Category Insurgency & Counterinsurgency

Hope you all have your taxes done….speaking of economics. Anyhow, picking back up on the Economics of Warfare posts by Dr. Spagat. The good news is that these blog posts by me apparently inspired (read: forced) Dr. Spagat to post all 20 of his excellent Economics of Warfare course lectures on his blog.

Starting an examination today of the thirteenth lecture from Professor Michael Spagat’s Economics of Warfare course that he gives at Royal Holloway University. It is posted on his blog Wars, Numbers and Human Losses at: https://mikespagat.wordpress.com/

The lecture looks in depth at one country, Colombia. Dr. Spagat has done a lot of work there, and even helped set up a non-profit to analyze the Colombian civil wars. These have been the bloodiest series of conflicts in the western hemisphere in the period after World War II. It was through his work on Colombia, and our related work on insurgencies, that we first became acquainted.

Slide two of his lecture starts with the statement that: “To get anywhere with empirical research you need to have a reasonably large number of data points. (This is a basic fact about empirical analysis that many students beginning research projects overlook)”

Actually, it is a basic fact that many in the Army and Defense operations research community overlook!!! I remember getting into discussion with a senior OR practitioner, a retired corporate president who once shared an office with Geroge Kimball of Morse and Kimball fame (Methods of Operations Research, 1951), who tried to make the argument that all you need to 15 good data points. This was at the time we were doing the Bosnia Casualty estimate (see America’s Modern Wars, Appendix II). Needless to say, I strongly disagreed, especially as we were looking at “social science” type data.

The next line in Dr. Spagat’s presentation is: “So we need to ask ourselves — where are all of these data points going to come from?”

This is the issue, and quite simply, the gap that The Dupuy Institute has attempted to fill. For example, Dorothy Clark’s seminal study on Breakpoints (Force Changes to Posture) was based upon only 43 cases [Dorothy K. Clark, Casualties as a Measure of the Loss of Combat Effectiveness of an Infantry Battalion (Operations Research Office, Johns Hopkins University, 1954]. This is not a lot of data points, which of course, she understood. But, producing “data points” requires research, which takes time and money. There are some existing databases publically available that can help with some problems, but for many problems, there is simply not enough data points assembled for any meaningful analysis. There does not seem to be the mechanism in place to make sure that the Army or DOD has the data that it needs for all of its analytical work.

After starting page 2 with two rather significant statements, Dr. Spagat then goes into discussing Colombia in more depth. I will pick this up in a post tomorrow, as this blog post has already gotten long (and preachy).

The link to the lecture is here: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%2013.pdf

Examining the twelfth lecture from Professor Michael Spagat’s Economics of Warfare course that he gives at Royal Holloway University. It is posted on his blog Wars, Numbers and Human Losses at: https://mikespagat.wordpress.com/

This paper continues with discussion of the studies done by Fearon and Laitin (lecture 11, slide 5) and Collier and Hoeffler (lecture 11, slide 15) on civil wars. The lecture basically goes through and tests or challenges their papers in two areas: 1) ability to predict, and 2) causality.

Warning: This lecture may cause you to lose confidence in multi-variant regression models.

I already had. If you go to my book America’s Modern Wars, in Chapter 6 (pages 63-69) I propose a two-variable model of insurgency success or failure. I then tested the model back to the cases I used to make up the model and the model predicted the correct result in 53 out of 68 cases used (77.9%). The model predicted incorrectly in 15 cases, or over 20% of the time. Now, if I was at a blackjack table in Vegas, I would be pretty damn happy to predict the outcome of game almost 80% of the time. The problem I had is that I could not find a clear third variable. I could easily explain away why 7 of the 15 cases were incorrectly predicted, although they were for a variety of reasons; but I could not easily explain why the other 8 cases were incorrectly predicted. In three of the cases the model predicted a red win (insurgents won) when the blue side won (the counterinsurgents); and in four of the cases the model predicted a blue win when the red side won (page 67). There was clearly a third, fourth or fifth variable in play here, but I could never figure out exactly what it was, and it was probably multiple variables. This was the next step and would have been pursued further if we could have obtained further funding.

Of course, we could have just added three or more additional variables to the model and this would have certainly improved the fit….but what are we really doing? This is the point where I begin to loose confidence in adding more variables, so I choose not to.

Getting back to Dr. Spagat’s lecture, one person analyzed the two papers by Fearon and Laitin (called FL) and Collier and Hoeffler (called CH) as to their predictive value. They were not very good at prediction, and sometimes gave false predictions (slide 6). Note that the “false positives” outnumbered the correct predictions for the Collier and Hoeffler model.

On slide 10, Dr. Spagat shows the variables used in each model and how much each variable impacts the results. You will note that one to three variables in each model provide far more explanatory value than the rest of the variables. GDP sort of stands out in both models, although one uses GDP while the other uses GDP growth, which are very different values. There are also some odd variables in there (for example using “squared commodity dependence” in addition to using “commodity dependence” in one model).

Dr. Spagat then goes into the issue of causality, ending up with a discussion on rainfall. Unfortunately, the real world is more complex than the models. A regression model assumes that the inputs are “independent” variables and the output is a “dependent” variable. Yet, in the real world, there can be another variable out there that is influencing both the “independent” and the “dependent” variable. Also, the alleged “dependent” variable can sometimes influence the independent variables. This he discusses in slide 12 (“This is, while it is true that low or negative growth might cause conflict is also true that conflict might cause low or negative growth.”).

Note that Dr. Spagat does address using different measurements of variables in slide 22 (rainfall levels vice growth rates in rainfall). This is an issue. Does one use an independent variable with an clear value (like a GDP figure) or does one use the change in the value of the variable over time as the measure (like percent change in GDP)?

The link to the lecture is here: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%2012.pdf

This is a continuation of my previous post: Economics of Warfare 11. It is based upon a review of Michael Spagat’s lecture: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%2011.pdf

The Fearon and Laitin study on the “Determinants of Civil War Onset” (link to paper on slide 5) had eight factors that influenced that chance of a new civil war (they are listed on slides 8 and 10). One of them is “population (positive effect)”. This is summarized by Michael Spagat as “more people, more chances for war.” The Collier and Hoeffler paper came to similar conclusions on this (link to paper on slide 15). As Michael Spagat summarizes it “Population size is still positive and significant” (slide 16). And then there is a third paper by Hegre and Sambanis (like to paper on slide 23) that states that “country size (population and territory) is positively associated with civil war onset” (slide 24).

Now when we did our insurgency studies we also saw the same thing as it related to winning or losing an insurgency. It show up in our Iraq Casualty Estimate (see America’s Modern Wars, Chapter 1) and in subsequent analysis. Quite simply, insurgencies in large countries were often successful. For example, when we looked at all 10 insurgencies with foreign intervention where the indigenous population was greater than 9 million, the insurgents won 80% of the time (see page 47). In our original Iraq estimate for the cases we had, we found the insurgencies won 71% of the time in large countries (290,079-2,381,740 square kilometers…14 cases); 100% of the time in populous countries (population of 9,529,000 or higher…7 cases); and 78% of time in countries with a large border (which is probably related to country size). This is on pages 17-18.

Apparently size matters when it comes to violence. I am not sure of the cause-and-effect here. Obviously, there are many other factors at play. The Symbonese Liberation Army (SLA)…the people who kidnapped Patty Hearst…were a small urban insurgency operating in a large populous country (the United States). It fizzled quickly. So, it certainly does not mean any insurgency in any large place is an issue. But, there does seem to be some correlation here that continues to haunt analysis of insurgencies and the onset of civil war (which are two different subjects, but somewhat related).

Examining the eleventh lecture from Professor Michael Spagat’s Economics of Warfare course that he gives at Royal Holloway University. It is posted on his blog Wars, Numbers and Human Losses at: https://mikespagat.wordpress.com/

This lecture discusses analysis of cross-country datasets, correlations, and then discussed some problems with statistical testing in general. This is worth reading carefully in its entirety.

The datasets they are discussing in the first slides I assume are from the “correlates of war” (COW) dataset, a publically available data set that many in academia have used. We have never used it. When we created the MISS (Modern Insurgency Spread Sheets…now called DISS), we built them entirely from our own research.

He then looks at two different studies on the probability of conflict, one done by Fearon and Laitin (slide 5) and the other done by Collier and Hoeffler (slide 15). Even though they are based upon the same data, they produced somewhat different results (all, of course, to 90% or 95% confidence intervals). He summarizes the conclusions of the Fearon and Laitin study on slides 8 and 10 and the conclusions to the Collier and Hoeffler study on slides 16 and 18. It is worth comparing the differences.

Throughout this paper, he starts giving warnings about the problems with this analysis. First he discusses “story lines” on slides 12 – 14. This is important. One you have a correlation….then most people are clever enough to be able to explain why such a correlation exists, be it right or wrong.

But the part of the lecture that hit home with me starts with the statement that “These reported results may just have come out out that way by luck or chance.” (slide 20). The cartoon on slide 22 makes the point. Basically, if you test 20 different things, even if they are completely irrelevant, even to a 95% confidence interval; then with average luck you will get at least one correlation! Test enough things, and you will get a correlation. By the same token, add enough variables to your regression model and you will get a fit.

This is done all the time and I did discuss it in America’s Modern Wars, page 73-75 on a study done by CAA (Center for Army Analysis) in 2009 using our MISS. As I note in the book they identified 34 variables and then built a regression model based upon 11 of them and then boiled the final model down to four: 1) Number of Red Factions, 2) Counterinsurgent per Insurgent Ratio (Peak), 3) Counterinsurgent Developed Nation and 4) Political Concept. As their model was based on force ratio and political concept, it was similar to my regression model, except they added two more variables to the model. The problem is that one of those variables, “Number of Red Factions,” should not have been added. As I note in my book “In our original research we did not systematically and rigorously establish a count of factions for insurgency…. It should not have been used as a variable without further research.”

To continue from my book: “My fear it that this variable (“number of factions”) worked in their regression model because it was helping to shape the curve even though there is not a clear cause-and-effect relationship here. Also, because of the methodology they choose, which was establishing variables based upon statistical significance, as opposed to there being a solid theoretical basis for it, then I believe that statistically there should be around two ‘false’ correlations among those 11 variables.”

I end up concluding: “My natural tendency as a modeler was to make sure I had clearly identified cause-and-effect relationships before I moved forward. That is why my approach starts simply (two variables) and moves forward from there. It is also why I independently examined each possible variable in some depth. In addition, I reviewed and examined a range of theorists before proceeding (see Chapter Seventeen). I have had the experience of dumping lots of variables into a regression model, and lo-and-behold, something fits. It is important to make sure you have clearly established cause-and-effect.”

Anyhow, what Dr. Spagat warns of on slide 22 is what some people are actually doing. It is not a mistake made by grad students, but a mistake that a professional DOD analytical organization has done.

Enough preaching; the link to the lecture is here: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%2011.pdf