Hello, welcome to my website! 

My name is Yu Zhao (赵钰). 

I am a doctoral candidate in Wharton's Marketing Department and previously served as a visiting researcher at Indeed, Inc. within the Marketing Analytics Lab.

My research focuses on topics in the digital economy, privacy, and the economics of information. I investigate how the presence or absence of information impacts the digital economy, influencing both firm outcomes and consumer decisions. 

I use economic and statistical modeling and apply causal inference methods in my research. I apply computational linguistics methods to derive insights from textual (e.g., online search query) data. 

Click here for my CV.

Job market paper

with Pinar Yildirim, Isaac Dinner and Joseph Zucker

How does the presence of brand information influence the return to its marketing efforts? We examine this question utilizing a comprehensive data set from a major US job search platform. On the analyzed platform, brands can provide additional company information by claiming a company page and populating it with job- and firm-specific information. We employ matching and staggered DID methods to causally identify the effects of a firm’s information provision through claiming own company page. Results show that firms see increased jobseeker attention, with number of applications increasing by 3-5% after claiming their company pages. The effects of the company page are more pronounced on the less experienced applicants in the market, who may benefit more from information. Similarly, effects are higher for the firms for which there were less information on the platform before the company starts to manage its page. These findings are consistent with the idea that providing information makes a difference where it is lacking.

Working Papers

with Pinar Yildirim and Pradeep Chintagunta

Major revision at Management Science

How do privacy regulations in the market impact online search for products and information? This paper investigates the impact of the General Data Protection Regulation (GDPR for short) on consumers' online browsing and search behavior using consumer panels from four countries, United Kingdom, Spain, United States, and Brazil. We utilize recent advances in the synthetic control literature and apply the generalized synthetic control and matrix completion methods for causal inference. In terms of consumer browsing, we find that after GDPR, consumers have increased online activities: the average EU user browsed 16% additional domains compared to non-EU counterparts. As for online search, we find that the number of search terms submitted increased by 5.4% for EU panelists relative to their non-EU peers after GDPR, consistent with the idea of higher information friction. When searching for products, EU-panelists spent 9.7% more time browsing products, viewed 6.1% additional products in 3.4% more unique e-commerce sites relative to non-EU panelists. However, we do not find evidence of increases in the number of transactions. We also document decline in domain traffic through emails, and explore changes in online traffic through different marketing channels. Overall, the post-GDPR online environment may be more difficult for EU consumers to navigate through.

Preparing for revision submission (second round).

A conceptual framework on how privacy policies impact consumer's interactions with firms.

After enforcement of a privacy policy that allows consumers to easily opt-out from the firm's data collection and tracking, we have noticed a decline in a site's traffic resulting from direct marketing communications, such as emails. This decline is particularly evident among individuals who are on the margin of their interests in the website.

with Pinar Yildirim

Pay transparency laws, which recently went into effect in states like Colorado and New York, mandate employers to disclose relevant wage information to applicants in job advertisements. These new regulations affect the information environment in online labor markets, but how do they impact the posted wages, effectiveness of job ads, and the search outcomes? In this paper, utilizing a proprietary data set from a major US job platform, we document the effects of the CO and NYC wage transparency regulations. Reduced-form estimates indicate that, following the policy, employers disclosed wider wage ranges in job ads, possibly reducing the informativeness of ads. Analysis also indicates that the changes in posted wage ranges are heterogeneous across jobs with different demographic compositions. Linking the job ad data with data from the Bureau of Labor Statistics, we find that predominantly female-oriented jobs disclosed wider wage ranges in job ads, but the posted lowest wages declined after the mandate. These findings are consistent with the hypothesis that pre-policy, strategic disclosure of information in these job ads were a means to maintaining bargaining power of the employers.

with Ada Aka 

How much do consumers value their personal information? We answer this question by investigating consumers' online purchases following exogenous data breach events at e-commerce sites. Event-study analysis shows that consumers respond negatively to data breaches, while the declines are short-lived. Further analysis show that consumers respond more negatively towards data breaches involving financial information, and those data breaches with later notifications. These findings highlight the importance of timely communications between consumers and firms regarding privacy issues. We conduct experiments to causally examine whether and what types of information are more valued by consumers in the firms' notification of data breaches.

Preliminary analysis complete, experiments in progress.

Conference Presentations

Invited Talks