Token Engineering from an Economic Perspective

I have spent the last year working on a project with the goal of developing a tokenized ecosystem for the shared mobility sector. In the back of my head I knew that in order to come up with a proper solution I had to use the experience I gained during my work in the banking sector, the start up project, my former poker career and combine it with my education. I decided to do this in an academic environment at the University’s Research Institute and suggested to initiate a research stream of Token Engineering. Having a mathematical background with focus on economic modeling I had some ideas in mind how I could contribute to the research agenda. On October 1st, 2018 I started my engagement at the Cryptoeconomics Research Lab at the University of Economics in Vienna.

In one previous article Shermin Voshmgir, the director of our Research Lab, describes in detail the motivation behind the token engineering project and outlines how we aim to approach the research activities. She also includes a summary of one recent workshop we attended as well as an outlook of upcoming events. The most important takeaway from this description is that we will investigate to which extent existing economic and mathematical models can be used in tokenized economies. For more details I would like to refer you to Shermin’s article, this article however will concentrate on all underlying assumptions and the economic framework we use and further describe one model in more detail to demonstrate how we will conduct our research in the token engineering group.

Economic Framework

Economics as a scientific field has existed for at least 250 years now. It consists of many different schools and theories which are interrelated and interdependent. Many researchers and scientists from various economic schools have contributed to understanding human economic behavior. Capturing their theories in mathematical equations helped to analyze interconnections and mutual influences between different variables. Economics and Mathematics have already developed various models and approaches of formalizing economic motives and mannerisms of agents. Many results have been derived from these models and applied to real-world policy made though governments, regulators and institutions.

However, outcomes and solutions might contradict each other and heavily depend on the assumptions underlying the modelling framework. The popularity of all particular competing beliefs also hinges on the practical applicability in the real world and successful explanation of events observed in practice. An impressive overview map of different economic schools can be found here.

Impressive, huh? — General Guide of Schools of Economic Thought © Sergio A. Berumen

Having said this, a goal of the Research Lab is to investigate existing economic schools and understand the context and the motives driving them. The Token Engineering group will analyze similarities between societal modelling and cryptonetwork modelling, evaluate the potential of existing economic models to cryptographic networks and find out to what extent these economic models apply in cryptoeconomic applications.

Our take on this topic emphasizes the importance of combining different economic sub-fields as microeconomics, macroeconomics and institutional economics. These sub-fields of economics, are in our opinion a good framework to begin with because we believe they include tools to be suitable for understanding tokenized networks. Our approach builds on the assumption that tokenized systems and their stakeholders, tokens and governance structures resemble federal states consisting of laws, inhabitants, corporations, currencies and institutions. Most importantly, (microeconomic) behavior of actors is embedded into (institutional) restrictions and results in an influence of the (macroeconomic) development of the whole system.

To start, it’s good to list some of the properties of these fields:

• Microeconomics focuses on individuals. It studies particular markets & segments of the economy and their form and structure. Examples are consumer behavior, individual labor markets and the theory of firms.
• Macroeconomics focuses on aggregates. It studies the whole economy by looking at aggregate variables such as aggregate demand, national output and inflation.
• Institutional Economics: It takes into account that imperfections, irrationalities and information differences exist. It focuses on transactions, their costs and how institutional coordination can reduce them.

a beautiful slide

Therefore for understanding tokenized ecosystems we argue that it is helpful to combine the perspectives provided from the synergies of these three economic sub-fields.

In addition, the field of Monetary Economics covers one important aspect of tokenized systems as well — it investigates how the existence and the specific properties of a currency holds an economy together and how it influences the participants’ actions. Similar to currency and capital in our society, the token has an important role to play: it can serve as an incentive mechanism of the system. This native intrinsic control instrument is unprecedented in behavioral economic governance and makes tokenized ecosystems a very attractive tool for the design of decentralized networks. The token does this by combining properties of currencies and capital. This enables it to capture value creation within the network and thus will affect actions and mannerisms of network participants. You can read a comparative discussion of tokenized systems versus capitalized systems here.

The DSGE Model — an example

To demonstrate how we will conduct our research, we elected to start with an investigation of the Dynamic Stochastic General Equilibrium (DSGE) model. We chose this model for the following reasons:

• It combines Microeconomics and Macroeconomics in the form of microfoundations. This means, that the model specifies optimal decision rules of individuals interacting within the economy. Simply put, this model looks at actions, preferences and problems of individual participants in order to understand how their decisions will affect the global economy.
• Forms of these mathematical models are used by many central banks and play an important role in the formulation and communication of monetary policy. Most cryptonetworks use a form of monetary policy as well.
• It is used by the European Central Bank (ECB) to model the Eurozone. Since we build on the assumption, that a cryptographic system resembles nation states (or in this case a group of nations) it seemed plausible that we might find similarities
• It is dynamic; it includes time. This is done in the form of expectations about the future and by including intertemporal trade-offs into the decision making of individuals.
• It is stochastic; it includes uncertainty. However, actors are aware of the possibility of shocks and unknown outcomes
• It is a toolkit; DSGE models exist in the simplest forms but can be extended to very sophisticated models as well. This allows us to begin small and gradually include more and more elements to finetune the dynamics. We saw the potential that we might be able to identify selected tools to model cryptoeconomies.

The remainder of this section will summarize the most important elements. A fully detailed description of the model and its implications can be found here.

• Characteristics: The model is microfounded, it consists of representative actors who maximize their lifetime utility by choosing their behaviour and base their decisions on future expectations including intertemporal tradeoffs of utility.
• Actors: In the model described, the actors are households, firms and the government. Their goals are: maximize lifetime utility (for households), maximize lifetime profit (for firms) and steer the economy with respect to growth, inflation and cost of credit (government or monetary authority)

General structure of the modeled economy and their participants

• Structure of economy: This simple model assumes the following: There is only one final good in the economy (Y). The households (A) work for intermediate good producing firms (Bi) and receive wages (w) for their labor. Intermediate firms sell their intermediate good (Yi) to one final good producing firm. The final good producing firm (C ) assembles the intermediate goods to one final product. The household can buy this final good at a market clearing price (P).
• Motives and incentives of actors: 1) Households: They like consumption of the final good but dislike work. They want to maximize their subjective utility by deciding how much to work, how much to consume and how much to save for future consumption. 2) Intermediate firms: They want to maximize their profits. They have to compete for workers on the labor market thus increase wages if they want to attract workforce. They have to borrow money at the cost of the short-term interest rate. 3) Final good firm: They have to meet the market demand for the final good. In this model they do not have much decision power and only act reactionary. 4) Monetary authority: They want to secure growth and employment. They choose the short-term interest rate to steer the economy.

Based on this general framework, we subsequently investigate the household optimization problem and how it influences the economy. In mathematical form, the problem can be formulated like that:

(1): objective function with (2): budget constraints

The above equation (1) is the objective function of the problem. It describes that the household maximises (max) its expected (E) utility by choosing optimal paths for saving (B), consumption (C ), and hours of work (H). The term behind the Greek sigma ( Σ ) describes the utility of consumption (C ), and the term behind the integral ( ∫ ) describes the disutility of labor (H). This optimization problem is subject to budget constraints (2) which basically guarantee that households cannot consume more than they can afford, consumption has to be pre-financed by work.

So what do we hope to achieve? This model can help us to understand token systems by establishing connections to existing models. In the best case scenario we could find a model fully applicable in the crypto-world. More realistically we could use existing models and include adaptations and extensions to capture token dynamics. In any case we can utilize the same techniques, we might be able to apply similar formulas, we will find out what might work and what not and maybe we are lucky to find similarities in solutions and policy implications as well.

The similarities might be found on many different levels: in the architecture and general structure of the economy, the modeling and mechanics assumptions, the equations and mathematical formulas, the solutions and also the derived policy recommendations. Applying this to the crypto-world we are now equipped with many important questions we have to ask before designing or evaluation a network.

To continue our work we decided to take the following steps:

• For everyone interested in contribution to our scientific research, here is a working document where you can describe the scientific questions you are interested in, give hints in which direction the research could evolve or even suggest solutions.
• For everyone interested in contribution to practical implementations, here is a working document where you can describe your project and apply for a short presentation of your token model during one of our workshops.
• We will host a monthly token engineering workshop in Vienna beginning from November 13th this year. It will be the practical implementation of our research. In these regular meetups we will have the opportunity to present the progress of our work but also give all participants the chance to present their practical approach to the public and discuss their approach with us. Find the link to the meetup here and join the telegram group here.