Noah Smith writes a great piece on “what economics gets right” over at Bloomberg View today. In it, he criticizes the view of Alex Rosenberg and Tyler Curtain, who claim in a recent New York Times piece that economics “lacks the most important of science’s characteristics — a record of improvement in predictive range and accuracy.”
To counter their claim, Smith cites micro-level examples of economists making successful predictions: rates of use for public transit in San Francisco, Google’s use of game theory to maximize profits from sale of ad space (specifically, a model refined by microeconomist Hal Varian), and so-called “gravity models” of international trade that Smith says “do a very good job of predicting how much trade will occur between any two countries.”
But Smith also blames economists for Rosenberg’s and Curtain’s faulty perspective. He writes:
Why do Rosenberg and Curtain get it wrong? Part of the fault is their own — they just didn’t bother to do their homework. But part is the fault of economists, who don’t do a very good job of trumpeting their predictive successes to the world. Some economists still defend the idea that economic theory doesn’t need to make predictions in order to be useful, and instead merely has to give people a framework for thinking about the world.
That argument hasn’t carried much water with the general public, and rightly so. Without empirical support, you can’t really be confident that a theory is a good framework for thinking about the world. It’s likely that economists using bad theories will make bad policy recommendations, no matter how organized or internally consistent those economists’ mental frameworks.
The link behind “defend the idea” in the passage above is to a paper by James E. Anderson called The Gravity Model. I’m not about to read this entire paper, so I’m not sure why he linked to it. I actually think it might be a mistake, as it’s the same link as the previous link in the article.
But whether the link is correct or not, I propose that he might as well have linked to Hans-Hermann Hoppe or some other proponent of the praxeological method, which denies, on philosophical grounds, the legitimacy of a criterion for the usefulness of economic science that regards “predictive success” as the ultimate goal. Hoppe writes:
Just recall that according to its very own doctrine, neither a predictive confirmation nor a predictive falsification would help us either in deciding whether a causal relationship between a pair of variables did or did not exist. This should make it appear rather doubtful that anything is gained by making prediction the cornerstone of one’s philosophy.
Instead of predictive success, then, “praxeologists” like Hoppe emphasize a correct application of a priori knowledge toward expanding the analytical framework with which we can think correctly about economic issues or hypothesize about future economic outcomes. Theorists can make predictions on the grounds of that a priori framework, but their predictions’ accuracy—good or bad—doesn’t necessarily inform us of the legitimacy of the framework itself. A bad prediction simply means the principles of the framework have been misapplied.
I’m guessing it’s this type of thinking that Smith is referring to when he criticizes economists who “defend the idea that economic theory doesn’t need to make predictions in order to be useful, and instead merely has to give people a framework for thinking about the world.”
That said, don’t think Smith’s allegation that empirical support is necessary to prove that a framework is correct. I’m not the first one to day this—it’s a well-known criticism of empiricism from practitioners of the alternative praxeological method. While the need for empirical support might be true for the realm of the natural sciences, it’s not true for social sciences like economics.
But I’m not going to rehash that entire argument here. Instead, I simply want to defend the usefulness of refined mental frameworks—ones that don’t have as their object the making of predictions (that is, the type of frameworks Smith refers to)—toward helping people understand economics in the real-world.
To put it simply, even strictly-empirical economists rely heavily on sound mental frameworks that aren’t necessarily empirically provable. They rely on a priori truths from the very beginning of their analysis, when they posit that nothing in reality can be known to be one way or another prior to future experiences which may confirm or disconfirm a hypothesis (in other words, the foundational assumption of empiricism). And as Hoppe points out, economists hypothesizing about possible causes of certain phenomena is almost always colored by some a priori, “framework” knowledge. This is because according to strict empiricism, we cannot know with certainty whether something is a possible cause of something else. So if we want to explain some phenomenon, our theorizing would be unconstrained by a priori considerations of any kind, which leaves us with no starting point from which to launch an investigation. Everything can have some influence on anything. Most economists, then, do their analyses with certain principles or assumptions in mind—rules that have no empirical support, but are undeniable by their very composition (i.e. humans act to satisfy their ends).
Likewise, a proper framework disqualifies a range of propositions as influential. Hoppe writes:
[Praxeological propositions] restrict the range of possibly correct predictions. And it would do this not as an empirical theory, but rather as a praxeological theory, acting as a logical constraint on our prediction-making.19 Predictions that are not in line with such knowledge (in our case: the quantity theory) are systematically flawed and making them leads to systematically increasing numbers of forecasting errors. This does not mean that someone who based his predictions on correct praxeological reasoning would necessarily have to be a better predictor of future economic events than someone who arrived at his predictions through logically flawed deliberations and chains of reasoning. It means that in the long run the praxeologically enlightened forecaster would average better than the unenlightened ones
To say like Smith, then, that economists focused on developing a sound analytical framework through which people can view economic phenomena are misguided is blatantly unhelpful. The framework is a prerequisite for good empirical research. It’s the most “useful” thing about economics, and it’s postulates aren’t necessarily empirically verifiable. It’s the foundation upon which economic reasoning rests. If the mental framework is bad, then even a theory with a thus-far stellar track record of predictive success could lead us wildly astray.
To close, here’s Hoppe again:
To understand the logic of economic forecasting and the practical function of praxeological reasoning, then, is to view the a priori theorems of economics as acting as logical constraints on empirical predictions and as imposing logical limits on what can or cannot happen in the future.