In this paper we investigate the relationship between leverage and the level of economic activity in the United States, using quarterly data over the period 1951–2012. We address the question for five different measures of leverage—household leverage, nonfinancial firm leverage, commercial bank leverage, broker–dealer leverage, and shadow bank leverage—making a distinction between traditional banks and shadow banks, the latter being a consequence of financial innovation and deregulation in the financial services industry over the past 30 years. We investigate whether the relationship between leverage and the level of economic activity is nonlinear and asymmetric using slope-based tests as well as tests of the null hypothesis of symmetric impulse responses. Our results inform policymakers about the important distinction between traditional banks and the market-based financial intermediaries that have been at the center of the global financial crisis of 2007–2009. They also inform about the macroeconomic effects of the deleveraging process that began in 2008, as well as about the need for countercyclical macroprudential policies to reduce the procyclicality of the financial system.