Putting the design in mechanism design
Economic design fields — auction design, market design, mechanism design — have become success stories for economics, especially for their linkage to Silicon Valley tech companies. But economic education still lacks far behind when it comes to teaching design fundamentals. We try to change that.
The story begins about a year ago, when Cem F Dagdelen, founder of Curve Labs, and I met up for coffee. One of the ideas we tinkered with was how to teach economics “the Bauhaus way”. This idea triggered a year-long deep dive into the various approaches and philosophies to teach design — both with and without connection to economics.
A little backgrounder to motivate this talk: the story of Armando Ortega-Reichert. Armando was a graduate student in Industrial Engineering at Stanford in the 1960s. This is his dissertation, which he didn’t get published because he couldn’t afford the fees. He went back to Mexico, had a distinguished career in the civil service, retired, and is now in his eighties.
He would’ve faded completely into obscurity had his dissertation not popped up in the papers of at least six guys who would go on to win the Nobel Prize. We got all of them: auctions, markets, mechanism design. Today auction theory is a trillion dollar industry. In large part built on Armando’s work. Stanford should at least name a building after him.
But for all its successes. Teaching economics is still as far away from teaching design as one could imagine.
The same problem pops up on our side of the spectrum. If you look beyond the obvious reasons why cryptoeconomic systems fail, there’s a huge number where the whole conceptual basis never really worked out. This was a major reason why we started looking for new ways to convey economic design concepts.
Let’s do a very quick run-through of all the five-plus-one inspirations we looked at and used as sources for our concept.
First: the Bauhaus Vorkurs, the preliminary course all Bauhaus students had to take. For us this set the template: the idea that you develop a thorough conceptual understanding before you delve into the formal parts, while currently, conceptual understanding is considered optional.
Especially in the way it evolved over time from a purely philosophical to increasingly material, even materialistic focus. The simple reason: the Bauhaus needed to make money to survive.
Second, and this is the one that most of you should be most familiar with: the Bloomington School around Elinor and Vincent Ostrom, of Governing the Commons fame. One of the key things that sets the Bloomington school apart is the focus on design principles from the very beginning.
Third, the Carnegie School around Herb Simon. This is very much the epicenter of computational economics in the fifties and sixties. Eight Nobel Laureates and five Turing Award winners from that era, and Herb Simon as certainly the only person who founded a business school, an economics program, a computer science department, and a design school.
Fourth, the Stanford Economic Engineering tradition, which I mentioned before. This also includes Ken Arrow and some other famous names. The important thing here is how most of this originated outside the econ department.
And ultimately, Christopher Alexander’s Pattern Language. I’ll talk a bit more about this in a second. Mostly because we realized that this is the approach we want to follow. In part because it has already been successfully implemented in software development, as design patterns, even more successfully than in architecture where it comes from. The other one is that it is a very intuitive, self-guided approach. The polar opposite of economics.
But to quickly finish this off, the bonus inspiration is Brian Eno and Peter Schmidt’s Oblique Strategies, a card set of creativity triggers. The main reason to include them here is simply because we love the format.
If we put them all together, they line up wonderfully along this conceptual and mindset gap between design and economics. These were the main sources from which we extracted the expert knowledge for the project.
Why we want to use patterns as the fundamental information carrier. Here is the official definition by Chris Alexander. One major attraction is the last sentence: “brings life to a building.” In the same sense, we want to use this to design inhabitable economic spaces. Also, this point in the small print: “A pattern describes the core solution to a problem in such a way that you can use the solution a million times over without ever doing it the same way twice.” This already encapsulate the idea of repetition and variation we want to drive home.
So we used this to extract our own definition of a pattern which we plan to use to find and express expert knowledge.
Where are we now with this? For us, the pure research phase is over. We have about a semester’s worth of patterns, in a variety of shapes, and the first idea of a structure. So for us, it’s time to go public and also move into the experimentation phase, to see what sticks.
Just to give a first glimpse of the structure we came up with: the engine room. These are all economic engines, if you want. Allocation mechanisms and governance mechanisms. There’s about a dozen of those: selection filters, network flow, load balancers, and so on.
But we also want to go one layer down, to the constituent parts.
After all this wind-up, here are some teasers.
The fundamental pattern for a decentralized system.
One of the most frequently implemented sequence of activities.
No matter if it’s a Greek agora, a modern supermarket, or an online finance platform, all markets are made up of the same constituent structures.
And ultimately, we also want to incorporate as many patterns from famous people as possible. Herb Simon.
Joseph Schumpeter’s fundamental entrepreneurial conflict.
All of this just as a teaser of what’s to come. If I piqued your interest, if you’re curious, if you might even want to contribute, feel free to get in touch. There’ll be more soon. Thank you.
This is a draft version of my talk given at Funding the Commons Berlin September 2023. Much more about the patterns, the inspirations, intuitive economics, and future developments from now on at econpatterns.com.
