TPI’s Research Roundup is our semi-regular compilation of recent outside research of interest to tech policy nerds. If you’ve read a paper you think might be interesting to include in the next Roundup, feel free to send it to [email protected]
Hyunchul Kim and Gene Hyungjin Kim
Their question: Does bundling affect the cost of switching telecom providers?
Their answer: Analysis of Korean data suggests that bundling services increases switching costs, which they estimate to be about 65% of the monthly cost for the services. These switching costs can allow providers to leverage market dominance in one area into another.
Why does it matter? Bundling is common, making it important to understand its effects.
Apostolos Filippas and John Horton
Their question: How do people choose who to follow on social media?
Their answer: People wants to see good content, but also want other people to see their own content. Using data from econtwitter, the authors find that people use “attention bartering” to gain followers, where non-top users follow each other to gain audiences.
Why does it matter? How content competes for consumers’ scarce time is a key part of the attention economy.
Otto Kässi, Vili Lehdonvirta, and Fabian Stephany
Their question: How many people work on online platforms like Amazon Mechanical Turk?
Their answer: ˜10 million workers have ever worked an online job, and 2.7m have made at least $1000 or worked on ≥10 projects. They base their answer on data from 351 online work sites they consider to be “non-trivial” using Google Trends, Alexa ranking, and more.
Why does it matter? Online work is a new and growing segment of the labor market and we need to understand it in order to make good policy
Raviv Murciano-Goroff, Ran Zhuo & Shane Greenstein
Their question: How does non-priced open source server software cause mismeasurement in productivity and national economic statistics?
Their answer: The authors estimate that missing value for the open-source Apache and Nginx web servers using a massive data set of Internet Archive websites from over 200,000 organizations with more than 50 employees. They estimate that usage of Apache and Nginx generated $66M of value in server software in 2000, increasing to $125-315M in 2018, which was not captured in official data.
Why does it matter? National economic data do not generally capture zero-priced goods, such as open-source software. Sever software is just one type of open source software, and open source software is used widely. The large mismeasurement just from Apache and Nginx server software alone suggests that the total mismeasurement resulting just from open source software is much larger.
Not all authors are affiliated with TPI. We do not necessarily agree with everything, or even anything, in these papers, but find them interesting.