Part 4 of this series concluded the analysis of funding candidates, settling ultimately on a mandatory minimum fee as the most suitable choice. This post examines how an optimal level of funding might be determined in a sustainable network.
Introduction
A secure, usable Bitcoin network requires that its miners be funded. But exactly how much funding is enough? On the one hand, a network with zero funding would be wildly insecure and unusable. On the other hand, a network that required vast funding levels would struggle to compete against state-controlled competitors, where many of the systemic costs are implicit or otherwise hidden from end-users. Somewhere between these two extremes lies an optimal funding point, where a level of security is obtained that satisfies the security demands of network users, without overshooting security needs and imposing onerous costs. This post examines this issue primarily from a political perspective, questioning how a community of network users might coordinate to choose a funding level.
Hardcode and developer consensus
The first and simplest method is when developers write a certain funding level directly into the network’s consensus code. Indeed this is the system in use today, as both the funding level (currently roughly 10% of network value p/a) and mechanism (inflation) are hardcoded architectural decisions made by Satoshi themself, decisions honoured and upheld, virtually unchallenged, by the network community ever since. This method is also quite likely to play a role in Bitcoin’s future: in the absence of preemptive action, the declining inflationary reward will eventually result in an unacceptable degradation in network security, attracting the community’s attention. Under these circumstances, it is highly likely that core developers would issue an update implementing some method of mandatory funding.
One could argue that this method is a fairly good approximation of democratic decision making. Developers must implement changes that miners will accept, and miners should only accept changes that will sustain and improve the popularity of Bitcoin, because their continued profitability depends upon continued mass acceptance and use of the protocol. Relying on developers does however have downsides. Firstly, a system which relies on a group of developers reaching consensus and pushing version updates will always be slower to respond to changing circumstances than a system where the ability of the funding requirement to adapt dynamically is built into the protocol. The severity of this issue depends upon just how dynamic the security needs of the Bitcoin network are likely to be— an open question.
Secondly, smaller groups of direct decision-makers are always going to be easier to bribe, coerce, or otherwise influence than larger groups. Thirdly, and most worryingly, a small group of individuals trying to divine the preferences of network users at large recalls Soviet command-style economies, contravening economic theory that shows that marketplaces are most efficient when as many economic participants as possible can express their economic preferences and contribute in determining supply and demand levels of goods.
Direct democratic determination
How could as many network users as possible contribute to determining a funding level? A traditional democratic system of a central authority collecting votes from a population of registered voters, while enforcing a system of one vote per voter, is clearly completely and utterly unworkable. What would be the alternatives? A system where each transaction could count as a vote would also fail, among other reasons because miners could stuff blocks with transactions voting for a higher block reward. Similarly, any system which attempted to weight a vote based on transferred value would fail, because transferred value is a meaningless metric in Bitcoin.
The only fraud and malice resistant voting system is one in which votes are validated through the sacrifice of real bitcoin. In such a system, network users would be able to directly voice their opinion on whether the block reward should increase, stay the same, or decrease, by spending, and thus sacrificing, bitcoin to a special address, a voting output. Effectively, each block would allow an unlimited number of network users to express their security preferences and influence the future block reward. The money spent to the voting output should also logically be used for the funding reward itself.
An interesting corollary of this system is that it is by necessity a democracy of the wallet. There is no way to prevent network users from casting multiple votes, so it is simplest to let people pledge as much money as they wish to affect the vote, with the strength of a vote being directly weighted to the amount spent to the voting output. This has the advantage of forcing people to price their preference for security, putting their money so to speak where their mouth is. Unlike in a traditional modern democracy, users must believe enough in their security preference to be willing to sacrifice money for it.
Implications of direct democracy
Such a proposal runs contrary to the popular contemporary political sentiment that everyone should have an equal voice in political decisions, and that in consequence individuals should not be able to leverage discrepancies in wealth and relationships to influence politics. History demonstrates that the ideals this sentiment have consistently failed, often spectacularly. Among other evidence, popular political culture largely accepts and even embraces the critical importance of massive campaign contributions for electoral success. More ominously, the activities of wealthy entities to influence decision-making by offering monetary incentives to political agents are legion, and practically impossible to police.
Furthermore, in a climate that demonises direct monetary voting, and in the absence of a system which facilitates and coordinates low-level pledging of funds, these funding mechanisms are largely conducted through expensive personal relationships, cutting the majority of the population, which lacks the funds to justify or afford such engagement, out of political decision making. Systems which attempt to prevent monetary voting are actually counter-productive, and the most effective and fair solution is to facilitate as far as possible direct monetary influences of all magnitudes on political decisions. The failures of such systems are the basis of political symptoms such as regulatory capture, where political institutions seem to act primarily in the benefit of larger, more centralised economic agents, simply because only such large-scale entities can afford to develop the mechanisms of covert monetary influence necessitated by the political environment.
This proposal ventures into largely uncharted modern political territory. It requires a high level of trust that, absent attempts by a central authority to control outcomes, a whole community of network users can reach optimal decisions. A wide range of behaviour would be likely to ensue if such a system was implemented. One might expect voting syndicates to arise, most probably coordinated through popular wallets. One might expect to see attempts by hostile entities such as banks, nation-states or even anarchist groups to “crash” the system by either down or upvoting the reward. Happily even hostile actors would be forced to contribute funds to the security of the network. Finally, and interestingly, this proposal is compatible with previously discussed funding ideas such as altruism and assurance contracts.
Conclusion
This proposal should result in a system which sets network funding at a level corresponding to the net willingness of the network community to pay for security. One’s reaction to it is probably a litmus test for one’s confidence in the judgement abilities of the wider populace. If one does not believe that such a system is feasible, then one fundamentally does not trust the network community to be capable of making the right decisions for itself. Unfortunately attempts to “protect” populaces from themselves by confining political agency to dedicated decision makers are notoriously and chronically vulnerable to hostile takeover, and thus an untenable candidate for a secure network.
The next post will sketch a basic funding model, and conclude this series with some general commentary on the issue of network security.
