Economics Design

Research

By

Game Theory and Token Economics

What does that mean? Say the example of Airbus and Boeing. Airbus wants to produce and Boeing has little incentive to. But as a airlines company (e.g. British Airways), I know I need a bigger plan to fulfill the demands. That is my outcome: for the superjumbo jet to be produced.
Through game theory analysis, I can see that the probability of superjumbo jets happening is quite slim.

What can I do?

Some mechanisms to change the behaviours and outcomes:
  • Maybe I can pre-order 100 superjumbo jets from Airbus and Boeing, so they know they have a confirmed demand for it
  • Lobby the UK government to increase subsidies for the R&D part, so both parties have less losses/risks. Maybe that might entice them to develop
  • Look for a new competitor that produces plans and ask them to work with Airbus
Those are just basic simple (& maybe crazy, not-practical) ideas. But you get what I mean. I can introduce mechanisms to change the behaviours and outcomes, so it suits an objective.

What’s the link with token ecosystems?

Similarly, we want to analyse the outcomes and build in some mechanisms and rules to ensure that the outcomes are in line with the objectives.
Remember I mentioned that every ecosystem has 1 main objective? The truth is, that can be challenging to achieve sometimes, when you have a decentralised ecosystem and everyone behaves differently.
However, you can change that through rules.
I want my ecosystem to have honest miners. So I can make the hardware expensive and difficult to get, to reduce cheats from entering the game.
We can do that through incentives (rewards) or disincentives (punishments). These act as the mechanisms and rules to ensure that the outcomes are in line with the objective of the ecosystem.

What rules and mechanisms?

There are many, but to make it easy, let’s call it governance. Governance, in my opinion, is the most important aspect in the token ecosystem.
There are many types of governance:
  • How can we get the true value of the participants so we can allocate the goods efficiently? (e.g. there are 2 transactions waiting to be validated. What mechanism or governance can I have in place, to prioritise which transaction to approve first?)
  • How can I ensure that my rules induce a game among strategic agents in order to realise a system-wide objective or social choice function, in an equilibrium of the game? (e.g. how can I reduce instances of pump-and-dump, since this is not an objective of my ecosystem?)
We are just scratching the surface of game theory and its application to token economics. At the end of the day, it forms part of the design model for the ecosystem to exist.

By

Macro vs Micro Token Economics

Macro vs Micro Economics

Economics is divided into 2 branches: macroeconomics vs microeconomics. Similarly in token economics, we can divide the various categories into these branches.

Macro Token Economics

Traditionally, macroeconomics looks at business cycles and economic growth. It means understanding how the economy, as a whole, behaves by looking at indicators. For example, inflation, price levels, income level, etc.

Business Cycles

This is where your governance comes in (see where I’m going?). For example, monetary policy and fiscal policy. These are movement or actions by the government to make sure that the economy is healthy. For example, we don’t want inflation to be too high or inflation to be too low.
Macroeconomics also includes analyzing how different sectors relate to each other, especially in times of international trade. For example, how does consumption, investment, savings and international trade affect each other? What policies (or rules) the government needs to have, to ensure that the economy is healthy and balanced?

Economic Growth

Economic growth is just growth of your economy or ecosystem. Here, we want to understand the factors that contributes to the long-run and income growth. For example, technology innovation helps economic growth as according to the Solow Growth Model.

So, what’s the link with my token ecosystem?

This link is in the design of your token.
Macro Token Economics
Macro Token Economics

Business Cycles

Although token ecosystems do not have a centralised government to respond to changes your ecosystem, there are rules that can be coded and embedded. For example, MakerDao uses various stability mechanisms like target rate feedback and risk parameters to govern the ecosystem.
We can also analyse the aggregate behaviour of the economy.
If the token function is a currency, there are plenty of literature to analyse the inflation model, interest rate, allocation of tokens, supply and creation or burn rates.

Economic Growth

Since token ecosystems are a form of business models, you have to consider the long-run growth and factors contributing to it. And it is also important to remember that these factors have to be endogenous, not just exogenous.
For example, if your business model is tokenising common data, what are the internal factors (endogenous) that will affect the long-run growth? Maybe proprietary IoT technology hardware, sale of data with confirmed partner, etc.
If the token function is a security, check out financial literature on asset valuation, profit distribution and other relevant factors.

Micro Token Economics

As the name suggests, microeconomics focuses on individuals actors. The 3 types of actors in the physical fiat economy are households (people), firms (companies that hire people), and government (sets rules for households and firms).
It means understanding how actors behave, as an individual, given the limited resources. For example, why you are happier drinking the first cup of tea than the 10th cup of tea in a row.
Microeconomics looks deeper into various allocation mechanisms of resources, methods to extract the right value from participants, and how people make decisions. We look at demand and supply, production theory (inputs to outputs), production costs, etc. The field is endless.

So, what’s the link with my token ecosystem?

This link is in the design of your market and mechanism.
Some fields in microeconomics that is applicable to token economy:
Micro Token Economics
Micro Token Economics
These are part of the factors to consider when designing a token ecosystem. There is not “right answer” or “perfect model” for the ecosystem. It depends greatly on the token function, use-case, business model, etc. The only way is to read, apply, test, improve.
In the next few videos and articles, I will share more about each field in details and its application to token economics.

By

Token Valuation Models

Token valuation models are used to evaluate token prices. But what’s wrong with them?
Token valuation models
2 things about token valuation:
  1. Token valuation is not as important as you think it is. It is important from an investor’s point of view, but from the token economics and design perspective, it is the last thing to consider.
  2. Tokens should be valued with endogenous, not exogenous variables.

For the longest time, I am struggling with the last design aspect of token economics: “Token Design”. I struggle because I’m trying to balance what is necessary (variables needed to be considered in designing the token) and what people want (how much will my token be worth).

MV=PQ Token Valuation Model is wrong?

Let’s address the elephant in the room.
No, it is not wrong. This model is known as the “Quantity Theory of Money”, under the category of “Monetary Economics”. It is used to price money. Money itself has no price, but it has a value. We can extract this value from such models.
In the physical-fiat world, you have central banks governing our money. How do they price our money? The philosophy came from the general idea that price levels is directly proportional to the money in the market. Even in the physical world, this theory is challenged by Keynesian economics. Because prices are sticky and velocity is not stable in the short-run, this is a flawed model. This model is useful in the long-run.

The link with Token Economics?

Okay, so what the hell does it have to do with token economics?

2 things: (1) model for currency and (2) the idea of velocity.

1) This model is meant to get the price of money.

👉🏻 Your token is not always money. It’s like trying to use a fruit knife to cut a slab of beef. Please use the right tools instead.

2) Focuses on exogenous (external) variables.

👉🏻 Token ecosystems are amazing because you get to design many endogenous (internal) variables of the token. By that measure, the token’s value should be defined from the endogenous variables, not the exogenous variables. This is why token ecosystems are DIFFERENT from the fiat world ecosystem. We have comparatively more control of the output, given the input.
In token ecosystems, velocity is an exogenous factor because you can’t control how quickly your participants change hands. It can be indirectly linked to network effects, congestion, even your incentive mechanisms. But if we have direct links to the value the token ecosystem can provide, why not use it?

Why do people still use it?

Frankly, after speaking to many people in the industry, it boils down to 2 things.
(1) It is an easy model.
(2) This is the only model in the market.

To which I’d like to comment:

Easy doesn’t mean it’s good

(1) Yes, it is a pretty easy and straight-forward model. But is is not sound. If you are using velocity to figure the price of token, how are you defining velocity? Velocity of USD, a random number, or? If you modeled an ecosystem and allow the systems to trade to get a “V”, how does this account for the different behaviors of the participants (behavioral economics) and how sure are you that the incentivisation model will create such behaviors (mechanism design)? Also, how well designed is your market to allow for network effects to take place? There are so many questions.

It’s not the only model

(2) This is not the only model in the market. This model is just the popular model available. I believe that Chris Burniske was the first person to mention this model in his article: Cryptoasset Valuations. This model might work for SOME projects, but not all. Not every token model that we see or token function is a currency. Some are utilities, some securities. So, this model doesn’t always work.

So what token valuation models then?

Good question.

Firstly, value tokens with endogenous factors.

Token ecosystems are designed with multiple variables to allow value in its design. Hence, it is important to determine the value of tokens through the various design mechanisms in place.

Some variables that can allow tokens to have endogenous value:

  • Expected value of funds
  • Dynamic price equilibrium (Athey et al., 2016)
  • Exchange rate of tokens to dollars reflect buyer’s willingness to buy (ex-ante, market clearing condition)
  • Scarcity of tokens included by pricing choices causes buyer competition that reveals consumer values
  • Rational expectation of exchange rate in next period by agents
  • Value of price commitments
  • Demand growth from demand of buyers
  • Savings function/incentives to save
  • Function of expected growth through demand of platform or money supply
  • Value to consumer
  • Network effects
  • Heterogeneity of user base
  • Platform productivity
Disclaimer: this is not an exclusive list. This is not applicable to every token business model. It varies from the token function. There is NO one solution that fits every ecosystem.

Secondly, allow the market to dictate the price.

This can be done with either Dutch Auctions or Vickrey-Clarke-Groves auction. Or other new mechanisms.
Dutch auctions are more for homogenous good while VCG is for heterogenous goods.
Dutch auctions are a great pricing mechanism as the market will determine the price (value to buyer) instead of seller (usually monetary focused). It is a fairer way to allocate value to everyone in the network than being hostage to the dictation of a few people. Dutch auctions is also used by government when selling bonds.
The literature is researched till here, utilities and some extent of securities. I’m interested in valuation models for crypto securities. What is the current model for non-crypto securities (bonds, stock, options). What additional considerations needs to be in place for securities in the crypto economy? How different does securities need to be, when shifting from the physical to digital marketplace.

Sources and further reading:

By

Token Ecosystem Design Process

Token ecosystem is all about design. The economics of tokenisation (aka tokenomics) is part of this entire process. This article shares the Token Ecosystem Design process stages, elements to consider during each stage and where economics of tokenisation (aka tokenomics) fits in.
Transparency is very important to me, hence I’d like to share the token ecosystem design process with you. (Unpopular opinion) There are too many bullshit “advisors” and people out there, who don’t know what they are doing. But if you are equally clueless, it’s difficult to tell who’s bullshitting you and who isn’t. (This is what we call, moral hazard in economics.)
So, this post can give you more ideas and information about the whole process, if you need help and what specific help you need.

Token Ecosystem Creation

Token ecosystem is the entire system where your tokens and users co-exist. If you think about it, it’s basically your business model.
Token Ecosystem Design process
Token Ecosystem Design process
The creation process comes in 3 segments: discovery, design and deploy. This is adapted from the 3 Ds of token ecosystem creation, created by Outlier Ventures. It’s easy to remember, and easier than a model I created previously.
  1. Discovery Phase: determine the particular characteristics of the business model or ecosystem and why a token is needed in the first place.
    1. Who: customer analysis, actors in the ecosystem, developers, users
    2. What: token business model
    3. Why: why it adds value to participants
    4. How: Smart contracts, boards/authority, other solutions
  1. Design Phase: making high-level design choices including, governance structures, the (mathematical) token model and its parameters. These need to be optimized for stakeholders’ incentives and the long-term sustainability of the associated ecosystem in order to avoid value leakage.
    1. Define Design space
    2. Define objectives and constraints
    3. Economics of Tokenisation Framework design
    4. Token network fit
  2. Deployment Phase: comprises of validating and stress-testing the parameters that were defined during the design phase, before fully integrating into the network
    1. Surveys/theories
    2. Experiments
    3. Simulations
    4. AB tests

How It Works

It’s as simple as it can get. It is the execution that is really difficult.
Token Ecosystem Design process
Token Ecosystem Design process
Step 1: Discovery stage. This stage is to understand the basics of your token ecosystem. A good way is to think of the various participants, the business model, what value you add and possible solutions.
(Spend 10% of the time here.)
Step 2: Design (1/2) stage. This stage is to fully understand and agree on the most important objective of the token ecosystem. Then, you can expand to 2nd most important objective, 3rd most important, … Also, define the design space and constraints. This is very important because it defines the environment in which the tokens and participants will co-exist in. It also defines the environment that we will be designing, in the next steps.
(Spend 10% of the time here.)
Step 3: Design (2/2) stage. This stage is where economics of tokenisation comes in (aka tokenomics). This is where the market design, mechanism design and token design comes in. The idea is to design each part individually, then combine them together and find a token network fit.
(Spend 60% of the time here, designing, editing and enhancing the previous model.)
Step 4: Deploy stage. This stage is to run various tests to test the token network fit in previous stage. Steps 3 & 4 are an ON-GOING CYCLE. That means, token network fit draft 1 —> deploy and run tests —> improve and create token network fit draft 2 —> deploy and run tests —> improve and create token network fit draft 3 … you get the idea.
(Spend a total of 20% of the time running tests and collating results.)

Where Does Economics of Tokenisation (Tokenomics) comes in?

The economics of tokenisation comes in at the last half of this process, steps 3 & 4. In step 2, we define the outcome that we want from the token ecosystem. In step 3, we will design the various environment and mechanisms to achieve the outcome in step 2. Then, we run tests in step 4 to see if the outcome is achieved!
If not achieved, return to step 3 to edit and test them again.
The economics of tokenisation is not a one-off service. It requires time to research, test and analyse the right model for the token. I wish it was as easy as a cooking recipe, where you can take an existing model out there, and edit it to suit your ecosystem. The beauty of token ecosystem is that it is highly customisable to the outcomes we want. We just have to spend a lot more time understanding what it takes to achieve the outcomes.
We all want answers, and we want them now. But the truth is that good things take time. I hope you allow sufficient time to design the right economics for your token ecosystem. It is not easy, but if we want to make real changes and define our future of tokenisation, we have to start creating sound economics right now.
Feel free to contact me or inquire about my crypto services. And if this information is good enough, I wish you all the best for your token ecosystem!

Further reading/Sources

Ps: token ecosystem design and creation is like baking the perfect macaron. You can have the steps and general ingredients needed, but it is the details that really determines the success of it. Over whip, undercook, not resting enough, and it’s destroyed.
Just like token ecosystem, the general steps is applicable to every project. To get the best success in the token ecosystem, you need knowledge and experience, that is less easy than reading an article.

By

7 Wonders of Cryptoeconomics

Economics is a huge topic, so Cryptoeconomics is not any different. Cryptoeconomics can be divided into the many concepts, theories and classifications.
Cryptoeconomics sounds complicated, but I will prove to you that it is not. In this article, I share the 7 wonders of economics concepts, the explanation, real world example and crypto example. I am basically taking existing economic concepts in the physical real world to the digital crypto world. Hopefully, that inspires you to understand the power of economics in blockchain, and inspire you to learn more.
Note: this is not an exhaustive list of economic concepts. I just chose 7, because it makes a nice title!
7 Wonders of Cryptoeconomics
7 Wonders of Cryptoeconomics

1. Network Effects

You hear people talking about network effects all the time, especially at conferences by economists or people in the space.
  • What’s that? Basically, the more people using the thing, the more valuable the thing is.
  • Example in real world: classic example is telephones. But it’s 2018, so we are not using that example. Look at WhatsApp. You can only text someone if they have the app. The app becomes more valuable (to me) as more people join. Now, I can text my international friends with WhatsApp because they have the app. On the other hand, my grandmother doesn’t have WhatsApp, so it’s not possible for me to text her. WhatsApp is more valuable (to the users) as more people use it.
    • Other communication apps: Blackberry Messenger, KakoTalk, Line, WeChat, WhatsApp, Facebook Messenger, iMessage.
    • Note: there is a geographical network effect in the different types of communication apps!
  • Example in crypto world: Pizza costs 10,000BTC in the past. Today, it’s a fraction of 1BTC. What changed? Many people are using BTC, joining the network, accepting/using BTC. In the past, BTC was not accepted by many merchants. Once you receive BTC, you need to know how to spend it and convert it back to usable fiat. It was not very valuable (to the user in early days). Today, many companies accept BTC. Some people get their salary in BTC too! The value has risen, as a result of more people using it.

2. Lock-In

Lock-in is a relatively new concept compared to the rest of economics. This came with the evolution of information and technology, where we created information economics.
  • What’s that? Lock-in is a way to get people to stay on your system. There are many ways to do that:
    1. Make it difficult to leave the system (non-monetary ways)
    2. Keep upgrading the system so they have a reason to stay
    3. Make it expensive to change the system
  • Example in real world: Did you know, 8% of the world is still using WindowsXP? Why does that happen, when there are better systems that Microsoft created? Simple, it is expansive to upgrade to Windows 10. Imagine how much it would cost to get the Windows 10 licensing fee for 100 people in your business. There are also other things like the enterprise system being built on WindowsXP, there is no real need to change the system to Windows 10 and other knowledge specific to the OS.
  • Example in crypto world: Ethereum is working on becoming “the” platform for blockchain systems. You have the EEA and Consensys supporting participants and businesses in the network. Can you think of other ways Ethereum increasing their lock-in effect? (Hint: smart contract)

3. Schelling Point

Schelling Point is a phrase people enjoy throwing around. It’s created by this economist, Thomas Schelling. It is used commonly in game theory.
  • What’s that? With zero communication, it is the natural focal point people tend towards.
    • Here are 4 numbers. Which number will you choose? 584376, 230873, 10000, 238176. Probably “10000” because it is a natural comfortable number people tend towards.
  • Example in real world: If I say we are all going to meet at New York’s train station at 12pm, which train station comes to your mind? Probably Grand Central Station.
    • Or, you are in a new country and you don’t speak that language. There are 2 restaurants there, 1 with a long queue and 1 with zero people inside. Which restaurant will you choose? Most probably the longer queue.
  • Example in crypto world: Similarly, imagine you are a miner. You have a block and you can add to either LongChain or ShortChain. You’ll naturally tend towards the LongChain. Same idea.

4. Signaling Theory

This is quite common with marketing websites.
  • What’s that? Signaling theory is like playing a game of charades. You have the let the other person guess your card without saying the words.
  • Example in real world: You want to show your future employer that you are smart and capable for the job. What do you do? Probably get a good degree in a good university and be the leader in student unions, etc. You can signal your traits through actions and activities.
  • Example in crypto world: You can pay money for your company to be published on CCN, Forbes, Hackernoon, etc. They are signals to the public that your company is so innovative, these big publishers are interested to report on them. That’s how you can win trust of the public.

5. Game Theory

Arguably, the most talked about issue when it comes to economics of tokenisation.
  • What’s that? Looking at all the options and possible consequences of taking those actions, what will your choice be. It is about analysing a situation, to predict the next move.
  • Example in real world: Airbus and Boeing are the 2 airplanes manufacturer (duopoly). In the last 90s, they wanted to create a superjumbo jet. They came into the agreement to develop it together (half the cost of expenses), but Boeing kept delaying it. Why? We can use game theory to analyse why.
    • Boeing is in the lead for producing aircrafts then. They also have the largest planes then.
    • If this is successful, Boeing has to share the success with Airbus. That also means they are no longer the company with the largest plane.
    • Current situation: leader in industry (more than half of industry). Possible future situation: either superjumbo jet fails (lose all investment) or share the market with Airbus (half of the industry).
    • Naturally, you can see that Boeing should not develop this superjumbo jet. In fact, they should delay the development of it.
    • AND THAT IS WHAT THEY DID! In the end, Airbus had to develop the superjumbo jet on its own (super expensive & takes a longer time).
  • Example in crypto world: We look at bitcoin vs bit-torrent. They are both peer-2-peer in nature, and require nodes to be on the platform. Yet, bitcoin succeeded where bit-torrent failed – incentives to stay in the system. In Bitcoin’s case, nodes are incentivised to stay on the system as they can earn block rewards (current incentive) AND transaction fees (future incentive). Bit-torrent, however, does not have such incentive mechanism to encourage participants to stay on the network.
    • Bit-torrent situation: stay or leave the network (nothing happens). Bitcoin situation: stay on the network (rewarded with block reward & transaction fees), leave the network (nothing happens).
    • Naturally, people want to stay on the network in Bitcoin to earn some BTC.

6. Monetary Economics

This is applicable to currencies (e.g. regular cryptocurrencies and stable coins).

  • What’s that? Monetary economics is how money is managed and governed.
  • Example in real world: FED in USD controls how USD is managed. Bank of England controls how GBP is managed. MAS controls how SGD is managed. Pretty much ~10 smart people in the organisations decide how the currency is managed. (Centralised)
  • Example in crypto world: The difference is that it is mostly decentralised. That means, main decisions are embedded in code and baked into the crypto-project. Depending on the mechanism design, people in the network can vote to change certain system. It gives more power to the people. In crypto world, we have different ideas for the supply of token/digital money. Some like it capped, some like it uncapped. BTC has a cap at 21 million, while others don’t, like MakerDao.

7. Principal-Agent Model

This model is a lot more complicated, but the idea is simple.
  • What’s that? Skin in the game. To motivate a behaviour, the most direct way is for the person to have skin in the game. It’s like placing a bet against yourself. You’ll definitely do the best you can, to benefit yourself.
  • Example in real world: If your water bills is included in your rent, you’re likely to waste water and have a bath every day. But if you have to pay utilities based on what you use, you’re gonna take a shower once every 4 days. You’ve got skin in the game!
  • Example in crypto world: Proof-of-stake is a validation mechanism. Basically, if you want to validate a block, add your name + some money into the pot. Then, it’s like a lucky draw, and they will pick a name from the pot. That person gets to validate. Now, if I have to stake money to validate a block, do you think I will validate a “bad block” and risk MY money? No, because I will only do things that will benefit me. This is a way to enhance skin in the game!

Sources/Further Reading: