Top

W2019-1 Prospects for U.S. Labor Productivity Growth: What the Penn-Wharton Budget Model’s Microsimulation Reveals

Download PDF

Authors: Jagadeesh Gokhale

Abstract: When making projections of key macroeconomic aggregates such as total output, earnings and payroll tax bases, the Social Security Trustees assume that future labor-productivity growth will continue to remain close to its historical average. The labor productivity projection derived from this assumption is applied to the projected worker population on a per-head basis to project the aforementioned variables. However, assuming labor productivity growth near its historical average implicitly assumes that all contributing factors will also grow close to their historical rates or changes in those factors total will be mutually offsetting. However, the future composition of workers by productive abilities will differ from the past, potentially causing inconsistency between projections of key macroeconomic aggregates and the underlying characteristics of the future population.

One solution is to project as many of the productivity-contributing elements as possible using micro data information and organize them under an aggregate production function framework. This approach forces the budget analyst to define all productive inputs consistently with underlying demographic projections. Under such an approach, labor productivity growth is an auxiliary output consistent with micro-data-based projections of future worker populations, their attributes, macroeconomic aggregates, and projections of finances for programs such as Social Security.

PWBM’s microsimulation-based projection of U.S. demographic and economic features yields labor productivity growth estimates that are significantly smaller during the next few decades compared with the close-to-historical average rate of labor productivity growth assumed by the Social Security Trustees. Subtracting PWBM’s average projected labor-productivity growth over the next 75 years (2018-92) from the Social Security Trustees’ assumed value of 1.68 percent per year yields a difference of 26 basis points. If the effects of changes in the population’s demographic attributes on labor productivity growth are excluded from PWBM’s projection, the difference from the Trustees labor-productivity growth assumption equals 32 basis points.


Key Words: Productivity growth, Microsimulation, Social Securty finances, Production function, Labor force participation, Labor Efficiency