This note conducts a simple test for myopia and liquidity constraints in aggregate US consumption. The test exploits the implications of myopia and liquidity constraints for asymmetry in consumption. Under myopia, the relation between consumption and predictable income does not depend on the sign or magnitude of expected income change. Under liquidity constraints, however, consumption should be correlated with predictable income only when income is expected to rise. Using quarterly postwar data, I show that consumption is far more sensitive to predictable income when expected income growth is low than when expected income growth is high. This "perverse asymmetry" is inconsistent with both myopia and liquidity constraints.