Forget for a moment about the value of the cryptocurrencies that you may or may not own. Instead of thinking of blockchains as investment bets or just cool technology, think of them as entirely new, and previously impossible, economic systems. Because that’s what they are.

Just like any economy, a blockchain requires that its designers define monetary policy* (inflation), fiscal policy (block size), taxation (fees), voting (governance/upgrades), and provide for the common defense (securing the network). Yet, unlike traditional economies, they offer the possibility of greater freedom and transparency because they avoid the problems of centralization and concentration of power.

That’s the good news. The bad news is that these new economies comes with extremely high risk.

One of the risks, ironically, is also one of the technology’s greatest strengths. As Elad Verbin points out in his post on Behavioral Crypto-Economics, “Blockchain systems are, by design, difficult to change once deployed.”

Mark Zuckerberg’s hallmark mantra “move fast and break things” does not apply here. If blockchain developers don’t start from an extremely well thought out design, they may very likely have doomed their project. Repairs and improvements to these systems are famously difficult. Protocols with billion-dollar valuations could disappear overnight. Things can get very acrimonious. Want evidence? See the Bitcoin block size debate.

Then there are … the people

As if designing a system without flaws weren’t enough pressure, blockchain creators face another big risk when developing these new economies: accurately predicting people’s behavior.

It is one thing to lay down the rules for an economy and encode them in software (i.e. “code is law”). But those rules are based on predictions of how people will behave in the economy — the value they will place on a currency or the level of incentive that will drive them to participate in the ecosystem, for example. And those predictions are notoriously hard to get right.

Just look at some of the decisions each of us make on a daily basis. We may vote for policies that go against our own economic interests. We make food selections that are at odds with our physical health. There’s no clear, codeable logic in much of our behavior.

In fact, an entire field of people study this very phenomenon, including Nobel laureates Daniel Kahneman and Amos Twersky as well as University of Chicago professor and former Clinton advisor Cass Sunstein (author of Nudge). It was Sunstein who discovered that changing the default setting from “opt-in” to “opt-out” on things such as organ donation on a driver’s license and 401k contributions at work could dramatically improve uptake.

Of course, once Sunstein discovered this, he only had to share the findings with a few, key central authorities who were able to institute the decision across the entire network. They did not need to poll every citizen in a state to get approval to change the Driver’s License registration process. The “fork” of the protocol, in this case, was relatively painless.

Blockchain system designers face greater challenges than Sunstein in implementing changes. This makes the accurate prediction of how individuals will behave in a given situation an absolutely critical, non-negotiable, component to crypto-economic system design. In fact, creating sustainable peer-to-peer value transfer systems at scale is simply unlikely to happen otherwise.

And this is where incentives come in. Properly designed, they can encourage people (like Sunstein did) to behave in desirable ways that benefit and grow the network.

Incentives, in blockchain land, come in the form of digital tokens. These tokens are the internal currency of the network. Their perceived value is important in keeping miners active in securing the network (what amount of block reward at what frequency and with what difficulty is enough to motivate them?) as well as in helping individual users assess the benefits they get from the network compared to alternative networks.

Getting incentives right is fundamental to network growth, reflected in increased token adoption that yields positive network effects. Once this flywheel gets started, it serves as the ongoing funding mechanism for future development. Without it, the network cannot achieve self-sustainability. The value of the community and the token is what incentivizes new members to join. If that value is off, new people don’t join and a death spiral begins.

You can see why getting those incentives just right it is so fraught with risk.

There’s still a lot to learn

A mastery of crypto-token economics is critical to the design of robust, sustainable blockchain economies. It involves multi-disciplinary skills, including mechanism design, game theory, behavioral economics, public policy, macro-economics, and a decent understanding of decentralized technology. The day will come when we have Ph.Ds emerging from top universities in the field of crypto-token economics. Today, however, there are only a handful of experienced practitioners in the entire world. As Aleksandr Bulkin wrote over a year ago– “Cryptoeconomics is hard.” No joke.

Without understanding the token economics of a project, investors, developers, and potential network contributors have no way to assess the long-term potential of a decentralized network. Along with team, technology, governance, and community, “token” forms part of the critical “T3CG Framework” I use to assess the likelihood of project success.

As excited as I am about the potential of blockchain systems, I am humbled by the realization that we have only just scratched the surface on how to build this systems properly. Blockchains that depend a lot on humans and their propensity to behave rationally — such as Augur, Gnosis, Steem, and Numeraire, for example — will face bigger challenges than blockchains with less complex economic rules, such as Bitcoin.

It may end up being the case that the super-hyped, human-powered blockchain systems of 2017 and 2018 are merely the case studies that future students of crypto-token economics will study to avoid flaws in system design.

For further reading: The whitepaper Economics​ ​of​ ​Initial​ ​Coin​ ​Offerings helped me understand fiscal and monetary policy of crypto-token economies. I highly recommend it.

Jeremy Epstein is CEO of Never Stop Marketing and author of The CMO Primer for the Blockchain World. He currently works with startups in the blockchain and decentralization space, including OpenBazaar, Zcash, ARK, Gladius, Peer Mountain and DAOstack.