The misaligned incentives of protocol developers

Today’s most popular protocols like ethereum and bitcoin are running into serious bottlenecks, the most serious of all being scalability. Transactions fees are at all time highs, with bitcoin averaging around $30 per transaction and ethereum averaging $3. To develop protocols that scale is a serious goal for the industry, and the protocol that nails it first will most likely gain serious advantages in terms of user base. Once this happens, consolidation will start. Businesses will start implementing blockchain into their business models, and network effects will start to develop. Once these successful blockchains have developed their moats, there will be a serious incentive problem that we will all have to reckon with; why would developers (or more specifically, early token holders with governance klout) continue to improve scalability?

The Incentive Problem

The fundamental way to value cryptocurrencies is by using the formula for fiat currencies. This is the MV = PQ equation where PQ is Price x Quantity i.e. total cost of computing power to maintain the network, M is value of all tokens and V is velocity of the token.

Based on this formula we should consider how the incentives of our developers are aligned, or put another way, how can developers increase the value of their network?

1. Increasing transaction throughput — The most obvious mechanism for increasing network value is by increasing the usefulness of the network, i.e. getting more people using it. This is true competition, blockchains will compete with each other on scalability, partnerships, UI/UX, among other things. This will gain them users and thus higher valuations. There is no incentive problem here, as the incentives of the developers and users of the network are quite accurately aligned.
2. Decreasing velocity — There was a great piece written about this here, and I will try to give a general overview so you can dig in deeper in the link if you prefer. Ultimately with utility tokens much of the value is derived from the velocity of the token. If there is no reason to hold the token, this could skyrocket the velocity and decrease the value entirely. This author uses the example of a concert ticket payment token, where consumers purchase tickets using the tokens but refuse to hold the token because of general price volatility. In this case, developers are incentivized to develop schemes that make holding the token more profitable. These include buy-and-burn mechanisms like Monaco has, staking mechanisms that allow interest rates to be earned on tokens, and balanced burn-and-mint operations like Factom uses. Ultimately this type of incentive is not necessarily misaligned with user incentives. Each of the different velocity reduction mechanisms has its own pros and cons, but overall they mostly just create another incentive type for the user that does not as a whole hurt them.
3. Holding network cost steady — Here is the major source of misaligned incentives between the user and the developer. This will only will apply to platform tokens i.e. ethereum, neo, waves, etc. in which the cost of the token is tethered to the price of operating the network. It can, however, be applied to markets in which the developers are also in a sense ‘miners’. This has been argued to be true for bitcoin, where many of the core developers are also working for blockstream, a company that plans to use network transaction fees to incentivize exchanges to use its network (more can be found on this here). What this means is that once a chain has built up significant switching costs around itself, it is incentivized to set aside scaling opportunities in favor of more user-growing activities. Scalability actually decreases the size of P, which in turn decreases the price of their token. Obviously at the stage we are in today, blockchains are immensely unscalable and as such the tradeoff in user gains that can be achieved by increasing scalability are worth it, but there may come a day when this is not true. We have already seen it to some extent with bitcoin, but unfortunately they have not yet solved the scalability issue and as such it may in fact end up being their downfall.

Many people liken Ethereum to the oil of the new web. Like Standard Oil, Ethereum has the same monetary incentives to keep gas prices high and cut them temporarily to price out competitors

Competitive Response

The idea has been brought up that in scenarios like these, the switching costs to fork the blockchain and ultimately zero out the monopoly rent that these networks may be incurring by not improving scalability may be so low that users can effectively fork the protocol and create one in which the most updated scalable protocols are used. I agree it is more likely than legacy businesses because of the open-source nature of these protocols, but I believe there are a few switching costs we have to take into account before accepting this as a given:

1. Mining Profitability — The concept of mining profitability as a switching cost is very significant, as miners are what provide security to a blockchain. A forked protocol must effectively communicate to its users that there will be another protocol equal in everything except more scalable, and then somehow convince miners to accept lower mining rewards in exchange for usability. This really only makes sense if the miners believe that this chain will entirely displace the original chain, as they can get in early and eat up a bunch of the block rewards before people get hip to the new chain. It has been quite easy and common to see mildly successful forks but I would argue that this is a product of the early state of the market, and that many of these currencies have developed virtually no moats around them as they are so young.
2. Scalability Improvements — In the event that a blockchain is not improving its scalability, the idea that some newcomer can fork the chain and improve the protocol is also not guaranteed. You can’t just fork their code, you actually have to be better than them. This new protocol will need a significant development team, and already be well-versed in the blockchain in order to understand and improve upon the current protocol. Even so, the incumbent will have the advantage likely with more funding, expertise, and integration with existing structures such as exchanges and banks. Similar to the ability of Carnegie Steel and Standard Oil to lower rates temporarily to cut out competition, the large chains may have the most significant scaling advantages they can turn on and off when necessary. The most likely ‘fork’ in this scenario actually is businesses moving towards other similar chains like NEO instead of creating their own version of Ethereum. What this means in practice is that scalability will become a competition between major chains, who likely will only improve scalability to the point where it is economically viable for businesses. Improving any further would cut into their token value. This may bring about different lawsuits in the US around acting as oligopolies, and it is doubtful if many chains will even take the profit maximizing path (for example, Vitalik Buterin seems to be very idealistic about the future of the blockchain protocol and himself has argued that scalability is the number one bottleneck to widespread adoption), but regardless it is a useful exercise to look at the direct monetary rewards for certain behaviors.

Conclusion

There is significant monetary reward in being able to hold off improvements in scalability of network protocols, similar to how there was profit to be made by monopolies like Standard Oil. The different in today’s world is that blockchains are theoretically more forkable than network effects like Standard Oil who had to invest in large capital requirements to maintain their networks, and maintained proprietary manufacturing capabilities. However, these are not non-existent in the blockchain world and we should pay close attention to how forks like Bitcoin Cash develop so we can get a sense of how successful competition in the blockchain space will be.