In a world of ledgers, who holds the memory?
This is the question I found myself asking as I read The Economist's recent warning that perpetual futures quietly drain approximately 10% of value from long positions every year. The publication โ read in the boardrooms of central banks, on the trading desks of systematically important institutions, and in the policy circles that write financial law โ had quantified a cost that the crypto industry has known about, resisted naming, and built entire business models upon for eight years.
This is what I mean when I say the industry has a memory problem. Not a technical memory โ the blockchain preserves every transaction in immutable detail. But an institutional memory, a cultural accounting of who pays, who earns, and who designed the ledger to make that the cost of doing business. The Economist did not discover an exploit. It audited a feature and found the feature's social contract wanting.
We code the trust, but we must audit the soul. And the soul of the perpetual futures market is now under external examination.
Let me be precise about the architecture in question, because the conversation that follows demands technical rigor.
Perpetual futures โ perps to the initiated โ are derivative contracts with no expiration date. They were popularized by BitMEX in 2016, which designed them to solve a real and persistent problem: traditional futures contracts require rolling. When a contract approaches expiry, traders holding exposure must close their position and reopen the next month's contract, generating transaction costs and basis risk. Perpetuals eliminate this entirely. You enter, you hold, you exit when you choose. In theory, they are the closest thing the derivatives world has to frictionless directional exposure.
The mechanism that makes this perpetual existence possible is the funding rate. This is a periodic payment between longs and shorts, calculated to keep the perpetual price anchored to the spot index. Most major platforms settle this every eight hours; some newer venues have reduced the interval to one hour. When the perpetual trades above spot โ evidence that leveraged buyers dominate โ longs pay shorts a premium proportional to the divergence. When the perpetual trades below spot, shorts pay longs.
The formula breaks into two components. There is a base rate, standardized at 0.01% per period on most venues. This functions less like a fee and more like an interest charge โ a minimal, deterministic expense for the convenience of leveraged exposure that never expires. Then there is the premium component, which scales with the gap between the perpetual and spot prices. In a perfectly balanced market, the premium component is zero, and the long simply pays the base rate.
The math is deceptively simple. 0.01% per eight-hour period. Three periods per day. 365 days per year. Without any premium, a long position carries an annual cost of approximately 10.95%. That is The Economist's 10%.
But the market is rarely perfectly balanced. In a rising market โ which crypto has historically been during its expansion phases โ leveraged buyers cluster on the long side, pushing the perpetual above spot, and the funding rate climbs. Longs pay more. I have seen funding rates in the range of 0.1% per eight-hour period during peak bull sentiment โ that is 0.3% daily, 109.5% annualized. Even what feels like a minor premium becomes an extreme recurring cost when expressed as an annual rate.
The Economist's estimate is therefore best understood as a floor โ the cost of existing in the product under neutral conditions. The actual experience of a leveraged long, particularly during momentum phases, is frequently far more expensive.
Now the market context. Perpetual futures are not a niche instrument; they are the engine room of crypto trading. Industry estimates put derivatives at roughly 70-80% of total crypto trading volume, and perpetuals constitute roughly 80-90% of derivatives volume. Binance, OKX, Bybit, dYdX, GMX, Hyperliquid โ the most active venues in the ecosystem organize their product offerings around perps. The funding rate is not the first thing a new user sees in the interface, but it is the mechanism that makes the entire product category function.
The Economist's timing deserves attention. This warning arrives in a period of regulatory hardening โ the UK's FCA has banned retail crypto derivatives outright since 2021, the EU's ESMA has imposed leverage restrictions across all crypto derivatives, and the US CFTC has pursued enforcement actions against offshore derivative offerings. A mainstream quantification of inherent product costs is precisely the kind of evidence that regulators use to justify broader intervention. The narrative arc from "this product quietly extracts value from participants" to "therefore participation should be restricted" has historical precedent across every leveraged derivatives market that predates crypto.
Let me now deconstruct the 10% figure with the perspective I have gained from auditing protocol designs and governance frameworks over the past decade. The Economist's warning is accurate in its accounting but conservative in its conclusion. The actual cost to leveraged longs, under realistic conditions, is substantially higher than 10% annually.
The first layer of understatement is operational. The funding rate is only one of four compounding cost vectors. Trading fees on major centralized venues range from 0.02% to 0.06% per side โ a round trip costs 0.04% to 0.12% before any theoretical edge is captured. For a trader who rebalances monthly, this adds roughly 0.5% to 1.5% annually. For active traders who move in and out several times a week, the annualized drag approaches 10% or more on top of funding.
Slippage is the second understated cost. On a liquid venue, a modest order might slip a few basis points. But on altcoin perps, which is where retail engagement concentrates, slippage of 20 to 100 basis points per entry is common. The larger the position relative to the order book, the worse the execution price. A retail trader entering a fifty-thousand-dollar position in a mid-cap token can reasonably expect 0.5% to 1% per entry. Combined with fee drag, the total operational cost of an average quarterly cycle is several percent.
Liquidation is the third โ and most destructive โ layer. When margin falls below the maintenance threshold, the position is forcibly closed, and the trader incurs a liquidation penalty on top of their loss. The penalty varies by platform and leverage tier but commonly ranges from 0.5% to 1% of the position's notional value. Liquidation is not a discreet risk that can be hedged away; it is the brutal enforcement of the product's leverage logic. And the higher the leverage, the more frequently liquidations occur. I have modeled this repeatedly: a 10x leveraged long with a 2% adverse move suffers a roughly 20% equity drawdown before liquidation triggers. The cost of being wrong is not linear; it is sharply exponential.
The second dimension of understatement is leverage amplification. This is the insight I find most important, and I have rarely seen it properly explained in mainstream financial commentary: funding is charged on notional exposure, not on the margin used to achieve it. A trader with 10x leverage is paying funding on ten times their actual capital. The annualized funding cost, measured against equity rather than notional, is effectively 100% per year if the position is held at full leverage without adjustment. True, no one holds for a year at full notional without some position management โ the volatility alone forces intermittent de-leveraging. But this arithmetic explains the classic puzzle: why do leveraged longs lose money in markets that end the year relatively flat? They are paying interest on capital they never actually owned.
A concrete example illustrates the point. A trader deposits $10,000 and opens a 10x leveraged long position โ $100,000 of notional. The base funding rate of 0.01% per 8 hours on that $100,000 notional is $30 per day. That is $10,950 per year โ more than the initial margin. Even if the market remains perfectly flat, the funding cost alone would erode the account to maintenance margin within a few months. Fluctuations in funding rate, occasional fees, and slippage accelerate the timeline. It is not accurate to say the trader loses 10% annually. It is more accurate to say the trader's window of survival is structurally constrained.
The third dimension is structural asymmetry. The perpetual futures market is not a neutral venue where longs and shorts meet as equals. It is a market engineered so that the most informed participants on the short side systematically harvest a yield from leveraged retail demand. When the market trends up โ and crypto historically trends up over multi-year horizons โ the perpetual price typically holds a premium over spot. This means the funding rate is persistently positive, and longs consistently pay. The recipients are institutional market makers and arbitrage desks who run basis trades: buy spot, short the perpetual, and collect the funding with market-neutral risk. For these participants, persistent positive funding is not a cost; it is a reliable income stream. It is, in effect, a yield instrument that the retail long subsidizes.
This dynamic creates a peculiar form of class stratification. The retail trader who buys a perpetual long because they believe in the asset's appreciation is paying rent to the professional class that provides the liquidity. The system calls this market efficiency. From the perspective of the payer, it functions more like a subscription fee for a market access privilege that is never fully disclosed.
What of the decentralized alternatives? dYdX, GMX, Hyperliquid, and other decentralized perpetual protocols have attempted to address some of these issues. Their funding parameters are transparent and on-chain. Fee structures vary, with GMX experimenting with zero-funding models that internalize costs in spreads and liquidity pools. But the transparency advantage cuts both ways: the underlying economics are fully visible, precisely because the costs are real regardless of whether a centralized or decentralized venue hosts them. A decentralized venue can make audits easier, but it cannot eliminate the fundamental structural cost of leveraged perpetual exposure.
There is also a deeper infrastructural dependency that few analyses acknowledge. Every decentralized perpetual protocol relies on a price oracle to determine the index price that anchors funding and liquidation calculations. We have seen how fragile this pipeline can be โ flash-crash events, oracle manipulation incidents, and latency arbitrage have triggered cascading liquidations across venues. When I analyze the risk surface of decentralized perps, I find that oracle feed latency is the true Achilles' heel. A fast, honest funding rate is meaningless if the price feed that determines it can be gamed. Some of the most prominent oracle providers claim decentralization while operating what are effectively permissioned node clusters with centralized update keys. If The Economist had dug one layer deeper, they would have found that the funding mechanism itself relies on infrastructure whose integrity is assumed rather than guaranteed.
Let me share something from my own experience. When I audited a DAO governance framework in 2017 โ a project I declined to be paid for, because I believed the community's security mattered more than the market's enthusiasm โ I discovered three critical vulnerabilities in their smart contracts. What struck me was not the technical flaws, but the fact that the project's governance architecture had been designed to maximize capital inflow while devoting minimal attention to downstream user risk. The incentives were misaligned from the first line of code.
I see the same pattern in perpetual futures. The product architecture is elegant because it solved a mechanism problem. It is ethically incomplete because it never solved the information and governance asymmetry that the mechanism creates. Retail participants shoulder the cost, retail has no seat in parameter-setting decisions, and exchange revenue grows as volumes and forced liquidations increase. Funding rate design, liquidation thresholds, insurance fund allocation โ every decision is made by parties whose revenue model benefits from trading volume rather than from long-term participant wealth. The Economist's warning names an effect; the cause lies in a governance structure that fundamentally excludes the people who pay.
The market impact of this warning demands analysis. Cryptocurrency derivatives markets are dominated by retail participation โ BIS research from 2022 suggests retail accounts for over 70% of crypto derivatives volume. When a mainstream, widely distributed publication quantifies an inherent product cost, the initial reaction is typically dismissive: they do not understand, we know what we are doing. But the cognitive shift is real. Over time, exposure to a credible, quantified structural cost changes risk perception. The Economist is not a crypto-focused outlet; it is read by the people who set the frame for financial public discourse. The phrase "quietly drain" will find its way into regulatory arguments, policy briefings, and advocacy materials for restrictions on leverage and disclosure requirements.
There is also a competitive dimension that the coverage will accelerate. Centralized perpetual exchanges โ with Binance Futures dominating at an estimated 50%+ market share, and OKX and Bybit holding substantial portions of the remainder โ will face increasing scrutiny of their fee structures and funding mechanisms. Decentralized venues, with transparent parameters, may gain relative appeal among users who value auditability over convenience. But the incumbent exchanges' liquidity advantage is formidable; users tolerate cost opacity when execution quality is superior.
The regulatory trajectory is the deep undercurrent. This warning arrives in a window where financial authorities are actively constructing frameworks for digital assets. The EU's Markets in Crypto-Assets Regulation, MiCA, is now the world's most comprehensive crypto regulatory framework. The US has pursued enforcement-first policies. The UK has banned retail crypto derivatives. A mainstream read that perpetual futures systematically disadvantage retail participants is exactly the type of analysis that authorities cite when implementing product restrictions. The combination of a prestigious institutional voice and quantitative evidence creates a policy moment. Whether this moment results in well-calibrated disclosure requirements or blunt participation bans will depend on how the industry responds.
Now we arrive at the uncomfortable counter-thesis. And I want to be honest about it: the 10% funding drain is not entirely an injustice. It is also a price โ a toll exacted for the deepest liquidity in the crypto market, and removing the toll without replacing the liquidity would harm both the market's function and the very participants the warning intends to protect.
Perpetual futures are the deepest, most efficient markets in crypto. They serve as the price discovery mechanism that anchors spot trading across every major exchange. The funding cost is, in effect, a liquidity subsidy. It compensates the parties who take the other side of leveraged demand โ who provide counterparty access, hedge risk, and maintain continuous order books. Without this compensation structure, the depth of the books would shrink, spreads would widen, and the price discovery function that anchors spot markets would fragment. The retail long's 10% annual cost is, in part, rent for a market so liquid that it can absorb institutional-size entry and exit without catastrophic slippage.
This is not to excuse opacity. It is to insist that the problem is complexity, not conspiracy. Deleveraging the market through blunt regulation would reduce the toll on participants, but it would also remove the liquidity that makes the ecosystem viable.
The second contrarian point concerns regulatory responses. The most likely policy reaction to The Economist's warning will be leverage restrictions and participation bans โ measures that follow the pattern of the FCA's retail ban and ESMA's leverage caps. But these tools are blunt. They privilege sophisticated, accredited investors who can access the product through exemption channels while pushing excluded retail participants toward offshore, unregulated venues with zero disclosure standards and unpredictable enforcement. The actual consequence of a well-intentioned ban is frequently a redistribution of risk, not its reduction. We saw this pattern play out in the early 2000s when derivatives restrictions pushed retail leverage into opaque channels; we saw it again when pandemic-era measures reshaped retail trading flows into unregulated digital venues.
The deeper structural problem, as I see it, is not that perpetuals carry a cost that is too high. It is that the crypto industry has failed to provide a retail-facing alternative. Where is the product for a saver who wants crypto exposure without leverage, without expiry-based rollover, without counterparty games? The ETF market has partially addressed this in the United States. Staking offers a carry-positive alternative โ earn yield by supporting network security rather than paying yield for leverage. But the derivative venues dominate the interface layer because trading volume generates revenue in a way that patient capital accumulation does not. The ecosystem has bet heavily on the trader's liquidity magnet and under-invested in products aligned with the saver's balance sheet.
There is a parallel here with the L2 competition narrative. The industry's dominant layer-2 frameworks โ the optimistic rollup stacks and the zero-knowledge stacks โ are not ultimately differentiated by their mathematical properties. They are differentiated by persuasion, by which framework convinces more projects to deploy first, by which ecosystem's momentum becomes self-reinforcing. Perpetual futures went through exactly this dynamic: the mechanism that captured the most venues and the deepest liquidity became the industry standard, not because it was the fairest design, but because it won the adoption race. Now the industry is locked into a cost structure that follows from that early victory.
The final contrarian insight is about duration. All leveraged products carry carrying costs. In traditional futures, the embedded cost appears in the roll โ the difference between the expiring and the next-dated contract. In options, it appears in the decay of time value. Perpetual futures simply made the cost explicit, periodic, and visible to anyone who reads the funding table. The anonymity of the cost in traditional products is arguably worse โ most equity option buyers have no idea they are paying theta every day. The Economist could write a similar warning about any leveraged financial product in existence. What makes crypto perps special is not the existence of the cost; it is the concentration of unsophisticated participants who are structurally disincentivized from checking the price of the privilege.
The contrarian position I keep coming back to is this: the perpetual futures market does not have an information problem; it has a design problem. Everyone with a month of experience knows the funding rate exists. The Economist's warning will be circulated, discussed, and incorporated into essays such as this one. And then it will fade into background noise, because the alternative products that would channel frustrated retail capital toward non-exploitative structures are still immature. Information, without a destination, merely travels.
Proof is binary; meaning is fluid. The Economist's 10% is a proof โ a precise, verifiable quantification of a structural cost. But the meaning we construct from it is up to us.
We are not moving money; we are moving belief. The belief that a leveraged long in a perpetual contract is a path to wealth has cost their holders billions โ not through market crashes, but through the quiet, persistent erosion that The Economist has now described in terms the mainstream cannot ignore. The ledger records every one of those transactions; the question is whether we will build a market where the ledger also records a fairer distribution of the costs and benefits of participation.
I have spent years auditing protocols, modeling governance failures, and building infrastructures for decentralized identity in an AI-adjacent world. I have never met a mechanism that was inherently evil; I have met dozens that were designed without empathy. The funding rate is not evil. But its presentation, its governance, and its absence of meaningful disclosure to the users who pay it every eight hours โ those are design failures. And design failures are fixable.
The future I hope to see is not one where perpetual futures are banned. It is one where funding rate trajectories are visible in onboarding flows, where position economics are modeled in the interface before a trade is placed, where decentralized governance includes the voices of the participants who currently subsidize the market's liquidity. It is a future where longing a perpetual is a choice made with open eyes, not a habit grounded in a hidden tax.
In a world of ledgers, who holds the memory? We do. The Economist has reminded us what that memory contains. Now we must decide what it will become.