The Impact of Legalized Abortion on Crime

May 4, 2012 § Leave a comment

The Impact of Legalized Abortion on Crime is a highly debated and controversial paper, by authors Steven Levitt and John Donohue, in which they defend their hypothesis that the legalization of abortion in the 1970s has led to large reductions in crime rates in the 1990s. “Unintended pregnancies are associated with poorer parental care, greater smoking and drinking during pregnancy, and lower birth weights. Consequently, the life chances of children who are born only because their mothers could not have an abortion are considerably dampened relative to babies who were wanted at the time of conception” (pg. 414). Therefore, one can conclude that the parents of unwanted children neglect and treat their children negatively, often leading to these children to turn to criminal activity. 

Levitt and Donohue found that, after a 20-year lag, legalized abortion would cause the crime rates to drop – “an increase of 100 abortions per 1000 live births reduces a cohort’s crime by roughly 10 percent” (pg. 414).  Levitt and Donohue found that the crime rate would have been 15-25 percent higher without legalized abortions in 1997.

 The Levitt/Donohue essay predicts that the benefits of legalized abortion would be enormous, possibly accounting for as much as a 50% reduction in crime with social benefit around $30 billion dollars annually.

 The Impact Of Legalized Abortion On Crime: Comment, by Foote and Goetz make three important observations on Donohue and Levitt’s paper: 1) their regression was constructed incorrectly, by comparing different “age cohorts within the same state and year”; 2) after a correction of this error through a per capita variable, the same regression yields less statistically significant results; 3) failure to use robust data, which would account for differing state trend crime rates. Through re-examination of the previous work, in addition to improving upon the foundation that had already been established, Foote and Goetz disprove the “compelling evidence that abortion has a selection effect on crime (pg 422).

Chapter 4 of Freakonomics “Where Have All the Criminals Gone?” discusses the impact the legalization of abortion has had on crime rates in the United States. Although there are many examples that are similar to the Levitt/Donohue paper supporting the causality of abortion and lowered crime rates, one needs to consider the fallacy of cause and effect: just because decreased crime rates occurred decades after abortion was legalized, does not mean they are directly linked, especially as there are other factors that may be coming into play. This is demonstrated by the analysis by Foote and Goetz: “there may be some reason that New York had both a higher crime rate and higher abortion rate than Utah had before 1985. Perhaps, New York’s urban density, its wealth, its demographic structure, or some other aspects of its culture offers New Yorkers more chances for interpersonal connections that lead to more crimes and to more unwanted pregnancies (pg. 417).

 The analysis by Foote and Goetz goes along the same lines as my blog from last week where I questioned the assumption that the increased crime rate was a direct consequence of more unwanted children.  I agree that there may be other factors than neglected and mistreated children such as higher density of population so more competition for limited social resources or perhaps larger class sizes so children do not receive the same attention at schools.  Or perhaps larger families made it more difficult for the families to make ends meet?  It would be interesting to test some of these other factors to see if it was really “unwanted children” or some other factor.






Where Have All the Criminals Gone?

April 27, 2012 § Leave a comment

Chapter 4 of FreakonomicsWhere Have All the Criminals Gone?” discusses the impact the legalization of abortion has had on crime rates in the United States. When abortion was illegal a significant number of unwanted children were being born, and there was a higher incidence of illegal activity among those children. Once abortion was legalized, fewer unwanted children were born, and crime rates dropped significantly.

The chapter does a convincing job of showing how the link between the legalization of abortion and crime rates is more than a simple correlation, but a case of causality. The effect of abortion on crime rates is made clear by comparing statistics in various states in which legalized abortion occurred earlier than in the rest of the country. In the period from 1988 – 1994, in states which had legalized abortion, violent crimes were down 13% compared to other states that had not legalized abortion as early. And from 1994-1997, murder rates were down a staggering 23% in states with legalized abortion, comparison to other states. 

If that was not convincing enough, the authors further evaluated the correlation between the states’ abortion rates and crime rates. They found a definite correlation between both. The states with the highest abortion rates in the 1970s demonstrated the largest decreases in crime in the 1990s and vice versa. The chapter also showed how states with high abortion rates have seen decreases in crime rates of about 30%. 

One possible area I would question the authors would be over their assumption that the increased crime rate was a direct consequence of more unwanted children who were more likely to engage in illegal activity.  Could higher crime rates be driven by higher overall birth rates so more competition for limited social resources?  Perhaps larger class size or larger families were a factor, where children did not receive enough nurturing or attention at school or in the home?  There is also an economic issue with larger families, in that it was perhaps more difficult for these families to make ends meet, which may have led children to turn to criminal activity as a solution to money problems?  It would be interesting to explore some of these other factors to see if the key reason was really “unwanted children” or some other consequences of the legality or illegality of abortion.

Life Expectancy

April 13, 2012 § 2 Comments

People are living longer.   Of course, this is a good thing, but it represents a major demographic shift.  By 2030, 45% of Japan’s population will be over the age of 60. Germany and Italy are not far behind and even China, with its one-child policy, will face this issue within the next few decades.  The shift towards an aging population has huge implications for national policies in healthcare, taxation, social security, production capacity, and other economic and social areas. Central to understanding the aging demographic is the life expectancy of a given population and how that may change over time.  My initial premise was that the increased wealth of a nation would mean an increased life span. The most interesting finding from my analysis was that while increased wealth is a major driver for increased life expectancy in underdeveloped/developing markets, once a certain wealth level is attained (in more developed markets) factors other than increased wealth begin to play a bigger role: distribution of income for example or cultural peculiarities, such as alcohol consumption in Russia, or prevalence of smoking become the main differentiator.  Governments should act on these drivers and help their people to achieve longer life.

Big Families

March 29, 2012 § Leave a comment

Chapter 5 of Poor Economics by Abhijit V. Banerjee and Esther Duflo seeks to answer the questions of why and whether the poor really desire large families.  Clearly this is an essential issue to be concerned with because for centuries, population growth has troubled policy makers and academics.

It is not simply an issue of lack of access to contraception; various other elements play a role in family size: economic considerations, family dynamics, and social norms, to name a few.  The chapter suggests “the most effective population policy might therefore be to make it unnecessary to have so many children” (PE, pg. 128). These policies may include effective social safety nets (health insurance and old age pension schemes, for example), or various other financial programs that would allow individuals to save for the later portion of their lives.

One statistic that I found interesting was from Matlab, Bangladesh, where between 1977-1996, half of the 141 villages in the area received intensive family planning outreach programs. This availability to family planning led to a 1.2 reduction in children from women between the ages of 30 – 35.

I would want to test the hypothesis that access to family planning results in a reduction in the average amount of children conceived by women.  The one issue that I would see with testing this hypothesis would be finding a variable that would measure the level of family planning present in a location.

Regression model: AvrgChildren = B1 + B2(FamilyPlanning)+ B3(D)+ U

B1: intercept; average amount of children when no family planning is present.

B2: slope; which measures the rate of change in average children for an additional unit change in Family planning.

U: error term

One possibility of an explanatory dummy variable could be whether or not a location is considered a patriarchal society (D: 1=patriarchal and 0=not patriarchal). This would allow us to examine whether or not the average amount of children is affected by male dominated societies, in other words, whether the mean values of children are different in patriarchal and not patriarchal societies. In order to determine whether the values are statistically significant would require running regressions as well as t-statistic tests.

A Statistical Analysis Of Life Expectancy

March 9, 2012 § Leave a comment

          A Statistical Analysis Of Life Expectancy Across Countries Using Multiple Regression by Professor Tony Smith has given me several new insights as to how to approach my research project on the positive relationship between GDP per capita and life expectancy.  The paper found that, in the models used, life expectancy was positively correlated to GNP per capita, population growth, fertility, enrollment, and access to safe water.  Akin to my thesis, Professor Smith hypothesized “the wealthier a country is, the more money its citizen will have to spend on healthcare, and correspondingly, the more likely they are to have time for leisurely activities and exercise” (pg. 4).

I believe it might be worth including additional variants of GDP in my regression model, those being: purchasing power equivalent dollars, GDP growth (to analyze the significance of absolute amount), rate of increased personal wealth, comparable purchasing ability. This would be in addition to inflation rates, as high inflation shows economic instability, and may have a negative effect on the stress levels and overall health of a population.

The paper also dealt with an important issue in plotting the distribution of life expectancies, which will yield two distinctly different distributions. But by including a dummy variable, that of Country Development (Developed Counties =1, Less developed = 0), this issue can be solved easily.


Argentina and Life Expectancy

March 2, 2012 § 2 Comments

The article “Argentina Losing Regional Leadership Position in Health,” by Marcela Valente, has helped me appreciate the significance of incorporating a measure of the level of equality (Gini coefficient) into my regression model.

Argentina has the largest expenditure in healthcare among Latin American countries, which has allowed them to build up significant amounts of infrastructure, acquire talented doctors and other health professionals, and various other health related resources.  With that in mind, how is it that within a few years, other countries, such as Brazil, will boast infant mortality rates superior to Argentina?

There is a need for sizeable expenditure in healthcare (Argentina currently spends 10% of its GDP), making it possible to effectively treat the majority of the population. However, in Argentina, “there is a lack of proactive health policies,” which leads to an inefficient health care system. The state needs to begin to utilize its resources to bridge the wealth inequality gap in the provinces. Clearly, their current policies are doing the opposite. For example: the country’s average life expectancy at birth is 75.4 years, but in the northeastern province of Chaco, current life expectancy is just 69.9 years.  Furthermore, in the Argentinian capital of Buenos Aires, the under-five mortality rate is under 8 per 1,000 births, whereas in the northern city of Formosa, it is a shocking 25 per 1,000 births.

Without policies and initiatives to close the wealth gap between rich and poor, the effectiveness of the healthcare system in Argentina will continue to decline.

The Economics of Moneyball

February 27, 2012 § 1 Comment

Bennett Miller’s film Moneyball (2011) recalls the story of Oakland Athletics (the A’s) General Manager Billy Beane, as he struggled to build a winning team on the lowest budget for players’ salaries in Major League baseball. Knowing he needed to find some competitive advantage without the budget to simply buy it, he turns the player selection process completely on its head. He employs statistical data analysis to rank candidates on their desirability for the A’s roster, calculating a numerical value for each player and getting the best ability with his limited budget.

This compelling film about a small team that went against all odds also manages to cover multiple economic topics: statistical analysis, entrepreneurship, market efficiency, and constrained maximizations problems.

The film chronicles a budget constraint problem:  the A’s were limited to a $40 million budget and were competing against teams with budgets in the $125 million range, such as the New York Yankees.  Billy Beane understood that the only way the A’s could succeed in a system where so much depended on wealth (which created a incredibly unfair advantage) was to innovate to find new ways to win.  By incorporating statistical analysis, the A’s could create a competitive team within their budget. The analysis allowed them to avoid biases in selection that would lead to over-valued, underperforming players.

Moneyball provides a definition of the economic term coined by Schumpeter “creative destruction” – which basically involves a process of radical innovation to revolutionize the existing approach and replace old and inefficient methods with superior and more effective ones.

Business enterprises and other institutions are dogged by inefficiencies so embedded that we consider them business as usual. For example, firms may find it difficult to be flexible with investments or initiatives that would allow them to fully tap the resources their front-line, low paid employees may have to offer. The Moneyball story shows that it is possible to break this inefficient cycle but that it may require a difficult competitive situation before even being considered. Companies and governments could reap tremendous advantage by getting ideas from all parts of their organization and encouraging change to create better products and greater productivity.  GE does this with their “workout” approach which lets large groups of employees spend a day brainstorming on ideas to address a business problem and then present their ideas to top management for an immediate “yes” or “no” – who better to ask then the people who face the issue every day?