Half of every payout on Hive goes to the people who voted rather than to the person who wrote. That is 23,995,866 HBD since the fork. This is where it went.
Every curation_reward operation the chain emitted from block 41,818,753 on 20 March 2020 through 15 August 2026. There are 375,528,104 of them, one per voter per payout. Every single one was paid in Hive Power, so none of it could be sold on the day it arrived. Adding up those 376 million payments at each day's feed price lands within 0.07% of the curation total the chain recorded on the posts themselves, which is the check that says the two sides of this ledger agree.
50,874 accounts have earned curation at least once. That is fewer than the 90,658 accounts ever paid for writing. Between them they collected 23,995,866 HBD against the 18,128,376 HBD that went to authors.
Those 40 accounts took 61.4% of all curation paid since the fork. The 40 biggest authors took 7.5% of all author rewards over the same six years and five months. Same chain, same period. The two halves of a payout are not concentrated anything like the same way.
| # | Curator | HBD | Votes paid | First | Last |
|---|---|---|---|---|---|
| 1 |  @appreciator | 2,143,856 | 275,500 | 2020-03-20 | 2026-08-15 |
| 2 |  @ocdb | 1,655,872 | 317,226 | 2020-03-20 | 2026-08-15 |
| 3 |  @blocktrades | 1,322,115 | 162,865 | 2020-03-20 | 2026-08-15 |
| 4 |  @alpha | 1,051,791 | 67,689 | 2021-06-15 | 2026-08-15 |
| 5 |  @smooth | 696,104 | 20,965 | 2020-03-23 | 2026-08-15 |
| 6 |  @rocky1 | 613,233 | 247,221 | 2020-03-20 | 2026-08-15 |
| 7 |  @buildawhale | 475,097 | 101,007 | 2020-03-20 | 2026-08-15 |
| 8 |  @blocktrades.com | 458,308 | 67,645 | 2021-07-12 | 2026-08-15 |
| 9 |  @curangel | 453,894 | 228,360 | 2020-03-20 | 2026-08-15 |
| 10 |  @theycallmedan | 439,789 | 151,662 | 2020-03-20 | 2026-08-15 |
| 11 |  @leo.voter | 428,714 | 178,778 | 2020-03-20 | 2026-08-15 |
| 12 |  @bdvoter | 421,476 | 117,435 | 2020-03-20 | 2026-08-15 |
| 13 |  @steemmonsters | 305,581 | 86,299 | 2020-03-20 | 2026-08-15 |
| 14 |  @trafalgar | 294,447 | 39,532 | 2020-03-20 | 2026-08-15 |
| 15 |  @threespeak | 276,701 | 91,556 | 2020-03-20 | 2026-08-15 |
| 16 |  @themarkymark | 269,451 | 101,020 | 2020-03-20 | 2026-08-15 |
| 17 |  @ecency | 256,329 | 547,838 | 2020-06-17 | 2026-08-15 |
| 18 |  @ranchorelaxo | 247,067 | 23,283 | 2020-05-14 | 2026-08-15 |
| 19 |  @tipu | 227,706 | 171,512 | 2020-03-20 | 2026-08-15 |
| 20 |  @altleft | 227,045 | 361,405 | 2020-09-03 | 2026-08-15 |
| 21 |  @sweetsssj | 188,956 | 70,649 | 2020-03-20 | 2026-08-15 |
| 22 |  @gtg | 187,886 | 27,114 | 2020-03-21 | 2026-08-15 |
| 23 |  @newsflash | 159,084 | 225,955 | 2020-07-22 | 2026-08-15 |
| 24 |  @abit | 154,782 | 54,622 | 2020-03-20 | 2026-08-15 |
| 25 |  @steempty | 142,515 | 37,510 | 2020-03-20 | 2026-08-15 |
| 26 |  @xeldal | 142,429 | 121,439 | 2020-03-20 | 2026-08-15 |
| 27 |  @liketu | 132,248 | 148,149 | 2021-12-21 | 2026-08-15 |
| 28 |  @usainvote | 128,199 | 109,385 | 2020-08-20 | 2026-08-15 |
| 29 |  @steemed-proxy | 128,057 | 235,197 | 2020-03-20 | 2026-08-15 |
| 30 |  @zingtoken | 127,319 | 22,545 | 2023-10-24 | 2026-08-15 |
| 31 |  @neoxian | 106,311 | 83,338 | 2020-03-20 | 2026-08-15 |
| 32 |  @azircon | 104,062 | 235,893 | 2020-03-20 | 2026-08-15 |
| 33 |  @resonator | 101,712 | 29,025 | 2022-02-25 | 2026-08-15 |
| 34 |  @upmewhale | 98,139 | 44,290 | 2020-03-20 | 2026-08-14 |
| 35 |  @adm | 96,263 | 31,119 | 2020-03-20 | 2026-08-15 |
| 36 |  @ocd | 95,751 | 250,307 | 2020-03-20 | 2026-08-15 |
| 37 |  @acidyo | 93,764 | 133,332 | 2020-03-20 | 2026-08-15 |
| 38 |  @lazy-panda | 92,821 | 51,721 | 2021-09-01 | 2026-08-15 |
| 39 |  @arcange | 91,117 | 644,018 | 2020-03-20 | 2026-08-15 |
| 40 |  @canadian-coconut | 89,258 | 62,947 | 2020-03-20 | 2026-08-15 |
| Slice of curators | Accounts | Share of all curation |
|---|---|---|
| Top 0.1% | 50 | 64.5% |
| Top 1% | 508 | 89.0% |
| Top 5% | 2,543 | 97.8% |
| Top 10% | 5,087 | 99.3% |
| Top 25% | 12,718 | 99.9% |
| Top 50% | 25,437 | 100.0% |
Curation is a return on stake before it is a reward for taste. A vote earns in proportion to the Hive Power behind it, so the table above is close to a ranking of who holds the most and votes with it, not of who found the best posts first. That is the design rather than a fault in it. The pool pays stake to point at things, the pointing is what readers see as a trending page. The question the chain has never really settled is whether that pointing has been worth what it costs.
One thing worth knowing when you read these numbers. Curation arrives as Hive Power, which means the top of this table is also a description of how stake compounds.
Built from Hive chain data on 16 August 2026 · blocks 41,818,753 to 109,070,000 · figures in HBD at payout time
