Evidence and Data and Economics

Mises believed that theory is useful for interpreting data. The only rejection is that of using the data to derive theory. But, if theory is used to understand real world events, then analyzing these real world events are obviously a fundamental part of economic science.

The above passage is from Jon Catalan: found here. In this post he defends the Austrian School, and specifically Mises, on the straw man that Austrians hate empirical analysis and thus find it irrelevant to economic science. Though, I can understand, for example, the critics questioning minimum wage conclusions when Austrians make claims that “it is just common sense that if you raise minimum wage, you increase unemployment,” thus they may be selective on what they choose to look at as empirical analysis, but to generally say that empirical analysis is useless in Austrian economics is a straw man. As Catalan pointed out, the simple fact that humans act is an empirical claim, or better yet, it can be supported by empirical analysis.

I also have distaste of this horrible attack on Austrian economics, there are elements which to need to be improved in this school but we sure wouldn’t know if this was how critics respond. I have distaste for it on two main reasons 1) it is somehow claimed that republicans are now Austrians and 2) it just so fixed into thinking of Austrian economics as a philosophy than as an actual economics school that can actually teach the mainstream something.

1) I do not get where people get this information that republicans are now influenced by Austrian economics (maybe it is this author that lacks viewing empirical evidence as important). Most republicans, and especially those who are politicians or have a big name in the right wing, are influenced by mainstream economics, namely by conservative New Keynesians and New Classicals. Even if we assume Austrian economics as a school that has similar views of that of Ron Paul, a good majority of republicans think of him as a crazy man.
And somewhat relevant, I also doubt that Bachmann reads Mises, even more so, on the beach.

2) More important than the first point, Austrian economics does indeed offer many things to the economics profession. I’ll present one thing, briefly but important, that the mainstream needs to learn (using economic reasoning and not a philosophical reasoning or ideological one etc). Data analysis itself is not sufficient to say anything about the economy. People from the Austrian school to the Shacklian Keynesians know that due to a transmutable reality, we must be skeptical of data to say anything about future states. The future is determined only by action done in the present, and in turn leaves us with the conclusion that the future always remains to be written. A common question may arise from this: if we are to be skeptical of data then how are we to say anything about macro concerns?

Maybe to an Austrian, this may sound ‘weird’. It is widely believed in the Austrian school that a divide of macro and micro is unnecessary. “Economics is just economics, what is this macro or micro economics that you speak of,” I heard said by a peer while I was at the Mises Institute last year. This is to mean that Austrians generally like to use similar, if not the same, tools to analyze ‘macro’ effects and ‘micro’, and thus when explaining macro, we try to incorporate as much of micro foundation as we can. In Ludwig Lachmann’s essay Macro Economic Thinking and The Market Economy, he gives a justification for using micro foundations to macro:

The more firmly a macro-economic argument is linked to its micro-foundation in choice and decision, the less it lends itself to statistical verification. Since the range of choice present to the minds of decision-makers defies statistical measurement, no theory linking observable events, like output quantities or prices, to choice and decision is, in this sense, ‘testable’. The circumstances influencing decisions find their mental reflection in plans. All economic action is, in the first place, the making and carrying out of economic plans. So long as there are no statistics of plans there is nothing to which the econometricians can correlate their measurements

Keep in mind, as also mentioned by Lachmann in this essay, that Joan Robinson said the same thing. The difference in her statement is that Robinson only reserves what she says to the neoclassical production function, while Lachmann extends this to explain economics as a whole. Here is Robinson:

Statisticians can find out in a rough general way, for a particular situation, the capital-output ratio in dollar values and the share of profit in the dollar value of net output, so that they can estimate the overall ex post rate of profit on capital. They cannot describe what was in the minds of directors of firms or on the drawing boards of engineers when the choices were made which led to the creation of the existing stock of capital equipment. Still less can they say what choices would have been made if the rate of profit had been different from what it is.

In macro, there needs to be a stronger basis for this kind of analysis but, at best, this will hardly be achieved any time soon, for the economics profession (macro and micro levels) is still lacking micro foundations based on subjectivism (based on decision and choice).

Jack Wiseman asked a very haunting question, given micro foundation considerations, to the mainstream:

… what becomes of positive economics once it is recognized that the basic econometric data (recorded prices etc) is, and always going to be, the outcome of mistaken predictions?

At best (I think), we can use Lachmann’s advice:

[it is]… impossible to discuss meaningfully policy measures designed to affect the magnitude of such aggregates without also discussing those changes in their composition which must accompany them.

And why is this?

Since these micro-economic actions are not necessarily repeated from day to day, even less from year to year, we have no reason at all to believe in the aggregative constancy of the macro-variables over time.


3 responses to “Evidence and Data and Economics

  1. Within the context of empiricism, if you were to ask where Austrians might improve it would be in the area of historical research. Robert Higgs is the only really good econometrician that comes to mind, but he was trained in this area. Free bankers have also been good at looking at the data. We can debate on the quality of his data, but Rothbard (America’s Great Depression) showed effort and ability. But, in general, most empirical studies have been pretty bad — just see any paper applying ABCT to the recent recession.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s