Assessing the severity of economic problems often requires choosing between different sets of analyses that reach disparate conclusions. While much lip service is paid to “evidence-based policymaking,” all too often it works the other way in Washington, with problematic “facts” serving as the basis for “policy-based evidence-making,” in the phrase of my colleague Richard Reeves. Instead of trying to discern why different analyses yield different conclusions, and whether one is better supported than another, partisans simply pick the results that support their beliefs. Sometimes the evidence really is unclear, and resolving the question requires further research. But other times, it is clear enough, and those committed to evidence-based policymaking should favor some conclusions over others.
Consider the state of the middle class. According to published Census Bureau figures, median household income (adjusted for inflation) was just 5 percent higher in 2011 than in 1979. But thanks to economists such as Cornell University’s Richard Burkhauser, the University of Chicago’s Bruce Meyer, and Notre Dame’s James Sullivan — all of whose work has recently been bolstered by Congressional Budget Office estimates — it has become irrefutably clear that, when properly measured, middle-class incomes actually rose by at least 30 percent between 1979 and 2007 (both business-cycle peaks), and possibly by 40 percent or more. The Census Bureau figures indicate only a 15 percent rise between these years.