{"id":2,"date":"2013-04-08T23:37:07","date_gmt":"2013-04-08T23:37:07","guid":{"rendered":"http:\/\/site.hannesullrich.com\/?page_id=2"},"modified":"2026-10-09T14:47:11","modified_gmt":"2026-10-09T14:47:11","slug":"sample-page","status":"publish","type":"page","link":"https:\/\/www.hannesullrich.com\/","title":{"rendered":"Research"},"content":{"rendered":"\n

Research Associate, DIW Berlin
Associate Professor of Health Economics, University of Copenhagen<\/p>\n\n\n\n

Research<\/h1>\n\n\n\n

Applied Microeconomics \u00b7 Digitalization \u00b7 Empirical Industrial Organization \u00b7 Health Care & Policy<\/p>\n\n\n\n

Publications<\/strong><\/h3>\n\n\n\n

Provider effects in antibiotic prescribing: Evidence from physician exits<\/a> (Open Access), with Shan Huang<\/a>, Journal of Human Resources<\/em>, Vol. 61(4), pp. 1159-1191, 2026.<\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 38\/2024<\/a>. In the press: Dagens Medicin<\/a> (in Danish)<\/p>\n\n\n\n

Complementarities between algorithmic and human decision-making: The case of antibiotic prescribing<\/a> (Open Access), with Michael A. Ribers<\/a>, Quantitative Marketing and Economics<\/em>, Vol. 22, pp. 445-483, 2024. Runner-Up, Dick Wittink Prize 2025<\/a>.<\/p>\n\n\n\n

Supersedes “Machine learning and physician prescribing: a path to reduced antibiotic use” and “Battling antibiotic resistance: can machine learning improve prescribing?”<\/p>\n\n\n\n

Columns: <\/span>DIW Weekly Report 19\/2019<\/a> (in English), <\/span>DIW Wochenbericht 19\/2019<\/a> and <\/span>Oekonomenstimme<\/a> (in German). AI-generated podcast<\/a> (by Google’s NotebookLM)<\/span>.<\/p>\n\n\n\n

Assessing the value of data for prediction policies: The case of antibiotic prescribing<\/a> (Open Access), with Shan Huang<\/a> and Michael A. Ribers<\/a>, Economics Letters<\/em>, Vol. 213,  110360, 2022.<\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 13-14\/2021<\/a>.<\/p>\n\n\n\n

Career prospects and effort incentives: evidence from professional soccer<\/a>, with Jeanine Mikl\u00f3s-Thal<\/a>, Management Science<\/em>, Vol. 62(6), pp. 1645-1667, 2016. Last working paper version<\/a>. Online Appendix: Download<\/a>. Reprinted in Learning in Labour Markets<\/em><\/a>, edited by Michael Waldman, Edward Elgar Publishing, Cheltenham, 2017.<\/p>\n\n\n\n

Column (in German): ZEW News 9\/2009<\/a>.<\/p>\n\n\n\n

Belief precision and effort incentives in promotion contests<\/a>, with Jeanine Mikl\u00f3s-Thal<\/a>, Economic Journal<\/em>, Vol. 125(589), pp. 1952-1963, 2015. Last working paper version<\/a>.<\/p>\n\n\n\n

Regulation of pharmaceutical prices: evidence from a reference price reform in Denmark<\/a>, with Ulrich Kaiser<\/a>, Susan Mendez<\/a>, and Thomas R\u00f8nde<\/a>, Journal of Health Economics<\/em>, Vol. 36, pp. 174-187, 2014. Last working paper version<\/a>.<\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 14\/2014<\/a>.<\/p>\n\n\n\n

Working Papers<\/strong><\/h3>\n\n\n\n

Machine predictions and human decisions with variation in payoffs and skill: the case of antibiotic prescribing<\/a>, with Michael A. Ribers<\/a>, Berlin School of Economics Discussion Paper Nr. 27 (download pdf<\/a>). <\/p>\n\n\n\n

Revise and resubmit at Journal of the European Economic Association<\/em>.<\/p>\n\n\n\n

We analyze how machine learning predictions can improve antibiotic prescribing in the face of increasing antibiotic resistance. Estimating a binary antibiotic treatment choice model, we find substantial variation in physicians\u2019 diagnostic skill and prescribing preferences, and show that providing machine learning predictions of bacterial infections can improve prescribing efficiency. However, reducing overall antibiotic use also requires appropriate physician incentives, highlighting potential misalignment between individual and social objectives.<\/p>\n\n\n\n

AI adoption by human experts: evidence from primary care physicians<\/a>, with Shan Huang<\/a> and Renke Schmacker<\/a>, DIW Discussion Paper Nr. 2168 (download pdf<\/a>). <\/p>\n\n\n\n

Revise and resubmit at Economic Journal<\/em>. <\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 38\/2026<\/a>. Research brief<\/a> at the National Center for AI in Society (CAISA).\u00a0<\/p>\n\n\n\n

We study how physicians respond to AI-generated diagnostic information compared with an established diagnostic test. In a nationwide experiment with Danish primary care physicians, we find substantially less belief updating in response to AI, frequent non-adoption, and systematic errors in processing AI-generated information. These frictions may increase antibiotic prescribing, highlighting the importance of training and information design for effective AI implementation.<\/p>\n\n\n\n

Off-platform tracking and data externalities: evidence from Facebook<\/a>, with Luis Aguiar<\/a>, Christian Peukert<\/a>, and Maximilian Schaefer<\/a>.<\/p>\n\n\n\n

Replaces earlier working paper Facebook shadow profiles<\/a> (DIW Discussion Paper Nr. 1998, download pdf<\/a>). <\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 29-30\/2022<\/a>. Media coverage in Tagesspiegel<\/a>, Der Standard<\/a>, Kronenzeitung<\/a>, Heise<\/a>, PC-Welt<\/a>, BR<\/a>, and Computerwoche<\/a> podcast (in German). <\/p>\n\n\n\n

We study whether platforms\u2019 off-platform tracking generates data externalities by allowing them to infer characteristics of individuals who do not disclose them. Using clickstream data, we show that Facebook can predict demographic characteristics of non-users from their browsing behavior using models trained on its users. While the GDPR substantially reduced Facebook\u2019s tracking ability, prediction accuracy declined much less, suggesting that restrictions on data collection may leave inference-based privacy risks unresolved.<\/p>\n\n\n\n

Returns to data: evidence from web tracking<\/a>, with Luis Aguiar<\/a>, Tomaso Duso<\/a>, Jonas Hannane<\/a>, and Christian Peukert<\/a>, DIW Discussion Paper Nr. 2091 (download pdf<\/a>).<\/p>\n\n\n\n

Column (in German): DIW Wochenbericht 27\/2023<\/a>. Media coverage in taz<\/a>, Table.Media<\/a>, finanzen.net<\/a>, and ntv<\/a> (in German). AI-generated podcast<\/a> (by Google’s NotebookLM)<\/span>. <\/p>\n\n\n\n

We study how the size and scope of web browsing data affect the quality of consumer profiling, for example for targeted advertising. We find decreasing returns to additional users and websites, but substantial gains from combining browsing and demographic data, favoring large platforms such as Google, Facebook, and Amazon. Data-sharing provisions may reduce this advantage and help smaller firms compete in consumer profiling.<\/p>\n\n\n\n

Assessing the impact of payment card fee regulation<\/a>, with \u00d6zlem Bedre-Defolie<\/a> and Minjae Song<\/a><\/p>\n\n\n\n

We analyze the effects of a hypothetical payment card fee regulation on bank profits, consumer welfare, and merchant welfare. We model consumers’ and merchants’ bank choices for debit card services, cardholders’ demand for card usage (conditional on bank choice), and how banks account for these in setting card fees to their customers. To estimate the model, we use bank-level data and survey data from the Norwegian debit card scheme, BankAxept. We conduct counterfactual exercises to analyze the effects of interchange fee regulations in the debit card scheme.<\/p>\n\n\n\n

A note on regressions with interval data on a regressor<\/a>, with Daniel Cerquera<\/a> and Francois Laisney<\/a>. Online Appendix: Download<\/a>.<\/p>\n\n\n\n

Former title: Considerations on partially identified regression models<\/a>, Centre for European Economic Research Discussion Paper 2012-024, BETA (Bureau d’Economie Th\u00e9orique et Appliqu\u00e9e) Working Paper No. 2012-07, Online Appendix: Download<\/a><\/p>\n\n\n\n

Consumer welfare and unobserved heterogeneity in discrete choice models: the value of Alpine road tunnels<\/a>, with Daniel Cerquera<\/a>, Centre for European Economic Research Discussion Paper 2010-095.<\/p>\n\n\n\n

Work in progress<\/strong><\/h3>\n\n\n\n

Career concerns and managerial risk taking: evidence from professional football, with Paul Bose<\/a> and Florian Sch\u00fctt<\/a><\/p>\n\n\n\n

Competitive pressure and antibiotic prescribing in primary care, with Temulun Borjigen<\/a><\/p>\n\n\n\n

The causal effect of antibiotic prescribing on population antibiotic resistance, with Shan Huang<\/a>, Michael Allan Ribers<\/a>, Barbara Juliane Holzknecht<\/a>, Jonas Bredtoft Boel<\/a>, Jette Brommann Kornum<\/a>, and Michael Pedersen<\/a><\/p>\n\n\n\n

My research repositories and social media profiles<\/strong><\/h3>\n\n\n\n

SSRN<\/a>
Google Scholar<\/a>
ORCID<\/a>
arXiv<\/a>
Bluesky<\/a>
LinkedIn<\/a>
Mastodon<\/a><\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Research Associate, DIW BerlinAssociate Professor of Health Economics, University of Copenhagen Research Applied Microeconomics \u00b7 Digitalization \u00b7 Empirical Industrial Organization \u00b7 Health Care & Policy Publications Provider effects in antibiotic prescribing: Evidence from physician exits (Open Access), with Shan Huang, Journal of Human Resources, Vol. 61(4), pp. 1159-1191, 2026. Column (in German): DIW Wochenbericht 38\/2024. In […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"page-no-title","meta":{"footnotes":""},"class_list":["post-2","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/pages\/2","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2"}],"version-history":[{"count":308,"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/pages\/2\/revisions"}],"predecessor-version":[{"id":871,"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=\/wp\/v2\/pages\/2\/revisions\/871"}],"wp:attachment":[{"href":"https:\/\/www.hannesullrich.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}