Essays in Social Network and Migration
Author(s)
Zou, Haoqian
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Migration is essential for labor markets, fostering innovation, knowledge transfer, and economic resilience by efficiently allocating labor resources. However, US worker mobility has declined from 17% in the 1980s to 8.4% in 2020. Understanding migration determinants is essential for explaining this decline and informing policy. Beyond economic factors and amenities, networks can help reduce information asymmetries and overcome market inefficiencies. A new challenge is the rise of social media platforms like Facebook, Twitter, and LinkedIn, which have transformed how people think, communicate, and make decisions. This dissertation consists of three essays, each addressing different aspects of how social networks influence migration and job location choices, and how these network effects vary across different places and populations. The first essay examines how Facebook social networks between metropolitan areas affect the location choices of American workers. The results from the conditional logit model show that individuals tend to migrate to metropolitan areas with strong social connections to their current residence. This influence is particularly significant among disadvantaged groups in the labor market, such as younger adults, females, non-whites, and individuals without a college education. The second essay focuses on the role of alumni networks in the school-to-work migration of PhD graduates in non-academic fields. Using a comprehensive dataset of approximately 300,000 PhD graduates from 145 R1 universities, derived from LinkedIn resumes, the study finds that alumni in managerial positions facilitate migration by providing support to new graduates, whereas alumni in junior positions may limit migration due to competition effects. The third essay expands on the second essay by extending the sample from PhD graduates to all college graduates and deepening the analysis from the state level to the metropolitan level. Utilizing over 10 million LinkedIn resumes of graduates from 273 R1 and R2 universities, the study finds that the most substantial impacts are observed from alumni connections within the same discipline, gender, or cohort, particularly in metropolitan areas with less available information but other appealing attributes. The study also finds that the role of alumni networks is more important for demographic groups facing competitive disadvantages, and graduates from institutions with well-established alumni cultures benefit the most. The conclusions and contributions of this dissertation are multifaceted. These three essays use different social media data to measure contemporary social networks, demonstrating that both general networks and professional networks significantly influence migration choices. Moreover, the results find that the impact of group social networks varies across different populations and regions, highlighting the specific mechanisms through which social networks shape migration decisions. Although the results are primarily descriptive, this dissertation introduces additional network variables—Facebook social connections and alumni networks—to measure the connections between metropolitan areas, providing more options than just physical distance as considered in the literature.
Sponsor
Date
2024-07-16
Extent
Resource Type
Text
Resource Subtype
Dissertation