Imperfect Information in Health Care Markets
News
- Exam dates are Feb 4, 14:30 (100 Hörsaal VIII) and March 17, 10:00 (100 Aula 2 100/01/Aula 2). Registration is open on KLIPS. We will try to grade the exams of the first date before the registration deadline of the second but we cannot guarantee that we will manage (we managed in previous years).
- Note that there will be a first exercise session on October 15. This exercise session is dedicated to repeating some mathematical concepts used in this course.
Course materials
Please note that I do not use ILIAS because (i) I did not agree with certain usage conditions concerning tracking and data utilization and (ii) I want to make the material I create publicly available. The material below might get updated over the course of the term.
- Screencasts
- derivation Rothschild-Stiglitz equilibrium
- how to get descriptive statistics and do regressions in LibreOffice Calc
- Data sets and instructions for case studies:
- exercises (as pdf) and datasets for the empirical lectures are here available for download
- some notes on how to work on the longterm care insurance case study and the deductibles case study (in Julia)
- An old exam is available.
- material created by other people:
- a video on (adverse) selection
- Notes
- notes (book project – not all parts are relevant!; password announced in lecture)
Lecture times:
- lecture: Wednesday, 10:00-11:30 in 100/U1/HS V
- exercise classes: Wednesday, 14:00-15:30 100/U1/HS VIII
Textbooks
The course is not based on a single textbook. The majority of topics is covered in (Zweifel, Breyer, and Kifmann (2009)) (library link to ebook). (Morrisey (2008)) (link to ebook) covers also many of the discussed topics but has an (almost entirely) empirical approach. Detailed references are given in the schedule below.
Prerequisites and background
I assume that you know some introductory microeconomics, e.g. chapters 1-6 "Intermediate Microeconomis" ("Grundzüge der Mikroökonomik") by Hal Varian. A working knowledge of high-school mathematics and basic statistics (distribution, expectation, correlation, regression) is also needed.
Detailed schedule
This is a plan and as every good plan it may be adjusted if necessary.
Intro (2 lectures)
- Choice, preferences, utilities, welfare, models
- mathematical prerequisites: functions, expected value of a discrete random variable, summation sign
- economic prerequisites: preferences, utility maximization, Pareto efficiency, welfare
- choice, preferences, utility
- choice under uncertainty and expected utility
- welfare
- models
- reading:
- Insurance demand
- mathematical prerequisites: inverse functions, derivatives, monotonicity as well as concavity and convexity
- certainty equivalent and risk premium
- drivers of insurance demand
- (coverage choice and state dependent utility)
- reading: (ch. Morrisey (2008), 3)
- supplementary reading: (ch. Eisenführ and Weber (2013), 9)
Selection (5)
- Selection with fixed coverage
- economic prerequisites: demand function
- model
- welfare consequences
- gender specific premia
- some evidence for selection
- reading: (Einav and Finkelstein (2011), 115–23), (ch. Zweifel, Breyer, and Kifmann (2009), 5.3.1 and 5.3.2)
- Screening with coverage: Rothschild-Stiglitz
- mathematical prerequisites: repeat implicit function theorem, concavity and convexity
- shape of indifference curves
- equilibrium
- comparative statics
- derivation of eq in screencast
- reading: (ch. Zweifel, Breyer, and Kifmann (2009), 5.3.3)
- supplementary reading: (Rothschild and Stiglitz (1976))
- Genetic Tests
- two kinds of risks
- in RS model
- supplementary reading: (Doherty and Thistle (1996)), (Lagerlöf and Schottmüller (2018))
- Premium risk, community rating and risk adjustment
- mathematical prerequisites: linear regression ("ordinary least squares") and R2
- premium risk model
- segue into risk adjustment
- discussion: how could health insurers attract a profitable clientele?
- reading: (ch. Zweifel, Breyer, and Kifmann (2009), 5.3.4)
- supplementary reading: (ch. Zweifel, Breyer, and Kifmann (2009), 7), (ch. Morrisey (2008), 6), (Behrend et al. (2007)), (van de Ven and Ellis (2000))
- Advantageous selection
- case study: selection into long term care insurance in the US
- fixed coverage model
- reading: (Finkelstein and McGarry (2006))
- supplementary reading: (Hemenway (1990)), (Fang, Keane, and Silverman (2008)), (Boone and Schottmüller (2017))
Moral hazard (3)
- The question of moral hazard and empirical evidence
- mathematical prerequisites: significance in statistical tests (e.g. t-test)
- slope of demand
- RAND and arc elasticity of demand
- Oregon
- welfare
- ex ante moral hazard
- reading: (sections Einav and Finkelstein (2018), 1,2 and 3.1)
- Treatment choice and the donut hole
- mathematical prerequisites: (continuous) distributions (density, distribution function)
- simple model of treatment choice
- donut hole
- out of sample predictions
- utilization management and gatekeeping
- reading: (Einav and Finkelstein (2018), 3.2 -end)
- Case study: moral hazard in NL
- diff-in-diff estimate for arc elasticity of demand
Physician-patient interaction (4)
- Supplier induced demand: theory
- density model
- some empirical evidence
- second wave of SID studies
- reading: (Zweifel, Breyer, and Kifmann (2009))
- supplementary reading: (McGuire (2000), 5), (Fuchs (1978); Gruber and Owings (1996); Krasnik et al. (1990))
- Supplier induced demand: empirics
- How Danish physicians react to incentives
- Case study: German hospitals
- Credence good model
- problems/assumptions and appropriate incentives
- discussion: DRG system like liability? implications?
- reading: (Dulleck and Kerschbamer (2006))