The article “The Impact of Legalized Abortion on Crime” by Donohue and Levitt supports the thesis presented in chapter four of Freakonomics, that the legalization of abortion was the reason for lower crime rates in the 1990’s. The article provides a detailed statistical analysis of the hypothesis offered in Freakonomics. Donohue and Levitt similarly to Freakonoimcs discuss the commonly cited causes for the lowered 1990 crime rates only to dismiss them and accept that Roe v. Wade was actually the cause for the drop in crime.

I was concerned that Freakonomics was not considering the possible difference in crime rates following the legalization abortions outside of major cities. However, this is discussed within the article and Donohue and Levitt find that the crime rate drop was consistent in rural areas. The only geographical areas that were not similarly impacted by the legalization of abortion were the five states where abortion was already legal.

Another issue that was worrisome in the Freakonomics chapter was how the authors discounted the effects of changes in policing, drug trade, and economy. Such changes would logical drastically effect crime rates. Donohue and Levitt also discuss such national trends finding that these trends “have been ongoing over two decades, and thus cannot plausibly explain the recent abrupt improvement in crime” (Donohue 380). However, I still take issue in such a dismissal. There could be a “tipping point” effect, such that these ongoing changes were not effective until there was a cumulative effect up to a certain point that resulted in a drastic change.

The second article “The Impact of Legalized Abortion on Crime: Comment” by Foote and Goetz finds fault with the statistical analysis of Donohue and Levitt. Foote and Goetz believe that a more accurate statistical regression would use arrest per capita term instead of total number of arrests particular to a cohort of young people. Using this model the relationship between the legalization of abortion and the decline in crime is not as strong.

The articles show that depending on the constraints and variables you use in your statistical regression you can conclude completely different relationships exist.  The possibility of such divergent statistical results shows that statistically proven facts may not actually be as accurate as readers may suspect. Therefore, you should always be skeptical of peoples work and conclusions.

I think that regardless of the accuracy of either regression results Freakonomics presents interesting ideas that are beneficial for any reader to consider.

The authors of Freakonomics determine that the decline in crime during the 1990s was a result of the legalization of abortion following Roe v. Wade. This conclusion is well backed up by the authors who first disprove many commonly accepted reasons for declining crime such as increased policing, economic development, and changing drug situations. The chapter also discusses the international effects of abortion, strengthening the already strong argument for abortion’s effect on diminished crime.

However, I do have some concerns with the authors sweeping assignment of such crime correlation and wonder if such a strong relationship would exist outside of the main cities discussed, such as in smaller towns where crime rates are already lower. The shrunken crime rates in New York and other cities is only  pronounced because large cities tend to have high crime rates.  But people receive abortions all over the country, even in areas where the crime rate is not high. Therefore, such areas would not experience such a dramatic decline in crime following the legalization of abortion. I think that the data is autocorreleated because of regional difference.

There could also be problems of heterscedasticity because not all the women getting abortion are women who are low-income and single mothers. There may be married women who have some other personal issue that causes them to get an abortion. The conclusion that the children of these mothers would have become criminals is not consistent with the reasoning for why children of low income or single mothers become criminals. There may also be time lag problems from when abortion was legalized to when the effects of  abortion are actually felt that are not be accurately accounted for.

Further, the conclusion that the authors reach about abortion and low-income families is not thoughtful and should include a discussion of the importance of social programs for single mothers and low-income families.

This paper evaluates the effects of third party signatures, a proxy for third party intervention, on peace agreements following armed civil conflicts. People often believe that involving an outside party to help mediate a conflict will results in a quicker resolutions, for that reason countries like the United States and boarding countries typically intervene to promote peace. International organizations, such as the United Nations, also typically try to resolve conflicts and mediate peace. The actual effect of such intervention is empirically positive and negative. On one hand a third party can expedite peace by pressuring both parties and helping to create a social atmosphere of mutual cooperation. However, third parties also have and promote their own agendas complicating matters. Consequently the effects of intervention are empirically ambiguous. After statistical analysis it becomes clear that average effects of intervention are twofold. First, the presence of a third party signature on a peace agreement increases the likelihood that a peace agreement will fail. Secondly, the presence of a third party signature on a peace agreement increases the time period between when the peace agreement was signed to when a peace agreement fails for all failed peace agreements. 

Chapter 5 of Poor Economics discusses the effects of family size and population growth on an economy.  The economic growth in a country and of a family is partially determined by family planning choices including education and birth control availability and social phenomena including disease control, cultural evolution, and government programs.

An interesting statistic in the chapter is that South Africa could be 5.6 percent richer in the future due to the reduction in population from the HIV/AIDS epidemic (107). This statistic reflects the theory of some experts that countries with higher fertility rates are poorer. Yet, some studies find that historically regions with higher populations grow faster than the rest of the world. Therefore, my hypothesis is that the decline in population due to HIV/AIDS will result in increased wealth in rural areas of South Africa. Alternatively, the decline in population due to HIV/AIDS in cities will result in decreased wealth and innovation.

To test this hypothesis you would need to develop a regression model that test the effect of the HIV/AIDS infection rate and the fertility rate on the wealth of a rural population. Such a model could be set up as shown below.

            Y1=a+ bX1+ bX2+ m

            Y1= wealth in rural areas

            X1= percent of population infected with HIV/AIDS

            X2=death rate

A dummy variable that might be useful to include in this regression is gender, which could be coded 0 for women and 1 for men. According to poor economics young men are less likely to be infected with HIV/AIDS than women of the same age therefore this dummy would account for such variation. The sign on b (positive or negative) for this dummy variable would tell us if the gender of the percent of the population infected with HIV/AIDS effects the wealth in rural areas differently. 

Armed Conflict and Peace Agreements by Lotta Harbom summarizes notable conflicts after World War II up until 2005 and general trends in conflict duration during certain years. Harbom finds that conflict has dropped sense WWII. However, categorization of conflict may be artificially deflated and inaccurate because only conflicts with at least 25 deaths are reported as conflict. Therefore, if a country in conflict has a year of relative peace, less than 25 deaths, they are no longer reported as a country in conflict in the dataset used. Secondly, the Harbom analyzes common political provisions in full peace agreements, partial peace agreements, and peace process agreements.

The article does not suggest any issues that I might have regarding the assumptions of the classical linear regression model. However, because the author uses the same data I will use to evaluate third party effects on peace agreements the author’s evaluation of the shortcomings of the data are very interesting. Harbom highlights that when the death toll drops below 25 the country is no longer reported as a country in conflict. In my regression I will have to control for this because such reporting would inflate the occurrence of peace when there is no third party and when there is no formal peace agreement. This will affect my regression results. Countries that are not reported as in conflict because of low intensity will likely fall back into intense conflict, 25 deaths or above, making it appear that peace was unsuccessful when it was never actually reached. This would lower the relative importance of having a third party signature on a peace agreement because countries would falsely appear to be reaching peace whenever intensity drops. On the other hand is would over report the importance of having a third party signature on a peace agreement when considering duration of peace after conflict resolution. Such reporting would also falsely lower the duration of peace when there is no third party.

The article also discusses different kinds of peace agreements and their relative success.  I had not considered that different types of peace agreements would have different levels of success. Therefore, this is something I will need to control for. I can adjust my model by determining which of the three types of peace agreements (full, partial, process) are most likely to have third party signatures and what, if any, effects third party signature have on the success of that particular peace agreements relative to the other two. I can do this by running in individual regressions for the three types of agreements.

Article Citation and Link:

Harbom, Lotta, Stina Hogbladh, and Peter Wallensteen. “Armed Conflict And Peace Agreements.” Journal Of Peace Research 43.5 (2006): 617-631. EconLit. Web. 9 Mar. 2012.

The article “EU: Syrian Leadership Will Be Held Accountable” by Robert Wielaard discuss the difficulties of Syrian intervene given the current Syrian civil war and atrocities committed by the Syrian government. For example, The EU has promised to step up sanctions and document the war crimes so that reparations can be made upon peace. On the other hand, the United Nations has not step in or condemned the atrocities because China and Russia do not support such actions. The political implications of third party intervention can be very contentious when members of international organizations have economics stakes in a country. Therefore, peace agreements can be tricky to reach and once they are reached there effects can be minimal.

My paper hypotheses that peace agreements that include third party signatures are less effective than peace agreement between warring parties that do not include third party signatures. This article shows that within third parties dissention can diminish their authoritative power and even be apologetic to the offending warring party. China and Russia are hindering the UN’s ability to be an authoritative condemning force capable of encourage Syrian government to consider discussing peace resolutions.

After the conflict in Syria has ended and conditions for peace are being negotiated, I believe that UN intervention will be undermined due to China and Russia. Even though the majority of the international community has condemned Syrian violence, they have no real power to intervene and stop it unless they invest troops and lots of money. This intervention could result in a temporary suppression of violence but not a real resolution.


Article Link:

Chapter four of Poor Economics discusses education in developing countries and the most effective way to establish schools. First the authors present two models for efficiently building schools. One model is supply focused, build a school and students will come. The second model is demand focused, wait until the citizens demand a school and then build it. These two models offer different methodology for the most effective way to get poor children to go to and stay in school. The chapter also discusses incentive systems for education and the implications of economic background, parental influence, and social expectations on student achievement.

Poor Economics focused on developing and very poor countries. However, even in America there is a disparity between the education the rich and the poor receive and obstacles to providing quality education to the poor. The New York Times article “Education Gap Grows Between Rich and Poor, Studies Say” by Sabrina Tavernise discusses the disparity between rich and poor student achievement based on the initial endowment of family resources, a phenomena that is also discussed in Poor Economics.  The educational problems faced by poor people internationally and poor people in the US are not that different. They have lower test scores, less education, and attend below average schools.

Both the article and the book attribute some of the difference in educational success to parental influence. The article cites that in 2007 American upper-income families were spending nine times as much on educating their children than lower-income families. The upper-income children as a result were scoring higher on standardized tests and completing more years of schooling. Tavernise concludes that parents investment in their children starting at an early age is a key factor in success.

Similarly, in poorer countries parents chose the level of investment they make in their children. This determines the level of education and encouragement a child will receive from their families. In the very poor countries discussed in the chapter there are financial incentive systems to try to encourage such investment. This also happens in America. Laws, scholarships, and tax breaks for parents with dependent children are all incentives for parents to invest in their children.

Taversine also discusses the social factors that contribute to the differences between the rich and the poor. For example difference in everyday life experiences such as time spent at cultural events and shopping malls make upper-income children more likely to be successful. Such social differences are similar to the confidence problem face by lower caste people discussed in Poor Economics. When people are aware of their shortcomings in social or educational situations they are less likely to excel and more likely to discount themselves and their intelligence.

It is very interesting and disheartening that similar educational obstacles exist in the wealthiest country in the world and in the poorest and most socially rigid countries. It shows that the educational solutions offered in Poor Economics may be insufficient to stop inequality and educational obstacles will perpetuate even in the ideal environment.



Article Link:


Thesis: Peace agreements in armed interstate conflicts are more likely to be reached with the intervention of a third-party. Third-party intervention has a positive effect on ending and encouraging peace between warring parties.

Topic: I want to examine what social, political, and economic issues lead to peace agreements. Primarily analyzing what kind of effects third-parties have on either speeding up the peace process or exasperating conflict and what factors lead to third-party intervention.


1.     UCDP/PRIO Armed Conflict Dataset Version 4-2011, 1946-2010:

Description: This is dataset has information on armed conflict where one of the engaging parties is the government of the state from the 1946 to 2010.

2.     UCDP Peace Agreement Dataset Version 1.p, 1989-2005:

Description: This dataset has information on all peace agreements signed between two warring parties in armed conflict from 1989-2005. It also includes information on third parties to the agreements.


Being socially aware of war and short-term conflicts is very important for developing a framework and context of the world we live in. There is always some kind of civil war or short-term violent conflict and understanding the results of similar past conflicts can help shape how people and governments intervene to shape the outcome of present conflicts for the better. For example, recently there have been very important short-term conflicts such as the revolutions in Egypt, Syria, and Libyan. Reflecting on similar historical revolutions can help determine how these conflicts will progress and how the internationally community can facilitate peace.

Historically the U.S. frequently intervened in conflicts in the Middle East, but chosen to shy away from conflicts in Africa.  Therefore, I want to see if there is a relationship between third-party intervention and resource availability of the country in conflict. The type of intervention may also be related to this. For example the U.S. maybe more likely to intervene when countries have plentiful or strategic resources, while the U.N. maybe more likely to intervene in conflicts with higher intensities and casualties. I think it will also be very interesting to look at the long-term effects of third-party intervention. Does intervention have any effect in preventing future conflict or promoting long-term peace in the area compared to countries that reached peace independently?

There are also policy implications that motivate intervention, which I discussed briefly in the above paragraph. For example, time period may contribute to interesting distinctions and relationships. Such as, during the cold war resources availability was probably less of a predictor of international and third-party intervention than fear of countries falling to communism. The first dataset includes information on government type that may shed light on some of these issues.

Estimating the effect of third-party intervention maybe difficult because third parties are judicial about when and how they intervene. It seems likely that intervention would only occur once peace was probable. Consequently, it may appear that third-party intervention may result in more solid peace agreements, even though the countries were already independently headed in that direction giving the third-party the confidence to become involved.  For this reason, third party intervention in peace agreements maybe artificially correlated to successful peace agreements. There is definitely information that I don’t have and will need, such as, external variables that I will need to control for.

Chapter three begins with a discussion of a famous quote by economist John Kenneth Galbraith who stated, “we associate truth with convenience” (86). This statement is the basis for the beginning of the author’s line of reasoning to conclude that statistics are easily manipulated. For example, the public will not challenge or doubt a statistic if it sounds reasonable or would be socially inappropriate to question. Therefore, individuals and groups have been able to manipulate statistics to their benefit without resistance.

For example, Mitch Synder exaggerated the presence of homeless in the US and women’s rights activist exaggerated and continue to exaggerate the incidence of sexual assault. Most importantly, the authors examine the exaggeration by polices departments in the 1990’s that over reported the means and weaponry of crack dealer in the 1990’s.  Police departments successfully painted a picture of crack dealers with state-of-the-art weapons, working one of the most profitable jobs in America. However, after seeing the manipulation of previous statistics the reader is immediately left questioning how crack dealers could actually have greater resources than police. The accuracy of this statistics is the premise for the discussion of the chapter.

To determine the condition and means of crack dealers the authors examines the research of Sudhir Venkatesh. Venkatesh, a scholar who researched the crack culture in Chicago, lived among a branch of a gang the “Black Gangster Disciple Nation.”  He researched the culture and hierarchal relationship of the crack industry for six years. He also was lucky enough to meet J.T. a college educated gang branch leader who took meticulous notes on the finances of his gang.

Through the documentation and budget reports that J.T. compiled over four years Venkatesh found that there was indeed a lot of money in dealing crack, if and only if you were a leader. J.T. for example was a leader and was earning about one hundred thousands dollars a year. J.T.’s bosses were earning five times as much as he was. However, these were not the individuals selling crack on the streets the police were referring to. The boys who were selling crack on the street earned slightly more than three dollars an hour, which was below minimum wage at the time. As a result, the corner neighborhood crack dealers, as the name chapter suggests, was living with their mothers without bottomless pockets and state-of-the-art guns.

According to Venkatesh the reported wealth of crack dealers by police stations during the period was only half true. Although, there were crack dealers making a lot of money the corner crack dealer certainly was not. The authors do not plainly say that the police department’s statistics were inaccurate, but as a reader who drew this conclusion I suspect most other readers would come to the same conclusion.

The authors then examine the prevalence of the effects of crack in African American community by comparing its evolutions from cocaine to the evolution of silk stockings to nylon stockings. Originally, silk stockings were a luxury only the very wealthy could afford like cocaine. After the advent of nylon stockings, stockings were available to the masses for very cheap similarly to crack. The advent of crack, however, had detrimental effect to African American communities totally stopping and reversing the progress made after the civil rights movement.

Statistics used:

1. Page 86: Mitch Snyder’s 1980 statistic that there are 3 million homeless Americans and that 45 homeless people die each second.

2. Page 88: Women’s rights advocates have exaggerated the incidence of sexual assault from one in eight to one in three.

3. Page 88-89: Atlanta police underreported crime in the early 1990’s because they want to host he Olympics and needed to be viewed as a safe city. (88)

4. Page100: In the Black Disciples the top 2.2 percent of the gang earned more than half the money earned by the gang.

These statistics help tell the chapter’s story by summarizing the argument that static’s are manipulated and crack dealers do not have bottomless wallets. The first two statistics illustrate that common statistics are manipulated and even made up in Mitch Snyder’s case, but are still accepted by the masses as fact regardless of their accuracy. The statistic in number three transitions the story to police departments and shows that even police departments, Atlanta in this case, have motive for manipulating static’s.

Atlanta’s underreporting makes the reader doubt the police departments claim that crack dealers had state of the art weapons and bottomless wallets in the next paragraph. The reader has been set up to be distrustful of the police claim because of the inaccuracy of the prior statistics. Finally later in the chapter the forth statistic disproves the financial situation of crack dealers reported by police. The majority of money in the crack industry goes to the leaders who never stand on the street corner of a ghetto, the conclusion that the reader has been waiting for and expecting while reading Venkatesh’s experience.

Levitt, Steven D., and Stephen J. Dubner. Freakonomics: A Rogue Economist Explores the Hidden Side of Everything. New York: William Morrow, 2005. Print.

In the third chapter of Poor Economics: A Radical Rethinking of the way to Fight Global Poverty authors Abhijit Banerjee and Esther Duflo present their thesis about health care for the poor and “low hanging fruit.”  Low hanging fruit is used to describe cheap medicines, preventive medicines, and low-cost actions that drastically benefit people’s health. Some of the low hanging fruits discussed are bed nets in India and other regions affected by Malaria and ORS salt for children with diarrhea. The poor underutilized these cheap but affective solutions to health problems and instead select more expensive medications once they are sick.

For example, the poor tend not to immunize their children or finish immunization shots once they have begun. Such immunizations protect the children and community. The most mind-boggling example of poor individual’s disregard for cheap preventative measures is not using chlorine to disinfect water. Clean water drastically affects the health and increased productivity of individuals resulting in higher life earnings.

The three preventative measures discussed all lead to healthier more productive lives for villagers because they prevent common illnesses. However, the authors found that the poor tend not to use these free and subsidized solutions. Instead the poor wait until they get sick and then choose to go to more expensive private doctors even when they have access to free public health clinics. There are many reasons for such a disregard for the low hanging fruits and use of private doctors. First, it may be rational. Private doctors are more likely to be in their offices during work hours than doctors at public clinics. Many poor people are also uneducated, which leads to misunderstandings about their options.

The authors, however, think that the main reason poor people are not utilizing the lower fruit is that they suffering from a universal ailment. Poor people like most individuals value their present time more highly than their future time. This leads to procrastination. In the present bringing one’s child to the doctor to be immunized takes time and commitment, but the reward of such commitment will not be realized until the future. Therefore, because the present is more valuable than the future many mothers will not bring their child to get immunized and instead will hold off until the future even though the rewards of such present actions will reap great benefits in the future.

The authors present some statistical evidence about preventative and cheap medication on page forty-two. The discussion is opened by a sweeping statement: “Of the 9 million children who die before their fifth birthdays…roughly one in five dies of diarrhea (42).”  This sweeping conclusion is followed by the introduction of statistical evidence on the effects of chlorine packages and ORS salt. Chlorine packages, which are cheap and easily obtained are only used by ten percent of families in Zambia, even though, they could potentially prevent childhood diarrhea caused by unsanitary water.  Also, ORS salt, which prevents dehydration the fatal symptom of diarrhea in children, is only used in one-third of children under five who have diarrhea in India. This statistical evidence has a great effect when it is used along with the initial sweeping statement.

The relationship the authors create between the one and five children who die from diarrhea and the minimal use of preventative and cheap diarrhea medication is not defended, however.  The author makes an assumption that of the one in five children who die of diarrhea, ORS salt and chlorine water sanitization was available to them, or to their families. Therefore, the families of these children for various reasons made a choice not to utilize these measures. Such a conclusion does not take into account that many poor children do not come from stable, intact families.

Throughout the chapter there was an assumption that it was the mothers role to get these medication. I wonder how many of the children who died from diarrhea have healthy mothers and fathers. The conclusions and causation implied about the effects Chlorine and ORS salt could have had on the “one in five children” is not fully explained. I think that the thesis would have been strengthened had the authors discussed family background and its effect on medication usage. For example, the statistic which claims only one-third of Indian children with diarrhea receive ORS salt could be symbolic of the fact that these Indian children did not come from stable homes with mothers to get the salt. Therefore, not using ORS salt maybe indicative of not having parental influence and not of poor peoples disregard for cheap medication.

Regardless the statistics reinforce the thesis that poor people are not using preventive and cheap medications. ORS salt and chlorine are shown to be cheap and highly effective yet are minimally used according to the cost-benefit analysis on page forty-two. The conclusions and broad notions about the effects and reasons such measures are being minimally used, however, may be beyond the scope of the present statistics.

The thesis of chapter three is fascinating and reminds me of similar discussion about AIDS. Although it is widely know that AIDS is a deadly disease, it is still being transmitted and people are not protecting themselves in some parts of Africa. Condoms are very cheap and often given our for free, yet people don’t use them and continue to be infected and transmit the disease. Overall, the discussion is very interesting and presents some very relevant data about the effects and under utilization of cheap medication and preventative measures.


Banerjee, Abhijit V., and Esther Duflo. Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. New York: PublicAffairs, 2011. Print.