What I Do.
Since I’m semi-frequently being asked about what my work entails, here is a short explainer of some of the major themes in my work put in context.
Industrial engineer turned economist
On my Linkedin page, and in most of my other self descriptions, I call myself ”industrial engineer-turned-economist” or some variant thereof.
Both professions deal with the deeply embedded mathematical layer of the economy, from two very different perspectives: industrial engineers operate inside companies, economists operate inbetween companies. It turns out that this combination is quite a rare one, and I am one of the few people doing both.
It also translates into a very unique perspective I have on economics in general and my focus, the (micro)economics of technological innovation, in particular.
“I design economic engines"
In a former life used to work in real industrial production environments, dealing with machine scheduling, material flow, routing, queueing, and other typical production and supply chain matters.
In 1995, at the dawn of the Internet age, I moved to the U.S., to two hotspots of the new era: Illinois (home of the web browser) and Berkeley (front row view of the dot-com boom and bust). And I was impressed enough with this new development that I decided to switch sides not once but twice: first from understanding companies to understanding markets, and second from the industrial economy to the information economy. In many digital realms, I was there from the very beginning.
But what ultimately sets me apart from most economists (outside maybe the Stanford OR group) is that I still view the economy from an engineering perspective. This is what my “I design economic engines" tagline is about.
Markets, auctions, matching algorithms, even elections and polls, are economic engines in the sense that they are abstractions which allocate scarce resources to achieve productive ends. How good they are at their job is ultimately the consequence of a myriad of design decisions.
Two of the better known engines, markets and auctions, have been the focal point of economic emerging design fields which have recently been awarded Nobel Prizes, as have the two most important economic design tools, game theory and causal inference. These are the tools I use, on a conceptional or formal level, every day.
UC Berkeley was at the time I was there an epicenter of the “empirical revolution” in economics, so I translate both theory and empirics to an operational environment.
Matching and filtering
My own focus is typically on matching markets — the economic engines behind most online platforms — and on the less heralded funnels, filters, and tournaments — the engines behind every selection process from hiring the best talent to picking the most promising startups.
I also — this is where my engineering background comes in — spend time on concatenating multiple economic engines. In many ways this is not different than putting different machines on a shop floor, except these engines are purely written in code and produce value chains like sales outreach funnels or supply networks.
Most of the time I work with early stage tech startups, so I usually help setting things up from zero to full automation. This also means I do a lot of the groundwork around setting up value chains. Plus I inevitably do a lot of all the other stuff than needs to get done in early-stage startups.
Pricing and pricefinding
Pricing used to be a comparatively simple and staid business. In the pre-Internet economy, the boilerplate economic exchange was the delivery of a physical product, a car or a breadmaker, for a single payment. Installment pay and leasing schemes were already considered sophisticated.
This changed drastically with the emergence of digital products and digital commercialization channels. Pricing strategy has become intricate and sophisticated, and a design task with dozens of interlocking parts, and an overall shift from pricesetting by committee to dynamic, automated pricing. The pricing engine of an e-commerce retailer like Amazon is an engineering marvel.
I do a lot of pricing strategy in the course of my work. It usually means discovering the touchpoints where the customer can recognize the value of the goods and services provided, and finding opportunities to learn about willingness to pay.
One of the biggest surprises entrepreneurs usually encounter is to learn that the demand curve from microeconomics class does not exist, but has to be painstakingly constructed by trial and error out of many observations in the market.
The only feasible way to start pricefinding is to derive a plausible starting point, and to fine-tune it. At the upper level this is what economics Ph.D.’s do for tech companies. I help startups take the steps to reach that level.
Ultimately, like with all engines, finding the best performance curves is to combine the right instincts with the right formal approach.
Technology scouting
I have a very long and pretty successful career discovering and assessing the economic potential of new technologies, right from the days when as students we urged the University of Illinois to patent a herky-jerky video streaming protocol that years later morphed into YouTube.
I’ve been involved in machine learning research since the early 1990s, and both my master’s thesis and dissertation contain machine learning components. In this I was certainly ahead of the economists by some twenty years. I’ve worked with decentralized systems since 2011, and my economic knowledge about governance mechanisms reaches further back.
This is why I like to work with technical founders and product-centric startups who understand the need to also engage with the market. If I am convinced of the economic potential of the technology and the follow-thru of the team, I am happy to support. Most of the startups I work with produce some kind of knowledge or information good and operate in the digital world, but this is not a requirement.
Ways to engage
Sometimes I am called in for a defined work package like a feasibility assessment for an emerging technology (I have a bunch of origin stories on that), but the engagements I prefer the most are the ones that require building trust because I get to learn about — and help solve — the things that don’t work.
This requires a willingness to open up and trust my judgment, and for the right startup I am happy to find the right arrangement to make a cooperation work. I am looking for startups that can operate on a very high clock speed and I am willing and able to do the same.
Oliver Beige is an economist (Ph.D. Berkeley) and industrial engineer (MSIE Karlsruhe, MBA Illinois), who gained most of his lifetime value from spending time in the southwest German technology belt, Silicon Valley, and the then-burgeoning Berlin startup scene. He’s worked for a few big companies (Mercedes, SAP, Bosch) and a few smaller ones, and is now mostly advising startups and tech companies on all kinds of knotty decision problems.
