Deciphering the fall and rise in the net capital share

Matthew Rognlie

March 2015

Abstract

In the postwar era, developed economies have experienced two substantial trends in the net capital share of aggregate income: a rise during the last several decades, which is well-known, and a fall of comparable magnitude that continued until the 1970s, which is less well-known. Overall, the net capital share has increased since 1948, but when disaggregated this increase comes entirely from the housing sector: the contribution to net capital income from all other sectors has been zero or slightly negative, as the fall and rise have offset each other. When decomposed into a return on fixed assets and a residual share of pure profits, the fall and rise of capital income outside the housing sector in the US owes mostly to the residual: it is not paralleled by fluctuations in the measured value of non-housing capital. This observation— combined with the theory of factor substitution, and simulation results from a multisector model—casts doubt on explanations of changes in the net capital share that rely on changes in the value of capital. There is greater support in the data for narratives that emphasize cyclical and trend variation in market power