The Quantitative Peace

Updates

Just a couple of links, comments, etc.

1. Six seismologists in Italy have recently been convicted and sentenced to six years in prison (each) for issuing a "falsely reassuring statement" prior to the 2009 earthquake in L'Aquila, Italy (BBC Article). The seismologists have been charged with manslaughter over the deaths of over 300 people in the quake, and have also been barred from holding public office ever again in the future. Yesterday's BBC Global News Podcast (Monday, October 22 AM) has some additional details on the situation behind this particular event—apparently there were a series of smaller quakes in the preceding month, but it sounds like such clusters of quakes are pretty common. Some commentators interviewed by the BBC have suggested that this could pave the way for politicians in Italy and other countries to impose new laws criminalizing what I can only label as "bad advice" from scientists, thereby making scientists less likely to offer advice to policymakers. I can't really speak to how likely I think this is, though my initial thought it "not very." Still, I would not have thought it very likely for six seismologists to be sentenced to prison terms under such circumstances in an advanced industrial democracy either. If such laws did become more prevalent, however, I guess I'd be more worried that one potential implication is that failure to come forward to begin with would suggest culpability as well. Damned if you do, damned if you don't?
2. Dan Nexon has recently announced a new blogging reception at ISA's 2013 meeting in San Francisco.  It looks as though nominations for the various award categories are open in the comments thread. Some great blogs already appear to be listed, like Phil Arena's and [email protected]. If you somehow read this blog, but not any of these others, you should probably go check them out.
3. The presidential foreign policy debate was last night, and though I don't have a lot to say at this point I do have a couple of brief thoughts. First, Romney definitely seemed more tentative last night than in either of the previous debates. As many others have noted, it seems like this is partially because there's just not a lot of daylight between Romney and Obama on many of these issues. This could also be the result of Obama simply having a greater edge on this subject area because of the job he's held for the past four years, combined with the fact that he's proved far less a lefty on foreign policy than many on the left would have liked, and many on the right had feared. Romney clearly tried to steer the conversation away from what I'll say was a more traditional discussion of security issues to the economy. That's fine, so long as he could have framed it in a clear and convincing way. It's hard to argue that economic policy is not intimately tied in with national security. However, I don't really think Romney did that effectively. His reference to Mike Mullen's earlier comments were helpful, but I think it's getting harder for the Republicans to argue that Obama has reduced our military to some impotent shadow of its former self. Obama's comment on bayonets and horses, while cute, does carry some truth.* Further, that Obama highlighted the fact that we spend more than the next ten top spenders combined on our military helped to further discredit the idea that our military is languishing. He could have increased that number quite a bit and only strengthened that position. In short, I think Romney was weaker here not just because Obama had an experiential advantage, but because even on national security the facts just don't play out as the Republicans would like. 

 

*I'm curious as to how much of this total is simply smaller support vessels that we just don't need anymore. The way to support and maintain ships is quite a bit different now than it was in the early 1900s. The elimination of battleships probably cuts down on this as well. The period Romney referenced is often identified with a significant buildup of increasingly large big-gun ships. 

 

 

"Kountry" Package for STATA

I may be late to the ball on this one, but in discussing a project that we're working on, fellow QP blogger Michael Allen recently pointed me to the "kountry" package for STATA. The package was created by Rafal Raciborski for the purposes of facilitating the creation of country codes and switching between codes provided by different sources (e.g. COW, IMF, etc.). The cite for the STATA Journal piece corresponding to the package is below:

 Raciborski, R. 2008. kountry: A Stata utility for merging cross-country data from
        multiple sources.  Stata Journal 8: 390-400

Anyone that has worked with country codes knows that this can be a pretty big pain. I've used it once so far, but it seems to work pretty well. Others may have been aware of this before now and might be aware of any bugs, problems, etc. If so, please add to the comments section any notes that you might want to make on the matter.

---

On another note, I just wanted to reiterate the earlier message that posting will continue to be sparse for a while longer. While we would all very much like to be blogging more regularly—especially in light of the recent debates, and the final upcoming foreign policy debate—everyone is currently engaged with other professional responsibilities. I believe this will prove to be time is well-spent, though, as everyone seems to be keeping busy and is pretty productive. 

Military Capabilities

Phil Arena has a very cool post up wherein he explores a potential alternative to the ubiquitous CINC score for measuring military capabilities. This strikes me as something that is long overdue.  And there are a couple of things that I like about the measure just from taking a quick glance at it. First, given that China just launched its first aircraft carrier, which is actually a refurbished Soviet carrier if I recall correctly, I think the notion that China has already surpassed the US in terms of military capabilities is a big pill to swallow. Second, just glancing at his figures for European powers during the late 19th century, Phil's measure looks more noisy, which in truth is kind of what I'd actually expect to see. The CINC measure makes there appear to be a lot more stability than I would expect to come along with a period where many of these European powers are vying for access to colonial possessions, foreign markets, etc. Furthermore, there were definite differences in terms of how each of these countries allocated their military resources. England certainly had the dominant navy during the late 1800s, but countries like France, Germany, and Russia had advantages in terms of their ability to field ground forces that the British just didn't possess.  

As Phil mentions, the measure may need some further development, but I still think this is really neat. 

Don't Call It A Comeback (This Civil War Has Been Here For Years)

Well, maybe not years. And LL Cool J references aside, while perusing my usual news outlets this morning I've found that in the wake of what is reported to be a fresh massacre, Syria is again a common theme in the headlines (see CNN, BBC, FP). More after the jump.

Math's PR Problem OR Changing our Educational System

I was listening to the BBC's Global News Podast this morning and found an interesting segment on the utility of mathematical modeling. Professor John Adam's book (which can be found here) looks at how one can model various aspects of urban life—pollution, traffic congestion, number of dentists offices, etc. I've not read the book, but Adam's message in his interview was something that I can sympathize with. In particular, Adam talks about the degree to which the educational process divorces math from any kind of applied usage. Learning math in this way, he goes on, is partially responsible for why more children/people are not as interested in mathematics as they might otherwise be. 

Sure, there are the ubiquitous "if a train leaves city A at rate X, and a train leaves city B at rate Z, when will their paths cross," kind of examples. But seriously, is that the best we can do? Now, I recognize that I'm far from being the strongest, or the quickest, when it comes to picking up a lot of the methods that I've been taught over the past few years. I, like many people I suppose, was not the biggest fan of math class in high school. Some of this was laziness on my part, but some of it was also the framework in which the subject matter was taught. Even in college it was far from clear that any kind of math beyond balancing a checkbook would play as big a part in my life as it currently does. I did well enough in these classes, but it was not until graduate school that I really had to seriously engage math and statistics.  

As it turns out, I've really enjoyed the quantitative side of my training here at Binghamton, and I'm far more willing to work at developing my understanding and skills now than I was when I was 16. Had I known in high school that I this stuff could be used to study things like war, human rights, etc., I might have cared a bit more. These are things that I've always been interested in. And this is perfectly rational behavior, right? If the expected utility of learning this material is zero, or even negative, then why expend the effort? Where is the payoff? If this is the case, then a major part of the reason that the expected utility of learning this material is so low is because the people that should be able to provide you with the relevant information to update your expectations simply do not do a very good job of that. In fact, I'd say that I didn't really get that at all until grad school. And being able to make that initial connection between math and statistics, and the substantive topics that I care about, has subsequently led me to develop a greater appreciation for the math itself that underlies the models I often use. 

I don't have any answers for how the educational system could be modified to address this issue. I also realize that you would have to find a way to incorporate applied mathematics into most subject fields, as not everyone has the particular set of interests that would make more math in history or government class a sufficient solution to the problem. Still, there must be a better way. With folks like Neil deGrasse Tyson going around trying to promote greater scientific literacy, and the recent attacks on NSF funding for political science, it seems like there is a space for us to modify how we educate our younglings and how science, math, history, etc., can be better integrated in this process. Maybe if people developed a better understanding of science as a process, or math as something with real world and substantive applications, rather than something that existed strictly from 1pm to 1:45pm during 8th period, people would be better able to develop a more robust appreciation for how mathematics and science fit into, and matter for, our daily lives. This is fairly obvious when phrased like that, but the fact is that it's not really what we do currently, is it?

Am I missing something?

Also—and this is a total aside—Brian Rathbun at the Duck of Minerva has a discussion of how Iron Maiden material can be used to teach political science. This is perhaps a good example of what I was talking about above, so perhaps this is not a total aside, but what we have here is essentially "applied awesomeness."

NSF Funding Amendment, Part II

Erik Voeten at the Monkey Cage with an update on the amendment that cuts NSF funding for political science programs. A version of the amendment passed the House with a vote of 218-208. You can see the breakdown of the votes here (thanks for Erik for so conveniently providing the link to the votes). Erik also notes a couple of important points: 1) This is not the end of the issue, and 2) No other discipline was singled out in the same way as political science. 

Other interesting facts:

1. Jeff Flake has a BA in International Relations (see his official House bio page).
2. Jeff Flake has an MA in Political Science (see his official House bio page).
3. As described above, political science appears to have been singled out. In 2011 political science programs were awarded approximately $21 million dollars. Keep in mind these are awarded amounts, and do not reflect the actual amount spent in a given year. The Duck of Minerva makes reference to actual annual expenditures being something closer to $9 million. 
4. Economics programs were awarded $55 million in 2011. If we assume a similar pattern applies, that means econ programs are getting somewhere around $23 million per year in actual appropriations.
5. Sociology programs were awarded approximately $15.5 million, working out to be about $6.7million per year.

Political science awards do not even come close to what economics programs are awarded each year. Consequently, I'm not entirely sure what the fiscal justification for this cut can be. To single out only political science in this way would seem to indicate the gripe is more substantive than fiscal. If not, then why not make a flat cut of 10%-20% across all BSES programs? Spread the cuts evenly? You could cut the annual econ awards by half and that would be the equivalent of cutting all of the annual political science awards AND some of the sociology program. So, what gives?

There is probably a better/more efficient source for summary information than what I used, but you can search NSF records of programs and awards here. My estimates of yearly expenditures are extremely rough. Again, it's important to note that these are awarded amounts, not what is actually spent each year. There is going to be overlap from one year to the next as programs are often funded over the course of a couple of years. There is also some degree of overlap between programs as well. Accordingly, the figures provided should be taken as rough guidelines/comparisons, as well as with a healthy grain of salt. If anyone has a link to actual annual expenditure statistics by program, as opposed to awards, please share. I did a quick search this morning but didn't find anything more helpful than this particular source. 

Hunger Games: Civil Wars, Victory, and Repression

The movie version of the book The Hunger Games (Paperback, Kindle) is being released this week; the movie is expected to draw a large crowd and is already drawing superb ratings from critics. As such, given that there are both political and economic themes that run in the book that are ripe for political scientists and economists, I figured I would touch upon at least one of those interesting threads related to the study of international relations.  

Now, what I am drawing upon is purely from the first chapter of the book and can be easily gleaned from the trailer of the film.  As such, this discussion should not be spoiling anything.  However, if you are attempting to see this movie without any background whatsoever, consider this your spoiler warning for the post.

The book takes place in a future dystopian world set in North America. During this time period, political organization is non-democratic and highly centralized.  North America is divded into 13 districts excluding the Capitol.  As Katniss, the protagonist (the novel is told through first person narration), explains in the first chapter (p. 18):

The result was Panem, a shining Capitol ringed by thirteen districts, which brought peace and prosperity to its citizens. Then came the Dark Days, the uprising of the districts against the Capitol. Twelve were defeated, the thirteenth obliterated. The Treaty of Treason gave us the new laws to guarantee peace and, as our yearly reminder that the Dark Days must never be repeated, it gave us the Hunger Games.

The rules of the Hunger Games are simple. In punishment for the uprising, each of the twelve districts must provide one girl and one boy, called tributes, to participate. The twenty-four tributes will be imprisoned in a vast outdoor arena that could hold anything from a burning desert to a frozen wasteland. Over a period of several weeks, the competitors must fight to the death. The last tribute standing wins.

In the above account, the districts mobilized to overthrow the central government (perhaps as revolutionaries or as separatists, it is not clear in thie passage) and the Capitol is able to persevere over all 13 districts.  1 district is obliterated and the 12 remaining districts are put in a worse position than they were prior to the conflict as the Capitol ramps up the repression of the districts by extracting tribute in the form of humans (the Capitol is quite repressive in other economic and political venues as well). Two children from each district are randomly selected to fight in the Hunger Games each year. The above passage is leading up to the point where we find out which children serve as tribute for the 74th annual Hunger Games. 

Since the government of the Capitol gets to play the obvious villain in this post-civil war story, we should be interested in the likelihood of some situation akin to this happening in reality.  That is, after a failed civil war attempt, are there times when the government uses its new position to further repress instead of adopting some form of accommodation to stave off a future rebellion.  While we are unlikely to see the specific scenario of the Hunger Games played out, increasing repression after a civil war is something that seems feasible.  We can think of civil war as a bargaining failure between two actors (rebels and the government) and, once a victor has been determined through the random number generator that is war, we are left with one party to reap the spoils.  It would make since that the a newly consolidated government would ramp up its extraction of resources beyond what it could achieve pre-war.

To understand increased repression after a conflict, I employed the Cingranelli-Richards data set (CIRI) and used their Physical Integrity Rights Index (henceforth refered to as either PI Rights or PIR) to look at their 9-point index as a proxy for government repression.  The index is composed of four 0-2 indicators of government respect or use of torture, disappearances, extra-judicial killing, and political imprisonment.  A score of 0 would be interpreted as little to no respect for physical integrity rights while a score of 8 is high respect for those rights.  The index runs from 1981-2010, so that helped to establish the sample of possible civil wars I could look at.  For civil wars, I used the Correlates of War intra-state war data (v 4.1) and employed the 19 conflicts that had system members as governments, started in 1982 or later, had ended by 2007 (a COW limitation), had over 1,000 battle deaths combined from each side, and the government had won the conflict (thus, excluding compromises, stalemates, and rebel victory).

To get a feel of the data and government use of repression, the first item I looked at was the PIR value one year before the conflict started and one year after the conflict had ended.  This produces the following scatter plot:

The line represents no change in the level of respect for PIRs while points below the line represent a decline and points above the line represent improvement.  So, in this instance, we have some cases in which the government represses more after a civil war.  Since there are only 13 dots for 19 cases, there must be some overlap.  As such, we can examine this as a change histogram as well:

The slim majority of cases are worse off with a few very large decreases in PIR values.  There are several states that show improvements as well.  However, a one year change may not be enough time for the government to adopt a new strategy for winning a civil war and the PIR value may be a hold-over from conducting the fighting in the civil wars.  Looking at 5 years ahead, the 18 cases (losing one observation due to temporal limitations) produces the following distribution:

Five years after the conclusion of the civil war looks generally better for those societies.  There still remains 28% of the sample that have a worse off situation, but 50% show an improvement.  Naturally, this is a small sample of intra-state wars and the possibility of a Panem type situation is certainly possible.  The 3 cases that show a change below -1 are Myanmar in 1982, Indonesia in 1992, and Ethiopia in 2000.  The Panem situation, if I understand the story correctly, has been ongoing for 74 years by the time of the first novel, no such analogue is available in the data given the limitations of the CIRI dataset.  Also, the amount of state capacity demonstrated by the Capitol in enforcing laws seems far superior to any of the examples above.  The strength of the state in the book is something that should decrease the probability of conflict by rational revolutionaries.

As an aside, one of the other side tidbits I have noticed is that all the characters that are in the Capitol have Roman names such as Cinna, Octvia, Caesar, and Plutarch (among others).  I am not sure if this has relevance beyond harkening to a city-dominated empire, is veiled allusion to the idea that the United states is becoming a new Rome, or some other significance I am not familiar with yet.  I do not know the full story of the Hunger Games, but I will likely be catching the first movie this weekend. 

There's Defense, and then there's Defense.

This was posted a couple of days ago by Steve Clemmons. It's a table of US defense expenditures compiled by Winslow Wheeler at the Center for Defense Information. What's cool is that it collects information on expenditures on defense and security-related spending more broadly (Department of Energy, Veterans Affairs, Department of Homeland Security, etc.). The fact that total spending on defense and security exceeds the figures that we typically cite out of the defense budget is not really big news, but I've not seen anything tabulated like this before. I'm sure some spending categories could be added or removed, but the overall figure is impressive still.

The DOD base budget still comprises the majority of spending, but it's interesting to see how the categories stack up against one another. Overseas contingency operations are a large part of what we pay to prosecute the wars overseas, and the figures for this category shown in the table have dropped quite a bit comapred to the last 5-6 years or so. Whatever gains we make in this category, I suspect, will be offset to some extent by inreased expenses associated with veterans' healthcare, retirement, etc. Consequently, this helps to illustrate how the costs of war, even from a strictly financial perspective, far exceed the end of the actual conflict. It's also not hard for me to imagine this figure topping $1 trillion in the next couple of years as supplemental spending packages are almost always passed in recent years after the fiscal year's budget has been approved. 

2010 CIRI Human Rights Data Project Report

The CIRI Human Rights Data Project, of which I am a part, released its 2010 data last weekend.  A report pointing out interesting patterns can be viewed over at the CIRI Blog.  One interesting bit:

The CIRI Index of Physical Integrity Rights measures a government's overall level of respect for four rights: torture, extrajudicial killing, political imprisonment, disappearance. The index ranges from 0 (no respect for any of these four rights) to 8 (full respect for all four of these rights). In 2008-2009, the world saw an overall average increase in these rights of .047. However, a reversal of this improvement was seen from 2009-2010, with it's overall average decline in respect of -.031...An important part of this overall decline in respect for physical integrity rights comes from a continuing degradation, globally, of respect for the right not to be tortured. For example, 2009-2010 saw 17 countries engage in more torture, while only 8 engaged in less torture...This trend in the greater use of torture is not a post-9/11 phenomenon, however. The chart below shows the increased use of torture beginning in the early 1980s. The extent of the drop in respect for this right differed by region and is seen to be particularly acute in Africa.

This is particularly interesting when combined with the fact that, according to CIRI, political imprisonment has become less common over the same time period.  Off the top of my head, I would probably attribute these changes to substitution of one form of repression for the other.  It seems likely that, as information on human rights abuses has become more widespread, governments have engaged less in the physical integrity rights violation that is the hardest to publicly deny (political imprisonment) and more in a form of repression that is much more clandestine (torture).  This seems especially plausible to me since one of the world's greatest publicizers of human rights abuses, Amnesty International, began its life as an organization primarily committed to supporting and publicizing the plight of "prisoners of conscience" worldwide.  Over time, the cost of engaging in political imprisonment may very well have become high enough to make other, less visible forms of repression more attractive.

Of course, there are many alternative explanations for these changes, and there is a lot of work that remains to be done in terms of developing theories of state repression.  This simply stands out as one of the many empirical puzzles that remain to be solved for those of us interested in studying government respect for human rights.  In the meantime, check out the CIRI Human Rights Data at http://www.humanrightsdata.org and the CIRI blog at http://humanrightsdata.blogspot.com/.

 

That's the Spirit!

In spite of the massive flooding in Binghamton, I'm finding some time to focus on something else.  

So hey, remember the debt deal?  Sure you do.  Well things are already off to a great start with that bipartisan commission.  Apparently Senator John Kyl is threatening to leave the bipartisan commission if defense spending remains on the table.  Kyl says that even Defense Secretary Panetta, and former Secretary Gates said enough was enough on defense cuts.  

In fact, I'm pretty sure that's not what Panetta said.  I'm pretty sure Panetta has said that up to $350 billion more would be feasible.  Beyond that and security might take a hit. As I understand it, Panetta was saying that beyond what was already cut as a part of the debt deal, an additional $350 billion over ten years could be cut.  

So what does Kyl get out of this?  Didn't he get the memo?  Bipartisanship is IN!  Oh, that's right...

If the 12-person congressional committee fails to come up with a plan to make those additional cuts, that would trigger $600 billion in defense-spending reductions. And Arizona is one of the top half-dozen or so states in the nation in terms of the defense industry's footprint.

The industry already has felt the ripple effect of military efforts to wind down the U.S. presence in Iraq. After leveling off in 2009, Defense Department contracts in Arizona in fiscal 2010 dropped nearly 19 percent, to about $10 billion.

Defense and aerospace manufacturers employ about 37,000 Arizonans who earned $3.1 billion in salaries in 2008, according to the chamber's figures.

The Army, Air Force, Marine Corps and National Guard also employ 45,000 military and civilian personnel at nine Arizona military installations.

First, I guess it's possible that Kyl really wants an excuse to be off the panel.  I really can't imagine defense cuts not being on the table in some form or another, and perhaps he's reluctant to be associated with the cuts that he know are inevitably coming down the line.  Would he really gain any political points for quitting?  Second, and alternatively, wouldn't Kyl be better off remaining on the panel to curb the potential losses that Arizona faces?  Yes, there may be cuts, but could he effectively argue that those cuts would have been worse for Arizona had he not been there to guard against them?  

Ultimately I guess this comes down to whether or not his threat to leave the panel is credible.  What am I missing here?