Category 4: How Russia Turned Its Miners Into Interruptible Load

Somewhere in the regulatory annex of Russia's Energy Ministry, a single classification did more damage to mining economics than any headline ban ever could. Miners were not prohibited. They were reclassified. Category 4. In February, the Russian government established a fourth tier of power-supply reliability. The first three are familiar to any grid engineer: hospitals and defense installations at the top; ordinary industrial and residential users in the middle. The new fourth tier is reserved for a single client type — crypto miners. The catch is mechanical, not rhetorical. Category 4 users can be disconnected at any moment, without prior consultation. No negotiation. No notice window. The code whispered secrets the whitepaper buried. That single phrase — "no prior negotiation" — is the entire story. Everything else is commentary. To understand why this matters, you need the grid architecture, not the press release. Russia legalized industrial crypto mining in 2024. For a sector that had spent years in a gray zone — cheap Siberian hydroelectricity, lax enforcement, abundant stranded gas — legalization read as validation. Miners celebrated. Analysts priced in "regulatory clarity." The narrative was simple: Russia wanted to formalize a sector it already dominated. Russia holds a substantial share of global Bitcoin hashrate, historically estimated in the low double digits, making it one of the largest mining jurisdictions on earth. The economics rest on one variable: electricity cost. Siberian hydro, subsidized regional tariffs, and proximity to stranded energy gave Russian miners a structural edge no amount of capital expenditure could replicate elsewhere. But the 2024 legalization carried a clause the celebratory coverage buried. Legal does not mean prioritized. And in winter, Siberia and the Far East face genuine power shortages — residential demand spikes, aging transmission strains, and the grid must decide who gets cut. The Energy Ministry's answer arrived in February 2025, formalized as a new reliability category. The framing was careful. The ministry stated the measure would "effectively utilize existing energy infrastructure without threatening residential supply." Read that sentence twice. The operative word is residential — because it tells you exactly who the miner is being ranked against. Here is the mechanical anatomy. In power-systems engineering, supply reliability categories rank users by importance. Tier 1 — hospitals, water treatment, defense — is shed last. Tier 3, the traditional floor, is shed first among ordinary users. The introduction of a Tier 4 is not a technical innovation. It is an institutional downgrade dressed as a regulatory category. The design logic is internally consistent, and that is what makes it dangerous. Miner load is highly elastic. An ASIC can be powered down instantly with no humanitarian consequence — no darkened operating room, no frozen water main. From a grid-balancing perspective, miners are the ideal demand-response resource. Ranking them below every other user is, purely as engineering, rational. But rational engineering and fair treatment are not the same thing. Between the lines of the ABI lies the intent. What Russia has constructed is a revocable-permission model. Miners are legal. They can physically connect to the grid — the policy even permits "technological connection" in power-deficient regions. But connection is not supply. This is the critical distinction the coverage missed. Technological connection means the wire exists. It does not mean the wire stays energized. So the miner receives a strange hybrid right: the right to plug in, without the right to stay on. In finance terms, this is operating-availability risk with no contractual hedge. In grid terms, it is interruptible load — a customer class that accepts lower reliability in exchange for lower tariffs, cut first when the system tightens. Now quantify the cost, because that is where abstraction bites. A mining facility is a heavy-asset business. Site selection depends on proximity to cheap power. Capital expenditure is sunk into immersion cooling, transformers, and long-term power-purchase agreements. The investment model rests on one assumption: uptime. A facility that runs 95% of the year has a computable cost-per-hash. A facility whose uptime is administratively determined — with no disclosed trigger, no frequency cap, no restoration protocol — has no computable cost-per-hash at all. For the individual operator, this is a slow-motion squeeze. Every megawatt-hour of interrupted load is a hashrate unit that will never be produced, a block reward that will never be captured, a machine that depreciates while it waits for the wire to return. In a bear market, when hashprice compresses toward the cost of production, uptime is not a performance metric — it is the difference between survival and insolvency. A policy that makes uptime administratively uncertain is, functionally, a policy that decides which miners live. The policy never defines what "energy shortage" means. Who declares it — the federal ministry or the local grid operator? At what threshold? For how long can a disconnection last? Is there an appeal? A compensation mechanism? The source material is silent on every one of these questions, and silence in a regulatory document is not an oversight. It is discretion preserved. The information gap is itself a finding. The policy discloses no restoration mechanism, no maximum switching frequency, no distinction between ASIC and GPU facilities, and no pathway for a miner to contest a disconnection. In an audited protocol, these would be flagged as missing invariants — the conditions under which the system is guaranteed to behave. Here, the invariants are absent by design. What cannot be measured cannot be challenged. Logic does not lie, but architects often do. And the architecture here is deliberate. The 2024 legalization and the 2025 reclassification are not contradictory. They are complementary. The state captured the mining sector into the formal economy, then retained unilateral power to throttle it. Legalization without supply guarantees is not a gift to miners. It is the acquisition of a control lever. Consider the sequencing. First, bring miners inside the regulatory perimeter, where they become visible, taxable, and countable. Then define their supply reliability below that of every other user, so that when the grid tightens, the state can shed them without legal friction. The miner becomes a grid buffer — a shock absorber the system compresses at will, with the political cover of "protecting residents." This is institutional centralization mapped onto physical infrastructure. The blockchain is decentralized. The electricity feeding it is not. Whoever controls the electrons controls the hashrate, regardless of what the consensus layer claims. Based on my audit experience, I have learned that the most consequential clauses are the ones written in the passive voice. "Users may be disconnected" is a sentence designed to hide an actor. Someone disconnects them. Someone decides when. The grammar conceals the discretion the substance grants. There is one more layer the headlines skipped. The disconnection authority does not just protect residents — it redistributes margin. When the grid sheds a miner, that power flows to whoever bids higher. The miner's loss becomes another user's gain, and the grid operator captures the coordination premium. Mining has quietly become the residual claimant of the Russian grid: the last in line for power, the first in line for curtailment, and the easiest to blame when the lights flicker. The transmission chain runs cleanly, which is how you know the design was thought through. Upstream, energy and grid policy produce a new reliability tier and a discretionary disconnection right. Midstream, mining operations absorb the shock: facility uptime becomes a non-controllable variable, cost-per-hash rises unpredictably, payback periods stretch. Downstream, regional hashrate and miner treasuries adjust — curtailed miners earn less, sell differently, or migrate. The crucial point is where this chain terminates. It terminates at regional hashrate, not global. Russian hashrate is significant but geographically substitutable. If Siberian facilities throttle, hashrate reallocates to Texas, to the Middle East, to wherever power is firm. The global network absorbs the shock. The individual Russian operator does not. So the policy is calibrated with precision: painful enough to force marginal miners out of power-deficient regions, mild enough to avoid a global hashrate crisis that would invite international scrutiny. It is a soft de-capacity tool — a way to squeeze weak operators out of contested grid territory without issuing a single ban. This is not a Russian peculiarity. Kazakhstan rationed miner power during its 2021–2022 shortages. Iran has repeatedly cut licensed miners during peak summer demand. Several U.S. states have debated curtailment frameworks. The global pattern is consistent: wherever the grid runs tight, mining is formally ranked below residential and critical demand. Russia's innovation is administrative — it turned that preference into an explicit, named reliability tier. A legal template. And notice what the policy does not do. It does not prohibit self-generation. It does not restrict off-grid operations. It does not touch stranded-gas or flare-gas mining. That is not an accident. The state wants mining to exist — the tax base, the regional employment, the hard-currency exposure. It simply wants it on terms where the miner, not the grid, absorbs the volatility. That is the trade being offered: operate, but hold the risk. Now the part the bears will hate. The bulls are not entirely wrong. The policy is not a prohibition, and reading it as one misses the actual mechanics. Russia explicitly permits technological connection even in power-deficient regions. It does not touch self-generation, off-grid mining, or flare-gas operations. The state is not trying to kill mining. It is trying to relocate the risk. This creates a genuine opportunity for operators who read the fine print correctly. Stranded energy — associated petroleum gas, curtailed hydro, surplus wind — becomes structurally more attractive, because it sits outside the grid's disconnection authority. A miner on its own generation is not a Category 4 customer. It is not a customer at all. The bearish consensus — "Russia is killing mining" — is the lazy read. The accurate read is that Russia is filtering mining, keeping the tax base while shedding the obligation to guarantee it power. Miners who build captive generation survive and gain relative advantage. Miners who depend on the public grid are now price-takers on a discretionary market. Read the function calls, not the press release. The function here is "shed load," and it is callable by the operator at any time. But it is also avoidable by anyone who stops being a load on that operator's grid. The lesson extends well beyond Russia. Every jurisdiction facing grid stress now has a template: legalize mining, then rank it last. The category is portable. The discretion is the point. For operators, the question is no longer "is mining legal here?" It is "who can switch me off, and what does it cost them to do it?" If the answer is "the grid, for free," you do not have a business model. You have a permission slip. Which raises the question the celebratory coverage never asked: if a state can revoke your power with a single administrative clause, in what sense was your operation ever decentralized?