We talked about the first 2 pillars of Tokenomics, which generally affects the incentives of the system (DApp) today. Now we are moving on to the more complicated part: how to ensure that the system (DApp) has specific properties in the long run.
Let’s make this simple.
In the physical world, when the economy messes up, the government can step in to help. Maybe the government can spend lots of money to help a company get out of debt. Or the government can print more money to encourage spending. Whatever it is, the government can react to existing problems.
Now, in the crypto space, things are a little different. There is no “central government” that controls anything, much less react to current issues. So….. we fucked? Nope. In the crypto space, we can incentivise people to behave in a right (rational) way. This can be done through game theory.
Hold up. What’s Game Theory?
I’ll explain game theory on a very high level.
Imagine there is a sealed cookie jar at home. Mum says that no one can eat the cookie. You and your sister both wants the cookie. What options are there?
You and sister both get the cookies.
You get the cookie.
Your sister gets the cookie.
No one gets the cookie.
So now, based on the options, what will happen next?
Both gets the cookie. Mum doesn’t find out because no one tells her.
You get the cookie. Sister does not. Sister tells mum about your bad behaviour. You gets punished.
Sister gets the cookie. You do not. So you tell mum about her bad behaviour. She get punished.
No one gets the cookie. Nothing happens. You get hungry.
So, what will you choose? You’ll probably choose option 1: where you and your sister both get cookies and no one tells mum.
Game theory is basically strategic decision making. You look at the options, consider the outcomes to the solutions and make the best decision. Important note: other person’s behaviour can affect the outcome.
Why should I care? Is this even important?
If game theory is looking at the outcome based on the game, mechanism design is designing the game based on the outcome (aka the opposite). Simply put: You want the participant to behave in a certain way. What rules can you have in your game to incentivise good actions, punish bad actions, and ultimately influencing their behaviours?
Real World Example
For example, a problem we face is bribery. Let’s look at the physical world first. If Mr President passes Bill BadForEnvironment, Mr RichOilGuy will pay Mr President £500,000. Do you think Mr President will pass the bill? Yeah, for sure. His incentive is that £500,000. It affects his behaviour. So what can we do? We can create a committee of 100 thought-leaders to evaluate the bill. Mr RichOilGuy has to bribe 51% of the committee for them to say, “yes this a good bill”. That is a very expensive bribe for him. And if he is caught bribing, there is also a punishment: jail time. He probably wouldn’t bribe 51% of people then, affecting his behaviour.
In the Crypto Space
Putting this in the crypto space, we also have committees of people who vote on issues. When we reach consensus, the issue is resolved. Bribery can also happen in the crypto space. We can affect his behaviour by:
Having a lot of people on the committee. So it is expensive to bribe. (De-incentive)
Randomly allocate users to be on the committee, so you don’t know who to bribe. (De-incentive)
To raise an issue to the committee of people, you have to put some tokens into the fund. If it turns out that you are a bad actor, you lose those tokens. (Punishment)
So, mechanism design is important because it affects the actions and behaviours of people in your network. This will affect the long-term sustainability of the project. Imagine if it is easy to bribe the president, the economy will not do well. Same with your project. It will not do well.
What should I take note of?
Unlike customer analysis and market design, mechanism design has many variables. Market design has 3 important variables to determine success (thickness, de-congestion, safety), but mechanism design doesn’t work that way. There isn’t a “formula” per say. So I will share a few categories to consider. Further on, we will look at case studies and examples.
3 categories to take note of:
These categories require a lot of explanations on its own, so we will dive into further explanations later.
How do I get started?
Now we understood the concept of mechanism design. How do we implement it? There are many solutions to this problem, and it really depends on the business model and product. There is no “one size fit all” solution.
Understand what the goal/objective of the system is. Start off with something simple, and expand on It later. If this is a game, what is the goal of the game?
Consider the information available, for participants to make an informed decision. If the information is known to all, one or none of the players?
What types of behaviours do you want to encourage?
What can you do to incentivise it?
How can you punish bad actors?