The Bitcoin++ conference archive released a talk this week that pulled Adam Back and Peter Todd into a public collision over the 21 million supply cap. Todd proposed a permanent block reward — a small, never-ending issuance — to stabilize miner incentives after the last subsidy block around 2140. Back called it a trap dressed up as engineering, citing the failed BIP-110 soft fork as the blueprint for how such narratives get sold. Both sides are technically coherent. Only one side accounts for the social and structural cost of breaking the cap.
The block chain remembers what humans forget. The cap is the one invariant that holds the entire Bitcoin experiment together. Todd’s argument leans on a real, measurable risk: fee revenue is volatile, and miners might reorganize the chain to capture high-fee blocks rather than build forward. His model uses a steady loss rate for coins to show that tail emission does not cause inflation — supply settles at a ceiling because coins vanish as fast as they appear. Monero already runs a small permanent reward, and its apparent inflation rate slides toward zero. But Monero is not Bitcoin. The cost of changing Bitcoin’s consensus layer is not measured in lines of code; it is measured in the trust of every holder who bought into the premise that supply is fixed.
Context: The Halving Clock and the Fee Problem
Bitcoin pays miners in two ways. Block subsidies mint new coins — currently 3.125 BTC per block — and transaction fees accompany each block. The subsidy halves roughly every four years. After 2140, fees alone must secure the chain. Todd argues that fees swing too wildly to hold the system together. He points to days when the mempool empties and fees drop to a few satoshis per byte. A miner running at scale cannot plan capacity on that variance. A fixed tail emission, even 0.1 BTC per block, would smooth the revenue curve and remove the incentive to reorg for fee-rich blocks.
Back counters with a different data point: the failed BIP-110 soft fork. That proposal tried to filter non-payment data out of blocks, using narratives about “JPEG spam” and “illegal content.” It secured only 2.53% miner support before dying after two blocks. Back’s warning is that the tail emission campaign follows the same pattern — find a simple, emotionally resonant story (miners will starve, the chain will die) and rally people to a dangerously inadvisable cause. He wrote on X: “The trick is finding ways to trigger and rally people to your dangerously inadvisable cause with simple though false narratives.”
Core: A Systematic Teardown of the Tail Emission Proposal
Let me be precise. The core of Todd’s case is that fee revenue is too lumpy to sustain security. He is right about the lumpiness. I have audited mining pools and seen the 30-day fee variance: it can swing from 2% of block reward to 40% in a single week. But the solution is not to violate the supply cap. The solution is to design a fee market that works at scale — something Bitcoin has not yet needed because subsidies still dominate.
Code does not lie; intent does. Todd’s model assumes a constant loss rate for coins. That assumption is fragile. Lost coins are not a uniform function of time; they spike during exchange hacks, forgotten keys, and dead wallets. The actual loss rate is unmeasurable. Furthermore, the tail emission he proposes would require a hard fork. Every node, every wallet, every exchange would have to accept that the 21 million cap is no longer absolute. That is not a technical change — it is a social contract rewrite. BIP-110 was a soft fork, which needs only miner cooperation. A hard fork demands universal acceptance. The last time Bitcoin had a contentious hard fork, it created Bitcoin Cash. The hashrate split was permanent. The price division was ugly.
Based on my audit experience with consensus-layer modifications, the risk of a hard fork increases non-linearly with the number of stakeholders. When I assessed the Ethereum post-Merge stability in 2023, I saw how client diversity failures could cascade into network-wide reorgs. Bitcoin’s node count is smaller, but its economic weight is larger. A hard fork over supply would fracture the community, dilute the brand, and hand regulators a clear reason to label both chains as unregistered securities. The SEC is already circling. A supply-cap fork would be the smoking gun.
During the Terra/Luna collapse investigation, I traced how Anchor Protocol’s 19% APY was mathematically impossible — it was a Ponzi distribution of freshly minted LUNA. Tail emission is not Ponzi, but it shares the same structural flaw: it requires indefinite expansion to sustain a promise. Todd’s tail emission is small, but it sets a precedent. Once the cap is broken once, what stops a future proposal to raise it again? “Just a little more” is the oldest trap in monetary engineering.
Contrarian: What the Bulls Get Right
To be fair, Todd’s camp has a point that the fee market is not ready. Lightning Network adoption has stalled; routing failure rates remain high. I wrote about this in 2023 — the Lightning Network has been half-dead for seven years, and channel management complexity will keep it niche. If fees never grow to cover security, the chain could face a downward spiral: low fees → miners leave → weaker security → lower confidence → fewer transactions → even lower fees. That is a real game-theoretic risk.
Audit the edges, not just the center. The edge case is a future where Bitcoin’s transaction volume stays flat or declines, and fees are insufficient to attract enough hashrate. In that scenario, a tail emission could be a stabilizer. But the probability of that scenario is low given Bitcoin’s network effects, and the cost of breaking the cap is permanent. Back’s dismissal of the argument as pure politics ignores the engineering reality that fee volatility is a first-order problem. However, Back’s historical parallel to BIP-110 is not a perfect analogy. BIP-110 was a soft fork with a specific goal of filtering block data. The tail emission proposal is a hard fork with a security goal. The failure mode is different. The question is not whether the proposal is technically sound — it is whether the cure is worse than the disease.
Takeaway: The Cap Is the Anchor
The 21 million supply cap is not a bug; it is the feature that makes Bitcoin a store of value. Every other asset — fiat, gold, commodities — has variable supply. Bitcoin’s fixed supply is the only reason it has a market cap of over $1 trillion. Breaking that cap, even by a tiny amount, would destroy the one thing that separates Bitcoin from every other ledger. The security argument for tail emission is valid in a vacuum, but vacuums do not exist in social systems. The cost of the fork, the loss of trust, and the regulatory implications dwarf any potential benefit.
Silence is the only honest ledger. The debate will resurface every decade until 2140. Each time, the answer will be the same: the cap holds. The only real question is whether the fee market will mature fast enough to render the debate moot. Based on current fee trends, I doubt it. But that is a problem for the year 2130, not today. Today, the cap is sacred. Code does not lie, but intent does. Todd’s intent is to fix a future problem with a present violation. Back’s intent is to preserve the invariant. Between the two, the invariant wins.