KAS Inflation Analysis · September 2026 · Supply growing, projected to keep growing
Kaspa is the cleanest supply story in the framework and it still reads positive: the Pressure Framework puts KAS at +0.68% over the trailing 90 days and +0.57% over the next 90. Mining is the only mechanism on the page — 188.6M KAS of sell pressure, 0 of buy pressure, no premine, no vesting, no treasury, no buyback and no protocol burn. What makes the forward number smaller is the emission curve itself: Kaspa cuts the block reward by 5.6% every 30.4 days rather than halving it every four years, so the reward fell from 2.59565436 KASa block to 2.18267645 KAS across this window and reaches 1.83540480 KAS by the end of the next one, under a hard ceiling of 28,704.0M KAS.
The verdict, in one paragraph
Against a circulating base of 27,687.0M KAS, the framework books 188.6M KAS of sell pressure and 0 of buy pressure over the trailing 90 days — a net of +0.68% — and projects +0.57% for the next 90 on the same published curve. The inflation monitor reads +0.71% for the same window, a gap of 0.03 percentage points, which is far inside the framework's 0.5pp tolerance, so this page ships with no data-conflict warning at all. That agreement is not luck: on a fair-launched chain with no allocation buckets to classify, an independent supply estimate and the chain's own coin count have almost nothing to disagree about. The label for KAS is a fair-launch miner on a decaying issuance curve: the only supply is new supply, and every quarter there is less of it.
Sell pressure: where new KAS comes from
All of it comes from mining, and the number is an integral rather than a rate. Kaspa's subsidy is set by a table inside the consensus code that multiplies the per-second reward by a fixed factor every 2,629,800 seconds — a 5.6%cut each time, which compounds to a halving every twelve steps. Because the reward is falling continuously, reading one month's rate and multiplying by three would be wrong in both directions, so the framework adds the curve up segment by segment across the exact block range the window covers. Three step-downs landed inside it, on Jul 5 2026, Aug 5 2026 and Sep 4 2026, taking the reward from 2.59565436 KAS a block to 2.18267645 KAS. Over 77,755,819 blocks that sums to 188.6M KAS, an average of 2.4257 KAS a block — so Sell #1, protocol inflation, is 188.6M KAS, or about 1.9M KASa day at today's reward. The block rate was measured rather than assumed: Kaspa runs ten blocks a second since the Crescendo upgrade of May 2025, and the chain's activity score advanced at 9.9994 per second at both ends of this window, straight through the Toccata hard fork of Jun 30 2026. Toccata added covenants, script pricing and on-chain proof verification to Kaspa; it changed neither the block rate nor the subsidy.
Sell #2, vesting unlocks, is 0, and on Kaspa that is structural rather than expired. There was no premine, no sale and no allocation to a team, an investor or a foundation, so there is no schedule for anything to come off. The chain proves it rather than asserting it: the node publishes exactly two supply numbers, the coins that exist and the ceiling, with no third bucket for anything withheld, and the count of coins in existence sits within two days of mining of the count treated as tradable. The genesis block created no coins at all, and the consensus code contains exactly one function that can create one — the block subsidy. Sell #3, Foundation and unscheduled unlocks, is 0 for the same reason: the founding company dissolved before mainnet and every group working on Kaspa since has been funded by donations, so there is no treasury wallet to watch. Sell #4, long-term locked or bankruptcy, is 0 — KAS has no bankruptcy estate, no trustee and no court-ordered distribution attached to it.
Buy pressure: where new KAS goes
Nowhere. Buy #1, programmatic buyback, is 0, and there is no mechanism by which it could be anything else: Kaspa has no treasury, no foundation balance sheet and no protocol revenue account, so nothing exists that could spend into the market. A promotional buyback-and-burn scheme circulates on social media using the Kaspa name; it is not a Kaspa mechanism and the framework does not count it. Buy #3, Foundation buy, is 0 for the same structural reason.
Buy #2, protocol fee burn, is 0, and this row was verified on both surfaces rather than assumed. Kaspa builds in no burn — transaction fees are paid to whoever mines the block, so fees recycle to miners instead of being destroyed. The community does use a burn address, a valid Kaspa address with no spending key, and it holds 11.2M KAS of historical sends. Every one of its 312 unspent outputs was pulled and summed rather than sampled, and the total matched the address balance exactly, so the enumeration is complete. Bucketing those outputs by block position, exactly three of them landed inside this window, worth 217.36 KAS between them — the largest a single 215.78 KAS send on Jul 15 2026. The second surface is the coin count itself, which on Kaspa is the sum of every unspent output and therefore can only rise; it rose by the mining reward at every read and never fell. Two independent surfaces, both read at both ends, and the measured removal is 217 coins out of 27.7 billion. Buy #4, new long-term lock, is 0: Kaspa is proof-of-work, so there is no staking to bond into, and the small amount of KAS sitting in Layer-2 bridge contracts is a custody transfer rather than a removal, because the same coins stay spendable on the other side.
Foundation and overhang
The team-controlled overhang on Kaspa is the smallest in this framework by a wide margin. No foundation holds KAS, no company holds KAS, nothing was funded at launch and no reserve was set aside, because the network began from an empty genesis and has only ever paid coins to miners. The framework does not treat the absence as an assumption: the reason Sell #2, #3 and #4 can all read zero is that the chain's own coin count and its tradable count are the same number, so there is nowhere for a withheld allocation to hide. Integrating the published subsidy schedule from genesis reproduces the observed coin count to within 1.2%, and that residual is a fixed offset carried by the chain's legacy ceiling constant rather than a growing discrepancy — the present-day mint rate measures slightly below the schedule, not above it.
Exactly one group-controlled wallet is enumerated, and its size makes the point better than its absence would: a community treasurers multisig holding 1,830.8 KAS — donated, already-circulating coins contributed by miners rather than an allocation. All eleven of its unspent outputs pre-date the May 2025 Crescendo upgrade, so nothing has arrived for over a year and nothing landed in this window; it is read from the chain at every rebuild. Two further items are watched without being counted. The Layer-2 bridges — Kasplex and Igra, the EVM rollups that use KAS as gas — hold KAS on behalf of depositors, together roughly 1.6Mdollars of value, which is custody rather than team supply and belongs with exchange balances. The other is the mining pools, which receive every newly created coin and are the real distribution channel on a chain like this one. The trigger sentence applies to all of them: if the treasurers' balance or either bridge's KAS balance falls between refreshes by more than user withdrawals account for, that outflow enters Sell #3 at the next refresh.
How KAS compares to other capped proof-of-work chains
KAS and Bitcoin sit in the same family — a hard supply ceiling, issuance only to miners, no allocation to anybody — and differ on the shape of the decay rather than its existence. Bitcoin cuts its subsidy in half once every four years, which concentrates the whole adjustment into a single dated event and leaves the emission rate flat for the 1,460 days in between. Kaspa spreads the same geometric decay across twelve small steps a year, so the rate is never flat and every quarter mints measurably less than the one before it. That is why Kaspa's forward reading is lower than its trailing reading by construction, 159.0M against 188.6M, while a mid-cycle Bitcoin projects the same number forward that it just recorded. It is also why carrying a trailing average forward is the specific trap on this coin: doing that here would have booked 188.6M instead of 159.0M, an over-forecast of 15.7%.
Against the uncapped continuous-emission chains, KAS is far stricter. A staking-linked L1 typically issues a policy-set percentage a year with a ceiling that is a document rather than a number in the code, and the number can be voted upward. Kaspa's ceiling is the subsidy table: a finite list of entries that ends in zero, with no governance layer able to extend it and no on-chain vote that could. Roughly 96% of all KAS that will ever exist has already been mined, so the entire remaining issuance is under 1,020M KAS spread across the decades to come.
The comparison that flatters KAS least is to the exchange tokens that run quarterly buybacks and fee burns. Those chains offset issuance with a demand-linked removal that scales with usage, and their inflation readings can go genuinely negative. Kaspa has no such mechanism at all — not a dormant one, not one that failed to fire, but none in the code, since fees go to miners by design. So KAS can approach zero inflation as the curve decays, and cannot cross it. On the sell side it is one of the cleanest coins in the framework; on the buy side there is simply nothing to count.
What to watch in the next 90 days
First, the three scheduled step-downs, which are the only things moving the number: the block reward drops to 2.06017223 KAS on Oct 5 2026, to 1.94454365 KAS on Nov 4 2026 and to 1.83540480 KAS on Dec 5 2026. Each is protocol-scheduled and needs no decision from anybody. Second, the block rate, which is the one thing that could break the forecast: a future acceleration hard fork of the kind Crescendo already delivered is designed to cut the per-block reward by the same factor it raises the rate, leaving issuance unchanged — but a rate change without that compensation would not, so the rate is measured at every rebuild rather than assumed. Third, DAGKnight, the consensus upgrade the roadmap targets next; on the evidence of Toccata it is supply-neutral, but each fork is read for a subsidy change before that is accepted. Fourth, the burn address, which is read in full at every rebuild — it took in 217.36 KAS this window and 977.38 KAS over the trailing year, so any real change in that pattern would show up immediately. Fifth, the Layer-2 bridges, the only contracts on Kaspa that hold KAS in size and the only place a genuine lock could ever appear.
Summary
The MrNasdog Pressure Framework reads KAS at +0.68% over the trailing 90 days and +0.57% projected forward: supply growing, projected to keep growing. The structural mechanism is mining and only mining — 188.6M KAS created in the window against a buy ledger of exactly 0, with no premine, no vesting schedule, no treasury, no buyback and no burn in the protocol. The key risk is the mirror image of the comfort: because there is no buy side at all, KAS has no mechanism that could ever take supply off the market, so the reading can only fall toward zero as the curve decays and can never turn negative. The ceiling is the genuine comfort — 28,704.0M KAS enforced by a finite subsidy table with about 96% already mined, and no governance path that could raise it.
MrNasdog Pressure Framework analysis of KAS, Metric 1 — Inflation. Data + explanation only. Not financial advice. Updated Sep 10 2026.