Abortion and Crime Reduction Revisited

4 May

In the article, “The Impact of Legalized Abortion on Crime,” Steven Levitt and John Donohue argue that with the legalization of abortion in 1970, the rise in the amount of mothers having abortions and reducing the number of unplanned children caused a drastic drop in the crime rate in the 1990s. Levitt and Donohue conclude that this lost generation of children, if born were more likely to be subject to neglect than if born into a planned family, reduced the number of children likely to engage in criminal activity. They support their argument in saying that states with the highest abortion rates in the 1970s and 1980s (the 1990s generation) witnessed the greatest crime reductions, and propose that legalized abortion appears to account for as much as 50 percent of the 1990s drop in crime. (Levitt and Donohue, 379) These are pretty lofty predictions.

In Chapter 4 of Freakonomics in which they propose the same theory, they compare it against other factors that may have affected the crime rate such as increased policing, gun control laws, and the strength of the economy. The article and the chapter reach the same conclusion in that they argue that legalized abortion had the most significant effect on the decreased crime rate of the 1990s. The main basis of their hypothesis is that of a cause and affect relationship, indicating that because there were less neglected children, thus less criminally inclined children, and thus less crime. This causal relationship provides the basis for a critique of their argument from Christopher Foote and Christopher Goetz.

 The main argument that Foote and Goetz produce against Levitt and Donohue’s hypothesis is against their data. They propose that Levitt and Donohue should have a used a per-capita variable as opposed to using age cohorts within the same state and year. Goetz and Foote show that with the per-capita variable in place the statistics showing that legalized abortions affect the drop in crime rate are less significant. They propose that Levitt and Donohue should have taken robustness of the variables into account so that the trends among certain states would have been accounted for.

I think the reconciliation between Chapter 4 and the two articles comes in the form of the economic theory and its validity behind their argument. Yes it makes sense to think that legalized abortion would reduce the number of bad childhoods, and that less bad childhoods would account for a drop in the crime rate. Furthermore it was even more beneficial to Levitt and Donohue’s argument that the time table fit perfectly, because the crime drop in the 1990s went unnoticed for some as cited in Chapter 4 of Freakonomics so they could just point to the legalization of abortion in 1970 as a prime, logical choice for the drop (considering the children not born during this time would be entering their prime crime committing age). However as Foote and Goetz point mere cause and effect is not enough to assume and that their regression as indeed lacking in important data.


Where have the criminals gone?

27 Apr

Levitt and Dubner answer this question not by where the criminals have gone, but argue that they were never born. The authors address the seven most publicized reasons for the crime drop in the United States beginning in the 1990s, and come to the conclusion that Roe V. Wade and the growing legalization of abortion was the biggest contributor to the crime drop. Their argument is based on the fact that children born into unhealthy situations are much more likely to become criminals, and this was avoided at a greater rate with the legalization of abortion.

Personally I really enjoy this book and think it is well constructed. Especially in this chapter, they do an impressive job of illustrating their points and arguments through real life statistics. For example, in laying out the effect of tougher gun control laws on the crime rate, I found their example concerning Switzerland’s gun control laws astounding. Every adult male is given an assault rifle for militia duty in Switzerland, and it remains one of the safest countries in the world. They present highly logical real life support for their arguments. 

With that being said, I wasn’t especially suspicious of any of the material, but I did get a small notion that they were understating the effect of the other factors although they do provide pertinent results and statistics. Holding their support of abortion as a crime deterrent in mind, the question I would ask them is whether they are pro-life or pro-choice. Not that it is significant or relevant to the story they tell in Chapter 4, I think it would be interesting to have a conversation about with an innovative thinking economist. Does the economic effect of abortion affect his opinion of it?

Sex, School Uniforms, and Sugar Daddies

30 Mar

In many developing countries, specifically India, population growth is a cause for concern among policy makers as they try to develop their country. What they have come to realize is that contraception is not the most effective way to monitor and influence the population growth rate, and that there are many other factors contributing to the family decision making processes including the family dynamics and even more so economic considerations.

One of the most fascinating topics I found in this chapter by Duflo and Banerjee was that of teen pregnancy in developing countries. Under the heading of “Sex, School Uniforms, and Sugar Daddies” the authors present the statistics of teen pregnancy rates in different countries, featuring an above 10% rate in the Ivory Coast, Congo, Zambia (all African countries). These countries are closely followed by four South American Countries with teen pregnancy rates been 8.2 and 9.2 per 100 teen women in Mexico, Panama, Bolivia, and Guatemala. What I found to be interesting and not so much surprising was that most of these countries with high teen pregnancy rates do not facilitate family-planning services for adolescent girls unless they have guardian consent. Especially given the statistic given that the introduction of family planning into numerous villages in Bangladesh featured a 1.2 reduction in the number of children a family has in this region, I would like to test the hypothesis that if teenagers were given unrestricted access to family-planning, would this then have a significant impact on the number of children people have in these developing countries, specifically the African countries with teen pregnancy rates above 10%.

This may be a very difficult variable to define since the problem in the first place is that the parents wouldn’t approve the family-planning in the first place which might alter their answer, but I think the best way would be to survey as many teen mothers in these countries as possible in determining if they had access to family planning, would they have taken advantage of it. With safer more effective contraceptive programs than the slogan “use a condom or you will die,” and barring any moral implications, I think that most of these teen mothers would have taken advantage of it and it would have  significant impact on the average number of children a family has. Especially if the message were spread that (even in the United States) there is evidence of children having more fruitful and prosperous lives when they are born into ready and willing families.

The model would look as follows:

AvgChildFam = B1 + B2FamPlan + B3TeenPlan + B4DumSTD

  • B1 represents what the average number of children would be per family without any family planning program whatsoever
  • B2 represents the change in the average number of children per family with a 1 unit increase in the family planning program (perhaps measured by number of women who take advantage of it)
  • B3 represents the change in the average number of children per family with a 1 unit increase in the number of teens of who are allowed unrestricted access to family planning.
  • By adding a Dummy variable accounting for the level of sexually transmitted diseases, perhaps earning a 1 if the level is above a certain percentage (I don’t know what would be high or low), I think this would also affect the average number of children in these regions in accordance with family planning because if the mother fully knows she will transmit the disease to the child she may be more likely to take advantage of family planning. We would know if it was statistically significant if the regression shows that there is a greater explanation in the average number of children with the STD dummy variable as opposed to it not being included.

The Effects of War Risk on U.S. Financial Markets

9 Mar


In a working paper published by the National Bureau of Economic Research entitled “The Effects of War Risk on U.S. Financial Markets,” Roberto Rigobon and Brian Sack undertook the question as to whether the risk of war and its affects, accumulated over the ten weeks leading up to the war with Iraq, had a significant impact on the movements of nine U.S. financial variables. Since the actual risk of war is impossible to quantify, they used a heteroskedastic estimator that allowed them to indentify the impact of war risk by determining a set of days on which the variance of war-related news was elevated. Based on their OLS regression including other variables that could affect the financial variables, they determined that the risk of war appears to have been a remarkably important factor in determining the financial variables movements in those ten-weeks, with the risk accounting for between 13 and 63 percent of the variances in their cumulative movements.


The most provocative idea that comes to mind after reading this paper is that the numbers alone cannot determine the impact of a war. Especially in America where so much of our economy is based on speculation, after reading this article I will have a more insightful view as to the results of my eventual regression, now knowing that the military spending and macroeconomic affects are not the only variables affecting the GDP during war. Throughout the United States conflicts there were bound to be different ways the news released “war news.” For example, maybe in WWII they only released news that would lessen the perception of war risk. By including this in my estimation of war affects on the GDP, different increases and decreases in the perception war risk would have different affects on the economy, making a general conclusion as to the overall affects of war on GDP all the more difficult.


The way this model and regression were done presents an issue regarding the assumptions of the classical linear regression model because it uses a heteroskedastic estimator, and the variance of the error terms in this model are not constant. Perhaps to fully answer this question I have undertaken would require the use of heteroskedastic estimators; perhaps I to come to an accurate answer I would need to identify a factor without actually quantifying it. I won’t be including this exact estimation in my regression, but it is certainly food for thought as to the eventual conclusion I will come to and how precise it can actually be.

To see the actual affect the risk of war had on the variables scroll to the bottom of the paper, the results are in the tables.



“Military Cuts and Tax Plan Are Central to Obama Budget”

29 Feb

The United States is in an awful amount of debt, with a national deficit around $15 trillion. Yes, trillion. According to an article published in the New York Times by Jackie Calmes, with the election coming up in November, Obama is making his final push in his budget request with tax increases on the rich and cuts in spending, specifically military spending. Through this he hopes to pay for priorities like education, public works, and clean energy. He envisions long-term initiatives with the end of the wars to spur new business and manufacturing activity.


Foregoing the political jargon associated with the possibilities of his budget request passing, military spending makes up about 25% of the federal budget, and about 4.8% of the Gross Domestic Product. The cost of the Iraq War alone is around $800 billion, and the war in Afghanistan around $500 billion. With the withdrawal of all troops from overseas, Obama plans to start “nation-building” at home, with (what Republicans are calling a gimmick) “military savings” to pay for a $476 billion program to modernize transportation. He also plans to raise $1.5 trillion over ten years mostly from taxes and closing tax breaks. Projections for 2013 show the budget deficit would be $901 billion, or 5.5 percent of G.D.P. The point is that if cutting the military spending, avoiding the $1.3 trillion expenditure can produce these types of results, then why do we spend so much in the first place?


The United States spends a vastly larger amount on military expenditure than most other countries combined. The question is, why? It is obviously negatively affecting the budget and thus the GDP, but that also raises the question as to why there is even a debate over whether wartime brings prosperity or not. Wartime must produce effects beyond the numbers, ranging from consumer confidence, expectations, the environment, everything. Especially recently, the wars overseas cannot be the only things charging the changes in GDP or the economy’s health in general. There must be other, arguably more effective factors at work in our economy, specifically consumer spending, investment, and imports and exports. The particularly daunting task before me is to account for these other things in my regression, and for specific events that may have caused changes in the GDP.



Top of the Class

24 Feb

Much like the low-hanging fruit dilemma, there is a two level problem in underdeveloped countries when it comes to education. More and more children are enrolling in schools, but the schools themselves are delivering very little results. Similar to the low-hanging fruit in absentee doctors, the teachers and also the students in these schools are often absent, and learning levels are very low. The question that Banerjee and Duflo present is whether it is based on supply or demand; supply in the government providing the services, or demand in the parents wanting to send their kids to school in order to see results.


Their answer is that it is both. Parents and teachers only expect their kids who graduate from secondary school to do well, while the other “regular” students are left behind. Duflo and Banerjee also contend that we know how to educate these children if only we decide to try. “We have to find a way to get children into a classroom, ideally taught by a well-trained teacher, and the rest will take care of itself.”


The supply side of the equation has been articulated the most by the United Nations, specifically in their Millennium Development Goals set up in 2000, and for the most part they have been successful in getting children into the classroom. In an article published in the New York Times published two days ago entitled, “India in a ‘Race Against Time’ to Meet Millennium Goals,” it is stated that they have already reached their target in boosting primary education and reducing H.I.V. But like Duflo and Banerjee contest, what is the point of this primary education if there is no worth to the schools. The executive secretary of the U.N. Economic and Social Commission for Asia and the Pacific is quoted as saying, “many of these goals (MDG) can still be reached with a redoubling of efforts.” Really? You have to double efforts?


Why the effort wasn’t put forth in the beginning in order to create a worthy education system is unclear. It is absolutely true that attendance in these schools has gone up, but for the most part to no avail in the quality of education. Duflo and Banerjee present startling statistics, saying that close to 35 percent of children in India in the seven-to-fourteen group couldn’t read a simple paragraph. The article simply states that the target, universal primary education, has been achieved with statistical basis. Obviously, Duflo and Banerjee are a little more convincing.





Wartime Effects on GDP

17 Feb

The basis of my research project will be wartime in the United States and its effect on the Gross Domestic Product. More specifically, does being engaged in war effect the United States’ GDP more drastically than in other countries? The bulk of my data will be retrieved from the Conflict Data Set provided by the Department of Peace and Conflict Research at Uppsala University in Sweden. It is a break down civil and inter-state conflicts throughout the world from 1946 to almost present day. The aim of my research is to show that wartime in the United States does in fact have a greater effect on its own respective GDP than other countries experience (respectively) when they are engaged in armed conflict.

Especially now in our country, witnessing the withdrawal from Iraq after a long, VERY expensive war, this is a pressing topic. We are now able to look back at this war, compare it to other wars, and make sound arguments about the true effects of wartime on GDP. Throughout recent history, wartime has been seen as a sort of catalyst for economic activity, particularly in the United States. This “myth” has supposedly been debunked, but if it is true, to what extent does the effect compare to that of other countries? Very few countries measure up to the United States and its military prowess even with everything calculated relatively, but which countries and how many countries would be best to use as a comparison? It is also a pressing issue because so much of it has to do with the government and their policies. One reason I chose this topic is because I find myself at a point where I am skeptical of government activities. The GDP may rise in time of war, but is it rising in efficient ways as opposed to exclusively military spending?

There shouldn’t be any information that I don’t have access to in this research endeavor, but there are an infinite number of variables that could factor into a changing GDP during a war. For example, gas prices are bound to change when engaged in a war with countries that produce the world’s major supply of it, in turn changing GDP. There is a plethora of data and theories based on wartime prosperity, but the difficult part of this will be obtaining the right data for the right countries in order to compare it successfully to the United States. It may be difficult to find the GDP’s of certain countries that prove vital to the comparison. Regardless of the validity of my thesis, I will absolutely have a better understanding of war time effects on economies, what parts of the economy it effects, and what policies the government employs in order to right the way (or doesn’t).

Link to Conflict Data