Research Assistant Exercise

A Sample of Research Assistant Work

Prof. Dickstein conducts research on health care markets. Some of his recent research projects include:

  • “The Impact of Market Size and Composition on Health Insurance Premiums: Evidence from the First Year of the ACA”
    • “Physician vs. Patient Incentives in Prescription Drug Choice”
    • “Efficient Provision of Experience Goods: Evidence from Antidepressant Choice”
    • “Price Setting in Medicare and Technology Adoption: Evidence from the RUC”

For a current project, Prof. Dickstein collected data for surgical procedures carried out in hospitals and “ambulatory surgery centers” for multiple regions/markets. The goal of the research is to analyze the market for hospital services and the incentives for entry of competing ambulatory centers.

We wish to study how competition evolved over time by comparing the number of procedures completed and total revenue earned in each market over time at facilities of each type—hospitals vs. ambulatory surgery centers.

There are a few data files needed for this exercise.

  1. data_exercise.csv is a dataset containing quarterly data for the years 1997 through 2012. This dataset contains one summary observation per facility per quarter.
  2. data_description.csv contains a description of the variables in data_exercise.csv
  3. fac_exercise.csv is a dataset containing a facility-level “for-profit” status indicator.

Steps to complete:

  1. Download http://pages.stern.nyu.edu/~mdickste/research/hospital_exercise.zip. Unzip the files.
  2. Open data_exercise.csv. We are only interested in examining hospitals and ambulatory surgery centers. Do not include any facilities labeled as other types beyond these two categories. Report the mean, median, standard deviation, min value, and max value of: number of procedures completed, total revenue, and the mean charge per patient. Report these summary statistics for each year and for each of the two facility types.
  3. The data begin in Q1 of 1997. Firms that existed at the time are considered “incumbents”; firms that entered after 1997 Q1 are considered “entrants” in the year and quarter in which they began having positive revenues, and incumbents in subsequent years. Create three new variables in the dataset to capture these designations. First, create an indicator variable that equals 1 for the periods in which a facility is an “entrant”. Second, create an indicator variable that equals one for the periods in which we consider the firm an incumbent. And third, create a count variable that contains the running number of years for which the facility has existed. Document exactly how you create these variables.
  4. Now you need to merge in data on the characteristics of each facility, including its for-profit status. These data are saved in fac_exercise.csv. Merge this into the dataset you've created in (3) above.
  5. We now want to analyze the effect of competition on prices. To do so, we need to create a count of the number of facilities operating in a region (region_id) in a given year and quarter. Create this variable and merge it back to the dataset you created in (4) as a new variable, “num_facilities”.
  6. Extra: Test how competition affects prices. Run a regression of the natural log of per-patient charges on a constant and the “num_facilities” variable you created in (5) (Useful stata commands, if using stata: gen, reg). Note: a regression is just finding a line of best fit through the data. Does competition have an effect on prices?

Please submit your script file (this is the name of the file containing your code) and your .log file (this file saves the output in a text file) to michael.dickstein@nyu.edu when you have finished your work.

Note that not everything in this write up will spell out exactly how to do each step. Be resourceful: look in the Stata/Sas/Matlab documentation files, use google... Please comment your code and be sure to note things that you were not quite sure about and the solution you chose. We are looking for work that carefully executes the assignment with clear documentation of the decisions made. If you can't figure something out, that is OK. Document your confusion, make a decision of how to proceed, and continue.

    •  

Take-Up, Drop-Out, and Spending in ACA Marketplaces (with Rebecca Diamond, Tim McQuade, Petra Persson)

The Affordable Care Act (ACA) reformed the market for individual insurance in the United States, eliminating insurance under-writing, regulating plan characteristics, and establishing marketplaces for consumers to select coverage. Using novel credit card and bank account micro-data on over 850,000 California account holders, we identify new enrollees in the California marketplace, and examine measures of their health spending and premium payments. Following enrollment, we observe dramatic spikes in individual consumers’ health consumption, with more transactions and greater out-of-pocket health care and drug spending. However, these enrollees rarely pay consistent monthly premiums: in the 2014 and 2015 open enrollment years in California, for example, only about half of enrollees pay all 12 months of premiums, with the sharpest drop-out observed after only one month of payments. We show that this drop-out behavior generates a distinct adverse selection problem and potential market unraveling, even absent differences in enrollees’ underlying health costs. We examine the pass-through of drop-out rates to premiums and consider alternative penalty designs to address this threat to market stability.

The Role of Experience in Physician Treatment Decisions: Evidence from the Introduction of Medicare Part D (with Marissa King and Tanja Saxell)

Abstract: How do physicians learn about new treatments? Using the setting of antipsychotic treatment choice, we measure the relative importance of two key pathways: observational learning, in which physicians update their knowledge from public signals, and learning-by-doing, in which the physician relies on her own private experience treating patients. To do so, we exploit two sources of exogenous shocks to physicians’ information. First, in 2007, regulators issued new guidance in the antipsychotic market, approving one drug as a secondary treatment for depression and warning that another posed a substantial risk of side effects. Second, in 2006, the introduction of Medicare Part D shocked the typical physician’s patient composition, with more patients obtaining private insurance coverage. Examining the time periods surrounding the drug advisories, we find physicians with greater patient volume and with more specialized training learn about product quality sooner. Public warnings primarily affect the decisions of the least experienced and least specialized physicians. Importantly, among physicians seeing few patients, recent graduates react more quickly and robustly to the advisories following their publication. We further show that exploiting variation in experience stemming from Medi- care’s insurance expansion is necessary to distinguish the effect of volume from unobserved factors, such as physician quality.

Accounting for Structural and Measurement Error in Binary Choice Problems: A Moment Inequality Approach (with Eduardo Morales)

Abstract: Many economic decisions involve a binary choice - for example, when consumers decide to purchase a good or when firms decide to enter a new market. In such settings, agents’ choices often depend on imperfect expectations of the future payoffs from their decision (expectational error) as well as factors that the econometrician does not observe (structural error). In this paper, we show that expectational error, under an assumption of rational expectations, is a source of classical measurement error, and we propose a novel moment inequality estimator that accounts for both expectational error and structural error in a binary choice model. With simulated data and Chilean firm-level customs data, we illustrate the identifying power of our inequalities and show the biases that arise when one ignores either source of error. We use the customs data to estimate the fixed costs exporters face when entering a new market.

DOWNLOAD
Online Appendix
Matlab code for simulation exercise