Every eight hours, a decimal shifts. 0.0001. Some interfaces render it as 0.01%. Most traders scroll past it the way they scroll past terms of service. That is the mistake. This tiny constant is the anchor rate inside the funding rate formula of every perpetual futures contract in the market. It executes three times a day. Three hundred sixty-five days a year. No volatility adjustments. No market conditions. No exceptions.
0.01% × 3 × 365 = 10.95%.
That is the number The Economist just placed in front of the global financial establishment. Perpetual futures quietly drain roughly 10% of value per year from long positions. The Economist called it a hidden cost. I call it the Perpetual Tax. And having spent years auditing the smart contracts and matching engines that run these markets, I can tell you the true figure is worse than their estimate.
The warning is not about a security vulnerability. It is about mechanism design. And mechanism design, unlike a hacked bridge or a stolen private key, cannot be patched with a code update. It has to be understood.
Let's start with the instrument itself. A perpetual future is a derivative with no expiration date. BitMEX invented the category in 2016. The design solves a structural problem that had plagued crypto derivatives until then: how do you keep a futures contract's price anchored to the spot market without a settlement event to force convergence? Traditional futures roll over. They expire. They force the basis to zero at settlement. Perpetuals never expire. So an alternative enforcement mechanism was required.
The answer is the funding rate.
Every funding period — usually eight hours, sometimes one hour on newer venues — longs and shorts exchange payments. The calculation has two components. The first is the base rate, an anchor typically set at 0.01% per period. The second is a premium coefficient that reflects the divergence between the perpetual price and the underlying spot index.
When the perpetual trades above spot, longs pay shorts. When it trades below spot, shorts pay longs. The system self-corrects: if too many traders are long and the price drifts upward, the funding rate rises, incentivizing shorts to enter. If too many are short and the price drifts downward, the funding rate flips negative, rewarding longs. In theory, it is an elegant price-discovery mechanism. In practice, it is a silent extraction engine.
Here is the crucial detail most coverage misses. In a perfectly balanced market — where the premium coefficient is zero and the perpetual price equals spot — the anchor rate still applies. Longs pay 0.01% three times per day. Unconditionally. That is not an incentive structure. That is a tax.
The locked-in annual cost of holding a long perpetual position — before fees, before slippage, before liquidation risk — is approximately 10.95%. This is the floor, not the ceiling.
Now let's quantify the full stack.
Cost Layer One: Funding Rate
The 0.01% anchor is the baseline. The premium coefficient scales it with market conditions. In a bull market with crowded longs, the rate extends dramatically. I have logged sustained funding rates above 0.08% per eight-hour period during the 2023–2024 rally — an annualized rate above 30%. In the 2021 mania, I recorded funding at 0.1% or higher for weeks. That is annualized above 109%. Retail longs, in their most euphoric state, were paying institutional shorts over 100% per year to maintain a directional bet.
Cost Layer Two: Trading Fees
Taker fees run 0.02% to 0.06% per open and close on major venues. Maker fees can be zero, even rebated. But retail traders are overwhelmingly takers. A trader opening and closing a position once per week pays roughly 2.6% to 6.2% annually in fees alone. Monthly rotation pushes the number toward 1% to 3%. Add that to the funding drain.
Cost Layer Three: Slippage
Slippage ranges from 0.05% to over 1% depending on liquidity and order size. Large positions in illiquid altcoins routinely suffer 2% to 5% slippage on entry alone. The market depth on most perpetual venues is concentrated in BTC and ETH; everything beyond the top two assets is a minefield. Slippage is a one-time cost per trade, but it compounds with frequency.
Cost Layer Four: Liquidation and Partial Reductions
The most destructive cost. When a position breaches its maintenance margin, the exchange’s liquidation engine force-closes the position at the market price, often in thin order books. The realized loss is amplified by fees and a liquidation penalty — typically 0.5% to 1.25% of notional on centralized venues, plus the insurance-fund deduction on some platforms. For a 10x leveraged position, a 10% adverse move wipes the entire collateral. The liquidation cost is not a percentage deduction; it is total loss. And even a partial reduction — some venues reduce a position when margin is insufficient rather than closing it entirely — realizes a permanent loss of capital that could otherwise recover with the price.
Combined Annual Carry: 15% to 50%+.
This is the number I want every retail trader to internalize. The Economist's 10% estimate is the theoretical minimum in a zero-premium, zero-fee, zero-slippage, zero-liquidation world. No one trades in that world.
Now let's trace the wealth transfer. This is where the analysis gets uncomfortable.
The funding rate is not a fee collected by the platform. It is a transfer between counterparties. When funding is positive, shorts collect from longs. Who sits on the short side of the perpetual market?
Market makers. Hedging desks. Institutional arbitrageurs running the cash-and-carry trade: buy spot, short the perpetual, collect the funding rate, remain delta-neutral. This is the least emotionally charged, most systematic flow in the entire digital asset ecosystem. It exists precisely to harvest the Perpetual Tax. And it harvests it from retail directional conviction.
The cash-and-carry trade is the most reliable institutional strategy in crypto because the funding rate structurally subsidizes the patient side. The retail long pays for the institutional hedge. Every. Single. Day.
I have examined the accounting tables of arbitrage desks that treat funding yield as their primary revenue line. Their entire business model depends on the perpetual market's persistent positive carry. They do not predict price movements. They do not need to. They simply collect the funding differential between the spot and perp markets while remaining directionally flat. The risk is minimal; the yield is steady; the counterparty is anonymous retail.
The ledger remembers what the wallet forgets. Millions of retail participants experience this drain as gradual account decay. Small funding debits scattered across hundreds of interfaces. No single transaction large enough to trigger alarm. Just entropy. Meanwhile, the institutional flow is precise, automated, and relentless.
This is not a hack. It is not even a bug in the conventional smart-contract sense. It is a structural wealth-transfer mechanism embedded in the product design itself. And it runs in the direction that rewards computational patience over emotional conviction.
Now multiply the problem by leverage.
The funding rate is calculated on notional position size, not margin. A trader with $10,000 in margin opening a 10x position controls $100,000 notional. At the 0.01% anchor rate, the funding cost is $10.95 per day — in notional terms. Annualized, that is approximately $10,950. Against the original $10,000 collateral, the annual cost exceeds the entire account. The position is mathematically insolvent within twelve months even if the price does not move. Even if the trader predicted the direction perfectly.
Leverage does not merely amplify market exposure. It amplifies carry cost proportionally. A 20x position doubles the cost-to-collateral ratio. A 50x position makes the funding burden impossible to sustain. And the liquidation engine — the permanent sentinel in every perpetual market — ensures that positions are force-closed long before the math reaches its natural conclusion.
I studied this interaction in detail after the May 2021 Bitcoin cascade. Long positions across centralized exchanges had been paying sustained above-anchor funding for weeks. The long basis was crowded; the premium coefficient was extended. When the drawdown came, it wasn't just a market event. It was the compression of an over-costed position matrix. The funding had been systematically enriching the short side for months. The liquidation cascade was the final settlement sweep.
The bull market paradox deepens the problem. In an uptrend, the perpetual price trades above the spot index. The premium coefficient activates. The funding rate extends upward. The crowd — all positioned on the long side — pays elevated rates to institutional shorts. The more convinced retail becomes in an uptrend, the worse the tax becomes.
The funding premium is essentially a sentiment tax on directional consensus. It measures crowd conviction and prices it accordingly. In the 2023–2024 BTC rally, funding reached 0.08% per period multiple times — an annualized carry above 30%. The retail long was not just betting on the trend; they were borrowing conviction at a rate that would be deemed usurious in any regulated market.
Let's put the time decay into explicit arithmetic. Assume $100,000 in a long perpetual position. Assume zero price movement — the crypto market does nothing for five years. Assume the funding rate holds at the theoretical 0.01% anchor and no fees apply. After Year One: $89,050. After Year Three: $70,700. After Year Five: $56,100. A 44% loss of capital in five years with no market movement at all. The position bleeds out without a single bad trade.
Now extend the horizon. At an annual decay of 10.95%, the half-life of a perpetual long position is 6.3 years. Most retail traders do not hold positions that long. But they do hold them long enough to feel the cumulative effect. And they rarely isolate the funding cost from their P&L because it is fragmented across dozens of individual transactions.
This is the structural unfairness The Economist's report gestures toward but does not fully articulate: the cost is real, recurring, and disproportionately borne by less sophisticated participants. The market structure feedback loop is vicious: retail provides the counterparty depth, institutional arbitrageurs extract the funding spread, and the platform captures the fees on every forced liquidation.
The ecosystem participants differ meaningfully in their market structure exposure. Based on industry-wide observations and public trading-volume data, Binance Futures commands roughly half of the global perpetual volume. OKX and Bybit account for another 20–30%. Decentralized venues — dYdX, GMX, Hyperliquid, and others — hold a smaller but growing share. Each venue implements the funding mechanism with slightly different parameters. The core economics remain identical.
I audited a decentralized perpetual protocol in 2021 with a novel funding formula that capped rates at zero. It was a fascinating engineering exercise: the team had recognized the retail drain problem and attempted to redesign around it. The result shifted costs into slippage and liquidity-pool risk instead. The trade-off was not a free lunch. It was a different flavor of the same structural challenge.
From a forensic engineering standpoint, the funding mechanism itself is technically sound. The formulas are deterministic. The state transitions are auditable. The economic incentive models are internally consistent. The problem is not the code — it is the design intent. The perpetual futures contract was designed as a short-term trading instrument. It was never optimized for long-term positioning. And yet, a significant share of its volume comes from participants using it precisely for that.
I have reviewed perpetual contract implementations from BitMEX's original WAVES codebase to modern DEX protocols. The engineering is consistently clean. The economic fairness is consistently not. You cannot audit your way out of a mechanism that was designed to extract carry from one side of the trade. You can only understand it and decide whether to participate.
Which brings me to the contrarian angle — the part of this story that most mainstream coverage gets wrong.
The Economist's warning is not merely journalism. It is a regulatory trigger. And that may be the market's only meaningful protection.
Every significant retail-facing leverage intervention in modern financial history has followed the same narrative path: a respected institutional voice quantifies the cost to retail; the regulator cites the study; restrictions materialize. The ESMA CFD restrictions of 2018 — which capped retail leverage and forced clearer risk disclosure — followed this exact template. The UK FCA's 2021 ban on crypto derivatives for retail consumers was grounded in consumer-protection narratives. The EU's MiCA framework now in force adds another layer. The phrase "customer harm" is the regulatory magic key. The Economist just provided the evidentiary lock-pick.
Stablecoin reserve requirements under MiCA are already crushing smaller issuers. CASP licensing costs are stratospheric. Leveraged perpetual products are the next target. The compliance cost of over-the-counter derivatives restrictions, when applied to crypto perps, will transform the market. Offshore platforms will face pressure; compliant venues will benefit; decentralized protocols will face an uncertain regulatory grey zone.
Here is the counter-intuitive insight: this is good for retail.
The warning gives retail traders information they were never given before. Information permits rational exits. The trader who reads this analysis and decides to deleverage — or to migrate from high-leverage perps to spot or regulated futures — has just saved years of compound decay. The regulatory pressure that follows will push exchanges toward disclosure. The disclosure will push more retail out of the negative-expectancy structure.
Code is law, but bugs are the human exception. The funding rate is not a bug. It is a design choice with clear mathematical consequences. But leaving retail participants financially illiterate about its cost structure — that is a user-interface bug in the largest unregulated leveraged market ever built. The Economist just delivered the patch notes.
There is a deeper governance problem beneath all of this. On centralized exchanges, the funding rate parameters are controlled by the exchange operator. They can be adjusted at will. On decentralized protocols, parameter changes pass through governance. But governance tokens are concentrated among early investors and large holders. The retail long paying the tax does not hold enough voting power to influence anything. I witnessed this dynamic directly in 2022 after a protocol quietly adjusted its funding ceiling — the governance vote passed with overwhelming whale support. The people who pay the rate had no meaningful representation. Decentralized governance is not automatically equitable governance.
The ledger remembers what the wallet forgets. The funding debits are recorded on-chain or in exchange databases, immutable and traceable. The cumulative extraction over years is visible to anyone who bothers to look. Very few retail traders look.
The warning's practical effect will follow a predictable sequence. The narrative will fade from headlines within weeks. The math will not. I expect a slow migration over the next 18 months: retail flow shifting toward spot markets, low-cost decentralized perps, or regulated futures with transparent fee structures. The offshore high-leverage venues will consolidate their remaining users — mostly algorithms, market makers, and professionals. The funding rate will evolve into a niche micro-market for institutional carry strategies. The warning, in other words, will accelerate the professionalization of the perpetual market — an outcome that is good for the market's long-term health and devastating for the casual retail speculator who refuses to adapt.
My view is shaped by two decades of observing these dynamics. In 2017, during the ICO peak, I spent eight weeks reverse-engineering a DEX's exchange contract and identified integer overflow vulnerabilities that the whitepaper never mentioned. In 2020, during DeFi Summer, I verified the invariant equations of a stablecoin swap protocol and discovered precision loss in amp coefficient calculations that undermined the protocol's economic assumptions under volatility. In 2021, I audited an NFT minting contract with absent access controls and produced a live exploit simulation. The pattern across all these cases is consistent: the marketing narrative and the technical reality diverged. The code was the only truth. The same principle applies to perpetual futures. The funding rate is in the code. The cost is verifiable. The tax is real.
So here is my recommendation as an engineer, not an economist: measure before you trade. Open the funding rate chart for any instrument you hold. Calculate your full carry — funding, fees, slippage, liquidation probability. Subtract it from your expected return. If the residual is negative, you are not holding an asset. You are renting a trade from a counterparty that understands the rent better than you do.
And if you decide to stay in the perpetual market, understand what you are. You are the funding provider. You are the carry yield. You are the other side of an institutional strategy that has calculated your cost structure more precisely than you have.
That will not change in this cycle, or the next one. The formula doesn't care about your conviction. It executes every eight hours. The only remaining question is whether you will treat the perp market as what it is — a talent transfer mechanism from the impulsive to the systematic — before or after the ledger proves it to you.