Tanja Saxell

Senior researcher,

PhD in economics, University of Helsinki, 2014

Visiting positions
NYU Stern School of Business, 10/2015

Stanford University, Economics Department, 2014-15

Contact information
VATT Institute for Economic Research
Arkadiankatu 7, P.O.Box 1279, FI-00101 Helsinki

+358403045574                                                                  tanja.saxell@gmail.com



My research interests are in 
empirical industrial organization and structural econometrics,  with a specific focus on healthcare markets. The goal of my current research is to study the efficiency of healthcare provision, and I aim to identify mechanisms leading to overuse of healthcare. More broadly, I develop methods to estimate demand for differentiated products with quality uncertainty. I am also interested in the economics of intellectual property rights, pharmaceutical innovation and competition.


Patient Mix and Physician Responses to New Prescription Drug Information: Evidence from Medicare Expansion (joint with Michael Dickstein and Marissa King)

Private Experience and Observational Learning in Pharmaceutical Demand

I introduce a structural model of demand for pharmaceuticals to analyze the interaction between private and social learning by physicians who do not know ex-ante how their patients respond to alternative drug treatments. Exploiting data on cholesterol drug prescriptions and physician-patient matches in a 10-year period, I study the impacts of policies that govern physicians' abilities to identify best treatments for their patients. I show that long-term treatment relationships improve the process of learning and lead to better treatment decisions in terms of efficiency by eliminating social learning from patients' previous doctors. I also find that equally good health care is not achieved with policies that compensates for lack of continuity in care by providing information on other physicians' past treatment choices.

Patent Duration, Breadth and Costly Imitation: Evidence from the U.S. Pharmaceutical Market (joint with Olena Izhak and Tuomas Takalo)

This paper investigates whether stronger (longer or broader) patents affect imitation and incentives to develop new, innovative drugs. We first present a simple model to formalize the predictions of the basic patent theory in our setup: first, extending the duration of patent protection increases generic entry prior to patent expiration and thus may provide low incentives for firms to develop new drug treatments; and second, broader patents decrease generic competition, hence enhancing the incentives for new drug development. We then explore the effects of stronger patents on early generic entry via Paragraph IV challenges in the US pharmaceutical industry where patent term extensions and adjustments compensate for delays in the commercialization, creating variation in patent duration across products. Our findings - based on both changes in patent law and the quasi-random assignment of patent applications to patent examiners - suggest that longer patents increase the probability of generic entry during patent protection and thus may distort the development of new drugs with long commercialization lags. We also document evidence which is consistent with pharmaceutical patents being too broad from the welfare point of view.

Experimentation and Learning in Pharmaceutical Demand

Buyers do not necessarily observe the quality of products in advance but may learn them through consumption. Such uncertainty creates incentives to experiment with new products to gain more information. At the same time, uncertainty makes risk-averse buyers reluctant to try new, less well-known products. Still, traditional demand models ignore uncertainty and learning. In this paper, I estimate a dynamic model of demand for cholesterol drugs that allows for learning through experimentation. The results suggest the average health effects of cholesterol drugs are heterogeneous across patients which creates uncertainty to medical decision-making. My analysis also identifies drugs that induce higher exit rates from the cholesterol drug therapy. I find that uncertainty and risk aversion make physicians unwilling to try new drug treatments for their patients. My results suggest that if doctors became more willing to experiment with their treatment choices, both the process of learning and efficiency in drug choices would improve.

Do Stronger Patents Protect Against Competition? Evidence from the Finnish Pharmaceutical Industry