3 ms·

It is a "testable hypothesis" which has been verified every single time a recession has happened, that the direct cause of a recession is a fall in aggreg

by temphn 14y ago

It is a "testable hypothesis" which has been verified every single time a recession has happened, that the direct cause of a recession is a fall in aggregate demand and all economists worth their salt ranging from Keynes to Friedman, from Marshall and Samuelson to Karl Marx would agree with that statement. Appreciate the polite discussion. In that vein, there are several issues with your basic premise. 1. Hayek won a Nobel, so by many definitions is an economist "worth his salt". As you may know (because he is absent from your list), Hayek does not agree with that statement. 2. The definition of "aggregate demand" is voodoo: http://en.wikipedia.org/wiki/Aggregate_demand#Components http://en.wikipedia.org/wiki/Aggregate_demand#Components What measurements could Keynes base his theory on in a time before computers? Did he really have access to Visa's databases of transactions to compute a measure of "consumer spending"? Did he have access to every VC and PE fund balance sheets to compute the amount of investment? Did he have a command line interface to JP Morgan's chart of accounts? Only today, when we can reliably measure the activities of millions of people, can we even begin to quantify these kinds of things, let alone reliably attribute direct causes. Cause must precede effect, effects must not occur without causes, cause/effect pairs must occur together many times, and (most importantly) effects must be predictably produced by causes. With real economies, we can't blow them up just to test hypotheses. With virtual economies, we have a chance of testing some of these things. More generally, the macroeconomy is not some fantastic thing invented out of whole cloth, it can and must be based on direct measurement of individual actors in the economy. For example: bpp.mit.edu is real economics. It computes macroeconomic quantities (like the rate of inflation) from direct microeconomic measurements on historical prices from the web. If you disagree with their analysis, you can replicate it from their dataset. I think this is our fundamental disagreement. I do not believe macroeconomics has a firm empirical foundation in the same way that physics does. Economics is best conceptualized as a many-body problem in biology, with highly non-fungible humans (100 Steve Jobs will greatly affect any reasonable simulation). We're just not there or even close in macro, and so the confidence of some of these people (especially Krugman) is just wholly unwarranted relative to the claims of a solid-state physicist. Indeed, many of them will deny that macro parameters can or should be derived from direct measurement of individual actors in the economy. By contrast, in physics, thermodynamics is firmly based on the microscale discipline of statistical mechanics.

photon137 14y ago
I agree with you that economics has nowhere near as much predictive power as physics (trained in physics, stuck in the financial-sector as a boring quant - I can testify to both). Nor it has infallible foundations (rational behaviour? game-theory based "utility" maximization? VNM utilities?) I also agree with you that aggregate demand as a macroeconomic quantity is hard to measure as are many macroeconomic variables. But nowhere does Hayek disprove the existence of aggregate demand (whether business cycles are caused by monetary policy or fiscal policy or some other factor is debatable - their end effect is always on aggregate demand). But statistical observations of the aggregate often lead to great discoveries and the foundation of a subject. Max Planck's enormous set of highly aggregate experimental data (without him knowing the mechanism by which light operates) led him to formulate the "quantum" hypothesis (ie light is quantized as photons) - which was the founding stone for quantum mechanics and all the great stuff that followed. We have such numbers in engineering - the Reynold's number, the Nusselt number etc. We don't know how they arise from chaotic microscopic behaviour but we do know these quantities exist. They have a profound effect on the onset of turbulence in fluid flow, onset of convection in heat-transfer and a whole lot more. The effects themselves can be measured and be tallied with a "Reynold's number" of say 3000 (which is again based on conduit dimensions, pressure gradient, viscosity - all measurable quantities but we don't how they tie up together to give rise to turbulence). Never stopped us from making pumps and engines and life better for the world. As an aside, Hayek's theory and the whole Austrian school of thought is the least empirical of all such branches. So you contradict yourself. Even Milton Friedman declared himself "an enormous admirer of Hayek, but not for his economics. I think Prices and Production is a very flawed book. I think his [Pure Theory of Capital] is unreadable. On the other hand, The Road to Serfdom is one of the great books of our time." (copied verbatim from Wikipedia). There is a difference between normative economics (which the political leaders and Paul Krugman propound - ie the "That's how it should be" class) and positive economics (which, as a subject, deals with discovering mechanisms of the economy and searching for empirically verifiable variables that can validate those mechanisms). I think you're confusing the two.
- temphn 14y ago
  There is a difference between normative economics (which the political leaders and Paul Krugman propound - ie the "That's how it should be" class) and positive economics Well, sure, but there can be no obligation to do that which cannot be done. Let's say you believe that the Federal Reserve's QE and money printing activities are like monkeying around with the number of shares in the cap table, that this nothing to do with the process of wealth creation, and that they downplay the effects of ruinous inflation on economies (viz. Weimar 1924, Argentina 2001). Then the positive and normative are tightly linked. You should not do X (normative) because X will lead to disaster (positive).

by temphn 14y ago

photon137 14y ago

I agree with you that economics has nowhere near as much predictive power as physics (trained in physics, stuck in the financial-sector as a boring quant - I can testify to both). Nor it has infallible foundations (rational behaviour? game-theory based "utility" maximization? VNM utilities?) I also agree with you that aggregate demand as a macroeconomic quantity is hard to measure as are many macroeconomic variables. But nowhere does Hayek disprove the existence of aggregate demand (whether business cycles are caused by monetary policy or fiscal policy or some other factor is debatable - their end effect is always on aggregate demand). But statistical observations of the aggregate often lead to great discoveries and the foundation of a subject. Max Planck's enormous set of highly aggregate experimental data (without him knowing the mechanism by which light operates) led him to formulate the "quantum" hypothesis (ie light is quantized as photons) - which was the founding stone for quantum mechanics and all the great stuff that followed. We have such numbers in engineering - the Reynold's number, the Nusselt number etc. We don't know how they arise from chaotic microscopic behaviour but we do know these quantities exist. They have a profound effect on the onset of turbulence in fluid flow, onset of convection in heat-transfer and a whole lot more. The effects themselves can be measured and be tallied with a "Reynold's number" of say 3000 (which is again based on conduit dimensions, pressure gradient, viscosity - all measurable quantities but we don't how they tie up together to give rise to turbulence). Never stopped us from making pumps and engines and life better for the world. As an aside, Hayek's theory and the whole Austrian school of thought is the least empirical of all such branches. So you contradict yourself. Even Milton Friedman declared himself "an enormous admirer of Hayek, but not for his economics. I think Prices and Production is a very flawed book. I think his [Pure Theory of Capital] is unreadable. On the other hand, The Road to Serfdom is one of the great books of our time." (copied verbatim from Wikipedia). There is a difference between normative economics (which the political leaders and Paul Krugman propound - ie the "That's how it should be" class) and positive economics (which, as a subject, deals with discovering mechanisms of the economy and searching for empirically verifiable variables that can validate those mechanisms). I think you're confusing the two.

temphn 14y ago
There is a difference between normative economics (which the political leaders and Paul Krugman propound - ie the "That's how it should be" class) and positive economics Well, sure, but there can be no obligation to do that which cannot be done. Let's say you believe that the Federal Reserve's QE and money printing activities are like monkeying around with the number of shares in the cap table, that this nothing to do with the process of wealth creation, and that they downplay the effects of ruinous inflation on economies (viz. Weimar 1924, Argentina 2001). Then the positive and normative are tightly linked. You should not do X (normative) because X will lead to disaster (positive).