The Steem reward system, Part 1: V-shares

@theoretical · 2016-05-31 04:40 · theoretical

This is the first post in a series about the Steem reward system. Please read my disclaimer. Scroll down to the "Conclusion" section of this post for a TLDR version.

The goals of this series are to:

  • Describe the nuts and bolts of how the reward algorithm works
  • Describe some of the big economic ideas of how the reward system provides a reasonable incentive structure
  • Make a few concrete suggestions for future hardforks to tweak the algorithms when improvements are possible

If you're interested in finding out exactly how a click on the "upvote" button, or the passage of time, translates into changes in pending payouts, follow this series!

As discussed in the whitepaper, content and curation rewards are 1 STEEM per block, or 3.875% per year, whichever is greater. The current implementation rolls both content and curation rewards into a single accumulator called total_reward_fund_steem(available with the cli_wallet command info, the get_dynamic_global_properties RPC, or at the steemd.com alternate UI).

So total_reward_fund_steem is the total reward available for all posts as of the current block. As of this writing (block 1872548 on 2016-05-29), total_reward_fund_steem is 3,748,054 STEEM. This is distributed according to a post's V-shares [1] (which are called rshares2in the code). There are currently about 176,134,568 septillion [2] V-shares in existence, so one septillion V-shares is worth approximately 0.0213 STEEM. As described in the "Allocation and Supply" section of the whitepaper, Steem are added to total_reward_fund_steem every block. When a post is paid, its V-shares and rewards are simultaneously removed from the pool, resulting in no net change of the current Steem per V-share.

Conclusion

Here are some points to take away from this post:

  • Each upvote "prints" V-shares, causing the upvoted post to get an increased percentage of total_reward_fund_steem and slightly decreasing all other posts' percentage.
  • The network "prints" a little bit of STEEM every block [3] and adds it to the reward without regard for the number of V-shares that have been issued, slightly increasing each post's payout every block.
  • The total_reward_fund_steem on July 4 will be worth well over $2 million at STEEM's current prices, meaning that if we only have 1000 actively posting users, it is entirely reasonble to expect each user will on average [4] get at least $2000 worth of STEEM.

[1] The reason I want to stop calling them rshares2 and start calling them V-shares will be revealed in a later post in this series.

[2] One septillion is 10^24 = 1,000,000,000,000,000,000,000,000.

[3] As discussed in the whitepaper, for every 1 STEEM printed to pay post rewards, 9 STEEM are printed to be divided among all holders of Steem Power, meaning that users who make a long-term commitment to the network by holding SP are protected to some extent from the inflationary effects of the "printing press".

[4] The payout per user will be unequal; there will be many users who get much less than $2000, or even nothing at all. This hypothetical situation is further based on the totally unrealistic assumption that the price of STEEM will remain the same over time. Remember my disclaimer.

Payout: 0.000 HBD
Votes: 96
More interactions (upvote, reblog, reply) coming soon.