Tech Industry Jobs for Econ PhDs — Part 1: the Landscape

This series is meant to help Econ PhDs understand the tech industry job landscape and prepare for the job hunting process.

I’m currently a 5th year PhD in Economics at Stanford. At the end of my 4th year I decided to look for jobs in the tech sector only (i.e. not going into academia). During my job hunting process I realized there wasn’t a guide to the tech sector specifically for Econ PhDs and decided to fill in the gap.

This will be nowhere close to a comprehensive guide; some people might not agree with all the content; but I hope it can be a useful starting point for people who just decided to make the jump.

Links to other parts of the series:

Part 2 — Interview Prep

Part 3 — the Job Hunting Process I (when to apply, how to find positions, networking, CV)

Part 4 — the Job Hunting Process II (the interview process, timeline management, wage bargaining)

Tech industry job landscape

There is a whole spectrum of jobs Econ PhDs can do from the most research oriented to the most business oriented:

Research organizations in tech

Examples: Microsoft Research; parts of the Core Data Science team at Facebook; parts of the Business Economics team at Uber

Pros: great if you want to keep doing research and publishing papers
Cons: not very close to the business, which means issues with

• Career progression (e.g. the chance that a major tech firm’s next CEO is directly promoted from its econ research team is low)
• Job security (e.g. Uber’s Business Economics team suffered severely during its recent layoff)

External facing research functions

Examples: Zillow Research; Uber’s policy research team; LinkedIn’s media/marketing research team

The difference between this and the previous category is that:

• the goal of the previous job type is to produce scientifically rigorous, potentially publishable research papers, which may or may not be useful for the business per se
• whereas for this type, the goal is to use ‘research’ (broadly defined) to help the business with its publicity/marketing needs. For example, producing summary-stats or yoy growth numbers to write up short news pieces; work with external parties (gov organizations, media) to fulfill their requests

Internal facing research functions

Examples: parts of Core Data Science team at Facebook; parts of Business Economics team at Uber; economics or behavioral science at Google; some economists at Amazon

This type of ‘economist’ works to help the business, but sits in a centralized team (i.e. not directly attached to a product)
Things such a team can be working on include:

• litigation, i.e. in-house Econ consulting
• work with product teams to design/evaluate/improve products, e.g. designing ads auction; evaluating driver incentive scheme

Economist/Data Scientist working with a product

Examples: some Economists at Amazon, some Data Scientists at Google, Data Scientists at Uber, Data Scientists at Quora, some Research Scientists at Facebook

This can be further separated into two ‘styles’:

• those who work as ‘advisors’: the PMs/engineers decide what to do with a product, but they would come to you with a specific request (e.g. show me the causal effect of X; what metric should we use) and you fulfill it and present it
• those who are in the ‘driver seat’: you have direct influence on what analysis or experiments to run, what changes in the product to make

Data/Product Analyst, Machine Learning Engineer, etc.

These are different job categories that are also directly attached to a product:

• ‘Analysts’: building dashboards, tracking metrics
• ‘Data Scientists’: doing experiments, building machine learning models
• ‘Economists’: Data Scientists who have some toolkits not all data scientists have (e.g. causal inference, forecasting, IO modeling, etc.)
• ‘Machine Learning Engineers’: putting machine learning models to production; serving as bridge between software engineers and data scientists

Caveats

There are no clear boundaries between many of the above ‘types’: a Economist/Data Scientist who sits on a product team can also be writing down scientifically rigorous IO models to evaluate the effect of Surge; a Research Scientist at the Core Data Science team might also be working on improving the platform that product teams use to run experiments.

Different firms call different things differently. Google used to call its data scientists ‘quantitative analysts’. Zillow call its data scientist ‘applied scientists’ and analysts ‘data scientists’. Evaluated by its job content, at Facebook, some ‘product data scientist’ are in fact ‘analysts’, whereas some ‘research scientists’ at Core Data Science are actually ‘data scientists’.

Different firms allocate work content to different functions differently. At Quora, data scientists do everything from metric definition to experimental design to building ML models. Whereas at bigger firms, such as Facebook, product data scientists wouldn’t be designing experiments as they have a centralized experimental platform that everyone uses. At Uber, data scientists have lots of influence on product decisions which is not true at many bigger firms such as Google, Facebook, Amazon.

Which tech firms hire economists?

• Amazon: one of the largest private employers of economists. At Amazon, there are two types of positions for Economists: centralized (e.g. BizDev) and embedded (attached to a specific product, e.g. AWS, device, HR, scout, etc.). Economists at Amazon are more of an ‘advising role’ than ‘in the driver seat’, i.e. often the PM/data science come to Economists with a specific request (e.g. ‘what is the causal effect of X?’) and the economist solve the problem and present the result
• Google: the ones specifically relevant for economists include the Economics team and the Behavioral Science team; what an econ phd might also be interested in is its Data Science team. Google’s econ team — if having headcount — might go to the AEA to hire; but its ‘Data Science PhD university graduate’ positions apparently won’t be at the AEA as it’s not targeted at econ phd per se. Data scientists (used to be called quantitative analysts) at Google are also usually attached to a specific product (e.g. Youtube, Google Home, etc.) and also is an advising role (i.e. product/engineering come to data scientists with specific requests)
• Facebook: the relevant teams include the Core Data Science team, Novi (the digital wallet that holds Libra), Infrastructure, etc. There are a lot of different teams within Core Data Science (CDS): some are more research oriented (e.g. using Facebook’s data to conduct research such as gender studies), and some are more relevant to Facebook’s main business (e.g. helping to design Facebook’s experiment platform). Note that, orthogonal to the above, there are a large amount of ‘product data scientists’ at Facebook that are attached to specific products (e.g. search, marketplace, etc.), but it’s rare to see econ phds joining Facebook as product data scientists. There are a few teams outside of CDS that hire econ phds, e.g. Infrastructure, which is responsible for solving optimization problems such as how many servers to use; Novi, which solves problems related to the launching of the new digital currency Libra.
• Uber: there used to be a Business Economics team that does research but they were hit hard in the COVID layoff (most of it gone; the remaining teams plugged into Uber’s main business); there’s also Data Scientists that are attached to specific products (e.g. driver incentives, rider pricing, surge, matching, etc.). With the question at its heart being an Economic one , a lot of Uber’s Data Scientists are econ phds, and Data Scientists also have a lot of influence at Uber: instead of being an advisor that fulfills PMs/engineerings’ requests, Data Scientists are often directly involved in the product design process and/or driving changes in the product.
• Zillow: there are also an Economics team and product-attached Applied Scientists (at Zillow they call analysts ‘data scientists’ and data scientists ‘applied scientists). The econ team work on (1) using Zillow’s data to write short news pieces for publicity (2) doing economic forecasting (3) legal/lobbying, etc. which might not necessarily require a econ phd. The applied scientists, e.g. at Zillow Offers (Zillow’s iBuyer business, where they directly buy and sell residential real estate to profit), are using Machine Learning and any other useful toolkits (e.g. econ/finance/stats) to design the purchasing, fee setting, renovation, and resale policy/strategies. Similar to Google’s Data Science team, Applied Scientists are not necessarily econ phd (e.g. many are stats and physics phds), but is something an econ phd can be interested in
• Wayfair: very similar to Amazon (also two sided digital market) and also goes to NABETEC and/or AEA to hire economists
• LinkedIn: again can be split into teams that do ‘research’ for publicity/policy reasons, and roles that econ phds can take that are directly attached to its products
• Stripe: again contains a ‘research’ team that uses Stripe’s data to do research and roles that econ phds can take that are directly attached to its products; Stripe also goes to NABETEC and/or AEA at times
• There are a large number of other firms that also go to NABETEC and/or AEA; you can also see a list of firm that hires economists in Susan Athey and Michael Luca’s paper
• There are an even larger number of firms that hire Data Scientists. There’s indeed slightly different culture at different firms: some prefer people with industry experience, some prefer people with a PhD (or has roles meant for PhD graduates). For example, Google has a new grad position titled ‘PhD university graduates’ whereas Netflix is famous for preferring people with industry experience

References

Economists (and Economics) in Tech Companies