Technology Policy Staff Networks at Top Tech Companies in the United States

I am taking the second half of the course, but am not doing this project.

Summary of project:

Using social network analysis, this project seeks to understand the social network of policy staff at the top ten technology companies in the United States: Apple, Microsoft, Alphabet, IBM, Intel, Cisco, Oracle, Amazon, Facebook, and Qualcomm.[1] These companies are influential in terms of technology policy, but also in arenas like tax, immigration, and employment policy. This project will look at how policy staff at these companies are connected to each other by school and former workplaces. What might those connections mean for who gets hired, policy positions, and policy outcomes?

Background:

As technology companies have become larger and their products more ubiquitous, there are numerous policy-relevant questions about the role of technology in society, such as who should technology companies be able to sell ads to during elections[2], who owns users’ data[3], and how do we ensure users’ privacy? Legislation and regulation on each of these topics is shaped by government staffers, who are influenced by lobbyists and policy staff from technology companies. Thus, it’s useful to understand who the staff are in policy jobs at technology companies, and their relationships with each other and past employment in government.

As far as I know, there is no social network analysis looking at this topic, with the exception of research by Miranda Bogen, which partially inspired this proposal.

Research questions:

Do policy and communications staff at top technology companies come from the same schools (i.e. Stanford) or the same former workplaces (i.e. the US Digital Service)?

Is the “revolving door” between technology companies and government more pronounced for any of the companies on the list?

Has the number of staff switching from technology to government, and vice versa, changed from the Obama era to the Trump era?

Hypothesis

I hypothesize that policy and communications staff share alma maters, in particular Stanford, and past government technology policy workplaces, like the U.S. Digital Service, Federal Communications Commission, and White House Office of Science and Technology Policy. I would also hypothesize that Google has the largest revolving door, as Miranda showed that a number of Google employees were appointed to work in the White House.

Plan for data collection and methodology

This would involve creating a new data set of public policy staff at the top ten technology companies. This data set would include information on each individual’s current workplace, past workplace since 2008 (the beginning of the Obama administration), university attended, gender, and level of seniority. I would gather staff information through company websites, media reports, and publically available information on opensecrets.org and LinkedIn.

Then, I would either create two two-mode data sets, one with companies and individuals, and one with companies and schools. For both data sets, years worked at or years attended would be an attribute. Then, I would look at where individuals overlapped, both at school and at work. Alternatively, I could create a valued data set, with the numbers indicating number of years at a shared workplace or school, and look at overlap that way.

Once I had a data set, and could look at overlaps, I would analyze the network for cliques, factions, directionality, and centrality measures. Do people come from the same schools and past workplaces? Are there a few people who are particularly well-connected, or is the network very dense? Do any companies have stronger “revolving doors” than others, and has that changed from administration to administration? Lastly, are there patterns that could indicate who might be influential in these roles in the future?

Challenges and limitations

The first major challenge is which companies to include on the list. I drew my list from the Forbes 2000 of largest companies by revenue, but there might be another better way to look at top companies.

The second, and larger, challenge is figuring out who all the policy staff are at each company. Each company has a policy department, but often staff do policy-relevant work, but don’t have “policy” in their title. So, from the outside, it might be hard to determine who is on the policy team. In an ideal situation, I would be able to do interviews with each company, and determine who was on the team. Otherwise, the data set would likely be incomplete.

Conclusions and potential future scope

This analysis would be very useful to technology companies looking to find the most influential policy team members, helpful for government looking to hire top technology policy talent, and equally useful to lobbying watchdogs and journalists to better understand and report on networks of power in the growing technology industry.

In the future, this analysis could be expanded to look at overlaps between the top 50 companies in the technology and telecommunications sectors. As large technology companies reach potential monopoly status, they will likely tap staff and draw on lobbying techniques from the telecommunications sector, who have relevant experience.

---

[1] https://www.forbes.com/global2000/list/3/#country:United%20States

[2] https://www.nytimes.com/2017/10/19/us/politics/facebook-google-russia-meddling-disclosure.html

[3] https://www.washingtonpost.com/news/the-switch/wp/2017/09/20/the-single-most-depressing-thing-about-the-equifax-breach/?utm_term=.991dac463e70