Marcel Garz

We’re happy to announce that the next media bias workshop will take place on March 8 and 9, 2023, at ifo Institute for Economic Research in Munich. Julia Cagé and Theresa Kuchler will deliver keynote speeches. The workshop is organized by the Research Network Economics of Media Bias. The program can be downloaded here.

In a joint project entitled “Competition, multimarket contact, and quality of local newspaper coverage”, Mart Ots and I investigate the role of competition between media outlets for news quality. We use a combination of content analysis by human coders and machine-learning techniques to construct a large-scale measure of quality of Swedish newspaper coverage. We then investigate how news quality is related to different forms of market structure and market overlaps between newspaper companies, using plausibly exogenous changes in multimarket contact for causal inference. The project is funded by the Swedish Competition Authority over three years between 2023 and 2025.

We’re happy to announce that the next media bias workshop will take place on March 31 and April 1, 2022, at Berlin Social Science Center. Alois Stutzer will deliver the keynote speech. The program can be downloaded here.

I joined a group of researchers that investigates media bias from a highly interdisciplinary perspective. The media bias group develops systems and large-scale datasets to automatically detect biased or unbalanced coverage, using methods from disciplines such as computer science, linguistics, psychology, and economics. The groups also devises new ways to visualize data, in an effort to better understand what types of bias appear in which context.

In a recent study, Greg Martin and I investigate how news about the economy influences voting decisions. We isolate the effect of the information environment from the effect of change in the underlying economic conditions themselves, by taking advantage of left-digit bias. We show that unemployment figures crossing a round-number “milestone” causes a discontinuous increase in the amount of media coverage devoted to unemployment conditions, and use this discontinuity to estimate the effect of attention to unemployment news on voting, holding constant the actual economic conditions on the ground. Milestone effects on incumbent US Governor vote shares are large and notably asymmetric: bad milestone events hurt roughly twice as much as good milestone events help.