Teaching

Statistical Tools for Causal Inference
This course covers the basic theoretical knowledge and technical skills required for implementing microeconometric methods of estimation of causal or treatment effects.
Goals of the class:
- Understanding of the basic language to encode causality,
- knowledge of the fundamental problems of inference and the biases of intuitive estimators,
- understanding of how econometric methods recover treatment effects,
- ability to compute these estimators along with an estimate of their precision using the statistical software R combined with latex using knitr.
In class, all the notions and estimators will be introduced using a numerical example and simulations.
The R code used to generate the results presented in class will be uploaded before class on the moodle webpage of the class, as well as the slides of the class. The files are publicly available here.
  1. The Two Fundamental Problems of Inference: slides, code, text.
    • Rubin Causal Model: the basic language to encode causality.
    • Treatment Effects: our causal parameters of interest.
    • The Fundamental Problem of Causal Inference (FPCI): the Treatment Effects of interest can NEVER be observed, even with a sample of infinite size (a very acute problem indeed!). What we can do instead is to use transformations of the observed data that, under certain assumptions, are equal to the Treatment Effect of interest when the sample size is infinite. 
    • The Biases of Intuitive Comparisons: We also learn that the intuitive comparisons that we use for causal inference (the before/after and with/without comparisons) are generally biased because of factors that determine both the outcomes of the program and who receives it. These factors are called confounding factors.
    • The Fundamental Problem of Statistical Inference (FPSI): in practice, sample sizes are finite. As a consequence, in each sample, our estimator differs from the Treatment Effect of interest. This phenomenon is called sampling noise. We will cover two useful statistical tools to help with this problem: gauging the size of the sampling noise ex-post; choosing sample size ex-ante to decrease sampling noise.
    • The perils of significance testing: specification search and publication bias. I suggest to NEVER use statistical tests and I explain why. I suggest to gauge sampling noise instead.
  2. Methods of Causal Inference: In this section, we learn the three sets of methods that are used by economists in order to suppress the influence of confounding factors and estimate Treatment Effects. For each estimator, we will cover identification (how it solves the fundamental problem of causal inference absent sampling noise), estimation (how to compute an estimator with a sample) and precision (how to gauge the sensitivity of our estimate to sampling noise with independently and identically distributed (i.i.d.) observations).
    • Randomized Controlled Trials (RCTs) solve for the problem of the confounding factors by allocating the treatment at random, i.e. independently of the confounders. We will cover the four most used RCT designs: randomization by brute force, after self-selection, after eligibility and encouragement designs. slides, code, text
    • Natural Experiments leverage on features of the implementation of the program that approximate the conditions of a RCT. We are going to cover the three most used natural experiment methods: Instrumental Variables (IV), Difference-In-Differences (DID) and Regression Discontinuity Designs (RDD). slides, code.
    • Observational methods try to measure the confounders and to account separately for their effects on the outcomes. Standard observational methods that we are going to study are OLS and Matching. I am also going to dedicate some time to more recent Observational Methods based on Machine Learning (ML). slides, code.
  3. Additional important topics
    • Power analysis: before implementing a given method, we want either to choose the sample size required to reach a pre-specified level of precision or to gauge the level of precision we might reach with a pre-specified sample size. slides, code.
    • Placebo tests: tests that we implement in order to check the validity of natural experiments and of observational methods. slides, code.
    • How to estimate precision when observations are not i.i.d. slides, code.
    • LaLonde tests: check whether observational methods and natural experiments can reproduce the results of RCTs. 
    • Analysis of diffusion effects.
    • Analysis of distributive effects.
    • Meta-analysis.


Empirical Environmental Economics

The aim of this class is to provide a broad overview of the empirical work in environmental economics.
The course will focus on the following questions:
- What evidence is there that externalities exist? We will talk about air pollution, water pollution, biodiversity and climate change.
- What is the evidence on the effectiveness and costs of environmental policies? We will look ar regulation, taxes, markets, subsidies, Payments for Environmental Services and nudges.

Here is the list of papers that will be discussed in class: (Numbers in parentheses refer to paper id in this database):

1. Existence of externalities
Air pollution:
Health:
Developed countries: Chay and Greenstone (7) ; Currie, Davis, Greenstone and Walker (16) ; Currie, Neidell (32) ; Bell, McDermott, Zeger, Samet, Dominici (73) ; Lavaine and Neidell (108)
Developing countries: Chen, Ebenstein, Greenstone and Li (17) ; Arceo-Gomez, Hanna and Oliva (25) ; Chang, Graff Zivin, Gross, Neidell (107) ; Jayachandran (254) ; Pitt, Rosenzweig, and Hassan (279)
Productivity:
Graff Zivin and Neidell (33) ; Hanna and Oliva (24)
Water:
Health:
Developed countries: Currie, Graff Zivin, Meckel, Neidell and Schlenker (27);
Developing countries: Ebenstein (255) ; Field, Glennester and Hussam (256) ; Kremer, Leino, Miguel, Zwane (257) ; Brainerd and Mennon (258).
Productivity:
To be confirmed
Climate
GDP:
Dell, Jones and Olken (187) ; Dell, Jones and Olken (186)
Agriculture:
Ricardian approach: Mendelsohn, Nordhaus, Shaw (204) ; Schlenker, Hahnemann, Fisher (205) ; Schlenker, Hahnemann, Fisher (206);
Panel approach: Deschênes and Greenstone (207) ; Fisher, Hahnemann, Roberts and Schlenker (208) ; Deschênes and Greenstone (209) ; Schlenker and Roberts (219);
Developing countries: Schlenker and Lobell (210) ; Guiteras (211)
World: Lobell, Schlenker and Costa-Roberts (220)
Adaptation: Burke and Emerick (223) ; Fishman (224);
Health:
Developed countries: Deschênes and Greenstone (20) ; Barreca, Clay, Deschenes, Greenstone and Shapiro (82) ; Deschenes, Greenstone and Guryan (246);
Developing countries: Burgess, Deschenes and Donaldson (242) ; Kudamatsu, Persson and Stromberg (248) ; Hajat, Armstrong, Gouveia and Wilkinson (245)
Productivity:
Cachon, Gallino and Olivares (238) ; Graff-Zivin and Neidell (235) ; Jones and Olken (237)
Conflicts:
Hsiang, Burke, and Miguel (268)

Agriculture and biodiversity
Temporal correlations:
Chamberlain, Fuller, Bunce, Duckworth and Shrubb (126)
Spatial correlations:
Doxa, Bas, Paracchini, Pointereau, Terres and Jiguet (137)
Panel data:
Bonthoux, Barnagaud, Goulard, and Balent (133)

2. Impact of policies
Regulation:
Air:
CAAA: Chay and Greenstone (6) ; Greenstone (9) ; Auffhammer, Bento and Lowe (292) ; Greenstone (12) ; Greenstone, List and Syverson (22) ; Walker (293) ; Chay and Greenstone (8) ; Hanna (98) ; Ryan (104)
Driving restrictions: van Benthem (23) ; Davis (68)
Gasoline content: Auffhammer and Kellogg (72)
Automobile standards: Reynaert (294) and references therein.
Private and public controls: Duflo, Greenstone, Pande, Ryan (19) ; Duflo, Greenstone, Pande, Ryan (295) ; Stoerk (308)
Water:
CWA: Keiser and Shapiro (106)
Superfund: Currie, Greenstone and Moretti (15) ; Greenstone and Gallagher (14)
India: Greenstone and Hanna (26)
Biodiversity
Parks: Andam, Ferraro, Pfaff, Sanchez- Azofeifa and Robalino (157) ; Sims (158)
Climate:
Internalities and regulation: Alcott and Taubinsky (41) ; Allcott and Wozny (42)
Kyoto protocol: Aichele and Felbermayer (57, 56) ; Grunewald and Martinez Zarzoso (58) ; Almer and Winkler (55) ; Voia (in class)
Markets:
Air:
SO2: Murphy (296)
NOx: Deschênes, Greenstone and Shapiro (21) ; Fowlie, Holland and Mansur (75) ; Fowlie and Perloff (74) ; Carlsson, Burtraw, Cropper, Palmer (77)
Developing countries: Montero, Sanchez, Katz (66)
Biodiversity:
To be confirmed
Climate:
EU ETS: Wagner, Muuls, Martin, Colmer (96) ; Calel and Dechezleprêtre (303) ; Dechezleprêtre, Gennaioli, Martin, Muûls and Stoerk (304)
Taxes:
Air:
Fowlie, Knittel and Wolfram (285)
Water:
To be confirmed
Biodiversity:
To be confirmed
Climate:
Martin, de Preux, Wagner (95) ; Aghion, Dechezleprêtre, Hemous, Martin, and Van Reenen (81)
Subsidies:
Air:
Hanna, Duflo, Greenstone (18)
Water:
Ashraf, Berry and Shapiro (282) ; Berry, Fischer, and Guiteras (277)
Biodiversity:
EU PES: Chabé-Ferret and Subervie (112) ; Pufahl and Weiss (185) ; Kleijn, Berendse, Smit and Gilissen (118) ; Kleijn and Sutherland (120) ; Kleijn, Baquero, Clough, Diaz, De Esteban, Fernandez, Gabriel, Herzog, Holzschuh, Johl, Knop, Kruess, Marshall, Steffan-Dewenter, Tscharntke, Verhulst, West, Yela (119)
Costa Rica and Mexican PES: Alix-Garcia, Sims and Yanez-Pagans (152) ; Alix-Garcia and Sims (151)
Climate :
Forest PES: Simonet, Subervie, Ezzine-de-Blas, Cromberg, Duchelle (114) ; Jayachandran (113) ; Jack (115)
Weatherization: Fowlie, Greenstone, Wolfram (1) ; Allcott and Greenstone
Peak pricing: Ito (2)
Efficiency subsidies: Houde and Aldy (3) ; Davis, Fuchs and Gertler (287)
French car feebate: D’Haultfoeuille, Givord and Boutin (305) ; D’Haultfoeuille, Durrmeyer and Février (306)
Nudges:
Air:
To be confirmed
Water:
Ferraro and Price (41)

Biodiversity:
To be confirmed
Climate:
Social comparison: Allcott and Rogers (37)
Norms: Ito, Ida, Tanaka (50)
Reputation: Yoeli, Hoeffman, Rand and Nowak (52)
Information: Jessoe and Rapson (53)
Defaults: Baylis, Cappers, Fowlie, Spurlock, Todd, and Wolfram (307)

ċ
Lecture_0_slides.Rnw
(64k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:25 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:25 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:25 AM
ċ
Lecture_1_slides.Rnw
(62k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_2_slides.Rnw
(147k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_3_slides.Rnw
(99k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_3_text.Rnw
(98k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_4_slides.Rnw
(65k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_5_slides.Rnw
(33k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
ċ
Lecture_6_slides.Rnw
(14k)
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Ċ
Sylvain Chabé-Ferret,
Jun 21, 2018, 1:26 AM
Comments