10. Tokenomics

Prices are important not because money is considered paramount but because prices are a fast and effective conveyor of information through a vast society in which fragmented knowledge must be coordinated.1

—Thomas Sowell

Designing systems of incentives to underpin blockchain networks is sometimes known as tokenomics—a blend of, you guessed it, “tokens” and “economics.”

While tokenomics may sound like something entirely new, it’s novel only insofar as it applies old concepts to the context of the internet. Conceptually, there’s nothing that groundbreaking. Tokenomics mostly just encompasses economics. (Practitioners also call the discipline protocol design, but I avoid this to prevent confusion with early internet-style protocol networks.)

Blockchain networks didn’t invent the idea of virtual economies with built-in, or native, currencies. Games have had virtual economies for years. In the 1970s and 1980s, arcades started swapping out2 the usual coin-operated games with cabinets accepting proprietary tokens. As arcades expanded, grew in popularity, and added more games, they often raised the price of their tokens. Old tokens would still be usable, so if you bought a bunch of tokens early and held on to them, you would have a lower effective cost to play games than others.

A more sophisticated version of the same idea exists today within video games. Eve Online, which has been around since the early 2000s, is probably the game most famous for its virtual economy. Eve has millions of players who trade and battle3 across a fictional galaxy called New Eden. The game maker, CCP Games, publishes a data-rich monthly economic report about conditions within the game, such as what market prices fictitious ores like “veldspar,” “scordite,” and “pyroxeres” are fetching. The studio takes its economy, based on so-called InterStellar Kredits, very seriously. It made headlines in 20074 when it hired a respected PhD to run its in-game monetary policy.

Eve’s success inspired a generation of followers, from simple mobile games like Clash of Clans to hard-core games like League of Legends. These games all have in-game currencies and ways players can earn and spend those currencies. The game makers create demand for their digital currencies with fun experiences that attract millions of players, who then use the native currencies to buy in-game virtual goods. Demand drives up the currencies’ value; ebbing interest sends it downward.

Tokenomic designs in blockchain networks build on lessons learned in video games. A blockchain economy, like any healthy virtual economy, should balance the supply and demand of native tokens to fuel sustainable growth. A well-designed token economy helps the network flourish. The right incentives turn users into communities of owners and contributors.

But incentives must be designed with intention, otherwise there can be unintended consequences. “Incentive structures work, so you have to be very careful of what you incent people to do,”5 as Steve Jobs once observed about corporate incentives. Sounding a note of caution, he added, they can “create all sorts of consequences that you can’t anticipate.”

Faucets and Token Supply

A common metaphor for thinking about the design of token economies is to imagine tokens as water that flows through the plumbing of a house. Sources of supply are “faucets,” which provide water, and sources of demand are “sinks,” which drain water.

The network designer’s first goal is to balance the faucets and sinks so water doesn’t over- or under-flow. Faucets that are too strong can lead to more supply than demand—and therefore downward price pressure. Sinks that are too strong can lead to less supply than demand—and therefore upward price pressure. Without the right balance, token prices can swing too far up or down, causing a bubble or crash. Such events distort incentives and decrease a network’s utility.

Much of the discussion in the last chapter concerns faucets: token grants for developers, bootstrapping networks with token rewards, air-dropping tokens to early users, and other activities. Ideally, faucets optimize for positive behaviors that grow the network. They should incentivize software developers to build new features and experiences, and they should turn other participants, such as creators and users, into a community that’s motivated to cultivate and grow the network.

Common examples of faucets include the following:

Faucets are a powerful tool for building networks. The token incentives they distribute can help overcome the bootstrap problem, recruit early contributors, fund ongoing development, share upside with a broad community of users, and keep networks secure. They are analogous to land grants in early cities that align incentives and encourage real estate, business, and other development.

Sinks and Token Demand

The best sinks tie token demand to network activity, thereby aligning the token price to the network’s usage and popularity. Useful networks generate more token demand, while less useful networks generate less demand.

Sinks that charge fees for network access or usage are known, aptly, as access or fee sinks. You can think of them as the digital equivalent of highway tolls, collecting just enough for network upkeep. Ethereum and certain DeFi networks take this approach. The Ethereum network has a maximum capacity and can run a limited amount of code at any given time. To avoid overload, the network charges for computing time. (Recall that Ethereum behaves like a public computer, reminiscent of time-sharing mainframe computers from decades ago.)

The cost of compute on Ethereum is called gas. The price of gas is a small denomination of the native token ether, which varies according to supply and demand. The Ethereum network collects some of its gas fees to buy and “burn” (read: destroy) tokens. This collection and burning activity reduces the token supply and, in theory, increases the price of ether (assuming constant demand). Similarly, DeFi networks like Aave, Compound, and Curve take fees and save the proceeds in their network treasuries, which can later be redistributed via faucets. All of this happens automatically, powered by immutable code embedded in each blockchain network.

Another common sink for base-layer blockchains is a “security” sink that rewards token holders for “staking,” or locking up tokens in validators. As discussed earlier in “Blockchains,” validators are computers that maintain the security of a network by verifying the validity of proposed transactions. Staking is the process by which users lock tokens in code-enforced escrow accounts. If a validator behaves honestly, it gets rewarded with more tokens. In some network designs there can also be a penalty if the validator behaves dishonestly. Staking is a double-edged sword: it creates both a sink, which locks up (and sometimes confiscates) tokens, and a faucet, which rewards honest stakers with tokens.

Security sinks have pros and cons. On the pro side, they promote network security. The more money that’s at stake, the more secure the network and its applications. As applications running on the network become more popular, more people pay to use them and network revenue goes up. This puts upward pressure on the token price, which raises the staking rewards. This, in turn, encourages more staking and, thus, improvements in network security.

On the con side, security sinks can be expensive. Designed with built-in faucets to reward staking, they can counteract demand pressure by adding to the token supply, potentially depressing prices. This is why blockchain networks like Ethereum combine access sinks with security sinks, and why their communities fine-tune token inflows and outflows to ensure balance. Too much of one or the other can throw a system out of whack.

The last common type of sink we’ll cover here is “governance” sinks. Some tokens give users the power to vote on changes to the network. Users will buy tokens, so the thinking goes, to gain more influence. The incentive to vote gets people to acquire and hold tokens, taking them out of circulation, therefore generating demand for tokens and a resulting sink. Governance sinks can, however, suffer from free-rider problems. Free ridership happens when people don’t vote. They might skip out because they think the outcomes of the vote don’t matter, or because they believe an election will swing their way regardless of whether they participate. Governance tokens help keep networks democratic, but they are unlikely to sustain token demand all by themselves.

Well-designed sinks correlate with network usage. As usage increases, more tokens drain away, which creates upward price pressure. Upward price pressure increases the value of token rewards used for security, software development, and other constructive activities. Designed correctly, sinks create a virtuous cycle.

Badly designed faucets and sinks can, however, fuel a speculative environment that destroys the spirit of the community. Some blockchain communities focus almost exclusively on token prices. Paying excessive attention to prices is a bad sign—a hallmark of casino culture. Well-designed token incentives focus communities on constructive topics, like new applications and technology improvements. The quality of a project’s discussions often reveals the health of its community.

Tokens Can Be Valued Using Traditional Financial Methods

A common argument against blockchain networks is that tokens are purely speculative and have no intrinsic value. Newspaper columnists routinely refer to them as scams. Warren Buffett branded them “rat poison.”6 Michael Burry, the contrarian trader of The Big Short fame, has labeled them “magic beans.”7 The implication is that networks that depend on tokens can’t be useful. It’s all just a speculative mirage.

Cherry-picking the worst examples of an emerging technology to dismiss a promising new industry out of hand may make for catchy headlines, but it is a disingenuous form of criticism. The railroad wasn’t worthless just because many unviable railroad companies fed early stock market mania. When automobiles made their debut, they were considered impractical, inefficient, and life threatening. The early internet featured content that was silly, offensive, even dangerous, and many people who thought they knew better regarded the industry as either unserious or, at the other extreme, morally hazardous.

Understanding new technologies takes work. Critics who focus on the bad while dismissing the good fail to foresee the long-term potential of disruptive innovation. While it is true that there are plenty of poorly designed tokens driven purely by speculation (see: most memecoins), this is not true of all tokens. What the criticisms miss is that software is a highly plastic medium and that almost any economic model that can be dreamed up can be implemented in software. An honest assessment would look at the details of token designs instead of generalizing from a few bad ones.

There are plenty of well-designed tokens that have sustainable sources of supply and demand. Take Ethereum, for example. Recall how the system collects fees on transactions, or network usage, and how it uses these funds to buy and burn tokens, thereby taking them out of circulation. Reducing the supply of tokens can increase the value of existing ones, benefiting token holders. All of this happens automatically as part of the system’s transparently encoded rules, with no company making decisions about the process behind the scenes. The design makes sense.

Put another way, Ethereum generates the token equivalent of cash flow. The more applications written for Ethereum, and the more those applications get used, the greater the demand for computing time and for Ethereum’s native token. The supply of ether varies but, generally, after all faucets and sinks are accounted for, has stayed relatively flat (in the past it increased slowly; recently it has been decreasing). This means the price of ether should roughly correlate with the popularity of applications built on the network. By studying the cash flows and burn rates of blockchain networks, you can value tokens like Ethereum’s using traditional financial metrics such as price-to-earnings ratios.

Ethereum shows what good token design can be, but it is not the only well-designed blockchain network. Others include DeFi networks that use similar models. The tokens these networks collect as fees go toward funding network activities, such as buying and burning tokens, or distributing money to token holders. If you understand a system’s faucets and sinks, you can evaluate its tokens. Access and fee sinks generate network earnings, minus any costs. The price of tokens times the supply (with some discount rate applied to future token issuances) yields market capitalization. All of this is standard finance.

Compare what we’re talking about with real estate. Blockchain networks that charge for access have characteristics similar to property holdings. “Price-to-rental” is a common valuation metric in realty, for example. You can calculate it by dividing home price by annual rent. The answer may inform whether you choose to buy or rent a house, or to live in or rent a house you own. That you can always rent out real estate and generate cash flow provides a model for valuing the assets. In the same way, you can apply fundamental analysis to blockchain networks to determine a fair value for tokens.

Whether tokens have value reduces, primarily, to whether they will have long-term demand. This depends, in part, on their economic design. The faucets and sinks of a blockchain network need to be designed such that network popularity converts into sustained token demand.

Of course, this raises a trickier question: Will the network be popular? It’s impossible to know. Some networks will succeed, and some won’t. This I can say for sure, though: the ones that succeed will offer useful services that attract users to the network.

A reasonable skeptic might doubt the viability of a specific network or whether the world needs blockchain networks at all. Maybe the internet has enough networks. Maybe corporate networks are sufficient and will keep winning, either because users are too locked in already or because they’ll always outcompete blockchains in areas like user experience. That’s not my view, but it’s a valid stance for a critic to assume. What’s unreasonable is to say that tokens are based on fanciful economic theories. Tokens are not magic beans. They are assets used to power virtual economies, and they can be valued using traditional financial methods.

Financial Cycles

Speculation exists everywhere property can be bought or sold, from equities and commodities to real estate and collectibles. Markets have always had speculation, and they always will. Tokens are no exception. Economic agents are excitable, especially in the presence of a promising new technology, business, or asset.

In her 2002 book, Technological Revolutions and Financial Capital, Carlota Perez, an economic historian, describes how tech-driven economic revolutions follow predictable cycles.8 First there is an “installation phase,” involving an “irruption,” or tech breakthrough, followed by a “frenzy” of speculation. Then comes a market crash: the bubble bursts. After that, a “deployment phase” takes place, including a period of “synergy,” where the new tech gets adopted. Finally, industry consolidates and reaches “maturity,” making once-groundbreaking inventions routine. And so, in fits and starts, capitalism progresses.

Another way of looking at the course of tech innovation9 is through the “hype cycle,” a management framework the consulting firm Gartner made popular starting in 1995. Gartner’s model builds on the work of other thinkers,10 like the economist Joseph Schumpeter, known for his theory of creative destruction. The model illustrates how when a new technology arrives, the excitement it generates can catalyze a financial bubble (the peak of inflated expectations). A crash usually follows (the trough of disillusionment). Then there’s a long period of productive growth as the technology gets broadly deployed (the slope of enlightenment).

The hype cycle has played out many times across many technologies, including railroads, electricity, and automobiles. Take the internet, for instance. Dot-com mania climbed toward its “peak of inflated expectations” in the 1990s. A slew of overpriced IPOs came out of that era, but so did several legitimate and wildly successful companies. After the early 2000s’ “trough of disillusionment,” two steady decades followed along the “slope of enlightenment,” bringing internet valuations back to new highs, this time driven by fundamentals. Any skeptic who dismissed dot-coms as magic beans would have missed out on the successes of Google, Amazon, and others.

Blockchain networks have already gone through multiple boom-and-bust cycles, each one bigger than the last. Some of the initial excitement was grounded in genuine tech breakthroughs. In 2009, Bitcoin pioneered the concept of a blockchain. In 2015, Ethereum expanded on the concept, creating a general-purpose programming platform. Both were technological advances that marked a classic period of irruption, in Perez’s terms. As often happens, market excitement then got ahead of itself. The technological reality didn’t support the outsized returns investors and entrepreneurs sought, at least not immediately. Crashes, sometimes precipitated by shocks, like macroeconomic events or the collapse of a prominent project, ensued.

One could argue that blockchains, more than other technologies, exacerbate speculative cycles because their key innovation is to enable digital ownership. When you own something, you can do what you like with it, including buying or selling it. If we lived in a world in which you could only rent houses, and one day someone invented a way for you to own houses, speculative real estate markets would almost certainly pop up. Smart policy and regulation can help tamp down speculation (a topic I discuss further in “Regulating Tokens”), but speculation also tends to quiet down naturally as people learn how to value new technologies based on fundamentals.

My colleagues and I have studied the ups and downs of token markets, and we call the pattern we’ve observed the price-innovation cycle. Token markets follow the same cyclical patterns economists have long studied, discussed above. New innovations kick off a period of interest and activity, which generates enthusiasm and price increases. This attracts more founders, developers, builders, and creators to the industry. If the market crashes because expectations overinflate, builders stick around and keep working on new ideas. Their efforts incubate further advances that eventually renew the cycle. As of the time of my writing, we’ve been through at least three cycles, and we expect the trend to continue.

Speculative manias don’t just characterize tech revolutions; they often enable them. Many emerging technologies are resource intensive and depend on big capital inflows to fund the infrastructure required for the next rollout phase. Railroads require enormous amounts of steel production and spike-driving labor. Electricity flows only as far as a power grid can carry it, and automobiles roll only as far as roads can take them. The dot-com boom built out massive broadband infrastructure, which would later be essential for the industry’s growth. Speculative investments don’t always go to waste.

Blockchains need big investments too. They require tooling and infrastructure, and the networks and applications that are built on top of them require capital to fuel growth. Big Tech corporate networks spent tens of billions of dollars to scale to billions of users. Networks that intend to compete with them will require similar sums. A little exuberance, whether rational or irrational, goes a long way.

I expect the markets around blockchain networks will follow the same trajectory that markets around other technologies have followed throughout history. Over time, fundamentals will drive token prices just as they drive prices in other markets. Speculation will cool, replaced by a more sober evaluation of the sources of token supply and demand. To quote an old Wall Street adage attributed to Benjamin Graham,11 the father of value investing: markets are a voting machine in the short term and a weighing machine in the long term.

In other words, assets with actual substance or weight—fundamental value, in finance-speak—have the best prospects over the long term. The implication is that it may be prudent not to let any short term razzle-dazzle distract you. Just because something wins a popularity contest today doesn’t mean it will age well.