future-proofing-the-economics-of-blockchains-pts-1--2

TL;DR

A transformation in society is taking place and layer-1 computational blockchains are the substrate. While the economic fortunes of layer-1 computational blockchains may wax and wane, the goal of the collective behind them should be to transition their economic stake from the infrastructure layer into the application layer. This leap will take time, and most blockchains will not make the leap. Coordinating the labor and care that goes into the work of producing a blockchain that can make the leap is not trivial. We propose some principles so that blockchains can successfully cross over to that shore when the day comes.

PART I - The Opportunity

Crypto is an experiment in digital scarcity. New digital resources have emerged, such as coins, tokens, NFTs and more. We're here to talk about layer-1 blockchain network "gas" coins. Through one lens, the liquid and instant settlement of these resources makes them look like cash. From another perspective, they look like deeds, rights, memberships, or even a cooperative business that defies our legal definitions of companies and partnerships. More generally, gas coins mediate access to one of most exciting social technologies in recent memory: Smart contracts.

The computation needed for smart contracts is not orthodox computing power. Credibly neutral replicated state machines are the technical artifacts that are required and currently the quantity demand exceeds the quantity supplied. Current crypto economics are not necessarily well configured.

We're not here to wax hyperbolic about the power of smart contract platforms but, for new entrants, suffice it to say that the blockchain as a ledger is only one facet of the innovation. At the baseline, smart contracts allow for the composing of many interactions of value, both native and non native to the digital realm. The collective ingenuity of workers in this space is engaged by the possibilities of programmable monies, firmly binding agreements, durable memberships, and all manner of automation of value. Beyond the horizon are even more experimental use-cases: Prediction markets, new forms of voting, new forms of identity and pseudo-identity, the integration of the real world and virtual reality in the metaverse. More generally, it’s obvious that the world is moving online and will continue to do so. The power of blockchains is that they let people create secure property rights and identity systems that amplify the power and utility of the online world.

The early experiments in this space (e.g., payments, decentralized finance, asset tokenization) are each trillion dollar markets; this certainly explains in part why the crypto space has boomed. However, the financial exuberance around blockchains contains the seeds of its own destruction. The promise of the blockchain will be fulfilled precisely when and if blockchains are cheap, abundant and commoditized. The froth we see in the market for blockchains is based on the assumption that these infrastructures can capture the value the ecosystem provides; but the ecosystem will produce tremendous value only when the infrastructure becomes a global public good accessible to everyone at low cost. The cheaper the rails the more value will ride the rails. This may appear paradoxical, but the implication here is that the rails don't reflect the value of what is riding upon them.

Layer-1 platforms of the next decade will begin to see more acceptance, but they will also see challenges to the unit economics of providing the compute resource. This article highlights what some of those challenges may be, and how the community which holds native gas coins - the "digital asset collective" - can future-proof the economics of the blockchain.

Cautionary Tales From Infrastructure Businesses

Not all blockchains will be successful. While transformative, a number may end up being more staid than they appear at the moment. There's a risk that the colorful and exciting vision of the metaverse escapes the actual economics of operating the blockchain infrastructure. Indeed, the most successful blockchains will likely become utilities.

Infrastructure businesses often boom early but over time tend towards steady but normal profit levels. In the United States, the opening of the Baltimore and Ohio railroad in 1827 began a boom in railroad construction that would last for over 60 years. The railroads were tremendous investments, but their real value was in opening up hundreds of millions of acres of farmland, thus lowering the cost of food and creating national sales markets which let every good enjoy economies of scale. As railroad construction boomed, however, competition eroded railroad profits and value was transferred to consumers and bystanders. As the railroads became infrastructure, the growth in value shifted towards those who were able to leverage the structure for growing businesses. In other words, as the railroads became infrastructure, you didn’t want to be a rail operator, you wanted to be Sears, Roebuck and Co.

You may think financial "rails" are different. They aren’t. You may think you have time, after all, the railroad bonanza lasted for a century. You don’t. The rate of change is speeding up.

There is lots of money to be made in the heyday of an infrastructure play, but the opportunity can vanish fast, especially in the age of digital abundance. By way of comparison, the window of opportunity for Internet Service Providers in the early 90s shrunk by a factor of 10. What took 60 years for railroads took 6 years for ISPs. When the government's Arpanet opened up access to the general public, a new land rush took place. Everyone who was operating bulletin board services began jockeying to provide dialup access to the Internet. The hottest skills on the market were not only software programmers, but also electrical engineers for wiring up banks of telephone switches. Real estate near points of access to phone exchanges went for premium.

Where are the names of the early ISPs today? Your equity in Compuserve, The World, Prodigy, or Earthlink did not earn you any stake in Google, Amazon, or Facebook. Over the next decade, the telecom engineers did not receive the same life-changing stock options their web developer peers did. The one notable exception was AOL, which avoided obscurity by buying Time Warner, and eventually being gobbled up and then spat out by Verizon, an infrastructure company. The AOL story is noteworthy because they tried in many ways to make the leap from the infrastructure layer and capture the application layer (with the ultimately misguided tactic of creating a walled garden) .

For digital asset cooperatives to make the leap from billing for ledger access to providing consumer surplus, the economics of the blockchain of today need to be anchored on solid economic principles. The blockchains that succeed will be the ones that transfer the most value to the structures that build on top of them. Blockchains as infrastructure means creating user experiences where the blockchain technologies become transparent to the user. The boom may not last, but it can be the foundation for continued growth and utility. The railroads, after all, never went away and continue to be critical to the world economy.

In the coming decade blockchains will see a number of changes and they will cement their place in society as infrastructure. The value of access to the ledgers, and the flows to capital, will increasingly resemble traditional economic assets. Like dialup networks, transaction fees on blockchains will fall to negligible amounts given advances in high-throughput blockchains (which will become commodities because of open source software).

Your digital asset collective may want to be more than a reliable infrastructure business. Yet capturing value on the application layer of blockchain is less than obvious. The (regulatory compliant) mechanisms don't yet exist, but the collective can future-proof its layer-1 economics to create optionality to escape a lackluster infrastructure fate in a world where the fat protocol thesis (an idea that enterprise value accrues into a vertically integrated settlement layer) proves hollow. The value is likely to be modular and decentralized, not monolithic and bundled.

Part II - Principles for Future-Proofing Your Blockchain

For a time, exuberance will mask the realities, but eventually, the community of holders of a chain’s native asset will need to make a leap. When transaction fees fall off the cliff, the collective will need alternative business revenues; simply selling access to the ledger will no longer produce attractive returns. We think the blockchains that survive the coming crisis will be built on the following principles.

Don't Eat Your Seed Corn

In most protocols we observe a confusion between capex (capital expenditures), and opex (operational expenditures). Payment for security in the moment, an operating expense, is not equivalent to investment in future security, a capital expenditure.

To illustrate, imagine the blockchain as a physical notary business. As is the case in many countries in South America and Europe, private companies can acquire a concession to store and provide access to contracts, land records, birth certificates, etc. The business has a finite budget. It can invest in security guards to stand outside the entry, or invest in physical vaults and computer systems. As with any business, there are strategic considerations and ultimately it will have to employ a mix of both outflows.

Paying security guards is a certain kind of recurring outflow - an operating expense. The costs appear in the same part of the income statement as expenses like phone systems, administrative staff, executive travel, etc., that is, activities (expenses) deemed necessary for the business. Opex should not be confused with outflows of capital made for investment purposes; that is capital expenditures. Capex is different (for example it doesn't even appear on an income statement); it's part of the value of the company (balance sheet) which is transferred between assets. You are trading the cash assets of today for future productive resources, e.g., by investing in new factory machinery, delivery vans, document vaults, etc.

The capex/opex distinction is not an accounting gimmick---confusing them is the downfall of many a business tycoon. No amount of paying for security guards would obviate the need for an investment in a new vault. The owner of the notary needs to set aside cash, equity, debt - whatever instrument - to pay for the capex. Startups a la Silicon Valley, pay for initial capex (acquisition of talent, software, R&D) with equity. However unfair, the labor market does not often make errors, and the people who accrue the most equity from the market are leaders and engineers, not security guards.

In blockchain, paying for moment-to-moment security is done with transaction fees to miners/validators. In the absence of these transaction fees, there are subsidies to cover for the market cost of providing network security, lest the network risk losing its perceived integrity.

For much of the first decade of blockchain, the future investment in the protocol was done on the basis of the "enlightened self-interest" of the early security guards of a chain. The logic goes as follows: The security guards amassed such a large equity in the notary business, that they now were the primary stakeholders, and as such had the incentive to reach into their pockets to recapitalize the notary, who had gone delinquent in its capital investments. Paying security guards in the hope of them later contributing to paying for your vault is circuitous and risk-prone. If the coins are readily and easily tradable, the dominant strategy is to dump and move on to the next chain when you notice capex being under-capitalized.

A protocol needs to be careful that the opex does not eat up the capex. The more cash you spend on opex the less you have for capex. The equity you issue to security guards dilutes the engineers and that's before we consider intergenerational fairness. Time compounds these losses: When dilute your brother with a drop, you dilute his grandchildren in buckets.

We are not trying to be obtuse or sanctimonious, but rather point out a structural problem that could prove fatal to protocols that mis-invest their cash flow in a philosophical concept, rather than in building useful applications. A new protocol seeking to be lasting and fruitful needs to have a sustainable long term investment model. While solutions are difficult to come by, meaningful experiments are taking place.

A number of protocols approached this problem by dedicating funds (or "founder rewards") to "foundations", but this then started attracting regulatory scrutiny. A later variation saw some chains implementing "decentralized treasuries" to make decisions on work that needs to be done (e.g.,as in Tezos, Cosmos, et al). While there are still regulatory overhangs here, and a trend toward bureaucracy, this is directionally the right move.

Digital asset cooperatives need to stand firm and preserve the capex games they create. Issuing new coins -- like issuing new equity -- is an invisible diluent, and it mostly hurts those who are not minding it. There will always be relenting and incessant requests to spend more today on opex. Don't confuse your opex for capex. Don't eat your seed corn.

Produce consumer surplus

Generally speaking, blockchains should aim to maximize consumer surplus; that is, the difference between the value a consumer places on a good and its price. Maximizing consumer surplus means increasing the value of a blockchain to consumers and keeping prices low so that the bulk of the tremendous potential from blockchains flows to consumers. Ensuring that gains flow to consumers requires competition and a blockchain design that doesn’t create artificial rents or bottlenecks that can be exploited by rapacious actors.

It’s widely acknowledged that the current payment infrastructure is slow and expensive, especially for international transactions. In contrast, a million dollar transaction can clear across a blockchain in minutes at a price of pennies. The claim is true, but it rings false when transaction fees on popular blockchains are high and variable, with spikes of $10, $100, or more, not uncommon for a single transaction.

A usable blockchain integrated with the real world must produce fees that are low. Low and consistent is ideal. Low and variable is ok. High and variable, however, is a problem. This is both a user experience and a negative network effect problem.

Consumer surplus is maximized when every consumer who values a good at more than its cost is able to purchase. In a competitive, well-functioning market, price (P) approaches the marginal cost (MC) of production. Consumers who value the good more than its price purchase the good and when P=MC. It follows that every consumer who values the good more than its cost purchases the good. If the price were above MC, too few consumers would purchase and if the price were below MC (say because of subsidies or non-price allocation) too many consumers would purchase. P=MC is the ideal. (There are, of course, well known exceptions to deal with cases of externalities and large fixed costs. We focus on the base case for clarity.)

Suppliers would prefer P>MC, which happens when markets are monopolized or otherwise broken. The US medical system, for example, is dominated by rents and bottlenecks that push P>MC and which have been exploited by the Shkrelis of the world. US housing markets are similarly broken by zoning and regulations that prevent building even in places where prices are well above the costs of production.

To fulfill their promise, blockchains must onboard billions of people into a new, lower cost financial system (as a first step!). Onboarding billions of people will happen only when P=MC, that is, when the price of using a blockchain falls to its true cost of production. To get there, blockchains have to be designed to operate at their maximal technical limits and not be throttled back in order to create rents. Blockchains must also surface information and not incentivize the creation and exploitation of information asymmetries. Everyone must have access to a blockchain on an equal footing.

In addition to keeping prices close to marginal cost, blockchains should be designed to increase value to consumers. Blockchains, as with other platforms, can be designed to maximize eyeballs, or information collection, or surveillance--techniques which can increase producer profits. In the short run, profits can attract investment and customers, but in the long run, a blockchain built for producers leaves a dissatisfied public only slightly better off than before.

Maximize the correct resource

Blockchains have attracted attention because the sector has produced outsized financial gains. These gains, however, are merely the promise of future value. As the technology matures, financial gains will diminish and gains to consumers will grow. We want to build the future in which consumers devote an ever-larger share of their time and contribution to the globally connected online world.

Security and decentralization are important for the bootstrapping and running of the network. As we put it earlier, the utilities must work reliably and we want a six-sigma blockchain. Most of today’s crypto industry revenues, however, flow to trading and mining as built-in economics, which leaves little room for rewarding the builders who make the things people love. We want to create organic incentives that benefit the builders and the users -- the economic actors rather than the security and rule enforcers.

A blockchain collective should aim to allocate funding to the building of an open and expressive space where people have the capacity to organize themselves around their shared interests, activities, outcomes, etc, and instantiate that as software, games, and economies. To make things, people have to choose to invest their labor in a protocol, and we want the protocol to be able to reward them for that labor and investment (rather than a venture fund which owns their equity, whereas the protocol is a community that generates public goods).

Whether one calls it the global village, cyberspace, or the metaverse is immaterial. Moving forward, the key idea is to assemble and reward the people who generate value in the new world.

Financial incentives are one method of attracting time and attention but are not the only nor always the best method. Paying fruit pickers per fruit will increase the number of fruits picked per hour but the fruit may be picked too early or too small. Thus, even in a simple task such as fruit picking, financial incentives must be combined with other methods of encouraging productivity such as monitoring or profit-sharing.

The key problem with financial incentives is that you get what you pay for but what you can pay for is not necessarily what you want. As a result, financial incentives must be used with care especially for complex, multi-dimensional tasks where monitoring and measuring are difficult. Said differently: algorithmic and programmatic distributions are very unlikely to be maximizing the correct resource.

Integrate with the world

It was natural for early innovations in the digital space to position themselves against the world. Most famously, John Perry Barlow offered A Declaration of the Independence of Cyberspace in which he declared:

Governments of the Industrial World, you weary giants of flesh and steel, I come from Cyberspace, the new home of Mind. On behalf of the future, I ask you of the past to leave us alone. You are not welcome among us. You have no sovereignty where we gather.

For better or worse, Barlow was wrong. Governments have power even in Cyberspace because people want to integrate their real lives and online lives. Similarly, if a blockchain is to remain relevant, it must integrate with the real world. The metaverse, so to speak, is not only a digital simulacrum, but also an online world synthesized with the quotidian. Integrating with the real world means considering the existing rules of the road.

Designing for the environments people live in is a question of user-experience. With everything UX, there are tradeoffs. While securities laws need to change to keep pace with new technologies, like it or not, the regulators are an agent in your game. While it is possible to design games that flaunt regulators, this exposes less adventurous users -- the next billion -- to unnecessary duress. Following existing law, especially US law, restricts some of the economic mechanisms which would be effective. The absence of ambiguity however should not be seen as a limitation; on the contrary, it should allow the protocol to be practicable and useful to future denizens of the metaverse. This is not an admission of defeat, it is actually an aggressive stance, positioning for exponential growth.

For the same reasons, economic mechanisms must be understandable. Mechanisms (i.e., the rules for which rewards are given) are what guide users, validators, and investors across language and other barriers. Good mechanism design incentivizes the crowd to "do the right thing". People need to know what activity they are to do, and what they should expect as a result. As is well documented in cognitive science research, humans are limited in their ability to navigate optionality, they place a lot of value on labels instead of mechanics, have non-obvious responses to price information, are notoriously bad at planning for the future, etc. etc. Given that context, if your incentive model doesn't match peoples’ intuitions, you should expect erratic, random behavior from the majority of your players and exploitation by the minority of informed insiders.

Sometimes the best incentive is to make doing the right thing easy and obvious.