"Mergers and R&R Revisited"

Bronwyn H. Hall

ABSTRACT

Using a newly constructed dataset containing approximately 6,000
United States publicly traded manufacturing firms that existed at some
time between 1976 and 1995, I reexamine a series of findings about the
relationship between restructuring activity and R&D investment during
the 1980s, extending them to 1995. First, hazard rate models are used
to explore the determinants of exit via acquisition, going private
transactions, and bankruptcy. Then I use of the Rosenbaum-Rubin
propensity score methods to analyze the outcomes of mergers between
two publicly traded firms controlling for the probability of that they
will merge. The focus of interest is on changes in R&D intensity and
in productivity post-merger.

The results of the analysis are the following: 1) between 1980 and
1995, employment in these firms fell from approximately 18 million
employees to 13 million employees, with most of the decline taking
place by 1990. 2) This shrinkage was achieved by internal shrinkage in
the sectors associated with the rust belt (primary metals,
automobiles, etc.) and by exit (often via leveraged restructuring or
going private) in the low to medium technology sectors. The high
technology sectors (pharmaceuticals, computing equipment, etc.) did
not experience employment reductions. 3) Prior to 1987, firms that
exited by going private had a substantially lower R&D-to-sales ratio
than other firms, but after that date, there is only a small
difference in R&D intensities between those who go private and the
others. More preliminary results are that: 4) Contrary to earlier
findings that R&D was unchanged by mergers between two publicly traded
firms during the 1980s, there is a slight hint that R&D may have
declined after merger in the 1990s, although the result is not
significant. 5) Mergers have no measurable effect on the productivity
of the resulting firm at this level of analysis.