Thoughts on healthcare economics

Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty (2011)

★★★★☆

By: Abhijit V. Banerjee and Esther Duflo

This was actually a fantastic read. I thought it really hit the sweet spot between being a pop economics book and being quite academic. For each topic, I really liked how the authors clearly stated their opinion but also acknowledged the other side of the argument in each case. The studies cited were also clear, and if there were any weaknesses in the analysis, the authors always stated them while arguing that there are still good conclusions to be drawn. I could tell that Banerjee and Duflo are true researchers. I didn't have to agree with every one of their arguments, but I understood their points clearly and could reiterate their arguments. I thought the titles of many sections were eye-catching, and enjoyed the overall organization of the book as well. However, it didn't feel boring, and like a laundry list of development papers. Each section had a cohesive theme, which made it an exciting and fulfilling read. I can see why it's a must-read for any student interested in development economics.

2. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything (2005)

★★☆☆☆

By: Steven Levitt and Stephen J. Dubner

I know I'm the last person on earth to read this, but I just put it off. Reading it now, I was actually very surprised to realize it had not aged well. I think the chapter on the link between abortion and crime could be particularly interesting now, as Roe v. Wade has been overturned, but generally I found the arguments tenuous and not very academically grounded. It didn't feel like each chapter had multiple papers that could support the arguments. It felt like often, the authors were making some big leaps based solely on their own research with no sense of the literature as a whole. I also think it felt incredibly disorganized. The authors state in the beginning there is no unifying thread throughout, but I think that the randomness of the topics wasn't satisfying to read. They introduce a lot of their findings in the introduction and then repeat many parts later. They also jump between topics quickly. I found myself getting confused at times as to whether I had accidentally skipped parts of the book. Lastly, I thought all the snippets from the New York Times and other papers talking about how "ah-mazing" Levitt is as an economist and a thinker were a little awkward. It felt a bit self-indulgent although I know it was likely Dubner's take on Levitt as a person. I felt it was unnecessary and awkward.

I think it was a good book in 2005 to introduce readers to the breadth of economics. I know many probably still think of economics as solely focused on business and finance. I think this book probably did a great job of disabusing people of that notion. However, I think it potentially made it seem like economists just draw connections based on correlations when that is exactly what I think we strive not to do. This is obviously a personal opinion, and I think the book is great in terms of what it did do: getting people interested in broad questions.

Next up for mini-reviews!

Bottle of Lies: The Inside Story of the Generic Drug Boom by Katherine Eban

I was watching this video by Market Power on youtube to see where I could find more interesting economics books. I'm currently reading through the old canon, but honestly I do want to read some books there are a bit more specific to my tastes as a PhD student interested in health economics.

To my surprise, he suggested that people really interested in academic nonfiction subscribe to university press newsletters. What a fantastic suggestion!

I immediately went to the Harvard and Yale University Press websites just to see what was there and I found a fascinating list of books. I decided to make a quick list of the books that might be most interesting to health economists (and as a future reference for myself!)

Obviously, these are just some of the initial ones I found interesting from the two press websites i visited, so I highly recommend anyone reading this to go take a look for yourself and find something specific to your interests! I'll definitely continue keeping tabs!

Harvard University Press

The Next Shift: The Fall of Industry and the Rise of Health Care in Rust Belt America by Gabriel Winant

The Cigarette: A Political History by Sarah Milov

The Age of Addiction: How Bad Habits Became Big Business by David T. Courtwright

Visualizing Taste: How Business Changed the Look of What You Eat by Ai Hisano

Ancestral Genomics: African American Health in the Age of Precision Medicine by Constance B. Hilliard

Maladies of Empire: How Colonialism, Slavery, and War Transformed Medicine by Jim Downs

Healthy Buildings: How Indoor Spaces Can Make You Sick—or Keep You Well by Allen, Joseph G. and Macomber, John D.

Yale University Press

Risky Business: Why Insurance Markets Fail and What to Do About It by Liran Einav, Amy Finkelstein and Ray Fisman

Diabetes: A History of Race and Disease by Arleen Marcia Tuchman

Disorder: A History of Reform, Reaction, and Money in American Medicine by Peter A. Swenson

American Contagions: Epidemics and the Law from Smallpox to COVID-19 by John Fabian Witt

Sarah Eichmeyer and Jonathan Zhang find that where you go for healthcare matters in a variety of ways!

"Assignment to a PCP who prescribes opioids at a 2.54 percentage point (pp) higher rate (equivalent to the average difference between a 90th and 10th percentile prescriber within a facility) increases the probability of long-term opioid use by 20% (or 0.43 pp) and development of opioid use disorder by 4% (or 0.035 pp)"

Eichmeyer, Sarah, and Jonathan Zhang. “Pathways into Opioid Dependence: Evidence from Practice Variation in Emergency Departments.” American Economic Journal: Applied Economics 14, no. 4 (October 1, 2022): 271–300. https://doi.org/10.1257/app.20210048.

06/10/2023

Jeff Ferris, Senior Manager and Economist, Amazon

Shilpi Mukherjee, Senior Economist, Keystone Strategy

Gustavo Suarez, Assistant Director, Division of Research and Statistics, Federal Reserve Board

Shalyce Tyson, Human Resources Specialist and Recruitment Coordinator, U.S. Census Bureau

Moderator: Emma Zetterdahl, Senior Manager, Economics Research, Spotify Advertising

This was actually a really helpful panel, hosted by mostly economists and people really close to economics research. I liked that they had a mix of people in consulting, tech and the government, and they gave a good picture of industry research in general. Specifically, attendees learned what skills private industry and the government are looking for in an economist.

Consulting and tech seem to be looking for causal, IO and time series at the moment. There is room for micro and macro theorists but they often tend to have some special skill that is related to working with data. They were clear that working with datasets seems to be the daily responsibility of most newly hired economists in industry and government. Machine learning and working with data are great skills to have now, and should be highlighted briefly either on the CV or with a link to a project on github. There was some discussion on making sure we show we are clearly practiced at thinking about economics problems and not just working with data, but I think the key takeaway is to be good at both.

Causal and structural estimation are critical since places in industry tend to have big data, and there is a lot of scope for different analyses. Structural is important to sort of pare down the question. Causal is the most generalizable.

For PhD versus MA, the opportunities are slightly different. Many times the economist role seems to be defined for someone who has a PhD. There are roles for MA students (e.g. survey statistician at the Census, data scientist at Amazon), but the idea for the economist position is to hire someone who has done some rigorous thesis work.

So what does the job look like for a new economist? In consulting, you'll be working with a supervisor on some project and performing some analysis that is assigned to you. This tends to be very technical and usually involves working with data. As you work your way up, you begin identifying the problems and choosing the problems. At the top, you decide broader direction in terms of which types of analyses work and which don't. In tech, you'll be handed a variety of tasks with little ambiguity to simply execute until you work up. Once you have more responsibility, you yourself might start scoping projects that have some more ambiguity. The more leadership you have, the more ambiguous the problems become. As you become a leader you become someone who chooses the direction.

The general consensus seems to be that as a fresh economist you'll be working on problems that have been identified and defined by someone else. Once you become a leader, you become the one to define the problems.

Hiring process: (1) In tech there are two phases. First, a phone screen, walking through a set of technical problems. Next, some more technical questions, questions on aptitude for leadership and how you approach problem solving. Don't bring up your job market paper as much for tech. They want to see how well you can approach a new problem (2) At the census: Submit resume and get certificate. Interview. (3) In consulting, specifically at keystone. First, there is a technical portion and you often discuss your job market paper. Next, there is an on-site interview. You might look over some technical problems with some company leadership. There is a behavioral portion as well. (4) At the Fed, you often start with AEA interviews. From there you receive flyouts. You present your job market paper, talk to the other economists there. They want to see your technical skills and communication abilities.

Card et al. study the impacts of childbirth in a hospital with high cesarean delivery rates versus low. Cesarean rates vary across hospitals in the US from 19 to 48 percent. That's huge variation! They use a distance based IV strategy to classify mothers living near high C-section hospitals versus low.

There is some reverse causality worry here since giving birth by C-section is often due to complications during labor, but using this IV should account for that. They have a strong first stage with a higher rate of cesarean deliveries among mothers that live closer to high C-section hospitals. They also show that patients living closer to high C-section hospitals don't seem to be significantly different from those living near low C-section hospitals.

They find some interesting evidence. C-section births provide benefits in that infants are born healthier and are less likely to be readmitted. However, they are still susceptible to respiratory problems.

Just a new space for me to put my thoughts about my research or any fun things I read.

I think better on paper and I forget things I don't write down.