Good piece at Marginal Revolutions on the new Nobel Prize winners in economics, David Card, Joshua Angrist and Guido Imben. They won the prize for developing techniques to do one of the hardest things in social science, isolating the effect of a single change on a system with many, many variables. The trick is that rather than trying to list all the possible confounders and "correct" for them, you search for some sort of control group and set up, after the fact, a "natural experiment." Like this:
The obvious way to estimate the effect of the minimum wage is to look at the difference in employment in fast food restaurants before and after the law went into effect. But other things are changing through time so circa 1992 the standard approach was to “control for” other variables by also including in the statistical analysis factors such as the state of the economy. Include enough control variables, so the reasoning went, and you would uncover the true effect of the minimum wage. Card and Krueger did something different, they turned to a control group.
Pennsylvania didn’t pass a minimum wage law in 1992 but it’s close to New Jersey so Card and Kruger reasoned that whatever other factors were affecting New Jersey fast food restaurants would very likely also influence Pennsylvania fast food restaurants. The state of the economy, for example, would likely have a similar effect on demand for fast food in NJ as in PA as would say the weather. In fact, the argument extends to just about any other factor that one might imagine including demographics, changes in tastes, changes in supply costs. The standard approach circa 1992 of “controlling for” other variables requires, at the very least, that we know what variables are important. But by using a control group, we don’t need to know what the other variables are only that whatever they are they are likely to influence NJ and PA fast food restaurants similarly. Put differently NJ and PA are similar so what happened in PA is a good estimate of what would have happened in NJ had NJ not passed the minimum wage.
As MR points out, this wasn't new in 1992, but has roots going back to the 1840s. But this approach has become much more sophisticated lately, and people are using it in many clever ways, thanks in part to these folks.
Here is Paul Krugman at the NY Times explaining that these techniques have in general made economics more liberal:
Overall, then, modern data-driven economics tends to support more activist economic policies: Raising wages, helping children and aiding the unemployed are all better ideas than many politicians seem to believe. But why do the facts seem to support a progressive agenda?
The main answer, I’d argue, is that in the past many influential people seized on economic arguments that could be used to justify high inequality. We can’t raise the minimum wage, because that would kill jobs; we can’t help the unemployed, because that would hurt their incentives to work; and so on. In other words, the political use of economic theory has tended to have a right-wing bias.
But now we have evidence that can be used to check these arguments, and some don’t hold up. So the empirical revolution in economics undermines the right-leaning conventional wisdom that had dominated discourse. In that sense, evidence turns out to have a liberal bias.