Archive for the ‘Corruption’ Category

Mathematics in Economics

November 17, 2007

I am entirely too impressed with myself at the moment. 

I have been working since this morning on trying to tease out a model from my various notes and ramblings on corruption, the topic of my economics thesis. Suffice to say that I spent a good portion of my time standing in front of a whiteboard jotting down equations, solving for n, binomially expanding, etc. 

I’ve been told (or is the word warned?) by my professors that mathematics is a tool of economics, but it is not required to do economics.  I buy that; however, had I not my knowledge of mathematics and the ability to apply it to economics, I would find it extremely difficult to formalize what would otherwise be a model weaved together by notions that often escape the notice or care of economists.  Economics has a language, and mathematics is a very important part of that lexicon. 

It is much easier for me to go to my professors, or other economists, with a set of descriptive equations, point at n, and say “That’s the state-civil society dissonance factor.  That’s what we need to find a way to measure.”  I am capable of weaving an analytic story, but why say with 1000 words what you can illustrate with a couple equations? 

Mathematics may only be a tool in economics, but more and more I find it to be an absolutely essential and integral tool for understanding and expanding on my field. 

Proposal Accepted

September 10, 2007

Today, Becca and I had our first meeting with my thesis committee:  Humphrey, Greenlaw, and Hansen.  I admit feeling a bit worried about how it would turn out because I have never had a class with Hansen before, and I had a feeling he would be the person that needed the most convincing.

In a strange twist, Hansen effectively talked us into doing precisely what we proposed to do:  develop a methodology.  Not to say he and the other professors did not help us narrow down the topic a bit more, rendering it more manageable; rather, the qualitative difference between what was proposed and what was decided would be a worthy thesis topic was not very significant. 

The thesis proposal, in its nearly final form, did not spend a great deal of time developing a model because that was never the intended focus of my thesis on political corruption.  The intention was to develop a schematism for translating the heavily descriptive work of anthropologists into the language of, at the very least, political economy, for the purposes of working out the ways in which social networks come into conflict with the state.  Humphrey, my main advisor for the thesis, seemed to think that the basis for desiring to develop this methodology needed more clarity, and so suggested that I spend more time formally developing and critiquing the model that characterizes the mainstream literature. 

It would appear, however, that this distracted unnecesarily from the main point: an alternative model is useless if it can not be tested, and since alternative models of corrution based on social networks appeared to lack an associated methodology, to even talk about testing  is premature.  We were proposing to develop that methodology, but with half the pages of the proposal devoted to developing models, the focus of the thesis was rendered unclear.  It struck Becca and I both as funny that in the end, Greenlaw, and Hansen especially, cared a lot less about models and a lot more about methods, and effectively talked us into focusing on the latter (our original plan), rather than the former.    

I have been warned, as mentioned before, that developing methodology does not provide the fame and fortune that an ambitious economist might seek.  Certainly my economics education these past four years has not explored alternative approaches (not so much a complaint – methods are rightly secondary to concepts in undergraduate work, methinks, and traditional methods are more than adequate for most of the problems that undergrads will take up).  Still, there is something to spending a class period or two to demonstrate that great strides are not always made in the synthesis of new concepts, but also in new approaches to familiar problems.  Stagnation of or dogmatic commitment to either concepts OR methods will doom any quest for knowledge.     

Search Terms for the Interdisciplinarian

August 15, 2007

Over the past few years, I have found that one of the hardest parts of research is devising a set of search terms.  I do not know if I am in the minority in this or not; I do tend to pick topics that flirt with the lines between different disciplines and my own, and on that line new ideas are often fleshed out and new words/phrases are invented or reinvigorated with a certain meaning in order to express those ideas. 

First, the obvious:  Why do you need a good set of search terms?  Perhaps the better way to phrase that question is “How do you take your research topic/question and decide what the key terms are?”  This is not always as straightforward as it might at first seem.  For instance, my topic is corruption, but more specifically it is a critique of the anticorruption agenda on the grounds that it ignores important sociohistorical and politicoeconomic intricacies in developing and third world nations.  At most out of that previous sentence, search terms could include:  corruption, critique, “anticorruption agenda”, development, and some play on the words sociohistoric/politicoeconomic/intricacy. 

The problem is that, in my humble opinion, such a critique is necessarily interdisciplinary.  This means that I am attempting to recognize and label ideas for which I have no specialized terminology or training.  I am, in effect, trying to reinvent the wheel.  If you place those words into a search database, you will get a long, very mixed set of search results, and going through them one by one to discover the key is a painstaking and inefficient process.  So you need a good set of search terms in order to acquire a strong, concise set of resources, which will define your starting point in your research. 

What, then, is a poor undergrad student to do in this situation?  Your partner the anthropology major is out of the country and your professors, as interested as they may be, are not in positions to help you redefine your set of search terms.  So you’re stuck staring at a Google Scholar page that has returned 1500 results on your search terms.  I think once you get to this point, you have to start improvising; find some roundabout way of getting at the sources you REALLY want.  Possible strategies:

1) Wikipedia:  Regardless of how you or your professor feels about the veracity of the entries in this online encyclopedia, it is often a good a first approximation of the issue at hand, and the fact that it is web-based means any theory worth its salt will probably end up on it.  Perhaps one of those theories comes close to describing your intended topic. 

2) Find an author:  This may seem a silly idea in light of the above discusssion; after all, how can you have a good author if you can’t run a good search?  Well the idea is that you don’t necessarily need THE seminal work on your topic.  What you need is someone who is making at least SOME of the claims that you would like to make/study.  You can get a source like this by going through articles/books you read for older classes or research projects ( NEVER DELETE/THROW AWAY OLD SOURCES FOR OLD PROJECTS).  Anyway, once you have that author that kinda sorta gets it, run her name through Google Scholar and see what comes out the other end.  She may have written something closer to what you want, or you might find someone who has cited her and who comes closer to what you’re trying to say.  Read that paper and if its close enough, devise some search terms based on that paper.  If it is closer but not yet perfect, then search the author again, or search using the new author that cited the original paper.  This is a process of elimination that can be very effective if you’ll take the time to read the articles/books you turn up.

3) One final possibility, probably often looked over by undergrads, is doctoral and masters theses.  Perhaps I’m just an optimist, but I tend to think doctoral/masters candidates are willing to be a little more daring than the establishment they are seeking to join.  Thus, a search on corruption (or some other topic) in dissertations may turn up a slightly more radical set of results than on an established academic search engine.  Also, because institutions are offering more and more specialized degrees, you may get lucky and find a department somewhere devoted entirely to the study of your topic.  Chances are that they have students who have written dissertations that are just what you are looking for, but not necessarily available on your local dissertation search engine. 

By the way, it turned out that the best search terms for me were:  networks, authority, legitimacy, corruption.

Who knew? 

Research as a Knowledge Creating activity

August 1, 2007

Some time ago I read a post at Pedablogy concerning the creation of knowledge and the way it is communicated to students.  This notion of knowledge creation fascinates me on a philosophical level, but it also has important practical implications, especially for my senior research thesis. 

As scientists, we subscribe to the relevant version of the scientific method, and I may be granted that the scientific method provides a general framework of how we match a priori theorizing with a posteriori empirical observation.  When the theory is shown to predict the observation, and that within some degree of tolerance (which can be fairly variable), and when those experiments are repeatable and the results verified, then at some point it becomes knowledge.  I have found it to be extremely tempting to develop theory a priori with the intention of testing it through simulations.  It is tempting because data collection is not only costly, but also because data is often difficult to acquire and full of holes or open to charges of imperfections.  Thus in economics it has become common in my experience to develop a clever model, then bask tautologically in one’s cleverness by running a simulation within the confines of the model without any accountability to the real world. 

I for one would prefer not to fall into this temptation, because I think ultimately it leads to obscurity.  However, I am also keenly aware of the fact that the difficulty of providing empirical evidence is that, simply put, data often does not exist.  In these instances, data sets must be somehow painstakingly created and models adapted for use with that data.  These data sets are often imperfect, but I think they are an important development in the social sciences.  A good example may be Transparency International’s Corruptions Perceptions Index (CPI).  While I think there are many reasons to worry about the underlying assumptions upon which the CPI is founded, I really admire the attempt to create such an index and a similar one at the World Bank.  Corruption and its effects are terribly difficult to measure because they are illicit in nature; no one will simply own up to their crime.  Indeed, this is a major obstacel in any study of illegal or taboo behaviors.  So if you want to measure and analyze any black market with any sort of accuracy, you have to be very original in your approach; that is, you need to have a new methodology.

And this is precisely the point:  Methodology.  Your methodology is where you decide what data to collect, how to collect it, and how to apply it to your a priori theory.  It is in your methodology that all your assumptions come to the fore.   It is (or at least it can be) the place where the most creativity and imagination is required because it is precisely here that pure method (mathematics and survey methodology), abstractions about the world (a priori assumptions and theory), and data all must be fit together into a cogent theory. 

 I really believe it is in the methodology that the most reform must take place.  In the absence of data in the form of dollars and cents, economists need to think about alternative methods for collecting, synthesizing, and evaluating data and models.  I think this is where I would most like to focus my attentions in my honor’s thesis.  I have been warned by a person who shall remain nameless that few economists have ever achieved fame and fortune for their work in methodology.  This may be true now, but I wonder if that trend will continue.  In any event, my overwhelming desire is to increase my own level of knowledge; being recognized for that is just the cherry on top. 

The Cancer of Corruption

June 28, 2007

Last weekend, my research partner and I agreed that we would work collaboratively on a thesis that deals with corruption, especially in the third world.  Political corruption, which gets defined in numerous ways, receives a great deal of blame (not necessarily unwarranted) for the failure of many aid programs, as elites and authorities siphon funding and resources into their own coffers for personal use. 

So we make a rather radical suggestion:  Suppose that corruption, as we call it, is integral and functional within the system; that is, were a “corrupt” official or institution to be suddenly removed from the equation, this would create a gap that would actually act to interrupt otherwise positive-growth activities.  Furthermore, corrupt practices are manifestations of the universal phenomenon of a person or organization that uses its vested power to prevent access to resources unless payment is made to them.  Under this perspective, whether you are paying a bribe or an income tax, you are participating in the same generalized system where access is restricted to those who pay the bribe/tax/fee/etc.

Taxes as glorified bribes?  In a sense, yes.  If you don’t pay your taxes, do you still get access to the system?  No.  In extreme cases you go to jail.  Analogously, if you don’t pay the armed officer on the road in Kenya a bribe to get through his roadblock, will you still get your goods to market?  No — or yes but only with a great deal of effort (travelling off road, which will take longer, be more dangerous, and make market activities less profitable). 

We grant there are differences, but the analogy is too valid to be ignored. 

The possibilites for research are three:

1) Fit corrupt practices into a more general theory of state formation, such as the roving/stationary bandit model developed by Mancur Olson and other institutional economists.

2) Identify particular practices that are labeled as corrupt and demonstrate that they are not.

3) Develop a methodology for testing the theory utilized in 1).

 More on these later.