New revenue stream - Demand-based recurring fees aka Harberger Tax on NS domains

To start-off the discussion an initial write-up by Perplexity:

What are the pro’s and cons of a harberger tax on SuiNS domains?

Implementing a Harberger tax on domain names—where owners self-assess the value of their domain, pay an annual tax based on this valuation, and must be willing to sell the domain at that price at any time—has several arguments both for and against it.

Arguments For Harberger Tax on Domain Names

  • Reduces Monopoly and Speculation: Harberger taxes discourage domain squatting and monopolistic behavior by making it costly to hold domains without using them, as owners must pay a recurring tax based on their self-assessed value [1] [6].
  • Improves Allocative Efficiency: By requiring owners to set a sale price, domains are more likely to end up with those who value them most, improving overall market efficiency[7][3].
  • Encourages Productive Use: Owners are incentivized to use domains productively or sell them to those who will, rather than letting them sit idle[3][7].
  • Sustainable Revenue Model: For decentralized systems, Harberger taxes can provide a sustainable funding source while keeping assets liquid and available for new entrants[3][7].

Arguments Against Harberger Tax on Domain Names

  • Risk of Forced Sale and Instability: Owners may be forced to sell at any time, which can be disruptive, especially if the domain is critical for a business or project. The cost and confusion of switching domains can be high, and there are negative externalities if a domain changes hands unexpectedly12.
  • Vulnerability to Attack: Malicious actors could force sales or drive up taxes by placing high bids, potentially causing legitimate owners to lose their domains or pay exorbitant fees. For example, a scammer could outbid a legitimate company for a valuable domain14.
  • Psychological and Economic Costs: The constant threat of losing a domain can create stress and uncertainty, particularly for non-sophisticated users or small businesses15.
  • Community and Intangible Value: Domains often have value beyond their market price, such as branding or community trust. Harberger taxes may unfairly burden those who derive significant but non-monetary value from their domains4.
  • Complexity and Abuse Potential: Implementing and maintaining a Harberger system for domains is complex, and can be prone to abuse or nuisance bids unless carefully designed82.

Summary Table

Argument For Argument Against
Reduces monopoly/squatting Risk of forced sale/instability
Improves allocative efficiency Vulnerability to attack by bad actors
Encourages productive use Psychological/economic costs for owners
Sustainable revenue for protocols Undervalues community/intangible benefits
Complexity and potential for abuse

Conclusion

While Harberger taxes on domain names can address monopolistic inefficiencies and provide sustainable funding, they also introduce risks of instability, abuse, and unfair burdens on legitimate owners. Alternative mechanisms, such as demand-based recurring fees, have been proposed to balance these concerns13.

  1. Should there be demand-based recurring fees on ENS domains?
  2. https://gwern.net/harberger
  3. Harberger Taxes can be Crypto's Sustainable Business Model
  4. https://www.lesswrong.com/posts/MjBQ8S5tLNGLizACB/combining-the-best-of-georgian-and-harberger-taxes
  5. A Reverse Harberger Tax for Eminent Domain? – Outside the Beltway
  6. Vitalik Buterin At Radicalxchange Taipei On Harberger Taxes | Diode™
  7. https://www.aeaweb.org/conference/2017/preliminary/paper/iiDK3dnr
  8. https://news.ycombinator.com/item?id=42319461
  9. https://web3domains.com/vitaliks-harberger-tax-idea-for-ens-domain-names/
  10. Harberger Tax
  11. https://economics.yale.edu/sites/default/files/ownership_7_16.pdf
  12. https://chicagounbound.uchicago.edu/cgi/viewcontent.cgi?article=12668&context=journal_articles
  13. 1 and 2 Character .eth Names - General Discussion - ENS DAO Governance Forum
2 Likes

I find this incredibly fascinating. Thanks for sharing.

I would really like to see this implemented in digital worlds or for future blockspace.

The main issue with using this for SuiNS names is that anyone can easily purchase a name from someone else (which is the intent of having a harbinger tax). For example: someone named Joe Tesla owns tesla.sui and puts a lot of effort into maintaining that as his identity, but the company Tesla decides it wants to purchase that name. There really isn’t a a reasonable price that Joe would be willing to pay for annual name fees that the company Tesla would be unwilling to pay to get that name. This just creates a different monopoly based on wealth. However this would accrue more fees to the protocol. The pure implementation of the harbinger tax works best when purchasers are limited to the government, which has to establish need requirements prior to exercising eminent domain powers.

I liked Vitalik’s auction variant for names. See Fn. 6. It allows people the option of selling, while ensuring that annual fees are appropriately priced. Interested in what other people’s thoughts are on this.

1 Like

That is good article bro :3

Really interesting concept, but I think it could be incredibly dangerous if it were applied to names legitimately being used.

For example, Protocol A using a domain with Walrus, and all of a sudden that domain is bought and redirected to a scam clone.

Or, User A sending to a wallet via a domain that was unexpectedly transferred to a different owner.

I might feel differently if there was a reliable way to apply this only to unused domains, but I think it’d be way too easy to game that.

Very good article, worthy of serious study. The first impression is that this is a very creative method worth trying, but whether it can be accepted by the public is worth worrying about. After all, it is too complicated and professional for most people. Maybe it needs to be tested on a small scale in some way.

SuiNS domain name holders already pay renewal fees. It is unnecessary to introduce another mechanisms to pay for owning them. The philosophy of Sui is to make web3 as userfriendly as possible. This taxation model will make extremely cumbersome to own domain names, making the whole SuiNS digital identity concept less attractive.

I don’t think that domain name squatting is such a serious issue. It is true that domain names are not very liquid, but you should not deal with it by any confiscatory mechanisms.
The issue is that you have a hard time contacting domain name holders if you want to buy their name. You can currently bid on them on Tradeport, but they won’t know about it until they look at it directly.

If you make domain name ownership extremely inconvenient, you can also bury the NS token.