Partnered With:

Placeholder: reduce voting time to control governance costs for encrypted networks

Read this article in 13 Minutes

The agreement does not control borrowing rates in the secondary market, but it can affect "governance costs" by controlling the time it takes to complete the vote.

How Much Does A crypto-vote Cost of governance in encrypted networks?

Writing: Joel Monegro, managing partner of Placeholder
Translation: Jhonny & have spent

In encrypted networks (such as daos or PoS networks) where tokens provide users with certain voting rights, we might measure the cost of each vote on these networks by calculating the interest that would be paid to borrow such tokens from secondary lending markets to vote during the voting period.

The idea emphasizes the importance of time as a variable in governance, since the longer a person has to borrow in token to vote, the more expensive the interest is to pay. If this is true, we can use this insight to design stronger governance systems. Although the agreement does not control borrowing rates in the secondary market, it can affect "governance costs" by controlling the time it takes to complete the vote.

The ecosystem of up-chain and down-chain lending markets, where anyone can lend or borrow in tokens, is growing. The lending market is useful: you can earn interest by lending out tokens, or borrow them (paying interest) to get some leverage against your collateral or short them. But when some tokens, for the management of the network, can be in the process of governance decisions through the lending market "borrowing" (i.e. by borrowing the tokens to participate in the vote, which affects the governance decisions) : whether the agreement to upgrade the management decisions, is also a community financing proposal, or at a deeper level of consensus decision making.

We can explore this with a rough (imperfect) calculation, which we will use as an example the ZRX token, which has a vote in the governance of 0x. Approximately 65 million ZRX tokens participate in voting on each 0x governance decision at a price of approximately $0.2 per ZRX, which implies that the cost of supporting each 0x vote is approximately $13 million (note: 6,500,000 *0.2=1,300,000).But that figure is not the actual cost of the decision.

The annual interest rate for borrowing ZRX from Compound platform is approximately 3.30%, which means that for each 0x decision vote, an interest cost of approximately 590 ZRX (note: 6,500,000 * (3.30/365) = 590) will be charged for borrowing this amount of ZRX from Compound platform for 1 day to vote. This means that in a similar case, the cost of a 0x decision vote would be about $120 ($0.2 at ZRX price), or the marginal cost of each vote would be about $0.000018 (note: 120/6,5000,000=0.000018) [1]. This figure applies to both the voter who borrows in tokens (who does not have the tokens to vote) and the non-borrower (who already has the tokens to vote), since the non-borrower bears the same opportunity cost by using the tokens to vote (as opposed to lending them to others to earn interest).

These Numbers can be surprising, because they imply that the cost of influencing a vote is very low for a borrower with enough collateral (or credit). I'm not the first observed the problem, Tarun Chitra highlighted in the published articles, if borrowed tokens, income more than the use of the tokens, the benefits, so borrowing could threaten PoS system security (I think, however, the security of the system may be for scrip holders is more valuable than the extra income).

But this may be a feature, not a bug: first, because it allows us to measure the cost of voting more accurately and precisely; Second, because it means that the PoS protocol and DAO can influence the cost of governance by changing the length of their votes.

Secondary lending markets influence the cost of voting through token prices and interest rates.The cryptographic networks themselves do not control these factors, but they do control how long users must hold tokens if they want their votes to count. That is, cryptographic networks can make voting more or less expensive by influencing the total amount of interest paid by token borrowers (that is, users who bring in their tokens to vote) (this applies to both token borrowers and non-borrowers, as described above).

The cost of 0x governance is so low in part because the votes are counted in a very short period of time (0x governance votes are votes at a particular block height) and the tokens are transferable before or after this polling time.

But if the vote takes longer, the cost will be higher. For example, in Decred's certificate of interest layer, users are required to obtain a "voting ticket" by locking a certain number of DCR tokens for about 28 days before voting. That means it will be much more expensive to influence Decred's governance by borrowing DCR tokens, since you have to pay a month's interest to vote (along with the cost of collateral and the risk of price volatility).

Controlling the cost of governance by changing the length of time tokens must be locked at the ballot box may be an elegant solution to many problems.Systems that use tokens to vote, such as 0x and MakerDAO, can emulate Decred and increase the cost of governance by locking tokens for N days before the end of voting and returning them to voters after the end of voting. It might be possible to have a grievance process in the meantime, or maybe you're more creative and make the lockup a little more flexible, such as asking for a longer lock-up period if a governance decision is important or controversial.

There is also a view that increasing the cost of voting necessarily increases the value of voting because, in an economic sense, equilibrium cost = price = value. Therefore, the higher the cost of governance, the higher the value of governance decisions must be to maintain balance. But simply by making voting more expensive, it does not necessarily bring additional governance security. If the cost of governance is too high (that is, the cost of decisions is higher than the value of decisions), then a lower turnout is likely; If governance costs are too low, the system can be easily compromised.

The above assumption has a number of practical problems that make it difficult to be a reliable and accurate measure of the cost of network governance. For example, lending rates vary according to liquidity at the supply and demand end, and relatively few categories of tokens have active secondary markets that we can analyze. Moreover, none of this addresses Tarun's concerns.

I don't worry too much, partly because the market will actively respond to the change of liquidity, and I think that a successful network will find a new balance to meet the security requirements of Tarun - if we are in the process destroys some of the system (dare I say another aspect is that people can lend and borrow voting rights in fact is probably a good thing, but I will leave this argument till later).

Now,I just want to emphasize the point that we can manipulate governance costs by changing the time it takes to complete the vote.Such calculations can be difficult to make with great precision when markets are still immature. But if we measure the right things, we can get away with it. At least we can look at time differently.

[1] the biggest hole in this calculation is that in current cryptocurrency lending (where loans are over-collateralized), you would have to have more than $1.3 million in cryptocurrency collateral as an individual to get that amount, which is an expensive barrier for the individual. We will need to do additional calculations to add the cost of the collateral to this calculation. Moreover, the warning assumes that one person will try to influence the vote in such a way that the impact would be so great. But nonetheless, if you have access to such large amounts of collateral, the cost to you of influencing governance may be very low. This can be a problem because it gives too much power to the rich: the more wealth you have, the cheaper it is for you to influence governance. So be sure to treat these Numbers as imperfect: but my point is to emphasize the impact of time, not the precise cost of 0x governance.

One last thing -- I do want to point out that the "low" cost of governance is not necessarily a bad thing. Just because it's cheap doesn't necessarily mean it's unsafe. The figure in the 0x example may seem uncomfortably low, but $120 May be the correct cost of the decision under discussion. My view is that the 0x governance decisions have so far been less controversial and therefore do not require higher governance costs (again, this is a crude and incomplete calculation). Remember, there are so many things about social systems, both in the lab and in the real world, that we can't predict.

Source:https://www.placeholder.vc/blog/2020/1/7/how-much-does-a-crypto-vote-cost