The political economy of imprisonment in affluent western democracies

Over the last three decades, the busiest line|of empirical research on imprisonment has treated incarceration rates as a function of macroeconomic trends. This research builds on a Marxian theoretical foundation laid in the 1930s by Rusche and Kirchheimer (Rusche 1978; Rusche and Kirchheimer 1968), who argued that the function of prisons in capitalist societies is to manage surplus labor. Since the 1970s, dozens of studies have tested the deriv-ative hypothesis that the imprisonment rate moves inversely with the business cycle, rising during recession and declining during economic expansion. Most results confirm the expected association (Chiricos and DeLone 1992), and many scholars believe the “Rusche– Kirchheimer hypothesis ” (henceforward RK) is amply supported. Others are skeptical, argu-ing that tests of RK are based on narrow sam-ples or misspecified statistical models (e.g., D’Alessio and Stolzenberg 1995; Parker and Horowitz 1986; Sutton 2000). If an empirical association exists between recession and impris-onment rates, it is by no means clear how it should be interpreted. Scholars in the RK tra-dition have suggested that recession may be linked to imprisonment rates as a result of increased crime (unemployed people steal), by the legitimating function of the state (unem-ployed people make revolutions), or by the agency of official decision makers (judges give harsher sentences to unemployed people). So far we have no way to sort out these alternative interpretations, since “the research has left many if not most of the key theoretical issues unex-amined ” (Chiricos and DeLone 1992: 432). I find three chronic weaknesses in the RK lit-erature that call not just for skepticism, but fo