How Do Economists Use Data

Many people believe at that place has been no progress in economic science, but that isn't true. For example, 1 of the most important questions in the 1970s and 1980s was whether monetary policy could be used to stabilize the macroeconomy. One popular theoretical model, known equally the New Classical model, unsaid that monetary policy could not touch on output and employment, and hence was of no utilise in trying to first cyclical fluctuations in these variables.

Economists call this "money neutrality." A competing theoretical model, the Keynesian model, asserts that coin is non-neutral. In these models, budgetary policy is a useful tool to stabilize fluctuations in output and employment.

The economics profession was split between these 2 theoretical camps, and there was bully passion on both sides. This left monetary policymakers in a quandary. If money was neutral, the best policy was to simply stabilize interest rates and the money supply to whatever extent possible. Only if money was non-neutral, then interest rates and the money supply should exist changed in response to macroeconomic weather condition.

How tin this question be settled? How can we determine which model is right? By testing their implications against real world data using econometric techniques. If, in real world data, changes in monetary policy touch on output and employment, then money is not-neutral, and if it doesn't, information technology's neutral.

When these tests were performed, the bear witness pointed to not-neutrality, and today the passionate debate over this issue is all just over. There are still a few economists who, despite the overwhelming bear witness to the contrary, believe in neutrality, but they are a small minority.

Thus, progress in economic science depends upon the use of econometric techniques to exam hypotheses and distinguish betwixt competing theoretical structures

What is econometrics?

Econometrics is a set of statistical tools that allow researchers to test economic theory against existent world data, and to forecast the future of the economy. For example, a item theory might hypothesize that when the government spends more than through deficit spending, it drives involvement rates college. This hypothesis could then be taken to the information, and tested using econometric techniques.

If the information are inconsistent with the hypothesis, so that is evidence that the theory is wrong. If it passes the test, we cannot say for sure that the theory is right, information technology may fail along other lines, but it does provide support for the particular theory under consideration.

But why practise nosotros have a divide discipline called econometrics? Isn't it just statistics? Can't economists just adopt the statistical tools and techniques used in other disciplines?

The essential departure betwixt statistics and econometrics is the disability of economists to perform laboratory experiments where the effect of one variable on another can be examined while holding "all else equal."

Economists must rely on historical information as it comes to them -- they cannot, for example, re-run the macroeconomy again and again and examine how well various policy interventions might work. Ideally, for example, for research purposes we would run an experiment in which the Great Recession happens again and again. The we could meet how well diverse policy interventions perform while holding all influences except for that policy constant (or do the same intervention again and once more to shine out whatsoever randomness in the upshot that might misconstrue the findings).

Thus, while data from laboratory experiments is commonly confined to two variables and the experiments can be performed repeatedly to ensure a unmarried observation is not a statistical fluke, economists must apply historical data -- a unmarried ascertainment -- where all else is definitely not equal.

Backslide this

For this reason, economists use a technique called multiple regression analysis. Essentially, what this means is that the issue of the treatment on the outcome is examined by including a (sometimes large) set of controls to account for all of the variables that cannot exist held constant.

For example, in the example above, it would be important to include all of the other variables too authorities spending that might influence interest rates. Information technology is not sufficient to just examine the correlation between those two variables. Failure to include these controls (which are generally not needed when laboratory experiments are performed since they are held abiding) can crusade all sorts of problems, from bias in the outcome of the tests to a failure to isolate the particular relationship the investigator is interested in.

If the researcher includes the correct set of controls and accounts for other statistical properties of the empirical model, this technique works fairly well. But there is one important caveat, something that is particularly problematic for testing macroeconomic theories. Nearly macroeconomists know the information on GDP, employment prices, interest rates, productivity and so on fairly well. So it is non very useful to build a theoretical model to explicate these information, and and then test to come across how well it fits. Of class the model would fit. Afterward all, why build a model that is inconsistent with the data you already know almost?

And that is the fundamental -- to employ data researchers did not know nearly when the model was built. Testing models against data that is revealed only after the model is built is the best fashion to practise this. That is particularly useful when, as with the Great Recession, the economy departs from historical norms (Notably, most macroeconomic models failed during this menses, and the race is on to build new models that can account for such macroeconomic outcomes.)

Considering economists take no pick merely to use non-experimental data when testing theories against the actual outcomes of the economy, the techniques can get very complicated. In improver to the inclusion of a big set up of controls, in that location are also issues involving the statistical properties of the model.

Simply in the cease, the goal is the same: to find a way to distinguish good models from bad ones, and move the profession forward. These complications brand progress slower so it would be if economists could exercise their analysis in a lab. Only although the progress is slow, and sometimes hard to come across, there is progress nonetheless.

CBS MoneyWatch contributor Mark Thoma is an economist at the University of Oregon. He also writes the blog Economist's View.

Marker Thoma

View all articles past Mark Thoma on CBS MoneyWatch»
Mark Thoma is a macroeconomist and time-series econometrician at the University of Oregon. His research focuses on how monetary policy affects the economy, and he has worked on political business bike models. Marker is currently a fellow at The Century Foundation, and he blogs daily at Economist's View.

Twitter