Do Computers Make Output Harder to Measure?

In recent years, U.S. productivity growth accelerated sharply in manufacturing, but has remained sluggish in the most computer-intensive service industries. This paper explores the possibility that information technology is generating output that is increasingly hard to measure in non-manufacturing...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of technology transfer 2001-10, Vol.26 (4), p.295-295
Main Authors: McGuckin, Robert H, Stiroh, Kevin J
Format: Article
Language:English
Subjects:
Business services
Computers
Decomposition
Economic growth
Economic theory
Growth rate
Information technology
Manufacturing
Output
Productivity
Service industries
Studies
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In recent years, U.S. productivity growth accelerated sharply in manufacturing, but has remained sluggish in the most computer-intensive service industries. This paper explores the possibility that information technology is generating output that is increasingly hard to measure in non-manufacturing industries, which contributes to the divergence in industry productivity growth rates. Our results suggest that measurement error in 13 computer-intensive, non-manufacturing industries increased between 0.74 and 1.57 percentage points per year in the 1990s, which understates annual aggregate productivity growth by 0.10 to 0.20 percentage points in the 1990s. This adds to an estimated 0.22 to 0.30 percentage point error from the increasing share of aggregate output in these hard-to-measure industries. Thus, increasing measurement problems may understate aggregate productivity growth by an additional 0.32 to 0.50 percentage points per year in the 1990s and play an important role in understanding recent productivity trends at the industry level. [PUBLICATION ABSTRACT]
ISSN:0892-9912
1573-7047
DOI:10.1023/A:1011170416813