession of ownership changes before Justin Sun acquired a controlling stake in 2022. Since then, its history has included founder investigations, significant layoffs, and the kind of institutional turbulence that usually precedes retreat from the spotlight. Instead, HTX has chosen aggression: marketing budgets as defense, incentive campaigns as moats.
The "Trade to Earn" model is not novel. Binance operates Launchpool, Bybit has experimented with trading mining, and the broad category is now a mature playbook in exchange growth. What distinguishes this campaign is the pairing of a TradFi derivative menu with unusually deep subsidies. Users trade perpetuals tracking US equity indices and commodities. The platform rebates up to 110% of trading fees, which renders the marginal cost of trading negative for any participant who can reliably measure their own volume.
The official narrative frames this as a positive cycle: trading volume generates fees, fees fund $HTX repurchases, repurchases feed a deflationary burn, and reduced supply elevates token value. Approximately 1.8 billion $HTX tokens were reportedly destroyed during the campaign. That framing demands skeptical accounting. A burn figure is only meaningful when weighed against supply. $HTX circulation is measured in trillions. The scarcity discount implied by 1.8 billion burned tokens is almost negligible against that denominator, unless ongoing demand is enormous and durable. The campaign's own rewards may be increasing supply on the other end.
There is also the regulatory dimension. The perpetual contracts on NVDA, MSFT, and QQQ function, in substance, as leveraged CFDs on equities โ a retail product that major jurisdictions including the United States and the European Union restrict heavily or prohibit outright. The exchange operates from offshore bases, but the product itself exists in a legal gray zone that has historically attracted enforcement attention. These are not side considerations. They are central to evaluating what this campaign actually is.
Begin with the balance sheet. Where does the money come from? An exchange that rebates more than it collects in fees is not generating income. It is deploying capital. The critical question is which capital. The answer determines whether the destruction of 1.8 billion tokens is deflationary or merely theatrical.
If the rebate pool is funded from a dedicated promotional reserve, the arithmetic is neutral โ a marketing expense reported in the ordinary course of business. If the rewards are funded from newly issued $HTX, the supply reduction attributable to the burn is partially offset by emission. The published materials do not disclose this. Based on my audit experience โ six months in 2017 examining Golem's governance token mechanics, and three weeks in 2020 simulating impermanent loss on Uniswap to understand liquidity behavior โ I have learned that what a protocol omits about token flows is usually the most informative part of its documentation. Absence is a data point. When a project is silent on reward provenance, the default assumption must be that rewards come from treasury inventory or new issuance, not from organic revenue.
The conversion math is likewise instructive. A campaign that returns 110% of fees must be cross-subsidized by something: future user acquisition, expected token appreciation, or a broader strategic objective. None of these appear in the "positive cycle" narrative. The narrative presents the burn as the engine, but the actual engine is the external subsidy. A flywheel that requires continuous injections of outside capital is not a flywheel; it is a treadmill with a marketing budget. To call this a positive cycle is a category error. The loop does not generate its own energy; it consumes it.
Follow the flow of value, and you find the real beneficiaries. Market makers with low-latency infrastructure can generate enormous transaction volumes at near-zero directional risk. For them, negative fees are a direct transfer from the exchange's budget into their profit and loss statement. The daily prize pool adds to that. Retail users, by contrast, are drawn to the campaign by the promise of high APR and subsidized trading, and in many cases they become the counterparties against which algorithmic flow trades. The incentive structure does not distribute risk evenly. It concentrates reward at the top of the sophistication curve and concentrates risk at the bottom. This is not a judgment about intent. It is a description of the incentive geometry.
There is a broader competitive reading. The exchange landscape is unforgiving. Binance commands roughly half of spot volume, OKX and Bybit carve out their own derivative-market niches, and every one of them can replicate a "Trade to Earn" mechanic within weeks. The technical barrier to entry for this campaign is effectively zero. There is no novel protocol, no audit trail, no verifiable on-chain mechanism. It is a marketing department's product, and marketing departments copy each other quickly.
An exchange that relies on outsized subsidies to attract users is broadcasting weakness as much as it is projecting strength. If HTX's user base were growing organically, it would not need to pay users to trade at a loss. I want to be precise about what that means. Paying users to trade is a standard customer acquisition strategy in any industry. The problem arises when the acquisition cost is embedded in the token's value narrative, because then the token price becomes a function of continued marketing spend. That is fragile. When the subsidy ends, the volume exits, the buyback fund shrinks, and the narrative collapses under its own weight.
And then there is the regulatory architecture. Calling these products "crypto perpetuals" does not change their economic substance. A perpetual tracking NVDA offers leveraged exposure to a US listed security, settled in stablecoin, offered to global retail users from an offshore entity. Regulators in the United States and the European Union have consistently signaled that retail CFD products on equities and indices are unacceptable without rigorous licensing. Derivatives offered through unregistered venues have attracted enforcement actions before. The fact that this product is denominated in a tradable digital asset does not move the analysis; if anything, the leverage and the global distribution channel make scrutiny more likely.
This is where the behavioral layer enters. The campaign attracts a specific profile: arbitrage syndicates, farming groups, and speculative retail chasing yield. These are not inherently durable user relationships. Loyalty built on subsidy is transitory; when the rebate terminates, so does the relationship. The architecture being constructed here is not trust. It is temporary alignment of interest. In the void, we find the architecture of trust โ but in this void, we find manufactured incentives instead. Trust requires consistency across time. It requires the absence of selection bias in favor of sophisticated actors. It requires disclosure of where value actually comes from. None of those conditions are visible in this campaign.
The sentiment analytics reinforce the concern. The campaign generates social media heat disproportionate to its fundamental contribution. That ratio โ heat to substance โ is a reliable marker of narrative-driven price action rather than value creation. I have watched this pattern across multiple cycles: in the ICO era, in DeFi summer, in the NFT boom. The tokens that rallied most persistently were those where the narrative was backed by measurable, auditable flows. The tokens that spiked and decayed were those where the narrative carried the fundamental weight. $HTX, in this campaign, sits firmly in the second category.
It is also worth noting the timing. Running a heavily subsidized campaign during a bear market, when trading volumes are depressed across the industry, is a period-appropriate move. It buys attention. It builds the appearance of momentum. And it generates the kind of data that an exchange needs to decide whether to continue investing in a market segment. But it is a test, not a transformation. The campaign does not change HTX's competitive position. It does not improve its technology, its compliance posture, or its user retention strategy. It rents attention. The deeper issue is informational asymmetry. The market knows how long a subsidy lasts only when the exchange announces it, and no exchange announces the end date of its own narrative.
There is, however, a counterintuitive reading that most critics will miss. What if the campaign is not about users at all? Consider the alternative hypothesis: the primary target is order book depth. By subsidizing market makers through negative fees, HTX consolidates liquidity on its own books during a period when competitors are also fighting for thin flow. Deep order books attract institutional interest regardless of promotional mechanics. The token burn narrative, meanwhile, supplies a sentiment floor for $HTX ahead of the second phase, which may coincide with a larger token event or a derivatives expansion. Under that reading, the campaign is not an act of desperation. It is a rational expense designed to buy positioning and data.
The blind spot in the standard dismissal โ "this is unsustainable, therefore it is worthless" โ is the assumption that all exchanges compete for the same customers on the same metrics. There is a genuine cohort of traders in jurisdictions where access to US equities is restricted, for whom HTX's perpetual menu is a rare gateway to directional exposure on names like NVDA. Subsidies are a discovery mechanism. They reveal whether that cohort exists and at what cost it can be retained. The data generated during the first phase will inform a quieter, more strategic calculation than the headlines suggest. Chaos is just data waiting for a story; this campaign is an experiment engineered to manufacture exactly that data. The observable outcome โ whether phase two arrives with deeper subsidies or narrower parameters โ will tell us which story HTX believes.
Phase two is the relevant test. Track three signals: the published details of the new rebate schedule, on-chain data from $HTX burn addresses, and HTX's aggregate reserves across major stablecoins. Watch also for regulatory responses in the United States and the European Union; one enforcement letter can dismantle an entire product line. The broader question remains unanswered. Anyone with capital can subsidize volume. The measure of this campaign is what survives the subsidy. We build bridges in the silence after the noise. When the rebates end, we will see whether HTX built a bridge to a durable user base, or a toll booth on a road that leads nowhere.