Cardano fees covered just 0.7% of staking rewards as transactions fall 72%

A Fee-Reward Gap That Numbers Make Hard to Ignore

Cardano generated 3.3 million ADA in transaction fees over the 73 five-day epochs between September 1, 2025, and September 1, 2026 – while paying out 493.7 million ADA in staking rewards across the same window. That single ratio, roughly 1 ADA in fees for every 150 ADA distributed to stakers, frames the network’s central economic tension more sharply than any roadmap projection can. The figures come from Bitquery’s full-chain count, which starts from Cardano’s first block and groups the latest data into the annual comparison period.

Fees covered approximately 0.668% of rewards – leaving the reward pool 149.6 times larger than what on-chain activity actually generated.

The gap is not a rounding error or a measurement artifact. Reserve emissions from Cardano’s 45 billion ADA fixed supply absorbed the difference during the period, functioning as a subsidy that keeps staking yields intact while organic fee revenue remains thin. That arrangement was always part of the design, but the question of when – or whether – fees grow enough to reduce that dependence is no longer abstract. Transaction data now supplies a concrete answer about the current trajectory, and that answer points in the wrong direction.

Transaction Volume Has Fallen Sharply, and Bots Are Filling the Gap

Cardano averaged 90,294 transactions per day in 2022. From January through August 2026, that daily average had dropped to 24,869 – a decline of 72.46%. Every transaction represents a fee opportunity, so fewer transactions translate directly into less revenue regardless of where individual fee rates are set. The network’s throughput capacity is not the binding constraint here; demand is.

The composition of that smaller transaction pool shifted at the same time. Bitquery classified wallets sending at least 3,000 transactions in a single month as bots, excluding wallets whose behavior resembled exchange activity. Under that methodology, bots accounted for 11.5% of Cardano transactions in 2022. By 2026, that share had risen to 32.8%, with batchers – automated systems that bundle multiple user orders into single on-chain actions – forming the largest identified bot subgroup. That means real-person-initiated activity shrank as a proportion of a base that was itself already shrinking.

Interpreting these numbers requires some care. One bot can generate thousands of transactions, which inflates volume counts without representing discrete economic participants. A decentralized exchange batcher processing orders for dozens of users appears as one on-chain actor. Modern Cardano wallets can also associate multiple addresses with a single stake key, and a holder who delegates ADA without transacting simply does not appear in sending-wallet counts at all. The transaction decline is unambiguous as a measure of on-chain actions – it cannot, on its own, quantify how many individual users have left, stayed active, or shifted behavior. What it can quantify is fee revenue potential, and on that measure the trend is negative.

One Epoch’s Snapshot Confirms the Annual Picture

Epoch 654’s official data, drawn from Cardano’s epoch 655 supply page, offers a shorter measurement window that arrives at a similar conclusion through a different denominator. The epoch recorded 108,500 transactions and 33,855 ADA in fees. Spread across five days, that works out to roughly 21,700 transactions per day, or 0.251 transactions per second. Against the approximately 9.998 million ADA distributed in staking rewards for that same epoch, fees covered about 0.339% of the total.

The percentage differs from Bitquery’s 12-month figure because the two calculations use different reward denominators – the one-epoch snapshot is not directly interchangeable with the annual staker total. Both windows, however, place transaction fees at well below 1% of their respective reward measures. The consistency across measurement approaches removes the possibility that either figure is a statistical outlier.

Staking participation also declined over the longer window. The share of circulating ADA held in staking fell from 75.6% at the end of 2022 to 58.3% by the final epoch of Bitquery’s study period. That is a participation metric, not a security metric – the data does not include any direct measure of network security outcomes, so no claim that security has deteriorated is supported by the cited figures. What the staking-share decline does indicate is that fewer token holders are actively engaging with the network’s core incentive mechanism than were doing so four years ago.

The Reserve Is Running, and Its Contribution Is Shrinking by Design

Cardano’s epoch 655 data showed 6,126,859,027 ADA remaining in reserves – equal to 13.62% of the 45 billion ADA maximum supply. Under the network’s monetary policy, each epoch routes transaction fees plus 0.3% of the remaining reserve into a virtual pool, which then funds staking rewards. Because the reserve itself declines each epoch, the absolute size of that 0.3% contribution shrinks over time. The reserve was explicitly designed to provide a long runway during which applications, users, and transaction activity were expected to scale. Whether that scaling arrives before the reserve’s contribution becomes negligible is the open question the current fee numbers force into view.

Planned protocol upgrades are expected to give Cardano meaningfully higher throughput capacity, which could support significantly more on-chain activity if demand materializes. The throughput ceiling is not what produced a 72% transaction decline – that decline reflects demand conditions, not processing limits. Higher capacity would enable more activity but cannot by itself generate it.

The data does not forecast what happens next. It describes a network where, as of mid-2026, the staking economy runs almost entirely on a shrinking reserve pool, organic fee revenue remains below 1% of reward obligations, daily transaction volume is less than a third of its 2022 level, and automated activity now accounts for nearly a third of what on-chain volume remains. Cardano still holds over 6.1 billion ADA in reserve – enough to sustain current reward levels for years under present conditions. The more pointed question is whether the fee curve bends upward before the reserve’s structural contribution becomes too small to matter, and nothing in the current transaction data suggests that inflection is imminent.

At 0.251 transactions per second and 33,855 ADA in epoch fees, the math does not close without a demand shift that the last four years have moved away from, not toward.