As Professor Casey emphasized early in the year during our class, economic models are useful and often necessary to explain key concepts, but, at same time, they cannot provide complete answers to problems. Models are formed by creating assumptions that simplify the real complex nature of economic scenarios. Therefore, they should be used a tool for understanding, but not viewed as a vehicle of absolute truth because of the unrealistic conditions that provide basis for the model. Paul Krugman, in his article, “The Fall and Rise of Development Economics” propagates this viewpoint as well, as he examines the history of the field of study.
Krugman’s example of Fultz’s simple weather model of a dish-pan filled with water in contact with a heating element exemplifies the limitations of the model and the untrue, unrealistic assumptions, but, more importantly, the model offered insight into how the complex system behaves.
In the same manner, economics adopts this approach, but the discipline of development economics shied away from modeling under Myrdal and Hirschman who choose not to formalize their ideas. A major shortcoming of verbal persuasion is that the argument is presented in a way that it sounds as if must be true, while models show what might be true. Economic theory is expressed through models. In order to be taught, and thus to endure, theory must be accompanied by a model. Why then did the developmental economists of the 1950s abandon this approach? Krugman answers, “Almost certainly for one basic reason: high development theory rested critically on assumption of economies of scale, but nobody knew how to put these scale economies into formal models.”
Rosenstein Rodan’s “Big Push” paper was published in 1943, but was not formally modeled until 1989 by Murphy, Shleifer, and Vishny. These economists demonstrated that it was indeed possible to tell high development style stories in the form of a rigorous model by not being bogged down by the inability to model imperfect markets, but rather by simply taking chances and “daring to be silly”. Sometimes the greatest insight can be gained through the simplest of models, even if it takes time for this to be realized.
In many ways, Krugman’s piece on models and their relationship to the field of economic study dovetails nicely with the article we read last week by Dani Rodriks. In his article, Rodrik does a great job critiquing economic modeling. He concedes that models are a great starting ground for understanding complex, real-world issues but suggests that the results from these models must be interpreted in context. In the case of economic modeling, or any kind of modeling for that matter, the context is inherently an oversimplified version of the reality the modeling is seeking to represent. Krugman builds off of this notion when he discusses the ways in which development economists like Albert Hirschman turned their back on “mainstream” economics for precisely the reasons Rodrik’s discusses. Development economists, like Hirschman, realized that the world they were observing simply could not accurately be described within the confines of traditional economic models. Rather than trying, and ultimately failing, to apply inadequate models to their observations, many development economists simply neglected to include models in their publications. Without the tangibility that models provide, many early theories about economic development fell on deaf ears.
Krugman’s point about the narrowness of vision that occurs whenever one simplifies the world around him in order to model it was particularly thought provoking. Krugman suggests that economists like Alfred Hirschman were in many ways far ahead of the science of economics itself. They were disinterested in the monotony of thought surrounding competitive markets and constant returns. However, because they branched out from the filled in parts of the economic map, they were largely ignored. Hirschman and his contemporaries were treading in uncharted waters and did not have the tools needed to convey their findings to community of narrow sighted economists who were accustomed to seeing economic theory in tight models. However, while the findings of these early development economists were discredited and/or ignored, the process was an important step forward in the field of development economics, for as Krugman writes, “a temporary evolution of ignorance may be the price of progress, an inevitable part of what happens when we try to make sense of the world’s complexity.”
In all of my economics classes, at one time or another my professor has always highlighted the fact that simplistic economic models fail to capture the reality of the world because they assume that a change in one variable can occur while holding all other factors ceteris paribus. This represents one of, if not, the biggest flaws of economic modeling. Due to the reliance on this ceteris paribus assumption, economic models lose their predictive capabilities when multiple exogenous shocks occur at once.
Yet, despite the flaws of simplistic modeling, often times these models represent the best guess, or in other cases, they represent a general truth that cannot be explained any better. Take for instance Adam Smith's proverbial assumption that capitalism is driven by an "invisible hand". While this statement lacks any mathematical derivations, or any tangible variables to test , it precisely accounts for the equilibrium of supply and demand in an economy. Even 225+ years since his general assumption, scholars and students still understand the basics of economics from this simple model. Hypothetically, think of the ramifications if economists would have overlooked Smith's basic ideas because they lacked concrete mathematical or empirical support.
In a similar sense, it seems almost academically irresponsible that social scientists would dismiss the work of Hirschman purely for its lack of formalized rigidity. Because of this oversight, economists spent years trying to develop revolutionary new models to explain factors affecting development in low income countries. In reality they had the answer sitting right in front of them.
In essence, this paper harks back to a simple idiom "if it ain't broke, don't fix it". In this case, Hirschman's model provided a perfectly good explanation, but others attempted to revamp his ideas, and in doing so lead themselves to more questions and less of an understanding on the topic of developmental economics.
I agree with Katie that modeling of development economics was abandoned in part due to the growing complexity of models. In terms of Krugman's article, I thought it was particularly interesting that he followed up this point with the African mapping analogy. As time went on and European maps of Africa became increasingly accurate in scale and coastline shape, they became far less accurate in the heart of the continent. This was not because Europeans were learning less about mapping; it was because cartography was becoming a more accurate science. Imprecise and second-hand details were no longer good enough to include in maps. It took an extended period during which more precise and reliable data was gathered before maps reached modern accuracy.
Krugman considers the science of economic modeling to be very similar to cartography. During the middle of the 20th century, the science became more exacting and previous assumptions turned into dark areas while economists worked out acceptable ways to explain and model them. Because the field of development economics has tended to be a less mathematic economic subdivision, its modeling advancement has lagged. Development theories are based on so many assumptions, and cultural differences across developing countries make generalizations even more difficult. This makes comprehensive models extremely difficult in development economics, and requires economists to take chances ("dare to be silly").
Krugman's analysis of the notion that models are imperfect is spot on with what we have been discussing in class. Models are constrained by ability to make them and resources allotted. He points out that models will never be right as laws of physics are, but rather should be something that conveys insight. Therefore, models are necessary but often not sufficient and not applicable in all scenarios.
Krugman also talks about the idea that many ideas that are expressed through rhetoric do not become important in the grand scheme of things until models are applied to them. Often, an idea will emerge and no one will consider it until years down the road...for example the map of Africa concept. Moreover, during the modeling process, basic information can be lost and not regained until years later. He explains that "during the process of model-building, there is a narrowing of vision imposed by the limitations of one's framework and tools." "The problem is there is no alternative to models."
The form of economics of the times makes some ideas with potential seem meaningless. When people were talking about development economics in the 1950s, not only did they not have the modeling tactics needed, they also were subject to the constraints of the times-economists at the time focused on a perfectly competitive economy (no room for the idea of economies of scale).
Economic modeling is, like any other science, an imperfect system. Modeling attempts to compare different instances in the same manner, which in many ways is useful because it provides one form of a benchmark to go off of; however, it is not something that should be religiously followed, as different circumstances require different actions and no one formula should be regarded as the solution to all problems. One thing that particularly stuck out in my mind in Krugman’s paper was the analogy he presented about the maps in Africa over time—while coastlines and detail of the outskirts of Africa became more precise with the advancement of science and technology, the interior of Africa became less and less accurate, losing information about cultures and towns. This reflects how science can lure people into a false sense of knowing, when in reality what it provides is more information about one thing as opposed to general knowledge.
Furthermore, Krugman points out the simplicity of modeling when he comments that Rosenstein-Rodan’s model could “have been written as easily in 1955 as in 1989.” It seems silly to suggest that something that would have been extremely relevant in 1955 would have the same impact 35 years later, but yet modeling has a timeless factor. While in some ways being able to compare things historically is critical, it again raises the point of how different the world can become in the span of a few decades, and these should not be the only things that development economists rely on.
Krugman’s paper “The Fall and Rise of Development Economics,” when read with Rodrick’s “Growth Theory,” fully define the spectrum in which economic models should be used and applied in order to better understand the world. While Rodrik emphasizes the risks associated with hasty policy applications of economic models, Krugman emphasizes their utility in economic application, despite their inherent “falsities” of simplifying reality.
Krugman uses the example of the “high development” economists, who had a theory that economies of scale played an important role in determining whether returns to scale were increasing or decreasing. However, the prevailing attitude necessitated theory to be strictly mathematically proven. Since they lacked the tools necessary to mathematically substantiate their claim, Hirschman and company instead employed metaphors and narrative reasoning to explain this phenomenon. However, mainstream economists simply dismissed the claims made by Hirschman that economies of scale were important on the grounds that it wasn’t empirically modeled. Instead, Lewis’s two-sector model that didn’t include the hard-to-model economies of scale became the model of the times, shaping the “fall” of economics.
Krugman criticizes Hirschman on his impatience in modeling his observation, which was ultimately correct. By accepting the restrictions imposed on him by the academic community, he could have potentially formulated a model that explained the theory of increasing returns to scale, which would have subsequently shaped a different approach to development economics, which according to Krugman, would have been much more effective. More broadly, though, Krugman faults the academic community that dismissed Hirschman’s ideas due to the lack of mathematical formulation. By using the analogy of Africa’s mapping over the centuries, Krugman argues that by so vehemently restricting the type of information considered for inclusion into economic modeling, economists run the risk of leaving out potentially too much, and ending up with a model that doesn’t explain much of anything, which is inapplicable in solving the very problems it was made to deal with.
When reading this article I enjoyed the commentary about the imperfect model. In order to create a model, Krugman emphasizes that a economist or any scientist must make guesses about what is important and improve the model after the system is "learned". Since there are no alternatives to modeling we must use the models in order to improve our knowledge on the subject at hand while keeping in mind that the theories provided might not be complete or work in every situation. This reminded me of Rodrik's observation of Chinese development. The neoclassical economic growth model would not have predicted significant Chinese growth without more major changes in policy. However, China has grown rapidly from the late 1970's, demonstrating just what Krugman suggests. The ideas that we have are too far simplified in order to accurately predict what will happen in each developing country but they are a good starting point in the study of economic development as a whole.
Studying the history of economic models has benefited my overall understanding of each model brought up in class. It's been helpful to see where past economists went wrong and what they have done right. The trend in methodological trends Krugman describes may be thought of in terms of a clothing trend analogy. Clothes go in and out of style, and the ideologies of the time may have effects on that cycle. Maybe the clothing trends evoke a particular way of lifestyle that the ones who rejected it just could not understand due to historical implications, such as war. Yet the ones who revitalize the trend must associate themselves in some way with the originators. Krugman's theory is that the trend of economics' increasing mathematical style created an inability in some economists to understand the non-mathematical style of the high development theorists. Krugman himself cannot understand how such a "simple" model could have gone undiscovered for so long.
Despite the difficulties in using models to predict economic outcomes, Krugman emphasizes the "false feel of security" that people feel without a tangible model to manipulate.
Since this article is a rather short one, and very to the point, most of the subject matter has already been covered in the blog. It was refreshing to read an article written in such a relatively informal and humorous way. The author displays a sense of humor which contradicts the stereotype of dry economists.
As to the usefulness of models, I think it also important to point out that models and econometrics can be incredibly useful tools to testing the efficacy of development policies. Models are tools, and cannot account for everything; they are not the ultimate solution, but applied properly can lead economists to practical policy recommendations. We discussed the case of conditional cash transfers in Nicaragua, where Gitter and Barham used data from control and test communities to determine the effects of the transfers on household expenditures and education for children. Their modeling and regression analysis allowed them to determine that the policy was indeed effective in the case of Nicaragua. Modeling and econometrics can be useful, therefore, both to predicting factors which may speed the general development process and evaluating policy choices within a given country or community.
In this refreshingly colloquial essay, Paul Krugman illustrates the shortfalls (and really, the lack of) of modeling in development economics. He argues that the initial problems of modeling—before the 1940s—was a lack of understanding how to model economies of scale in high development theory. But as the 1940s ushered in era of rapid advancement in economic theory, development economists lagged behind the times. Krugman stated that these breakthroughs "led to a much improved level of understanding of some things, but also led for a time to an unwillingness to confront those areas the new technical rigor could not yet reach." As many others in the class have noted, Krugman's paper correlates nicely with Rodrik's essay from last week and our class discussions in general. Taking a theory with simple, ceteris paribus assumptions and forcing it onto an increasingly complex and ephemeral world makes for "lousy" policy. Krugman criticizes development economist, such as Albert Hirschman, whose overly simplistic models lead to many years of an "evolution of ignorance." Krugman is quick to point out that while models help substantiate theory, they are only as good as their underlying assumptions. Models and theories need to be robust and adaptable in order to be applied uniformly.
I agree with Amy’s notion that the article provided a nice change of pace in terms of both format and writing style. Although Rodrik’s piece was mainly prose and did not make widespread use of statistics or numerical tables, it still depended heavily on tables and analysis of existing economic models. I was actually a bit surprised by Krugman’s style and prose and found his overall message relatively easy to understand. The takeaway message that I understood from the article—namely that one cannot discount non-modeling approaches to economics just as one cannot base a development approach purely on non-quantifiable data and information—seems a bit obvious to me. Although not an economics major, it seems to me that “suggestive metaphors, institutional realism, interdisciplinary reasoning, and a relaxed attitude towards internal consistency” would be absolutely critical to studying the people and interactions that, in aggregate, make up an economy (Krugman 10). I think it is to the credit of Washington and Lee that our academic departments recognize the necessity of interdisciplinary studies and allow Humanities majors like me to study the economic implications of major historical and political events in our Humanities classes. These lines of seemingly non-economic inquiry and knowledge are what intrigue me most about development and are what I hope to learn more about as the semester continues. I was happy to see that these kinds of “economic” research and inquiry—at least in the eyes of a Nobel Laureate—are both usable and critical to development studies. Now if only I could convince potential employers of truth of that opinion…
Paul Krugman raises many important points when it comes to the utility of these economic models. As mentioned before by others, models provide a "tangible" way to simplify the complexities seen in the real world. These basic models enable us to gain a better understanding of the world, which is not to be mistaken with a complete, correct understanding of the world.
Studying the origins of these models, such as the Harrod-Domar model from the 1950s, makes it easier to understand the motivation behind such models. As we discussed in class, it made sense that the Harrod-Domar model showed that large sums of investment will create physical capital in countries. Since most of the world, especially Eastern Europe, was in shambles after World War II, the main way to restart productivity and growth was to build infrastructure.
As right as he was, his reasoning was insufficient. Certain countries in Europe had the necessary institutions in place so the investments did help bolster the economy. In developing countries, these institutions are lacking.
If we go by Rodan's Big Push Theory, then government intervention is necessary to initiate the "push." In many cases, such as Pakistan, the government is in constant turmoil and does not have the power needed to start such an initiative. If we keep going with this idea, the next logical step is to look into foreign aid; however, corruption will impede any progress that would be possible from foreign aid. What happens then?
Krugman brings awareness to the fact that there is so much uncertainty and policies cannot stem directly from these models.
I feel like I can't elaborate much more than everybody else, but I did have an interesting conversation with my roommate. She's a science major and was also talking about models and how great they are ceteris paribus, and how tricky it is to apply them to humans. While hers are more psychologically based, a comment she had from class was that "models are not right or wrong they are simply useful or not useful." I think this pretty much defines Krugman's paper. He acknowledges the importance of models to gain broader insights and make predictions, but not to determine specific policy decisions. While these models can be helpful there is no experiment/model that can be 100% right, or every 95% right. The only thing that can be achieved, is usefulness.
It is very interesting to see that the high development theory put forth by economists in the 40s and 50s, discredited for not being formalized with models, actually makes perfect sense and has a useful application for understanding problems of developing countries. Krugman points out that development economic theory might be in a better state had these ideas been formalized twenty to thirty years earlier. We might have had twenty or thirty more years to make important insights into what may work and what may not for developing economies. What would have been the direction of development economics if institutions had have adopted theories like the Big Push instead of constant-returns and perfect-competition models? Krugman points out that it is also important to remember that simplistic models don't often translate well to practical policy, so who knows how these models would have affected development economics had they been formed back when the insights were first put into writing. We can only assume.
I found the article both interesting and entertaining, and it provided a great complement to Rodrik's piece that we read last week. It is unfortunate that an entire field of economics fell by the wayside only because certain elements could not be encapsulated into a model. As many have already mentioned, we spend a lot of time talking about the limitations of models and how they are simplified versions of reality and often lack empirical credibility. One wonders where development economics would be today if these roadblocks had not been encountered. What we now know is that while modeling can encompass some aspects of the development story, it is not the complete picture. Each country has a different solution to the development problem that no model can predict accurately. Make no mistake, Krugman accurately points out that there is a place for models and that they can prove very useful, but as we all know, they have their limitations.
An inviting piece - I appreciated Krugman’s meteorology metaphor in describing the limitations but necessity of models in the social sciences. Using high development theory as an example, he attributes its fall from mainstream economics and then resurgence as a result of the ability for economists to formally model the theory. He believes that modeling in Economics is essential, but that broad minded individuals such as Hirschman believed that more insight was lost than gained. Krugman points out that, while that may be true initially, taking the first steps to build a simple model eventually leads to the ability to rediscover older knowledge using the new techniques. He cites as an example Norwegian scientists’ discovery confirming folklore about cloud formations before a storm. As Development Economics recovered from its slump, Krugman said it was done by “daring to be silly: by representing the world in a dish-pan, to get at an essential point.” Like Benjamin pointed out, this piece also links well to Dani Rodrik’s “Growth Strategies”. Krugman concludes that “a temporary evolution of ignorance may be the price of progress, an inevitable part of what happens when we try to make sense of the world’s complexity.” Rodrik takes that idea of complexity and points out that, maybe seeking one unified theory is done in vain. “They key is to realize that these principles do not translate directly into specific policy recommendations.”
This article certainly makes me think more about the use of models in economics. Having only taken econ 101 and 102 prior to this class almost everything I have learned about economics has been taught to me by various economic models. Even when we looked at real life events we always brought things back to the models. Thinking about it after reading this article I realize that the models we used are really incapable of fully predicting any outcome in the real world. The world is a complex place and there will never be a model that can consider all of the potential variables to be able to predict a definite outcome. This however doesn’t mean that models aren’t useful. For one thing they are necessary to teach economic principles to the economists of tomorrow. Additionally they can certainly be useful to economists in predicting the most likely outcome. In the end models are both our only hope and our greatest disappointment. They are one of our most powerful tools in striving for a better quality of life but at the same time will never be able to provide us with a definite answer of how to meet our end goal In the field of development economics models can be especially tricky. There are so many variables when looking at all of world’s economies that it is nearly impossible to choose the variables most important to include in a model to achieve satisfactory results. The truth is while there are many common reasons that a country may be underdeveloped there are so many other problems affecting individual countries that you almost need to come up with a solution for each country rather than trying to just say x amount of education + y amount of healthcare and nutrition + z amount of nutrition = economic growth and an improved standard of living. When looking at how tough it is to use models in development economics it starts to makes sense why one would just give up the models all together. The problem with throwing out the models is that the economic community and the world as a whole need the models as proof that what is being proposed will work.
There is not much more to be written about “Big Push Theory,” as we read it and discussed it in class Tuesday, but the article hit on the history of the theory. It began as a string of ideas with no model and assumptions about returns and market structure that didn’t fit with the mainstream philosophy of its time. It took 35 years to advance economists modeling capabilities in order to reexamine the idea of the Big Push. The “what-ifs” concerning development had Rosenstein-Rodan been capable of modeling his idea in the 1950s are endless, but as Krugman points out, they are still just “what-ifs.” We cannot change the past. We can only continue to work on further development and not making the same mistakes twice. The article we read brings up a point in the conclusion that I really admire. Both the ability to go out on a limb with one’s ideas, even if they are not completely clear, formulated, or entirely make sense yet, as well as the ability to learn from each other, regardless of the others education or experience is necessary to the development of any field of study. The local farmer may have the best solution to rural labor economics, even without any formal economics training. He certainly knows more about rural labor than the Harvard economics graduate, although sadly, we are almost always more inclined to believe the college graduate. Secondly, if we wait until we know everything about a subject, we will never be able to publish our work or succeed at anything. Krugman’s point is that we can never know everything. Our models, metaphors, and data, no matter how good, only carry us so far. If we had perfect models, there would be no need to keep working, and it would fit for everyone. We know that empirically, every situation/nation/economy is different. I think what Krugman is hoping to show is that just because one doesn’t have the ability to fully express his ideas, he should still be willing to put them on the table and collaborate and eventually they may become a fully developed theory which might be used to the betterment of the collective.
I enjoyed Krugman's analogy to the mapping of Africa as a good portrayal of model formation in development economics. He points out that while we start slow in trying to understand a certain topic or area, we become more confused with the access to new information and our ideas crowd together and we essentially obtain no more value than when we first started our approach. When applying to economics and all fields that utilize models, the same thing can be said of trying to tackle the whole idea at once. If broken down into smaller variables and attached with assumptions, we can generate a closer depiction but still not satisfy the big picture. Just like growth and development are hard to sustain, so are the ideas represented in the models of growth and development.
Krugman also explains that simplifying models does not always make them more understandable especially when their designers are highly intelligent over-thinkers. When we are forced to narrow our vision when devising the models, we have to be careful to not leave anything of substance out or the whole idea will present itself more as a loss rather than as a gain. But in reality, things are not as simple as the models make them seem and therefore we cannot obtain an accurate replication of how economies act and react.
Maybe the best representation and explanation is to see the economies in their physical form and gain insight through observation and experimentation rather than a simple graph. Complexities are going to remain complex as much as we try to simplify them and sometimes (as with the Africa map), the simplification process will cause more confusion than was already present.
The most interesting quote I take away from this essay is when Kruger writes about the Big Push model, “A model like this makes one want to go out and start measuring, to see whether it looks at all likely in practice, whereas a merely rhetorical presentation gives one a false feeling of security in one’s understanding” (Krugman 11). This quote highlights the clear differences between economists and politicians. Economists generally develop theories and collect data to support those theories, always asking the question, “Does the data support that claim?”. Politicians, as Krugman points out, base their ideas on what is in their self-interest rather than theories supported by data. While I understand that people, including politicians, are motivated by rational self-interest which is an assumption for most economic markets, it is still frustrating to hear politicians spew arguments and ideas with the conviction that they speak the undeniable truth. In reality, they are often making up stories that will get them reelected. These stories often make some logical sense, but are completely unsupported by the data.
The quote from Krugman made me think about earlier in the semester when we discussed how politicians, especially in the U.S., often ignore research done by economists and other social scientists and create policy based solely on politics. I realize I have gone off on a tangent, but these models are useless if they are never used to implement policy to encourage development.
Given that most of the material concerning economic modeling has been covered already, I figured I will point out the interesting analogy concerning Europe's ignorance in mapping Africa. In that elaborate analogy, Krugman talks in relates the progress of development economics with the progress and setbacks of mapping the not-so-easily traveled terrain of Africa. In that analogy, Krugman points out that as the standards of measurement improve, the standards of observations become much more strict. Whereby, limiting the available data. Similarly, Krugman argues that as economic modeling has improved in both mainsteam and development economics, less attention/thought has been given to theories that do not specify some sort of model. Without a model, the theory does not provide a model inherently does not provide a way to test the theory, nor a method, in which, to easily teach it. Though this argument certainly holds much weight when concerning the practicality of the situation. However, this leads me to wander that if by being so picky towards theories that provide models, we are holding back the progress of economics at least to an extent. By this, I mean that if attention is only being given to theories with models and all other theories are discounted outright or not even published, could we be missing some important discoveries that simply have not reached its final form? By not providing a format in which people struggling with completing their theories are we doing more harm than good? Assuming that there are not many outlets for a theorist without a model to discuss an incomplete idea with some of the brightest in the field, can we omitting would-be ground-breaking theories from our ongoing history of Development/mainstream economics?
As several individuals above have already expressed, I thought Krugman’s contrast of the physical science disciplines with the social sciences was especially interesting. Beyond the specific example he utilizes of Fultz’s dish-pan experiment, in more general terms, when analyzing a complex system it is almost always the case that reality must be sacrificed for specificity or vice-versa. I think the content nature of the social sciences, with its potential to more directly impact the sociological functions of daily interaction, contributes to the comparatively negative response to economic modeling. In the physical sciences, it is common practice for real-life problems to be modeled through the experimental method, where by controlling for all variables except the single aspect being studied, scientists can assess the influence a specific factor has on the larger situation. It seems like the more concrete, determined nature of the concepts and ideas in the physical sciences actually makes it more acceptable for those items to be controlled for and simplified in experimental procedures. Because economic principles are often so innately intuitive yet less tangible, we are less willing to sacrifice them for the simplification that is necessary in modeling complex systems. Ultimately, I, in agreement with many of the comments above, believe the most valuable conclusions from Krugman’s article are the needs for creativity and compromise in the methodological approaches to economic analysis. This need is reflected as Krugman writes, “economists were locked in their traditional models, non-economists were lost in the fog that results when you have no explicit models at all” (12). In other words, breaking the barriers that exist between different disciplines and their respective methods and ways of thinking is ultimately our best chance at garnering insight into the functioning of a complex system.
I like how Krugman explains the problem of high development modeling. For me, this helped me understand whey high development theory had such a lull. THese theorists' ideas made sense, but no on knew how to model the imperfect competition market structure of developing countries. Krugman's paper spends a lot of time discussing economies of scale and what they do for a developing economy and he recognizes that the Big PUsh model does address imperfect competition, but only under certain circumstances. However, this limit-pricing assumption makes any forward linkages impossible. He does criticize Hirschman for being an economist of style rather than substance in the end. I think Krugman's final message lies in the fact that it is difficult to reconcile economies of scale with a competitive market structure
It is interesting that oversimplified models that are based on so many untrue assumptions seem to provide the greatest insight into the economic processes of a country. Even though economic models fail to illustrate reality entirely, they still provide tangible concepts in order to better understand economic observations. It is human nature to be proud and maintain one's values, so it is no wonder the presence of assumptions becomes the subject of criticism when analyses of economic models threaten one's own beliefs. When concern arises over the simplicity of models and the assumptions they make, one can argue ignorance. The map of Africa is given as an example. First, information about the coastlines was lacking in the 15th century, while details about the interior were abundant. By the 19th century, the shape of the continent was known and outlined, but the interior landscape of the continent had "emptied out" and was largely ignored because it required more accurate and precise details of, say, the topography, for example, as improvements in technology and resources to gather that information became more abundant.
As Rodrik emphasized in the paper last week, it is still important to consider each individual country, when applying economic models, given the differences in culture and what Rodrik called "pre-existing institutional landscapes." This is the very reason that high-development economic models are hard to illustrate and apply.
I know this paper largely talks about the shortcomings and great successes of models, and that they should be analyzed with careful consideration. All of the talk about it just kinda made we want to dig as deep as possible when I finally got to the model at the end.
The output curves created from the modern and traditional sectors were drawn in a way that the modern curve was increasing at a much higher rate than the traditional curve. The relative wage curve was drawn in between the two curves so that it would be profitable to start modern production at a high enough level of output. If the modern curve was not growing at such a high rate, or the relative wage curve did not fall below the modern curve none of these stories would be true. The stories revealed from models can provide great insights, but we have to realize we are always one component away from the model painting a completely different picture.
The discussion of the imperfect model in the “Fall and Rise of Development Economics” helps put into perspective the reasoning behind the use of economic models. I was reading through some postings and found a quote that summed it up really well. I really liked the point Michelle’s friend made about how “models are not right or wrong they are simply useful or not useful”. It is not about getting something completely right. It must be accepted that economic models cannot be perfect. Rather they can be useful tools to guide thought and action.
I also found Krugman’s comparison between early development economics and European mapping of Africa interesting. It was a really intelligent comparison because of how well it fit and how easy it was to explain. Both increased their standards of understanding to the point that both the mapping and development economics dropped off for a period of time.
I really enjoyed the way this article light-heartedly explained the processes economists have gone though over the past century to uncover a model to explain the frame work of the economy. By breaking any model down, it is difficult to decisively create a model that incorporates all the variables society faces and create a broad over arching theory. Although the assumptions we make with models may make it seem overly simplistic, such as with the only resource being labor in the big-push model, they allow us to see the economy though one of the necessary inputs. His example with the dishpan made me think of middle school science experiments: limited tools and basic knowledge are available, yet one can still covey the main concept to a large audience. In many ways, devising these overly simplistic models can be very useful to the public more so than an a more elaborate model. The more people that understand basic economic framework can lead to a more productive society since it is better educated with making practical decisions. People will be both able to use political as well as economic insight to make informed decisions if they are able to understand the essential components that lead to economic growth. Thus, even the simplistic development theory models are useful to society. As Krugman says, economists must have a "willingness to do violence to the richness and complexity of the real world in order to produce controlled, silly models that illustrate key concepts."
This article seems to highlight the general state of economic models... they are just that. The models only go so far and often fail as a result of an outside shock. Within development economics models can only go so far to explain the state of a developing market. I liked Krugmans analysis of the disconnect between politicians and economists.
As Professor Casey emphasized early in the year during our class, economic models are useful and often necessary to explain key concepts, but, at same time, they cannot provide complete answers to problems. Models are formed by creating assumptions that simplify the real complex nature of economic scenarios. Therefore, they should be used a tool for understanding, but not viewed as a vehicle of absolute truth because of the unrealistic conditions that provide basis for the model. Paul Krugman, in his article, “The Fall and Rise of Development Economics” propagates this viewpoint as well, as he examines the history of the field of study.
ReplyDeleteKrugman’s example of Fultz’s simple weather model of a dish-pan filled with water in contact with a heating element exemplifies the limitations of the model and the untrue, unrealistic assumptions, but, more importantly, the model offered insight into how the complex system behaves.
In the same manner, economics adopts this approach, but the discipline of development economics shied away from modeling under Myrdal and Hirschman who choose not to formalize their ideas. A major shortcoming of verbal persuasion is that the argument is presented in a way that it sounds as if must be true, while models show what might be true. Economic theory is expressed through models. In order to be taught, and thus to endure, theory must be accompanied by a model. Why then did the developmental economists of the 1950s abandon this approach? Krugman answers, “Almost certainly for one basic reason: high development theory rested critically on assumption of economies of scale, but nobody knew how to put these scale economies into formal models.”
Rosenstein Rodan’s “Big Push” paper was published in 1943, but was not formally modeled until 1989 by Murphy, Shleifer, and Vishny. These economists demonstrated that it was indeed possible to tell high development style stories in the form of a rigorous model by not being bogged down by the inability to model imperfect markets, but rather by simply taking chances and “daring to be silly”. Sometimes the greatest insight can be gained through the simplest of models, even if it takes time for this to be realized.
In many ways, Krugman’s piece on models and their relationship to the field of economic study dovetails nicely with the article we read last week by Dani Rodriks. In his article, Rodrik does a great job critiquing economic modeling. He concedes that models are a great starting ground for understanding complex, real-world issues but suggests that the results from these models must be interpreted in context. In the case of economic modeling, or any kind of modeling for that matter, the context is inherently an oversimplified version of the reality the modeling is seeking to represent. Krugman builds off of this notion when he discusses the ways in which development economists like Albert Hirschman turned their back on “mainstream” economics for precisely the reasons Rodrik’s discusses. Development economists, like Hirschman, realized that the world they were observing simply could not accurately be described within the confines of traditional economic models. Rather than trying, and ultimately failing, to apply inadequate models to their observations, many development economists simply neglected to include models in their publications. Without the tangibility that models provide, many early theories about economic development fell on deaf ears.
ReplyDeleteKrugman’s point about the narrowness of vision that occurs whenever one simplifies the world around him in order to model it was particularly thought provoking. Krugman suggests that economists like Alfred Hirschman were in many ways far ahead of the science of economics itself. They were disinterested in the monotony of thought surrounding competitive markets and constant returns. However, because they branched out from the filled in parts of the economic map, they were largely ignored. Hirschman and his contemporaries were treading in uncharted waters and did not have the tools needed to convey their findings to community of narrow sighted economists who were accustomed to seeing economic theory in tight models. However, while the findings of these early development economists were discredited and/or ignored, the process was an important step forward in the field of development economics, for as Krugman writes, “a temporary evolution of ignorance may be the price of progress, an inevitable part of what happens when we try to make sense of the world’s complexity.”
In all of my economics classes, at one time or another my professor has always highlighted the fact that simplistic economic models fail to capture the reality of the world because they assume that a change in one variable can occur while holding all other factors ceteris paribus. This represents one of, if not, the biggest flaws of economic modeling. Due to the reliance on this ceteris paribus assumption, economic models lose their predictive capabilities when multiple exogenous shocks occur at once.
ReplyDeleteYet, despite the flaws of simplistic modeling, often times these models represent the best guess, or in other cases, they represent a general truth that cannot be explained any better. Take for instance Adam Smith's proverbial assumption that capitalism is driven by an "invisible hand". While this statement lacks any mathematical derivations, or any tangible variables to test , it precisely accounts for the equilibrium of supply and demand in an economy. Even 225+ years since his general assumption, scholars and students still understand the basics of economics from this simple model. Hypothetically, think of the ramifications if economists would have overlooked Smith's basic ideas because they lacked concrete mathematical or empirical support.
In a similar sense, it seems almost academically irresponsible that social scientists would dismiss the work of Hirschman purely for its lack of formalized rigidity. Because of this oversight, economists spent years trying to develop revolutionary new models to explain factors affecting development in low income countries. In reality they had the answer sitting right in front of them.
In essence, this paper harks back to a simple idiom "if it ain't broke, don't fix it". In this case, Hirschman's model provided a perfectly good explanation, but others attempted to revamp his ideas, and in doing so lead themselves to more questions and less of an understanding on the topic of developmental economics.
I agree with Katie that modeling of development economics was abandoned in part due to the growing complexity of models. In terms of Krugman's article, I thought it was particularly interesting that he followed up this point with the African mapping analogy. As time went on and European maps of Africa became increasingly accurate in scale and coastline shape, they became far less accurate in the heart of the continent. This was not because Europeans were learning less about mapping; it was because cartography was becoming a more accurate science. Imprecise and second-hand details were no longer good enough to include in maps. It took an extended period during which more precise and reliable data was gathered before maps reached modern accuracy.
ReplyDeleteKrugman considers the science of economic modeling to be very similar to cartography. During the middle of the 20th century, the science became more exacting and previous assumptions turned into dark areas while economists worked out acceptable ways to explain and model them. Because the field of development economics has tended to be a less mathematic economic subdivision, its modeling advancement has lagged. Development theories are based on so many assumptions, and cultural differences across developing countries make generalizations even more difficult. This makes comprehensive models extremely difficult in development economics, and requires economists to take chances ("dare to be silly").
Krugman's analysis of the notion that models are imperfect is spot on with what we have been discussing in class. Models are constrained by ability to make them and resources allotted. He points out that models will never be right as laws of physics are, but rather should be something that conveys insight. Therefore, models are necessary but often not sufficient and not applicable in all scenarios.
ReplyDeleteKrugman also talks about the idea that many ideas that are expressed through rhetoric do not become important in the grand scheme of things until models are applied to them. Often, an idea will emerge and no one will consider it until years down the road...for example the map of Africa concept. Moreover, during the modeling process, basic information can be lost and not regained until years later. He explains that "during the process of model-building, there is a narrowing of vision imposed by the limitations of one's framework and tools." "The problem is there is no alternative to models."
The form of economics of the times makes some ideas with potential seem meaningless. When people were talking about development economics in the 1950s, not only did they not have the modeling tactics needed, they also were subject to the constraints of the times-economists at the time focused on a perfectly competitive economy (no room for the idea of economies of scale).
Economic modeling is, like any other science, an imperfect system. Modeling attempts to compare different instances in the same manner, which in many ways is useful because it provides one form of a benchmark to go off of; however, it is not something that should be religiously followed, as different circumstances require different actions and no one formula should be regarded as the solution to all problems. One thing that particularly stuck out in my mind in Krugman’s paper was the analogy he presented about the maps in Africa over time—while coastlines and detail of the outskirts of Africa became more precise with the advancement of science and technology, the interior of Africa became less and less accurate, losing information about cultures and towns. This reflects how science can lure people into a false sense of knowing, when in reality what it provides is more information about one thing as opposed to general knowledge.
ReplyDeleteFurthermore, Krugman points out the simplicity of modeling when he comments that Rosenstein-Rodan’s model could “have been written as easily in 1955 as in 1989.” It seems silly to suggest that something that would have been extremely relevant in 1955 would have the same impact 35 years later, but yet modeling has a timeless factor. While in some ways being able to compare things historically is critical, it again raises the point of how different the world can become in the span of a few decades, and these should not be the only things that development economists rely on.
Krugman’s paper “The Fall and Rise of Development Economics,” when read with Rodrick’s “Growth Theory,” fully define the spectrum in which economic models should be used and applied in order to better understand the world. While Rodrik emphasizes the risks associated with hasty policy applications of economic models, Krugman emphasizes their utility in economic application, despite their inherent “falsities” of simplifying reality.
ReplyDeleteKrugman uses the example of the “high development” economists, who had a theory that economies of scale played an important role in determining whether returns to scale were increasing or decreasing. However, the prevailing attitude necessitated theory to be strictly mathematically proven. Since they lacked the tools necessary to mathematically substantiate their claim, Hirschman and company instead employed metaphors and narrative reasoning to explain this phenomenon. However, mainstream economists simply dismissed the claims made by Hirschman that economies of scale were important on the grounds that it wasn’t empirically modeled. Instead, Lewis’s two-sector model that didn’t include the hard-to-model economies of scale became the model of the times, shaping the “fall” of economics.
Krugman criticizes Hirschman on his impatience in modeling his observation, which was ultimately correct. By accepting the restrictions imposed on him by the academic community, he could have potentially formulated a model that explained the theory of increasing returns to scale, which would have subsequently shaped a different approach to development economics, which according to Krugman, would have been much more effective. More broadly, though, Krugman faults the academic community that dismissed Hirschman’s ideas due to the lack of mathematical formulation. By using the analogy of Africa’s mapping over the centuries, Krugman argues that by so vehemently restricting the type of information considered for inclusion into economic modeling, economists run the risk of leaving out potentially too much, and ending up with a model that doesn’t explain much of anything, which is inapplicable in solving the very problems it was made to deal with.
When reading this article I enjoyed the commentary about the imperfect model. In order to create a model, Krugman emphasizes that a economist or any scientist must make guesses about what is important and improve the model after the system is "learned". Since there are no alternatives to modeling we must use the models in order to improve our knowledge on the subject at hand while keeping in mind that the theories provided might not be complete or work in every situation. This reminded me of Rodrik's observation of Chinese development. The neoclassical economic growth model would not have predicted significant Chinese growth without more major changes in policy. However, China has grown rapidly from the late 1970's, demonstrating just what Krugman suggests. The ideas that we have are too far simplified in order to accurately predict what will happen in each developing country but they are a good starting point in the study of economic development as a whole.
ReplyDeleteStudying the history of economic models has benefited my overall understanding of each model brought up in class. It's been helpful to see where past economists went wrong and what they have done right. The trend in methodological trends Krugman describes may be thought of in terms of a clothing trend analogy. Clothes go in and out of style, and the ideologies of the time may have effects on that cycle. Maybe the clothing trends evoke a particular way of lifestyle that the ones who rejected it just could not understand due to historical implications, such as war. Yet the ones who revitalize the trend must associate themselves in some way with the originators. Krugman's theory is that the trend of economics' increasing mathematical style created an inability in some economists to understand the non-mathematical style of the high development theorists. Krugman himself cannot understand how such a "simple" model could have gone undiscovered for so long.
ReplyDeleteDespite the difficulties in using models to predict economic outcomes, Krugman emphasizes the "false feel of security" that people feel without a tangible model to manipulate.
Since this article is a rather short one, and very to the point, most of the subject matter has already been covered in the blog. It was refreshing to read an article written in such a relatively informal and humorous way. The author displays a sense of humor which contradicts the stereotype of dry economists.
ReplyDeleteAs to the usefulness of models, I think it also important to point out that models and econometrics can be incredibly useful tools to testing the efficacy of development policies. Models are tools, and cannot account for everything; they are not the ultimate solution, but applied properly can lead economists to practical policy recommendations. We discussed the case of conditional cash transfers in Nicaragua, where Gitter and Barham used data from control and test communities to determine the effects of the transfers on household expenditures and education for children. Their modeling and regression analysis allowed them to determine that the policy was indeed effective in the case of Nicaragua. Modeling and econometrics can be useful, therefore, both to predicting factors which may speed the general development process and evaluating policy choices within a given country or community.
In this refreshingly colloquial essay, Paul Krugman illustrates the shortfalls (and really, the lack of) of modeling in development economics. He argues that the initial problems of modeling—before the 1940s—was a lack of understanding how to model economies of scale in high development theory. But as the 1940s ushered in era of rapid advancement in economic theory, development economists lagged behind the times. Krugman stated that these breakthroughs "led to a much improved level of understanding of some things, but also led for a time to an unwillingness to confront those areas the new technical rigor could not yet reach."
ReplyDeleteAs many others in the class have noted, Krugman's paper correlates nicely with Rodrik's essay from last week and our class discussions in general. Taking a theory with simple, ceteris paribus assumptions and forcing it onto an increasingly complex and ephemeral world makes for "lousy" policy. Krugman criticizes development economist, such as Albert Hirschman, whose overly simplistic models lead to many years of an "evolution of ignorance." Krugman is quick to point out that while models help substantiate theory, they are only as good as their underlying assumptions. Models and theories need to be robust and adaptable in order to be applied uniformly.
I agree with Amy’s notion that the article provided a nice change of pace in terms of both format and writing style. Although Rodrik’s piece was mainly prose and did not make widespread use of statistics or numerical tables, it still depended heavily on tables and analysis of existing economic models. I was actually a bit surprised by Krugman’s style and prose and found his overall message relatively easy to understand.
ReplyDeleteThe takeaway message that I understood from the article—namely that one cannot discount non-modeling approaches to economics just as one cannot base a development approach purely on non-quantifiable data and information—seems a bit obvious to me. Although not an economics major, it seems to me that “suggestive metaphors, institutional realism, interdisciplinary reasoning, and a relaxed attitude towards internal consistency” would be absolutely critical to studying the people and interactions that, in aggregate, make up an economy (Krugman 10). I think it is to the credit of Washington and Lee that our academic departments recognize the necessity of interdisciplinary studies and allow Humanities majors like me to study the economic implications of major historical and political events in our Humanities classes. These lines of seemingly non-economic inquiry and knowledge are what intrigue me most about development and are what I hope to learn more about as the semester continues. I was happy to see that these kinds of “economic” research and inquiry—at least in the eyes of a Nobel Laureate—are both usable and critical to development studies. Now if only I could convince potential employers of truth of that opinion…
Paul Krugman raises many important points when it comes to the utility of these economic models. As mentioned before by others, models provide a "tangible" way to simplify the complexities seen in the real world. These basic models enable us to gain a better understanding of the world, which is not to be mistaken with a complete, correct understanding of the world.
ReplyDeleteStudying the origins of these models, such as the Harrod-Domar model from the 1950s, makes it easier to understand the motivation behind such models. As we discussed in class, it made sense that the Harrod-Domar model showed that large sums of investment will create physical capital in countries. Since most of the world, especially Eastern Europe, was in shambles after World War II, the main way to restart productivity and growth was to build infrastructure.
As right as he was, his reasoning was insufficient. Certain countries in Europe had the necessary institutions in place so the investments did help bolster the economy. In developing countries, these institutions are lacking.
If we go by Rodan's Big Push Theory, then government intervention is necessary to initiate the "push." In many cases, such as Pakistan, the government is in constant turmoil and does not have the power needed to start such an initiative. If we keep going with this idea, the next logical step is to look into foreign aid; however, corruption will impede any progress that would be possible from foreign aid. What happens then?
Krugman brings awareness to the fact that there is so much uncertainty and policies cannot stem directly from these models.
I feel like I can't elaborate much more than everybody else, but I did have an interesting conversation with my roommate. She's a science major and was also talking about models and how great they are ceteris paribus, and how tricky it is to apply them to humans. While hers are more psychologically based, a comment she had from class was that "models are not right or wrong they are simply useful or not useful." I think this pretty much defines Krugman's paper. He acknowledges the importance of models to gain broader insights and make predictions, but not to determine specific policy decisions. While these models can be helpful there is no experiment/model that can be 100% right, or every 95% right. The only thing that can be achieved, is usefulness.
ReplyDeleteIt is very interesting to see that the high development theory put forth by economists in the 40s and 50s, discredited for not being formalized with models, actually makes perfect sense and has a useful application for understanding problems of developing countries. Krugman points out that development economic theory might be in a better state had these ideas been formalized twenty to thirty years earlier. We might have had twenty or thirty more years to make important insights into what may work and what may not for developing economies. What would have been the direction of development economics if institutions had have adopted theories like the Big Push instead of constant-returns and perfect-competition models? Krugman points out that it is also important to remember that simplistic models don't often translate well to practical policy, so who knows how these models would have affected development economics had they been formed back when the insights were first put into writing. We can only assume.
ReplyDeleteI found the article both interesting and entertaining, and it provided a great complement to Rodrik's piece that we read last week. It is unfortunate that an entire field of economics fell by the wayside only because certain elements could not be encapsulated into a model. As many have already mentioned, we spend a lot of time talking about the limitations of models and how they are simplified versions of reality and often lack empirical credibility. One wonders where development economics would be today if these roadblocks had not been encountered. What we now know is that while modeling can encompass some aspects of the development story, it is not the complete picture. Each country has a different solution to the development problem that no model can predict accurately. Make no mistake, Krugman accurately points out that there is a place for models and that they can prove very useful, but as we all know, they have their limitations.
ReplyDeleteAn inviting piece - I appreciated Krugman’s meteorology metaphor in describing the limitations but necessity of models in the social sciences. Using high development theory as an example, he attributes its fall from mainstream economics and then resurgence as a result of the ability for economists to formally model the theory. He believes that modeling in Economics is essential, but that broad minded individuals such as Hirschman believed that more insight was lost than gained. Krugman points out that, while that may be true initially, taking the first steps to build a simple model eventually leads to the ability to rediscover older knowledge using the new techniques. He cites as an example Norwegian scientists’ discovery confirming folklore about cloud formations before a storm. As Development Economics recovered from its slump, Krugman said it was done by “daring to be silly: by representing the world in a dish-pan, to get at an essential point.”
ReplyDeleteLike Benjamin pointed out, this piece also links well to Dani Rodrik’s “Growth Strategies”. Krugman concludes that “a temporary evolution of ignorance may be the price of progress, an inevitable part of what happens when we try to make sense of the world’s complexity.” Rodrik takes that idea of complexity and points out that, maybe seeking one unified theory is done in vain. “They key is to realize that these principles do not translate directly into specific policy recommendations.”
This article certainly makes me think more about the use of models in economics. Having only taken econ 101 and 102 prior to this class almost everything I have learned about economics has been taught to me by various economic models. Even when we looked at real life events we always brought things back to the models. Thinking about it after reading this article I realize that the models we used are really incapable of fully predicting any outcome in the real world. The world is a complex place and there will never be a model that can consider all of the potential variables to be able to predict a definite outcome. This however doesn’t mean that models aren’t useful. For one thing they are necessary to teach economic principles to the economists of tomorrow. Additionally they can certainly be useful to economists in predicting the most likely outcome. In the end models are both our only hope and our greatest disappointment. They are one of our most powerful tools in striving for a better quality of life but at the same time will never be able to provide us with a definite answer of how to meet our end goal
ReplyDeleteIn the field of development economics models can be especially tricky. There are so many variables when looking at all of world’s economies that it is nearly impossible to choose the variables most important to include in a model to achieve satisfactory results. The truth is while there are many common reasons that a country may be underdeveloped there are so many other problems affecting individual countries that you almost need to come up with a solution for each country rather than trying to just say x amount of education + y amount of healthcare and nutrition + z amount of nutrition = economic growth and an improved standard of living. When looking at how tough it is to use models in development economics it starts to makes sense why one would just give up the models all together. The problem with throwing out the models is that the economic community and the world as a whole need the models as proof that what is being proposed will work.
There is not much more to be written about “Big Push Theory,” as we read it and discussed it in class Tuesday, but the article hit on the history of the theory. It began as a string of ideas with no model and assumptions about returns and market structure that didn’t fit with the mainstream philosophy of its time. It took 35 years to advance economists modeling capabilities in order to reexamine the idea of the Big Push. The “what-ifs” concerning development had Rosenstein-Rodan been capable of modeling his idea in the 1950s are endless, but as Krugman points out, they are still just “what-ifs.” We cannot change the past. We can only continue to work on further development and not making the same mistakes twice.
ReplyDeleteThe article we read brings up a point in the conclusion that I really admire. Both the ability to go out on a limb with one’s ideas, even if they are not completely clear, formulated, or entirely make sense yet, as well as the ability to learn from each other, regardless of the others education or experience is necessary to the development of any field of study. The local farmer may have the best solution to rural labor economics, even without any formal economics training. He certainly knows more about rural labor than the Harvard economics graduate, although sadly, we are almost always more inclined to believe the college graduate. Secondly, if we wait until we know everything about a subject, we will never be able to publish our work or succeed at anything. Krugman’s point is that we can never know everything. Our models, metaphors, and data, no matter how good, only carry us so far. If we had perfect models, there would be no need to keep working, and it would fit for everyone. We know that empirically, every situation/nation/economy is different. I think what Krugman is hoping to show is that just because one doesn’t have the ability to fully express his ideas, he should still be willing to put them on the table and collaborate and eventually they may become a fully developed theory which might be used to the betterment of the collective.
I enjoyed Krugman's analogy to the mapping of Africa as a good portrayal of model formation in development economics. He points out that while we start slow in trying to understand a certain topic or area, we become more confused with the access to new information and our ideas crowd together and we essentially obtain no more value than when we first started our approach. When applying to economics and all fields that utilize models, the same thing can be said of trying to tackle the whole idea at once. If broken down into smaller variables and attached with assumptions, we can generate a closer depiction but still not satisfy the big picture. Just like growth and development are hard to sustain, so are the ideas represented in the models of growth and development.
ReplyDeleteKrugman also explains that simplifying models does not always make them more understandable especially when their designers are highly intelligent over-thinkers. When we are forced to narrow our vision when devising the models, we have to be careful to not leave anything of substance out or the whole idea will present itself more as a loss rather than as a gain. But in reality, things are not as simple as the models make them seem and therefore we cannot obtain an accurate replication of how economies act and react.
Maybe the best representation and explanation is to see the economies in their physical form and gain insight through observation and experimentation rather than a simple graph. Complexities are going to remain complex as much as we try to simplify them and sometimes (as with the Africa map), the simplification process will cause more confusion than was already present.
The most interesting quote I take away from this essay is when Kruger writes about the Big Push model, “A model like this makes one want to go out and start measuring, to see whether it looks at all likely in practice, whereas a merely rhetorical presentation gives one a false feeling of security in one’s understanding” (Krugman 11). This quote highlights the clear differences between economists and politicians. Economists generally develop theories and collect data to support those theories, always asking the question, “Does the data support that claim?”. Politicians, as Krugman points out, base their ideas on what is in their self-interest rather than theories supported by data. While I understand that people, including politicians, are motivated by rational self-interest which is an assumption for most economic markets, it is still frustrating to hear politicians spew arguments and ideas with the conviction that they speak the undeniable truth. In reality, they are often making up stories that will get them reelected. These stories often make some logical sense, but are completely unsupported by the data.
ReplyDeleteThe quote from Krugman made me think about earlier in the semester when we discussed how politicians, especially in the U.S., often ignore research done by economists and other social scientists and create policy based solely on politics. I realize I have gone off on a tangent, but these models are useless if they are never used to implement policy to encourage development.
Given that most of the material concerning economic modeling has been covered already, I figured I will point out the interesting analogy concerning Europe's ignorance in mapping Africa.
ReplyDeleteIn that elaborate analogy, Krugman talks in relates the progress of development economics with the progress and setbacks of mapping the not-so-easily traveled terrain of Africa. In that analogy, Krugman points out that as the standards of measurement improve, the standards of observations become much more strict. Whereby, limiting the available data. Similarly, Krugman argues that as economic modeling has improved in both mainsteam and development economics, less attention/thought has been given to theories that do not specify some sort of model. Without a model, the theory does not provide a model inherently does not provide a way to test the theory, nor a method, in which, to easily teach it.
Though this argument certainly holds much weight when concerning the practicality of the situation. However, this leads me to wander that if by being so picky towards theories that provide models, we are holding back the progress of economics at least to an extent. By this, I mean that if attention is only being given to theories with models and all other theories are discounted outright or not even published, could we be missing some important discoveries that simply have not reached its final form? By not providing a format in which people struggling with completing their theories are we doing more harm than good? Assuming that there are not many outlets for a theorist without a model to discuss an incomplete idea with some of the brightest in the field, can we omitting would-be ground-breaking theories from our ongoing history of Development/mainstream economics?
As several individuals above have already expressed, I thought Krugman’s contrast of the physical science disciplines with the social sciences was especially interesting. Beyond the specific example he utilizes of Fultz’s dish-pan experiment, in more general terms, when analyzing a complex system it is almost always the case that reality must be sacrificed for specificity or vice-versa. I think the content nature of the social sciences, with its potential to more directly impact the sociological functions of daily interaction, contributes to the comparatively negative response to economic modeling. In the physical sciences, it is common practice for real-life problems to be modeled through the experimental method, where by controlling for all variables except the single aspect being studied, scientists can assess the influence a specific factor has on the larger situation. It seems like the more concrete, determined nature of the concepts and ideas in the physical sciences actually makes it more acceptable for those items to be controlled for and simplified in experimental procedures. Because economic principles are often so innately intuitive yet less tangible, we are less willing to sacrifice them for the simplification that is necessary in modeling complex systems. Ultimately, I, in agreement with many of the comments above, believe the most valuable conclusions from Krugman’s article are the needs for creativity and compromise in the methodological approaches to economic analysis. This need is reflected as Krugman writes, “economists were locked in their traditional models, non-economists were lost in the fog that results when you have no explicit models at all” (12). In other words, breaking the barriers that exist between different disciplines and their respective methods and ways of thinking is ultimately our best chance at garnering insight into the functioning of a complex system.
ReplyDeleteI like how Krugman explains the problem of high development modeling. For me, this helped me understand whey high development theory had such a lull. THese theorists' ideas made sense, but no on knew how to model the imperfect competition market structure of developing countries. Krugman's paper spends a lot of time discussing economies of scale and what they do for a developing economy and he recognizes that the Big PUsh model does address imperfect competition, but only under certain circumstances. However, this limit-pricing assumption makes any forward linkages impossible. He does criticize Hirschman for being an economist of style rather than substance in the end. I think Krugman's final message lies in the fact that it is difficult to reconcile economies of scale with a competitive market structure
ReplyDeleteIt is interesting that oversimplified models that are based on so many untrue assumptions seem to provide the greatest insight into the economic processes of a country. Even though economic models fail to illustrate reality entirely, they still provide tangible concepts in order to better understand economic observations. It is human nature to be proud and maintain one's values, so it is no wonder the presence of assumptions becomes the subject of criticism when analyses of economic models threaten one's own beliefs. When concern arises over the simplicity of models and the assumptions they make, one can argue ignorance. The map of Africa is given as an example. First, information about the coastlines was lacking in the 15th century, while details about the interior were abundant. By the 19th century, the shape of the continent was known and outlined, but the interior landscape of the continent had "emptied out" and was largely ignored because it required more accurate and precise details of, say, the topography, for example, as improvements in technology and resources to gather that information became more abundant.
ReplyDeleteAs Rodrik emphasized in the paper last week, it is still important to consider each individual country, when applying economic models, given the differences in culture and what Rodrik called "pre-existing institutional landscapes." This is the very reason that high-development economic models are hard to illustrate and apply.
I know this paper largely talks about the shortcomings and great successes of models, and that they should be analyzed with careful consideration. All of the talk about it just kinda made we want to dig as deep as possible when I finally got to the model at the end.
ReplyDeleteThe output curves created from the modern and traditional sectors were drawn in a way that the modern curve was increasing at a much higher rate than the traditional curve. The relative wage curve was drawn in between the two curves so that it would be profitable to start modern production at a high enough level of output. If the modern curve was not growing at such a high rate, or the relative wage curve did not fall below the modern curve none of these stories would be true. The stories revealed from models can provide great insights, but we have to realize we are always one component away from the model painting a completely different picture.
The discussion of the imperfect model in the “Fall and Rise of Development Economics” helps put into perspective the reasoning behind the use of economic models. I was reading through some postings and found a quote that summed it up really well. I really liked the point Michelle’s friend made about how “models are not right or wrong they are simply useful or not useful”. It is not about getting something completely right. It must be accepted that economic models cannot be perfect. Rather they can be useful tools to guide thought and action.
ReplyDeleteI also found Krugman’s comparison between early development economics and European mapping of Africa interesting. It was a really intelligent comparison because of how well it fit and how easy it was to explain. Both increased their standards of understanding to the point that both the mapping and development economics dropped off for a period of time.
I really enjoyed the way this article light-heartedly explained the processes economists have gone though over the past century to uncover a model to explain the frame work of the economy. By breaking any model down, it is difficult to decisively create a model that incorporates all the variables society faces and create a broad over arching theory. Although the assumptions we make with models may make it seem overly simplistic, such as with the only resource being labor in the big-push model, they allow us to see the economy though one of the necessary inputs. His example with the dishpan made me think of middle school science experiments: limited tools and basic knowledge are available, yet one can still covey the main concept to a large audience. In many ways, devising these overly simplistic models can be very useful to the public more so than an a more elaborate model. The more people that understand basic economic framework can lead to a more productive society since it is better educated with making practical decisions. People will be both able to use political as well as economic insight to make informed decisions if they are able to understand the essential components that lead to economic growth. Thus, even the simplistic development theory models are useful to society. As Krugman says, economists must have a "willingness to do violence to the richness and complexity of the real world in order to produce controlled, silly models that illustrate key concepts."
ReplyDeleteThis article seems to highlight the general state of economic models... they are just that. The models only go so far and often fail as a result of an outside shock. Within development economics models can only go so far to explain the state of a developing market. I liked Krugmans analysis of the disconnect between politicians and economists.
ReplyDelete