Economics and Empiricism

I know it has been a while since I've posted anything here. That's because I had been busy until the end of last month with studying for the Physics GRE. Since then, my life has been dominated by a combination of classes, my UROP, and graduate school applications. That also means that for the next several weeks, posts here will once again be fairly infrequent. That said, I read an article today in the New York Times by economist Raj Chetty, who would like to convince readers that economics is a science like any of the natural sciences. Having seen such arguments poorly presented in my economics textbooks, I figured this would make for a nice quick rant. But then as I thought about it more, I realized that this issue is actually a lot more subtle than I originally imagined.

As a physics student, I don't know if I can ever shake the gut feeling that economics somehow will always be "mushier" in some way compared to the natural sciences. But I need to make sure that there is solid grounding for that feeling, or else I should be casting away that feeling as well. So why do I feel that economics is "mushier"?

Let's take the following sentence: "I’m troubled by the sense among skeptics that disagreements about the answers to certain questions suggest that economics is a confused discipline, a fake science whose findings cannot be a useful basis for making policy decisions." I'm sure there are people who believe that having multiple opinions and interpretations of even raw empirical data alone makes economics untrustworthy. I'm sure this thought may have crossed my mind at various points as well, but I realize this isn't why I feel economics is mushy. As I think about, I realize for example that after electroweak symmetry breaking (i.e. the reason why electromagnetic and weak interactions become identical at certain energy scales, but why at the scales we usually observe the photon is massless while the W and Z bosons are massive) was discovered, a few different explanations were posited for the underlying cause. One was the Higgs boson, and the other was technicolor. After the discoveries of the past two years, of course, it is safe to say the Higgs boson is a much more likely candidate than technicolor, but there was a decades-long gap between the theory and experiment. Clearly this is an instance in physics where before further experiments could be performed, two competing explanations were jockeying for attention in the physics community. So I can't say that economics is mushy because there are disagreements among economists in their interpretations and explanations of the same sets of data. So why do I feel that it is mushy?

It's likely because of what happened with Reinhart/Rogoff. From what I understand, Reinhart and Rogoff published an influential paper which posited that a country whose debt-to-GDP ratio pushes above 90% will quickly be plunged from economic growth into a recession (or worse if the starting conditions are worse). As it turns out, a graduate student found an error in the calculations that led to this conclusion and showed that in fact no such debt-to-GDP threshold ratio causing a sharp reversal of macroeconomic fortunes exists. As far as I can tell, rather than backing down, Reinhart and Rogoff have been pushing their findings harder than ever. So I may think economics is mushy because there seems to be little drive at the upper echelons to maintain scientific skepticism and admit errors as they happen, especially when it comes to macroeconomics. Granted, it took a long time after the first cosmic microwave background measurements came in for the steady-state theory to be cast out of cosmology, but it seems like especially in macroeconomics, this is a more prevalent problem than in the natural sciences.

In fact, the author of the article briefly touches on this, but proceeds to pretend that this is not the reason why people don't believe that economics is a science. Yes, the article does go through in pretty good detail what advances have been made in empirical microeconomics and econometrics that should earn economics the respect given to other sciences. And after having taken the class 14.03 — Microeconomic Theory and Public Policy and having learned there a little about empirical microeconomics and econometrics, I'm actually rather inclined to agree (as I actually learned some of the details of some of the examples in this article).

The issue is that when most news articles quote economists, it's usually in the context of macroeconomics, as macroeconomic activity will by definition affect the largest segment of the readership, as opposed to microeconomic moves taken by certain firms, labor groups, or municipalities. Macroeconomics is a very young field with not a whole lot of data as yet. Additionally, there is a lot of disagreement among macroeconomic theorists about how to interpret macroeconomic events/trends, and too many of them are acting more for political gain than for the sake of advancing the field (see: Reinhart/Rogoff). It's really cool when the LHC performs experiments to further probe the Higgs boson (i.e. science), but honestly, people will care a lot more if their federal income taxes change (i.e. macroeconomics), and such a change cannot be effected in a controlled environment (in relative contrast to the Higgs boson experiments). Regarding that last point, macroeconomic models are too unreliable compared to scientific models of macroscopic systems that economists would love to emulate, because they very frequently fail to predict the outcomes of certain economic events or enacted policies. So it's the combination of the fact that macroeconomics is unreliable and that its scholars have the very real power to personally gain from their models being correct potentially at the expense of other people's livelihoods (where I've emphasized that last point in contrast to natural sciences) that makes myself want to push economics away from the natural sciences.

But this isn't fair to microeconomics. So what's the solution? I would personally split microeconomics and macroeconomics much more. In any case, the author says that more and more research that even affects macroeconomic decisions is coming from empirical microeconomics. In that case, I would say that news outlets should devote more space to highlighting empirical research done in microeconomics that clearly affects people's lives, such as those highlighted in the article. And maybe farther in the future, all economics will basically be microeconomics, and macroeconomics will become what classical thermodynamics has become in relation to statistical mechanics: a bulk limit that is only obtainable under very specific conditions.

Book Review: "Fermat's Last Theorem" by Simon Singh

(Sorry, this time I don't have a picture of the book, though the owner is the same relative who lent me the books The Undercover Economist and Mokshagundam Visvesvaraya.)
And with this, I have read all of Simon Singh's books. Yay! (I think.)
I really enjoyed this book as much as his other two books. The passages talking about the cracking of the Enigma encryption system were like déjà vu after reading The Code Book. I also enjoyed reading about all of the tension and drama surrounding the various developments and failures leading up to the proof of Fermat's last theorem (though there seem to have been an awful lot of people who took their own lives in the process, unfortunately). It reminds me very much of Big Bang, as the material is presented in a very accessible format, and the focus is more on the developments and characters involved (as the end result is already known from the start).
I would recommend this to anyone even remotely interested in math, history, and puzzles. For readers in the US, however, note that while the UK edition is called Fermat's Last Theorem, the US edition is called Fermat's Enigma.

Book Review: "Big Bang" by Simon Singh

(CC-BY-NC-SA Das U-Blog by Prashanth)
This was actually given to me as a gift by the same relative who has The Undercover Economist. I will also say that I have read The Code Book, one of Simon Singh's other books, but it has been a few years since I read it, so I will not be reviewing it any time soon.
Singh starts with an overview of creation myths and the use of reason and science to determine whether the sun goes around the Earth or vice versa. He continues with the heightening and resolution of the debate between the geocentric universe and heliocentric universe camps, and introduces the debate between those who believed in an eternal universe and those who believed the universe had a moment of creation. He brings in the developments of relativity and the atomic model, the measurement of the speed of light, and the rise and fall of the ether hypothesis to lay the foundations for the theory of the Big Bang. He also discusses the debate between whether the Milky Way is our only galaxy or whether there are many galaxies separate from the Milky Way and its subsequent resolution, and connects this to the use of spectroscopy to show redshifting of galaxies to bolster the case for the Big Bang over the eternal universe theory. He continues with further developments in atomic theory and the postulation of CMB radiation (radiation coming from the farthest reaches of the universe dating back to 300000 years after the Big Bang) to frame the debate now between supporters of the Big Bang theory and supporters of the eternal universe theory, which had by then morphed into the Steady State theory (which held that the universe was expanding, but galaxies were forming in the voids as expansion happened, so the overall composition of the universe remained constant). He concludes with the solution of the nucleosynthesis problem (how elements heavier than helium could be formed in the early stages of the universe), the detection of CMB radiation, and the detection of tiny density variations in the early particle soup as the final pieces of evidence necessary to secure the Big Bang theory as the correct one. In the epilogue, he discusses some missing parts of the Big Bang theory, as well as its religious/philosophical implications.
In a sentence, I really like this book. (That was the sentence. The rest is extra.) It's a great overview of all of the science that led to the acceptance of the Big Bang as the dominant theory regarding the origin and state of the universe, and on top of that, it talks about all of the characters involved in its development. It is a fairly large book, but any formulae used are clearly explained in layperson's terms, and there are a lot of helpful illustrations tied with such formulae. More interesting than that, though, is the presence of a brief, illustrated summary section at the end of each chapter (because each chapter is approximately 80 pages long). This really helps to remind one of what one has read at the beginning of the chapter and how it ties in to later parts of the chapter. I also like how it talks about the arguments between people without just focusing on the science, as this adds a more human element to the history of it all. I would recommend this to anyone even remotely interested in the subject.
Follow the jump to read the rest.