replication_conscription_crime.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Replication Study: Galiani, Rossi, and Schargrodsky (2011) Revisited"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Author: Benedikt Heinrich Müller\n",
    "\n",
    "Course Microeconometrics, Summer term 2020, Prof. Dr. Eisenhauer\n",
    "\n",
    "M.Sc. in Economics, University of Bonn\n",
    "***"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introductory Remarks \n",
    "\n",
    "In the following I replicate the findings of the 2011 empirical study \"*Conscription and Crime: Evidence from the Argentine Draft Lottery*\" by Galiani et al. (for exact citation see [section 1](#1.-Brief-Summary-of-Journal-Acticle)). Throughout I apply the same nomenclatur as did Galiani et al. (2011), i.e. the tables and figures are enumerated as in the original paper. That is to foster comparability. <br>\n",
    "Galiani et al. (2011) used Stata for their analysis. Since Stata was not available to me, I collected the labels of the variables by means of comparison of the .do-file and the results presented in the paper.\n",
    "\n",
    "Tables and figures denoted with an **E** are extended tables from the original article, and thus, bear the enumeration as in the original article. **A** denotes the tables and figures from the online appendix. Tables and figures that were not part of the article and were instead constructed by me are denoted by a **B**.\n",
    "\n",
    "This notebook is structured as follows. [Section 1](#1.-Brief-Summary-of-Journal-Acticle) introduces the data used, the research question, and the research design applied by Galiani et al. (2011). I draw on causal graphs and a potential outcome model to clarify the empirical strategy and potential pitfalls. [Section 2](#2.-Replication-of-Key-Findings) replicates the key tables and figures of the article. Furthermore I replicate some figures from the online appendix. [Section 3](#3.-Extensions) contains my own contribution to the empirical study. [Section 4](#4.-Critical-Assessment-of-Quality) concludes this notebook with a discussion of the quality of the study. To this end, I also draw on my own contribution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Content of this Notebook \n",
    "\n",
    "1. [Brief Summary of Journal Article](#1.-Brief-Summary-of-Journal-Acticle)\n",
    "\n",
    "    1.1 [Background & Data](#1.1-Background-&-Data)\n",
    "    \n",
    "    1.2 [Empirical Strategy](#1.2-Empirical-Strategy)\n",
    "\n",
    "2. [Replication of Key Findings](#2.-Replication-of-Key-Findings)\n",
    "\n",
    "    2.1 [Descriptive Statistics and Pre-Treatment Characteristics](#2.1-Descriptive-Statistics-and-Pre-Treatment-Characteristics)\n",
    "\n",
    "    2.2 [Main Results](#2.2-Main-Results)\n",
    "    \n",
    "    2.3 [Complementary Results](#2.3-Complementary-Results)\n",
    "    \n",
    "    2.4 [Online Appendix](#2.4-Online-Appendix)\n",
    "    \n",
    "3. [Extensions](#3.-Extensions)\n",
    "\n",
    "    3.1 [Figures](#3.1-Figures)\n",
    "    \n",
    "    3.2 [Extended \"Testing\" of Exogeneity of Instrument](#3.2-Extended-\"Testing\"-of-Exogeneity-of-Instrument)\n",
    "    \n",
    "    3.3 [Robustness Checks](#3.2-Robustness-Checks)\n",
    "\n",
    "4. [Critical Assessment of Quality](#4.-Critical-Assessment-of-Quality)\n",
    "\n",
    "[References](#References)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import packages.\n",
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "import numpy as np\n",
    "\n",
    "from scipy import stats\n",
    "from linearmodels import IV2SLS\n",
    "from auxiliary.functions_v6 import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Brief Summary of Journal Acticle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this replication study I draw on the following article:\n",
    "\n",
    "Galiani, Sebastian, Martín A. Rossi, and Ernesto Schargrodsky. 2011. \"Conscription and Crime: Evidence from the Argentine Draft Lottery.\" *American Economic Journal: Applied Economics*, 3 (2): 119-36.\n",
    "[DOI: 10.1257/app.3.2.119](http://dx.doi.org/10.1257/app.3.2.119)\n",
    "\n",
    "Galiani et al. (2011) aim at studying the causal effect of serving in the military on subsequent criminal behaviour. They use administrative data on the Argentine draft lottery and criminal records. Since failing medical examination before being drafted is endogenous, the authors instrument conscription by *draft eligibility*, a variable indicating whether a unit of observation was randomly chosen to join military service. The unit of observation is all individuals of a birth cohort sharing the same last three digits of the national ID.\n",
    "The authors find that the effect of conscription on developing a criminal record is positive and statistically significant.\n",
    "The novelty of their analysis is that they mostly draw on data from conscription during peacetime."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.1 Background & Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conscription in Argentina has been first introduced in 1902 an has been subject to a draft lottery until the abolishment of mandatory conscription in the 1990s. Galiani et al. (2011) use this random assignment to estimate the causal effect of serving in the military on subsequent criminal behaviour. The core part of the analysis uses data from birth cohorts 1958 to 1962 (= *core cohorts*). I start by presenting the particularities of the Argentine draft lottery."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The Argentine Draft Lottery \n",
    "The timing of the draft lottery is as follows (see section I of Galiani et al. (2011)).\n",
    "1. Draft numbers are drawn in a public session. For every combination of the last three numbers of the national ID of all men in the birth cohort there is a draft number drawn which is between 1 and 1,000 drawn.\n",
    "2. The results of the lottery are publicly announced.\n",
    "3. Medical examinations of all males in the birth cohort are conducted.\n",
    "4. The exact cutoff number is publicly announced.\n",
    "5. Draftees start service the following year.\n",
    "\n",
    "Draftees start service with 21 years (core cohorts of the analysis). The length of military service is one year for army and air force. In the navy one serves for two years. Draftees were assigned to the different military branches based on their draft number. The lowest numbers of draft eligible men were assigned to the army, the medium numbers to the air force and the highest to the navy. E.g. if the cutoff number was 300, the draft numbers from 300 to 700 were assigned to the army, the numbers from 701 to 840 were assigned to the air force, and the numbers 841 to 1,000 were assigned to the navy.\n",
    "\n",
    "To avoid confusion, note that in Angrist (1990) an individual was draft eligible if he had a number *smaller* than the cutoff. In the Argentine draft lottery the draft number needed to be *higher* than the cutoff.\n",
    "\n",
    "The draft age was changed from 18 to 21 in 1955. Thus, the cohorts of 1956 and 1957 were not drafted into the military. Data on draft numbers are available for the cohorts from 1929 to 1965. Later the system of the lottery changed, to the effect that Galinai et al. (2011) could no longer determine which draft numbers were draft eligible. The cohort of 1976 was of particular interest since this cohort was not drafted at all due to the abolishment of conscription in Argentina. This fact is later used to strengthen the plausibility of the assumptions imposed on the instrument.\n",
    "\n",
    "Deferment of military service was possible, but draft eligible status did not change and the individual was still in the cohort-ID group of the initial draft lottery. Some individuals did not need to join military service: \"Clerics, seminarians, novitiates, and any individual with family members dependent upon him for support were exempted from military service.\" (Galiani et al. 2011, p. 122)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### War and Peace \n",
    "The novelty of the study by Galiani et al. (2011) is that they not only estimate the effects of conspcription in wartime but also in peacetime. This is important out of two reasons: 1. the effects of conscription may differ based on whether one served during war or peace and 2. peacetime service is generally more common than wartime service.\n",
    "\n",
    "1. The effect of serving in the military may be quite different when exposed to actual combat. Those individuals envolved in combat may suffer from detrimental effects on their mental and physical health. These conditions may have a large impact on their subsequent criminal behaviour after serving in the military and may also diminish labour market outcomes.\n",
    "\n",
    "2. Conscription during peacetime is more often the case than conscription during wartime. Galiani et al. note that most countries in the world have citizens drafted into the military during peacetime. Thus, the effect of serving in the military in peacetime is the more relevant case when evaluating this policy.\n",
    "\n",
    "From April to June 1982 there was the *Falklands War* between Argentina and the United Kingdom. The cohorts of 1962 and 1963 served during this war. Because of data availability, the authors considered the cohorts from 1958 to 1962 for their main analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Data & Important Variables "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Galiani et al. use data provided by various Argentine administrative records.\n",
    "Because of data privacy issues, they do not provide the raw data of their analysis. One observation in the data set stands for all individuals in the same birth cohort *c* and share the same last three digits of the national ID *i*.\n",
    "\n",
    "\n",
    "In the following, I describe the most important variables used in the analysis.\n",
    "The measure of criminal behaviour is the variable *crime rate*, which is also the dependent variable.\n",
    "Let *n* be the number of individuals *j* in birth cohort *c* which share the same last three digits of their national ID *i*. Further, let $I(j)$ be an identity function returning *1* if individual *j* has a criminal record and *0* otherwise. Then, *crime rate* is defined as\n",
    "$$crimerate=\\frac{ \\sum_{j=1}^n I(j)}{n}$$\n",
    "Note, that $I(j)$ does not indicate which kind of or how many crimes individual *j* committed. <br>\n",
    "The authors want to isolate the effect of *conscription*, represented in the data set by *'sm'* (probably from Spanish *servicio militar*). *Conscription* is the share of individuals in *c* and *i* who served in the military. *Conscription* is probably endogenous and therefore instrumented for.\n",
    "The potential instrument is the variable *draft eligible*, denoted in the data set by the dummy *'highnumber'*, which returns *1* if the draft number of *c* and *i* is above the cutoff number. The variable *draft number* is also available.\n",
    "Potential control variables used are birth-cohort, origin (*argentine*, *naturalized*, and *indigenous*), and province dummies (federal states). The latter are denoted *dist* plus a number indicating one of the 24 provinces. The varible *malvinas* (Spanish name of the Falkland Islands) denotes the share of individuals of cohort *c* and ID *i* that served during the *Falklands War* in 1982 (for the main analysis only cohort 1962 served during the war). *malvinas* is interacted with *draft eligible* for the analysis.\n",
    "\n",
    "Crime rates distinguished by type of crime were collected during the period 2000 to 2005. Again, crime rates are calculated for each cohort-ID group for the cohorts of 1958 to 1962. According to footnote 13 in Galiani et al. (2011) types of crime are only reported for 37 percent of the cases. They state that draft number was not correlated with the types of crime missing. Out of confidentiality reasons they do not provide the raw data.\n",
    "\n",
    "The authors consider labour market outcomes as well. Unemployment and income data are computed by using self-declared occupation from the national ballot registry of 2003. Based on the occupation, they assign unemployment shares and average hourly earnings by using official household survey data from May 2003. Participation rates in the formal labour market is retrieved from the national social security data base. Thus, the labour market outcomes are observed 19 to 26 years after military service."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2 Empirical Strategy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Causal Graph Representation "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I start the discussion of the empirical strategy by presenting the causal relationship Galiani et al. (2011) assumed in their study. To this end, I use the graphical representation suggested by Pearl (2014).\n",
    "Causal graphs were created with [dagitty.net](http://www.dagitty.net/). The following notation is used.\n",
    "- Edges show the assumed causal relationship.\n",
    "- Arrows indicate the direction of the causal relationship.\n",
    "- Red edges indicate threats to identification.\n",
    "- Blue nodes are observed variables, grey ones are unobserved.\n",
    "- Red nodes denote compromised instruments.\n",
    "- The outcome variable is denoted by an **I** in a blue node.\n",
    "- The variable whose effect we want to study is denoted by a triangle in a green node.\n",
    "\n",
    "#### Causal Graph 1 - Simple Case\n",
    "<img src='files/dagitty-model_v2.png' alt=\"Alt text that describes the graphic\" title=\"Causal Graph 1\" />\n",
    "\n",
    "Galiani et al. (2011) want to estimate the effect of conscription on crime rates. The fundamental problem of this endeavour is that the variable *conscription* might be correlated with the error term of *crime rate*. This is, because *crime rate* and *conscription* could both be influenced by third variables. For example, the socioeconomic background could influence whether an individual fails the medical examination and thus, influences military service, but socioeconomic background is very likely to shape criminal behaviour. Back-door identification identifies the causal effect of *conscription* on *crime rate* by blocking all back-door paths. According to Winship and Morgan (2007), a back-door path is a path from *conscription* to *crime rate* that starts with a directed edge that points to *conscription*.\n",
    "There is one back-door path in causal graph 1: *conscription* <- *U* -> *crime rate*. Since *U* is a set of variables that are unobserved, we cannot close this back-door path by adjusting for it. We end up in a state where selection on unobservables is an issue. Selection on unobservables occurs if treatment effect heterogeneity and self-selection are present.\n",
    "\n",
    "If we instrument *conscription* by the exogenous *draft number*, given the causal relationship implied by causal graph 1, we can isolate the effect of *conscription* on *crime rate*. Exogeneity of *draft number* implies that there is no relationship between *draft number* and *crime rate* other than through *conscription*. In the language of causal graphs this amounts to blocking all back-door paths from *draft number* to *crime rate*. Since there are no back-door paths between those, we can estimate the causal effect of *conscription*.\n",
    "\n",
    "Since lacking data on labour market experience, Galiani et al. (2011) can estimate the overall-effect of conscription only; they cannot adjust for labour market experience. But since they are mainly interested in the overall effect, this is not a drawback as long as causality is as depicted by causal graph 1.\n",
    "\n",
    "#### Causal Graph 2 - Compromised Instrument: Violation of the Exogeneity Assumption\n",
    "<img src='files/dagitty-model(1)_v2.png' alt=\"Alt text that describes the graphic\" title=\"Causal Graph 2\" />\n",
    "\n",
    "Causal graph 2 differs from the first one, by introducing a relationship between *draft number* and *labour market experience*. The causal effect of *conscription* can now no longer be isolated, since the exogeneity of the instrument is compromised. This possible relationship was proposed by Rosenzweig and Wolpin (2000). They applied their criticism on the Vietnam draft lottery and the article by Angrist (1990). *draft number* could influence *labour market experience* by other channels than just through *conscription*, if individuals with a high probability of being drafted face worse labour market conditions. Employers could hesitate from employing someone who is likely to be drafted into the military and would then no longer be available for the job at hand. Whether this criticism is applicable in the case of the Argentine draft lottery, see [section 4](#4.-Critical-Assessment-of-Quality).\n",
    "\n",
    "#### Causal Graph 3 - More Complex Relationship\n",
    "<img src='files/dagitty-model(2)_v2.png' alt=\"Alt text that describes the graphic\" title=\"Causal Graph 3\" />\n",
    "\n",
    "Causal graph 3 introduces the effect of wartime conscription. Wartime military service could have other detrimental effects than peacetime military service. For example, experience of combat could cause mental disorders that have an effect on criminal behaviour. There is a second channel through which mental disorders could influence criminal behaviour. Individuals suffering from mental disorder have probably less success in the labour market than mentally healthy individuals. Since *mental disorder* as well as *labour market experience* are both unobserved, one could not hope to isolate the effect of wartime military service if the underlying causal relationship is as assumend by causal graph 3. *military service* influences *labour market experience* by itself but also through *mental disorders*. Thus, the effect of wartime military service cannot be isolated.\n",
    "\n",
    "Due to restrictions on data availability, Galiani et al. (2011) do not use the instrumental variable (henceforth IV) approach to identify the effect of wartime conscription. So, they never really estimate the causal relationship implied by causal graph 3. For a discussion of this, see the discussion of [table 5](#Table-5---Estimated-Impact-of-Conscription-on-Crime-Rates-for-Peacetime-versus-Wartime-Service-and-1-Year-versus-2-Year-Service).\n",
    "The directed edge between *U* and *mental disorders* is no threat to isolating the causal effect of military service since this path is no longer existent by instrumenting *military service*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Empirical Specification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the following, potential issues concerning the identification of the causal relationship and its crucial assumptions are discussed.\n",
    "\n",
    "Galina et al. (2011) want to study the effect of serving in the military on subsequent criminal behaviour. They want to estimate the following regression (p.127). Note that the notation presented here does not coincide with the one presented in the article. I adopt the notation used by Winship and Morgan (2007): $$Y_{ci} = \\alpha + \\delta D_{ci} + \\gamma_{c} + \\epsilon_{ci}$$\n",
    "where\n",
    "- $Y_{ci}$ denotes *crime rate*,\n",
    "- $D$, denotes *conscription*,\n",
    "- $\\delta$, is the ATE,\n",
    "- $\\gamma_{c}$, a cohort effect, and\n",
    "- $\\epsilon_{ci}$, an error term.\n",
    "\n",
    "Formally this amounts to  the following:\n",
    "$$E[Y_{ci}] = E[\\alpha + \\delta D_{ci} + \\gamma_{c} + \\epsilon_{ci}]$$\n",
    "\n",
    "If the assumption $E[\\epsilon∣D]=0$ holds, that is the variable of interest is independent of the outcome variable, we can successfully identify the causal effect of $D$ on $Y$ as the average treatment effect.\n",
    "But, in the above described setting (see causal graph) it is possible that being actually drafted into the military is subject to some form of selection. Recall that all individuals in birth cohort *c* needed to attend medical examinations.\n",
    "Failing the medical examination could be correlated with the socioeconomic background an individual has. For example, health and socioeconomic status could be associated. On the other hand, individuals with a high draft number, i.e. with a high probability of being drafted into the military, fail the medical examinations by purpose and are not drafted into the military. One could imagine that financially better off individuals are more able to do so than individuals from other socioeconomic backgrounds. This renders *conscription* endogeneous.\n",
    "\n",
    "Hence, we cannot ensure that $E[\\epsilon∣D]=0$.\n",
    "This endogeneity issue can be partially overcome by using an IV approach. A valid instrument needs to necessarily fulfil the following two conditions (this specification applies for the case of a binary instrument):\n",
    "1. Exogeneity: $E[\\epsilon∣Z]=0$\n",
    "2. Relevance: $E[Z∣D]\\neq 0$\n",
    "\n",
    "After some manipulation we arrive at the Wald estimator\n",
    "$$\\hat{\\delta}_{IV, Wald} = \\frac{E[Y|Z=1]-E[Y∣Z=0]}{E[D|Z=1]-E[D|Z=0]}$$\n",
    "\n",
    "The exogeneity condition cannot be tested statistically. One can only argue whether an instrument is uncorrelated to the error $\\epsilon$. Condition 2 can be tested by means of estimating the effect of the instrument on the endogenous independent variable $D$. The effect needs to be large and statistically significant. Otherwise we would end up with a weak instrument, which would render the IV estimator non-applicable.\n",
    "\n",
    "In what follows, I employ the Potential Outcome framework as presented in Winship and Morgan (2007, pp. 78) in order to clarify the empirical stragtegy the paper is based upon.\n",
    "\n",
    "#### Potential Outcome framework & LATE.\n",
    "\n",
    "Consider the potential outcome model presented by Winship and Morgan (2007, pp. 200):\n",
    "$$Y = Y^0 + (Y^1 − Y^0)D$$\n",
    "Let $\\delta = Y^1 − Y^0$, then\n",
    "$$Y = Y^0 + \\delta D$$\n",
    "$$Y = E[Y^0] + \\delta D + Y^0−E[Y^0]$$\n",
    "\n",
    "Imbens and Angrist (1994) defined variables indicating potential treatment assignment conditional on the instrument $D^{Z=z}$. These summarise all possible behavioural responses to the instrument and allow the definition of a latent variable $C$. The following is taken from Winship and Morgan (2007, p. 201):\n",
    "\n",
    "- Compliers $(C=c)$: $D^{Z=0}=0$ and $D^{Z=1}=1$\n",
    "- Defiers $(C=d)$: $D^{Z=0}=1$ and $D^{Z=1}=0$\n",
    "- Always takers $(C=a)$: $D^{Z=0}=1$ and $D^{Z=1}=1$\n",
    "- Never takers $(C=n)$: $D^{Z=0}=0$ and $D^{Z=1}=0$\n",
    "\n",
    "*Compliers* react to being exposed to the instrument by selecting into the treatment. *Always takers* select into treatment irrespective of the instrument; vice versa, *Never takers* never do so. *Defiers* would have selected into the treatment if not exposed to the instrument, but do not do so if exposed to the instrument. Being exposed to the instrument means $Z=1$ for the individual.\n",
    "\n",
    "Following Winship and Morgan (2007, p. 201), $D$ can be defined as:\n",
    "$$D = D^{Z=0} + (D^{Z=1} − D^{Z=0})Z$$\n",
    "Let $\\kappa = D^{Z=1} − D^{Z=0}$, then\n",
    "$$D = D^{Z=0} + \\kappa Z$$\n",
    "\n",
    "$\\kappa$ represents the causal effect of $Z$ on $D$. If $D$ varies for the individuals, so does $\\kappa$.\n",
    "\n",
    "Additionally to the analogous identifying conditions defined above, we have a third condition. Monotonicity requires that $\\kappa ≥ 0$ for all i or $\\kappa ≤ 0$ for all i.\n",
    "\n",
    "In the light of essential individual heterogeneity, the IV results cannot be interpreted as an ATE, i.e. $\\delta$ is not constant in the population. This is the case, since we can only make statements about the causal relationship of individuals whose probability of treatment changes according to the IV. In the empirical investigation at hand this means that the estimated effect is the effect on the compliers only, i.e. the individuals who join the military because they are coerced to do so by having a draft number higher than the cutoff. That is, the Wald estimator in this context converges in probabilty to $E[\\delta |C=c]$ (Winship and Morgan 2007, p. 202).\n",
    "\n",
    "This treatment affect is denoted the *Local Average Tretament Effect* (LATE), introduced by Angrist and Imbens (1994). The additional condition of monotonicity implies that there are either compliers or defiers in the sample, but not both. In the problem at hand this amounts to assuming that there are no draft-lottery defiers in the sample. A draft-lottery defier is a person that would have served voluntarily, i.e. if he wasn't assigned a draft number above the cutoff, but abstains from military service if he got a draft number above the cutoff. To avoid military service, if one was assigned a number above the cutoff, it is necessary to fail the medical examination, what might come with some effort. Moreover, defying would imply irrational behaviour.\n",
    "Thus, one can argue that there are no defiers in the sample, i.e. people who had served if service was voluntary, but who did not if they were drafted.\n",
    "Moreover, the treatment effect of the analysis is only provided for the draft-lottery compliers, since, given monotonicity, only these are the ones having an impact on the estimates of the effect on crime rates. Recall that treatment status did not change for *Never takers* and *always takers*. Thus, all the variation in estimating the causal effect in an IV setting is due to compliers (Winship and Morgan 2007, p. 203).\n",
    "Thus, $\\delta$ is not constant."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Replication of Key Findings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data preprocessing & load core data set 'df'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "constant, highnumber, conscription, crimerate, malvinas, navy, origin, cohorts, districts, hn_malvinas, df = get_variables()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.1 Descriptive Statistics and Pre-Treatment Characteristics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table B.1 - Descriptive Statistics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since table 1 of the paper provides merely an overview over the dat set used, I do not replicate table 1 but present the variables in more detail."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table B.1 contains some descriptive statistics. Column 1 was calculated by (mean of variable of interest) x (mean cohort size).\n",
    "\n",
    "The mean of *conscription* for the cohorts of 1958 to 1962 is 0.5031. This means that more than 50 percent of the male cohorts of 1958 to 1962 served in the military. Since the average size of a cohort is 236,656, this amounts to a total of 119,067 individuals of cohort c on average who served in the military. This underlines the importance of the study at hand. Since half of each male cohort served in the military the impact of military service on total number of crimes committed could be quite large.\n",
    "\n",
    "There are only about 4 percent always takers in the data set, given that there are no defiers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable B.1 - Descriptive Statistics of Selected Variables of Interest for Male Birth Cohorts 1958 to 1962\u001b[0m\n",
      "______________________________________________________________________________________________________________\n",
      "                           (1)             (2)             (3)             (4)             (5)\n",
      "                 Cohort size*Mean            Mean        St. dev.   Mean eligible     Mean exempt\n",
      "______________________________________________________________________________________________________________\n",
      "Variables: \n",
      "\n",
      "['highnumber']   \u001b[1m       165612   \u001b[0m\u001b[1m       0.6998   \u001b[0m\u001b[1m       0.4584   \u001b[0m\u001b[1m       1.0000   \u001b[0m\u001b[1m       0.0000   \u001b[0m\n",
      "\n",
      "['sm']           \u001b[1m       119067   \u001b[0m\u001b[1m       0.5031   \u001b[0m\u001b[1m       0.3049   \u001b[0m\u001b[1m       0.7002   \u001b[0m\u001b[1m       0.0437   \u001b[0m\n",
      "\n",
      "['crimerate']    \u001b[1m        16392   \u001b[0m\u001b[1m       0.0693   \u001b[0m\u001b[1m       0.0178   \u001b[0m\u001b[1m       0.0698   \u001b[0m\u001b[1m       0.0681   \u001b[0m\n",
      "\n",
      "['formal']       \u001b[1m        80158   \u001b[0m\u001b[1m       0.3387   \u001b[0m\u001b[1m       0.0470   \u001b[0m\u001b[1m       0.3371   \u001b[0m\u001b[1m       0.3424   \u001b[0m\n",
      "\n",
      "['unemployment'] \u001b[1m        42537   \u001b[0m\u001b[1m       0.1797   \u001b[0m\u001b[1m       0.0543   \u001b[0m\u001b[1m       0.1763   \u001b[0m\u001b[1m       0.1877   \u001b[0m\n",
      "\n",
      "['income']       \u001b[1m       751011   \u001b[0m\u001b[1m       3.1734   \u001b[0m\u001b[1m       0.2343   \u001b[0m\u001b[1m       3.1731   \u001b[0m\u001b[1m       3.1743   \u001b[0m\n",
      "\n",
      "['arms']         \u001b[1m          238   \u001b[0m\u001b[1m       0.0010   \u001b[0m\u001b[1m       0.0024   \u001b[0m\u001b[1m       0.0010   \u001b[0m\u001b[1m       0.0010   \u001b[0m\n",
      "\n",
      "['sexual']       \u001b[1m          170   \u001b[0m\u001b[1m       0.0007   \u001b[0m\u001b[1m       0.0021   \u001b[0m\u001b[1m       0.0007   \u001b[0m\u001b[1m       0.0007   \u001b[0m\n",
      "\n",
      "['whitecollar']  \u001b[1m          803   \u001b[0m\u001b[1m       0.0034   \u001b[0m\u001b[1m       0.0046   \u001b[0m\u001b[1m       0.0035   \u001b[0m\u001b[1m       0.0031   \u001b[0m\n",
      "\n",
      "['navy']         \u001b[1m        28304   \u001b[0m\u001b[1m       0.1196   \u001b[0m\u001b[1m       0.3245   \u001b[0m\u001b[1m       0.1709   \u001b[0m\u001b[1m       0.0000   \u001b[0m\n",
      "\n",
      "['hn_malvinas']  \u001b[1m        32233   \u001b[0m\u001b[1m       0.1362   \u001b[0m\u001b[1m       0.3430   \u001b[0m\u001b[1m       0.1946   \u001b[0m\u001b[1m       0.0000   \u001b[0m\n",
      "\n",
      "['enfdummy']     \u001b[1m        19585   \u001b[0m\u001b[1m       0.0828   \u001b[0m\u001b[1m       0.0205   \u001b[0m\u001b[1m       0.0863   \u001b[0m\u001b[1m       0.0746   \u001b[0m\n",
      "\n",
      "______________________________________________________________________________________________________________\n",
      "Notes: highnumber = draft eligible, sm = conscription, formal = formal job market participation, income = \n",
      "hourly earnings in Argentine pesos,arms = crimes involving usage of weapons, sexual = sexual attack, white-\n",
      "collar = white collar crimes (such as fraud, scams, extortion), navy = eligible for navy, hn_malvinas = eli-\n",
      "gible during Falklands War, enfdummy = failure rates of medical examination. The number in column 1 contains \n",
      "the mean number of individuals in the subgroup. It was calculated by using the mean size of cohorts from \n",
      "1958 to 1962, which is 236,656. Columns 4 and 5 contain means conditional on draft eligible and draft exempt \n",
      "cohort-ID groups.\n"
     ]
    }
   ],
   "source": [
    "table_B_1()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 2 - Differences in Pre-Treatment Characteristics by Birth Cohort and Eligibility Group"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table 2 reports characteristics that were not determined by the treatment.\n",
    "The results of the t-tests show that samples are balanced before the treatment, i.e. draft exempt and draft eligible individuals do not differ according to their origin (not-indigenous argentine, indigenous argentine, or naturalized citizens).\n",
    "Galiani et al. (2011) used the following assumptions regarding the t-test:\n",
    "- Samples drawn independently from population.\n",
    "- Variances are allowed to differ across draft eligible and draft exempt groups.\n",
    "\n",
    "I apply the same set-up for the t-test. Nonetheless, my results differ in magnitude. Where the authors' test statistics are all quasi zero, mine are different from zero. The most important result is though, that we cannot reject the null hypothesis of equal means. That is, pre-treatment origin is balanced across the draft eligible and draft exempt group. This can be seen by considering the p-values, that are all way above the conventional level of at least 10 percent.\n",
    "\n",
    "Note that table 2 in Galiani et al. (2011) shows standard errors, where the table presented here shows the p-values directly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 2 - Differences in Pre-Treatment Characteristics by Birth Cohort and Eligibility Group\u001b[0m\n",
      "\u001b[1mDifferences by Cohort (draft exempt - draft eligible)\u001b[0m\n",
      "______________________________________________________________________________________________________________\n",
      "Cohort                                 1958            1959            1960            1961            1962   \n",
      "\n",
      "______________________________________________________________________________________________________________\n",
      "Argentine-born, not indigenous\u001b[1m     -0.07953   \u001b[0m\u001b[1m     -0.13760   \u001b[0m\u001b[1m     -1.05497   \u001b[0m\u001b[1m     -0.18741   \u001b[0m\u001b[1m      0.26607   \u001b[0m\n",
      "\n",
      "                                    0.93667         0.89060         0.29183         0.85140         0.79028   \n",
      "\n",
      "Argentine-born, indigenous    \u001b[1m     -0.14981   \u001b[0m\u001b[1m     -0.15394   \u001b[0m\u001b[1m      0.71760   \u001b[0m\u001b[1m     -0.47173   \u001b[0m\u001b[1m      0.64254   \u001b[0m\n",
      "\n",
      "                                    0.88104         0.87771         0.47327         0.63726         0.52077   \n",
      "\n",
      "Born abroad, naturalized      \u001b[1m      0.70539   \u001b[0m\u001b[1m      0.15141   \u001b[0m\u001b[1m      0.93374   \u001b[0m\u001b[1m      0.73305   \u001b[0m\u001b[1m     -0.87927   \u001b[0m\n",
      "\n",
      "                                    0.48126         0.87970         0.35083         0.46380         0.37957   \n",
      "\n",
      "______________________________________________________________________________________________________________\n",
      "Notes: P-values are shown below test statistics. The level of observation is the cohort-ID number combination.\n"
     ]
    }
   ],
   "source": [
    "table_2()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 3 - First Stage by Birth Cohort "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table 3 presents first stage regression results, that is, the endogenous variable is regressed on the instrument. Regression estimates of *draft eligible* are highly significant, indicating a strong instrument. All point estimates of draft eligible are well above 0.6. This implies that the probability of serving in the military was more than 60 percent larger for individuals with a draft number above the cutoff compared to the individuals with a draft number below the cutoff. Thus, we can rule out problems that arise due to a weak instrument.\n",
    "\n",
    "Note that some of my findings differ slightly from the ones presented in the paper. The point estimate for the constant in the paper is 0.0421 with a standard error of 0.0008. The significance level does not change and also the point estimate for draft eligible is the same. This is the Stata code by Galiani et al. (2011):\n",
    "> areg sm highnumber if cohort > 1957 & cohort < 1963, robust absorb(cohort)\n",
    "\n",
    "I absorb the cohorts by including cohort dummies directly in the OLS regression. I included n-1 of the cohort dummies in order to avoid multicollinearity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 3 - First Stage by Birth Cohort\u001b[0m\n",
      "Dependent Variable: Conscription\n",
      "________________________________________________________________________________________________________________\n",
      "Cohort             1958-1962            1958            1959            1960            1961            1962                   \n",
      "                         (1)             (2)             (3)             (4)             (5)             (6)                   \n",
      "________________________________________________________________________________________________________________\n",
      "Draft Eligible \u001b[1m       0.6587***\u001b[0m\u001b[1m       0.6279***\u001b[0m\u001b[1m       0.6210***\u001b[0m\u001b[1m       0.6505***\u001b[0m\u001b[1m       0.6972***\u001b[0m\u001b[1m       0.6853***\u001b[0m\n",
      "\n",
      "                      0.0012          0.0033          0.0027          0.0018          0.0017          0.0019   \n",
      "\n",
      "Constant       \u001b[1m       0.0323***\u001b[0m\u001b[1m       0.0578***\u001b[0m\u001b[1m       0.0389***\u001b[0m\u001b[1m       0.0377***\u001b[0m\u001b[1m       0.0556***\u001b[0m\u001b[1m       0.0343***\u001b[0m\n",
      "\n",
      "                      0.0017          0.0030          0.0008          0.0008          0.0011          0.0007   \n",
      "\n",
      "Observations           5,000           1,000           1,000           1,000           1,000           1,000                   \n",
      "Method                   OLS             OLS             OLS             OLS             OLS             OLS                   \n",
      "________________________________________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation is the cohort-ID number combi-\n",
      "nation. Column 1 includes cohort dummies.\n",
      "*** Significant at 1 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_3()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 Main Results "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the following I present the main results of the empirical study. The following results are key since they deocument the statistically significant relationship between crime rates and military service."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 4 - Estimated Impact of Conscription on Crime Rates "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For table 4 I can replicate the same figures as the authors did. There are some differences, though, that can be attributed to rounding. Percent changes for columns 1, 2, 4, 6, and 7 differ by 0.01 to 0.02 percentage points. Point estimates and standard errors do not differ.\n",
    "The results in columns 1, 2 and 5 to 7 are results from intention-to-treat regressions. The outcome of interest is directly regressed on the instrument. This means that we also consider individuals who were draft eligible but did not actually serve in the military. We do not get the causal effect of military service on crime rates.\n",
    "\n",
    "Thus, the key columns are columns 3 and 4 which identify the average treatment effect of serving in the military on crime rates for compliers. The effect is independet of including origin and province dummies. Due to the randomisation in draft eligibility this can be expected. The effect is small but statistically siginficant at the 1 percent level. The percent change associated with the coefficients of Two-Stage Least Squares (2SLS) including controls is 3.97. This means that the probability of a complier man developing a criminal record increases by 3.97 percent due to serving in the military, on average.\n",
    "\n",
    "The OLS percent change in crime rates due to *conscription* in the online appendix is 3.32. Given we successfully identified the causal effect of conscription through the IV approach, OLS results are biased downwards. The difference is 0.65 percentage points. The direction of bias supports the argument that individuals that on average more often commit crime had a larger propensity of being able to avoid military service. This does make sense if we assume that people who take criminal measures to avoid being drafted are also more likely to commit crimes later in life. Note, that we are observing crimes committed in the adult justice system only."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 4 - Estimated Impact of Conscription on Crime Rates \u001b[0m\n",
      "________________________________________________________________________________________________________________________________\n",
      "Cohort             1958-1962       1958-1962       1958-1962       1958-1962       1929-1965       1929-1955       1958-1965                \n",
      "                         (1)             (2)             (3)             (4)             (5)             (6)             (7)                \n",
      "________________________________________________________________________________________________________________________________\n",
      "Draft Eligible \u001b[1m       0.0018***\u001b[0m\u001b[1m       0.0018***\u001b[0m                                \u001b[1m       0.0006** \u001b[0m\u001b[1m       0.0003   \u001b[0m\u001b[1m       0.0012***\u001b[0m\n",
      "\n",
      "                      0.0006          0.0006                                          0.0003          0.0004          0.0004   \n",
      "\n",
      "Conscription                                   \u001b[1m       0.0027***\u001b[0m\u001b[1m       0.0027***\u001b[0m                                                \n",
      "\n",
      "                                                      0.0008          0.0008                                                   \n",
      "\n",
      "Percent change \u001b[1m         3.94   \u001b[0m\u001b[1m         3.98   \u001b[0m\u001b[1m         3.92   \u001b[0m\u001b[1m         3.97   \u001b[0m\u001b[1m         1.58   \u001b[0m\u001b[1m         0.70   \u001b[0m\u001b[1m         2.61   \u001b[0m\n",
      "\n",
      "Controls                  No             Yes              No             Yes              No              No              No                \n",
      "\n",
      "\n",
      "Observations   \u001b[1m         5000   \u001b[0m\u001b[1m         5000   \u001b[0m\u001b[1m         5000   \u001b[0m\u001b[1m         5000   \u001b[0m\u001b[1m        34904   \u001b[0m\u001b[1m        26976   \u001b[0m\u001b[1m         7928   \u001b[0m\n",
      "\n",
      "Method                   OLS             OLS            2SLS            2SLS             OLS             OLS             OLS                \n",
      "________________________________________________________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation is the cohort-ID number combination. All models\n",
      "include cohort dummies. The models in columns 2 and 4 include controls for origin (naturalized or indigenous) and dis-\n",
      "trict (the country is divided in 24 districts). In 2SLS models, the instrument for Conscription is Draft Eligible. Percent change\n",
      "for 2SLS models is calculated as 100 × Estimate/mean crime rate of draft-ineligible men. For intention-to-treat models, percent  \n",
      "change is reported as 100 × Wald estimate/mean crime rate of draft-ineligible men, where the Wald estimate is calculated as ITT  \n",
      "estitimate/(p1 − p2), where p1 is the probability of serving in the military among those that are draft-eligible, and p2 is the  \n",
      "probability of serving in the military among those that are not draft-eligible (since we do not have information on compliance   \n",
      "rates outside the cohorts of 1958 to 1962, in all cases we use the compliance rates for this period).                            \n",
      "*** Significant at 1 percent level.\n",
      " ** Significant at 5 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_4(df=df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 5 - Estimated Impact of Conscription on Crime Rates for Peacetime versus Wartime Service and 1-Year versus 2-Year Service"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table 5 presents intention-to-treat regression results of eligibility for service in the navy and the impact of the Falklands War. Columns 1 and 3 conduct the analysis for cohorts 1929 to 1965, whereas columns 2 and 4 use cohorts 1958 to 1965.\n",
    "I was able to replicate their results. But there are differences in the significance level. In the original paper *draft eligible* was statistically significant at the 10 percent level in column 1 and significant at the 5 percent level in column 3. The coefficient on draft eligible in column 1 is 0.00001 larger than in the original paper.\n",
    "\n",
    "There is an error in the do.-file provided.\n",
    "> areg crimerate highnumber navy if cohort > 1956 & cohort < 1966, absorb(cohort) robust\n",
    "\n",
    "So, they do the analysis for the cohort 195**7** to 1965, although they state that the cohort of 1957 was not drafted at all. The above figures do not differ from the one's presented in table 5 column 4. This is probably due to the handling of NaNs. If rows containing NaNs are dropped in Stata then the rows satisfying cohort==1957 are dropped entirely, since they have NaNs for *highnumber*. That is why the number of observations does not change either.\n",
    "\n",
    "As Galiani et al. (2011) state on p. 131, from the 202,500 men from cohorts 1962 and 1963 doing military service, only about 12,500 fought in the Falklands War. The coefficient on *navy* is positive and statistically significant.\n",
    "Galiani et al. (2011) use the latter finding to strengthen their hypothesis that military service increases crime rates through forgone labour market experience. The implicit assumption of this interpretation is that navy service has essentially the same impact on crime rates as service in the air force and the army, except for duration. Recall that service in the navy is two years instead of one year compared to army and air force. This assumption may not hold. The relationship between navy service and crime rates could be due to other differences as well. \n",
    "\n",
    "I thought of extending their analysis by including a dummy for air force or army eligibility. However, this data was not provided by the authors. *navy* is just a dummy indicating whether a cohort-ID-group was eligible for service in the navy or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 5 - Estimated Impact of Conscription on Crime Rates for Peacetime\u001b[0m\n",
      "\u001b[1mversus Wartime Service and 1-Year versus 2-Year Service\u001b[0m\n",
      "________________________________________________________________________________\n",
      "Cohort             1929-1965       1958-1965       1929-1965       1958-1965                \n",
      "                         (1)             (2)             (3)             (4)\n",
      "________________________________________________________________________________\n",
      "Draft Eligible \u001b[1m      0.00048   \u001b[0m\u001b[1m      0.00088*  \u001b[0m\u001b[1m      0.00054*  \u001b[0m\u001b[1m      0.00098** \u001b[0m\n",
      "\n",
      "                      0.0003          0.0005          0.0003          0.0004   \n",
      "\n",
      "War Eligible   \u001b[1m       0.0015*  \u001b[0m\u001b[1m       0.0011   \u001b[0m                                \n",
      "\n",
      "                      0.0009          0.0010                                   \n",
      "\n",
      "Navy Eligible                                  \u001b[1m       0.0007** \u001b[0m\u001b[1m       0.0011*  \u001b[0m\n",
      "\n",
      "                                                      0.0003          0.0006   \n",
      "\n",
      "Observations   \u001b[1m        34904   \u001b[0m\u001b[1m         7928   \u001b[0m\u001b[1m        34904   \u001b[0m\u001b[1m         7928   \u001b[0m\n",
      "\n",
      "Method                   OLS             OLS             OLS             OLS\n",
      "________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation\n",
      "is the cohort-ID number combination. War Eligible is a dummy that takes the value\n",
      "of one for the draft eligible from cohorts of 1962 and 1963. Navy Eligible is a  \n",
      "dummy variable that takes the value of one for those ID numbers eligible to serve\n",
      "in the Navy. All models include cohort dummies.\n",
      "** Significant at 5 percent level.\n",
      " * Significant at 10 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_5()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.3 Complementary Results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Galiani et al. (2011) extend their analysis by shedding light on the channel of labour market experience as a driver of higher crime rates for draftees."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 6 - Estimated Impact of Conscription on Crime rates, by Type of Crime"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The authors use data on types of crime at the cohort-ID level to analyse the effect of military service on types of crime and present the results in table 6. Recall that data on types of crime were collected for the period 2000 to 2005. Thus, the long-term effects of conscription on criminal behaviour is assessed.\n",
    "I was not able to replicate the same percent changes as the authors. They differ quite drastically. Point estimates and standard errors coincide exactly.\n",
    "\n",
    "The authors interpret the statistically significant coefficients in column 2 and 7 as further support for their channel of interest. They argue that individuals who can accumulate less labour market experience due to military service have a harder time to get a job that satisfies their needs. Therefore they commit \"white collar\" and \"property\" crimes that yield materialistic gains.\n",
    "A counterargument would be that drug trafficking yields also money. The effect, though, is not statistically significant.\n",
    "\n",
    "Interestingly, the coefficient on *conscription* associated with crimes that involve use of weapons is not statistically significant. There could exist a channel through which crime rates are increased by combat training during military service that involves training in the usage of weapons. Individuals who have served may be more inclined to use weapons since they received training. This hypothesis is not supported by the rsults in column 1 of table 6."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 6 - Estimated Impact of Conscription on Crime rates, by Type of Crime\u001b[0m\n",
      "________________________________________________________________________________________________________________________________\n",
      "Dependent Var.       Weapons        Property   Sexual Attack          Murder          Threat     Drug Traff.    White Collar                \n",
      "Cohort             1958-1962       1958-1962       1958-1962       1958-1962       1958-1962       1958-1962       1958-1962                \n",
      "                         (1)             (2)             (3)             (4)             (5)             (6)             (7)                \n",
      "________________________________________________________________________________________________________________________________\n",
      "Conscription   \u001b[1m      0.00013   \u001b[0m\u001b[1m      0.00082** \u001b[0m\u001b[1m      0.00013   \u001b[0m\u001b[1m     -0.00007   \u001b[0m\u001b[1m      0.00022   \u001b[0m\u001b[1m     -0.00009   \u001b[0m\u001b[1m      0.00064***\u001b[0m\n",
      "\n",
      "                     0.00011         0.00034         0.00009         0.00010         0.00014         0.00014         0.00021   \n",
      "\n",
      "Percent change \u001b[1m         0.14   \u001b[0m\u001b[1m         0.11   \u001b[0m\u001b[1m         0.20   \u001b[0m\u001b[1m        -0.08   \u001b[0m\u001b[1m         0.14   \u001b[0m\u001b[1m        -0.07   \u001b[0m\u001b[1m         0.21   \u001b[0m\n",
      "\n",
      "Observations            5000            5000            5000            5000            5000            5000            5000                \n",
      "Method                  2SLS            2SLS            2SLS            2SLS            2SLS            2SLS            2SLS                \n",
      "________________________________________________________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation is the cohort-ID number combination. All mo-\n",
      "dels include cohort dummies. The instrument for Conscription is Draft eligible. Percent change is calculated as 100*Estimate/mean\n",
      "dependent variable of draft-ineligible men.\n",
      "*** Significant at 1 percent level.\n",
      " ** Significant at 5 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_6()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Table 7 - Estimated Impact of Conscription on Labour Merket Outcomes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table 7 does not coincide with table 7 from the article, since I omitted the OLS intention-to-treat estimates from replication. The table shows 2SLS estimates of the effect of conscription on labour market outcomes, 19 to 26 years after the cohorts participated in military service. They are not statistically significant at any conventional level. This finding is in contrast to the findings in table 6. Although labour market outcomes are no longer significantly different for individuals who served and individuals who did not, the effect on crime rates is still significant for some types of crime. Both types of crime and labour market outcomes were retrieved during roughly the same period. Labour market outcomes were observed for 2003/2004 and the crimes from types of crime were committed in the period 2000 to 2005.\n",
    "\n",
    "This points in the direction that effects of military service on criminal behaviour are longer lasting than the effects on labour market outcomes. There are two caveats of this finding. For the cohorts 1958 to 196**5** the authors find statistically significant effects on earnings and participation rates in the formal labour market (all small and negative), but not for the cohorts 1958 to 1962. This effect could be due to the detrimental effects of wartime military service, since cohorts of 1962 and 1963 served in the Falklands War. This cannot be tested by an IV approach since tretament status is not available for the participants of the Falklands War. The second caveat is that the authors' measures of labour market outcomes could be very noisy (see section 1.1)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable 7 - Estimated Impact of Conscription on Labour Market Outcomes\u001b[0m\n",
      "_______________________________________________________________________________________________\n",
      "Dependent Var. Formal job market   Unemployment rate            Earnings                    \n",
      "_______________________________________________________________________________________________\n",
      "                       1958-1962           1958-1962           1958-1962                    \n",
      "                             (1)                 (2)                 (3)                    \n",
      "_______________________________________________________________________________________________\n",
      "Conscription   \u001b[1m         -0.00222   \u001b[0m\u001b[1m          0.00078   \u001b[0m\u001b[1m         -0.01688   \u001b[0m\n",
      "\n",
      "               \u001b[1m           0.0022   \u001b[0m\u001b[1m           0.0009   \u001b[0m\u001b[1m           0.0106   \u001b[0m\n",
      "\n",
      "Percent change             -0.65                0.41               -0.53   \n",
      "\n",
      "Observations                5000                5000                5000   \n",
      "\n",
      "                            2SLS                2SLS                2SLS                    \n",
      "_______________________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation is the cohort-\n",
      "ID number combination. Participation in the formal job market is as of 2004. Unemployment rates \n",
      "and earnings are as of 2003. Earnings are hourly earnings in Argentine pesos. All models include\n",
      "cohort dummies. The instrument for Conscription is Draft Eligible. Percent change is calculated \n",
      "as 100 × Estimate/mean dependent variable of draft-ineligible men.\n"
     ]
    }
   ],
   "source": [
    "table_7_IV()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.4 Online Appendix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the online appendix I replicate the failure rates of medical examination and the relationship between the conditional probability of serving in the military and the draft lottery numbers. Both are done for the core cohorts of the analysis 1958-1962. They are interesting since they show the behavioural response to the varying probability of being drafted into the military."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "years = list(range(1958, 1963, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Figure A.1 Failure rates of Medical Examination as a Function of Lottery Draft Number by Cohort"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure A.1 documents an interesting pattern. Recall that for the cohorts 1957 and 1958 mandatory military service was suspended. In the first panel of figure A.1 for the cohort of 1958 there is no clear pattern of failure rates. They are quite high for the entire sample. Only for the cohort of 1960 another pattern arises. Now failure rates increase until the proximity of the cutoff and stay on the former high level ever after. One could interpret these patterns as learning. In 1958 all individuals failed the medical examination by purpose. Only the later cohorts learn that individuals with low draft numbers do not need to put any effort into failing medical examination to avoid military service."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "binned_plot(df=df, bin_num=200, ylim=[0.04, 0.115], years = list(range(1958, 1963, 1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Figure A.2 The Relation Between the Conditional Probability of Serving in the Military and Draft Lottery Numbers by Cohorts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In figure A.2 rates of military service conditional on the draft number are presented. It is apparent that, although some volunteer for the military, most do not. Conscription rates increase abruptly at the cutoff numbers. Although this figure is not key it serves as further evidence of the effect of draft eligible on military service, i.e the relevance of the instrument. This further supports that *draft eligible* is a strong instrument for military service."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "figure_A_2(df=df, bin_num=200, ylim=[0, 0.9], years = list(range(1958, 1963, 1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Extensions "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Local IV would have been an appealing extension to this model. Unfortunately this is possible only if one observes the strength of the instrument on selecting into the treatment. This is not the case here. The instrument *Draft Eligible* is a dummy variable, thus there is no way of getting a measure of how large the incentives are to actually selecting into the treatment."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.1 Figures"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Figure B.1 The Relation Between Crime Rates and Draft Eligibility by Cohorts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure B.1 documents crime rates for draft exempt (on the left) and draft eligible groups (on the right). There are some data points that could be outliers and thus compromise the results of the analysis. There is no clear pattern in the data by mere eye-balling, though."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in years:\n",
    "    df[df.cohort == i].plot.scatter(x='highnumber', y='crimerate', title=f'Crime Rates by Draft Eligiblity for Cohort {i}', xticks=(0, 1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Figure B.2 The Relation Between Crime Rates and Conscription Rates by Cohorts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure B.2 presents the relationship between crime rates and conscription shares. Clusters on the left are the ones with a draft number below the cutoff, whereas clusters on the right represent ID-groups with draft numbers above the cutoff. As in the figures above there are no apparent relationships from mere eye-balling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in years:\n",
    "    df[df.cohort == i].plot.scatter(x='sm', y='crimerate', title=f'Crime Rates by Conscription Rates for Cohort {i}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.2 Extended \"Testing\" of Exogeneity of Instrument"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Galinani et al. (2011) provide suggestive evidence of the plausibility of the exclusion restriction of the instrument *draft eligible*. Of course there are no statistical tests that really test the exogeneity of the instrument. They present some evidence that further supports the claim of exogeneity of the instrument."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.2.1 Core cohorts: 1958 - 1962"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Galiani et al. (2011) use the draft-exempt ID groups for each core cohort, i.e. they take only the observations with a draft number below the true cutoff. Then they create fake cutoff numbers, assigning each of the ID groups below the true cutoff to fake control and fake treatment groups based on the fake cutoff number. They test whether crime rates differ among the two fake control and fake treatment group.\n",
    "\n",
    "For the fake cutoff number they use the median of the numbers below the cutoff. Galiani et al. (2011) do not include the results of their tests neither in the paper nor the online appendix.\n",
    "I extend the analysis by varying the fake cutoff number."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table B.2 reports t-test statistics for each cohort an different deciles used for generating the fake cutoff number. The fake cutoffs are computed as follows. Only draft exempt groups are considered, i.e. only cohort-ID groups below the real cutoff number. For these draft exempt groups a fake cutoff is generated. T-test compare the means in crime rates for the cohort-ID groups below and above the decile. Galiani et al. (2011) used the median only. Thus, their results are shown in decile 0.5, since they used the median. Using the median is an arbitrary choice. I extended their analysis by considering a grid of deciles.\n",
    "\n",
    "For the median, there are no statistically significant differences in crime rates. For cohort 1959 though, there are significant differences in crime rates for deciles 0.1 and 0.2. For cohort 1961 there is one significant difference for decile 0.3. Significant results though, share not the same sign. In general, there is no clear pattern in the direction of the difference.\n",
    "\n",
    "Individuals with higher draft numbers have a higher probability of being drafted. Cutoff numbers were announced only after the medical examination. Thus, the lack of a clear pattern of differences in crime rates on the basis of draft numbers on draft exempt ID-groups acts as evidence in favour of the exogeneity of the instrument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable B.2 - Differences in Crime Rates and Fake Eligibility Group by Birth Cohort\u001b[0m\n",
      "____________________________________________________________________________________________________\n",
      "\u001b[1mDifferences by Cohort (fake draft exempt - fake draft eligible)\u001b[0m\n",
      "____________________________________________________________________________________________________\n",
      "Cohort                 1958            1959            1960            1961            1962   \n",
      "\n",
      "Decile     0.1\u001b[1m       0.1818   \u001b[0m\u001b[1m       2.3410** \u001b[0m\u001b[1m      -0.0177   \u001b[0m\u001b[1m      -1.4336   \u001b[0m\u001b[1m      -0.0228   \u001b[0m\n",
      "\n",
      "                     0.8574          0.0243          0.9859          0.1586          0.9819   \n",
      "\n",
      "Decile     0.2\u001b[1m      -0.2419   \u001b[0m\u001b[1m       2.4489** \u001b[0m\u001b[1m      -0.3628   \u001b[0m\u001b[1m      -1.4350   \u001b[0m\u001b[1m      -0.2820   \u001b[0m\n",
      "\n",
      "                     0.8099          0.0160          0.7175          0.1539          0.7786   \n",
      "\n",
      "Decile     0.3\u001b[1m       0.5846   \u001b[0m\u001b[1m       1.5795   \u001b[0m\u001b[1m      -0.6737   \u001b[0m\u001b[1m      -1.9650*  \u001b[0m\u001b[1m       0.4975   \u001b[0m\n",
      "\n",
      "                     0.5603          0.1159          0.5014          0.0506          0.6194   \n",
      "\n",
      "Decile     0.4\u001b[1m       0.7488   \u001b[0m\u001b[1m       0.6690   \u001b[0m\u001b[1m      -0.8631   \u001b[0m\u001b[1m      -0.5604   \u001b[0m\u001b[1m       0.3945   \u001b[0m\n",
      "\n",
      "                     0.4552          0.5041          0.3889          0.5756          0.6935   \n",
      "\n",
      "Decile     0.5\u001b[1m      -0.4574   \u001b[0m\u001b[1m       1.2007   \u001b[0m\u001b[1m      -0.7181   \u001b[0m\u001b[1m      -0.9209   \u001b[0m\u001b[1m      -0.1360   \u001b[0m\n",
      "\n",
      "                     0.6480          0.2308          0.4732          0.3578          0.8919   \n",
      "\n",
      "Decile     0.6\u001b[1m      -0.4041   \u001b[0m\u001b[1m       1.1454   \u001b[0m\u001b[1m       0.2865   \u001b[0m\u001b[1m      -1.1303   \u001b[0m\u001b[1m      -0.2261   \u001b[0m\n",
      "\n",
      "                     0.6867          0.2531          0.7747          0.2593          0.8213   \n",
      "\n",
      "Decile     0.7\u001b[1m      -0.6472   \u001b[0m\u001b[1m       0.0171   \u001b[0m\u001b[1m      -0.2856   \u001b[0m\u001b[1m      -0.7896   \u001b[0m\u001b[1m       0.3604   \u001b[0m\n",
      "\n",
      "                     0.5188          0.9864          0.7755          0.4308          0.7190   \n",
      "\n",
      "Decile     0.8\u001b[1m      -0.3472   \u001b[0m\u001b[1m      -0.5452   \u001b[0m\u001b[1m      -0.2667   \u001b[0m\u001b[1m       1.2387   \u001b[0m\u001b[1m       0.6367   \u001b[0m\n",
      "\n",
      "                     0.7297          0.5867          0.7902          0.2183          0.5257   \n",
      "\n",
      "Decile     0.9\u001b[1m       0.2559   \u001b[0m\u001b[1m       0.9490   \u001b[0m\u001b[1m      -0.1248   \u001b[0m\u001b[1m       1.4031   \u001b[0m\u001b[1m       0.2898   \u001b[0m\n",
      "\n",
      "                     0.8004          0.3481          0.9013          0.1682          0.7736   \n",
      "\n",
      "____________________________________________________________________________________________________\n",
      "Notes: P-values are shown below test statistics. The level of observation is the cohort-ID number \n",
      "combination. Fake eligibility status computed by using only real draft exempt cohort-ID groups.\n",
      "Deciles refer to the different cutoffs generated.\n",
      "** Significant at 5 percent level.\n",
      " * Significant at 10 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_test_fake_cutoff_1(df=df, draft_status=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3.2.2 Cohort 1976 "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table B.3 presents t-test statistics and p-values. The differences in means for pseudo draft eligible and draft ineligible individuals are tested for statistical significance. Galiani et al. (2011) use the fact that the cohort of 1976 was assigned draft numbers but never actually served in the military. The draft numbers were assigned on May 27, 1994, but on August 31, 1994 conscription was abolished (ibid, p. 131). The authors employ the cutoff number of the cohort of 1975 and find no effect on crime rates. I extend their analysis by using a range of cutoff numbers. No statistically significant differences in crime rates occur when applying a random cutoff. In years with conscription, individuals with a high draft number could expect to be drafted. This could have detrimental effects on labour market opportunities, since employers are not willing to invest in an individual that is likely to be drafted. The cohort of 1976 did face the draft lottery but was not drafted at all. If there were effects on crime rates other than the effect through military service, cohort of 1976 should have a crime rate significantly different for high vs low draft numbers. Since this is not the case, this provides suggestive evidence for the exogeneity of the instrument.\n",
    "The authors argue that this ruled out all concerns that the instrument has an independent effect on crime rates.\n",
    "\n",
    "Garano (2010) states that conscription was abolished because a soldier was murdered on May 6, 1994 (p. 174). It was abolished by presidential edict in August 1994. One could argue that this was not a sufficiently long period for independent detrimental labour market effects to occur. Moreover, since the murder on May 6, 1994 is stated as a reason for the abolishment of conscription, individuals may have anticipated that conscription could be abolished. However, the anticipation of the policy change relies on many assumptions and this line of argument can be refuted. Thus, the claim of exogeneity of the instrument can be defended."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable B.3 - Differences in Crime Rates and Fake Eligibility Group for Cohort 1976\u001b[0m\n",
      "____________________________________________________________________________________________________\n",
      "\u001b[1mDifferences (fake draft exempt - fake draft eligible)\u001b[0m\n",
      "____________________________________________________________________________________________________\n",
      "Cohort                 1976   \n",
      "\n",
      "Decile     0.1\u001b[1m      -0.9195   \u001b[0m\n",
      "\n",
      "                     0.3596   \n",
      "\n",
      "Decile     0.2\u001b[1m      -1.4758   \u001b[0m\n",
      "\n",
      "                     0.1411   \n",
      "\n",
      "Decile     0.3\u001b[1m      -1.2678   \u001b[0m\n",
      "\n",
      "                     0.2054   \n",
      "\n",
      "Decile     0.4\u001b[1m      -0.4686   \u001b[0m\n",
      "\n",
      "                     0.6394   \n",
      "\n",
      "Decile     0.5\u001b[1m      -0.6568   \u001b[0m\n",
      "\n",
      "                     0.5115   \n",
      "\n",
      "Decile     0.6\u001b[1m      -0.4335   \u001b[0m\n",
      "\n",
      "                     0.6648   \n",
      "\n",
      "Decile     0.7\u001b[1m      -0.7631   \u001b[0m\n",
      "\n",
      "                     0.4457   \n",
      "\n",
      "Decile     0.8\u001b[1m      -1.0518   \u001b[0m\n",
      "\n",
      "                     0.2937   \n",
      "\n",
      "Decile     0.9\u001b[1m      -0.7436   \u001b[0m\n",
      "\n",
      "                     0.4586   \n",
      "\n",
      "____________________________________________________________________________________________________\n",
      "Notes: P-values are shown below test statistics. The level of observation is the cohort-ID number \n",
      "combination. Fake eligibility status computed by using only real draft exempt cohort-ID groups.\n",
      "Deciles refer to the different cutoffs generated.\n",
      "** Significant at 5 percent level.\n",
      " * Significant at 10 percent level.\n"
     ]
    }
   ],
   "source": [
    "table_test_fake_cutoff_2(df=df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.2 Robustness Checks "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Table E.4: 2SLS for each Cohort Separately"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Table E.4 presents 2SLS estimates of military service. I conduct the analysis of table 4 column (3) for each cohort from 1958-1962. The effect of military service could vary with the cohorts considered. In table 4 of Galiani et al. (2011) this is taken care of by including cohort dummies. I extend their analysis since the 2SLS estimates are the key piece of evidence that military service has a detrimental overall effect on crime rates.\n",
    "\n",
    "My results of the effect of military service are similar to those presented by the authors. It is apparent though, that the coefficient on *conscription* varies over cohorts. Moreover, the effect is not statistically significant for all cohorts. Only for the cohort of 1962 the coefficient associated with *conscription* is statistically significant at the 5 percent level. This could be due to wartime conscription. Percent change reacts accordingly since it depends on the point estimates. The magnitudes of the effect are comparable to the pooled scenario but with some fluctuations. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mTable E.4 - Estimated Impact of Conscription on Crime rates, for each Core Cohort Separately\u001b[0m\n",
      "\u001b[1m(Dependent Variable: Crime Rate)\u001b[0m\n",
      "_________________________________________________________________________________________________\n",
      "Cohort                  1958            1959            1960            1961            1962   \n",
      "\n",
      "                         (1)             (2)             (3)             (4)             (5)                \n",
      "_________________________________________________________________________________________________\n",
      "Conscription   \u001b[1m      0.00129   \u001b[0m\u001b[1m      0.00312   \u001b[0m\u001b[1m      0.00326*  \u001b[0m\u001b[1m      0.00152   \u001b[0m\u001b[1m      0.00375** \u001b[0m\n",
      "\n",
      "                     0.00249         0.00192         0.00177         0.00167         0.00181   \n",
      "\n",
      "Percent change \u001b[1m         1.89   \u001b[0m\u001b[1m         4.68   \u001b[0m\u001b[1m         4.82   \u001b[0m\u001b[1m         2.20   \u001b[0m\u001b[1m         5.44   \u001b[0m\n",
      "\n",
      "Observations            1000            1000            1000            1000            1000                \n",
      "Method                  2SLS            2SLS            2SLS            2SLS            2SLS                \n",
      "_________________________________________________________________________________________________\n",
      "Notes: Robust standard errors are shown below estimates. The level of observation is the cohort-\n",
      "ID number combination. The instrument for Conscription is Draft Eligible. Percent change is cal-\n",
      "culated as 100 × Estimate/mean crime rate of draft-ineligible men.\n",
      "** Significant at 5 percent level.\n",
      " * Significant at 10 percent level.\n"
     ]
    }
   ],
   "source": [
    "extension_table_4()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Critical Assessment of Quality "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Galiani et al. (2011) use exogenous variation in draft eligibility status to construct an instrument for military service. They successfully identify the average effect of military service on crime rates for draft-lottery compliers. They further investigate the effect of military service on labour market outcomes and do not find a large impact of military service. \n",
    "\n",
    "There are some caveats, though.\n",
    "The same criticism by Rosenzweig & Wolpin (2000) applied to Angrist (1990) could also be applicable to this study if there is a direct effect from draft numbers to labour market opportunities. However, the authors present convincing suggestive evidence in favour of the plausibility of the exclusion restriction.\n",
    "\n",
    "One identifying assumption of the LATE is monotonicity, which means that there are no defiers in the sample. The authors do not argue why defiers are improbable. This amounts to a mere assumption that is not discussed. However, the monotonicity assumption is plausible.\n",
    "\n",
    "The authors consider a deeper mechanism of conscription on crime rates and investigate the channel of labour market outcomes.\n",
    "The authors state that rates of white collar crimes and crimes against property should be larger if conscription had a detrimental effect on labour market outcomes (Galiani et al. 2011, p. 133). The same should apply to drug trafficking, though, since it also serves a pecuniary purpose, I would argue. The coefficient on *conscription* in table 6 column 6 is not statistcially significant, though.\n",
    "\n",
    "Also the effect of eligibility for service in the navy is questionable as support for the labour market channel. There could be other detrimental effects of serving in the navy. The argumentation of the authors relies on the implicit assumption that service in the navy is the same as air force and army on subsequent criminal behaviour except for duration.\n",
    "\n",
    "Another line of criticism concerns their measures of labour market outcomes that rely on self-reported occupation in the ballot registry. From these they infer unemployment rates and income. This could be a noisy measure since it relies on self-reported data. Furthermore, individuals begin military service at the age of 21. With 21 years they may have already an occupation. Thus, the effects on labour market prospects could be small.\n",
    "\n",
    "I could replicate most of the tables and figures presented in the paper, with some exceptions. The key pieces of evidence are replicated and correctly represent the figures in the original paper.\n",
    "\n",
    "In summary the authors succeed in identifying a relevant instrument and present strong suggestive evidence for the claim of exogeneity of the instrument."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## References "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Angrist, Joshua D. 1990. \"Lifetime Earnings and the Vietnam Era Draft Lottery: Evidence from Social Security Administrative Records.\". *American Economic Review*, 80 (3): 313-336.\n",
    "\n",
    "Galiani, Sebastian, Martín A. Rossi, and Ernesto Schargrodsky. 2011. \"Conscription and Crime: Evidence from the Argentine Draft Lottery.\" *American Economic Journal: Applied Economics*, 3 (2): 119-36.\n",
    "[DOI: 10.1257/app.3.2.119](http://dx.doi.org/10.1257/app.3.2.119)\n",
    "\n",
    "Garano, Santiago. 2010. \"The Opposition Front against Compulsory Military Service: The Conscription Debate and Human-Rights Activism in Post-Conscription Debate and Human-Rights Activism in Post-dictatorship Argentina dictatorship Argent.\" *Genocide Studies and Prevention: An International Journal*, 5 (2): 174-190.\n",
    "\n",
    "Imbens, Guido and Joshua Angrist. 1994. \"Identification and Estimation of Local Average Treatment Effects.\" *Econometrica*, 62 (2): 467-75.\n",
    "            \n",
    "Pearl, J. 2014. \"Causality.\" Cambridge, England: *Cambridge University Press*.\n",
    "\n",
    "Rosenzweig, Mark, R., and Kenneth I. Wolpin. 2000. \"Natural \"Natural Experiments\" in Economics.\" *Journal of Economic Literature*, 38 (4): 827-874.\n",
    "\n",
    "Winship, C., and S. L. Morgan. 2007. \"Counterfactuals and causal inference: Methods and principles for social research.\" Cambridge, England: *Cambridge University Press*."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}