Final sample consists of 6398 patents from 472 firms granted from 1984 through may 2007.
Buyouts of corporate divisions are most common, followed by private-to-private deals (investments in independent unquoted entities), secondary deals (firms that were already owned by another private equity investor), then public-to-private deals
Robustness Checks:
*Concern one: Private equity investments for which there was already an existing investor, patents may be double-counted. Employs these patents only the first time they appear then drops them. Results are little changed
*Concern two: Only measures citation count during the 3 years after the award. Using a longer window increases accuracy but decreases sample size. Repeats the analysis through the end of the second calendar year after the patent grant and after the fourth year and finds that results are quantitatively similar.
*Concert three: In divisional buyouts corporate parents may retain best patents and only give low quality patents to the PE backed division. This may lead to an apparent increase in quality in the patents applied for after the award. Addresses this issue by using the longer window for patents above and by rerunning cross tabulations and regressions with divisional buyouts excluded from the sample. Key results are little changed by this shift.
Variables:
*FA/TA Financial assets (net)/Total Assets (net)
*TA/TAg Total Assets (net)/Total assets (gross)
===Kaplan 1991===
@article{kaplan1991staying,
title={The staying power of leveraged buyouts},
author={Kaplan, Steven N},
journal={Journal of Financial Economics},
volume={29},
number={2},
pages={287--313},
year={1991},
publisher={Elsevier}
abstract={This paper documents the organizational status over time of 183 large leveraged buyouts completed between 1979 and 1986. By August 1990, 62% of the LBOs are privately owned, 14% are independent public companies, and 24% are owned by other public companies. The percentage of LBOs returning to public ownership increases over time, with LBOs remaining private for a median time of 6.82 years. The majority of LBOs, therefore, are neither short-lived nor permanent. The moderate fraction of LBO assets owned by other companies implies that asset sales play a role, but are not the primary motivating force in LBO transactions.}
filename={Kaplan (1991) - The staying power of leveraged buyouts}
}
Implications for the reasons LBOs occur and sources of value in LBO transactions. Describes characteristics, timelines, and stats of LBOs that return to public ownership.
Data:
183 large leveraged buyouts between 1979 and 1986 collected from Securities Data corporation or Morgan Stanley and Company. Post-buyout info obtained from Lotus' Datext databases, Nexis database, Wall street journal articles the year the LBO was completed, and financial reports filed with the SEC
Variables:
*number of LBO's
*total debt to total capital (book value)
*total debt to initial deal value
*interest expense to operating income
*inside equity ownership fraction
===Roden & Lewellen 1995===
*Warning signs: large cash flows, acquisition activity, low growth prospects
abstract={This paper documents the organizational status over time of 183 large leveraged buyouts completed between 1979 and 1986. By August 1990, 62% of the LBOs are privately owned, 14% are independent public companies, and 24% are owned by other public companies. The percentage of LBOs returning to public ownership increases over time, with LBOs remaining private for a median time of 6.82 years. The majority of LBOs, therefore, are neither short-lived nor permanent. The moderate fraction of LBO assets owned by other companies implies that asset sales play a role, but are not the primary motivating force in LBO transactions.}
filename={Kaplan (1991) - The staying power of leveraged buyouts}
}
Implications for the reasons LBOs occur and sources of value in LBO transactions. Describes characteristics, timelines, and stats of LBOs that return to public ownership.
Data:
183 large leveraged buyouts between 1979 and 1986 collected from Securities Data corporation or Morgan Stanley and Company. Post-buyout info obtained from Lotus' Datext databases, Nexis database, Wall street journal articles the year the LBO was completed, and financial reports filed with the SEC
Variables:
*number of LBO's
*total debt to total capital (book value)
*total debt to initial deal value
*interest expense to operating income
*inside equity ownership fraction
===Van de Gucht 1998===
@article{van_de_gucht_predicting_1998,
title = {Predicting the duration and reversal probability of leveraged buyouts},
abstract = {We examine the probability that a firm will return to public status following a leveraged buyout (LBO) transaction and for those LBOs that will eventually reverse, we examine the factors that impact the timing of the reversal. These two dimensions of the reversal decision are studied by estimating standard and split population hazard models for a sample of 343 LBO transactions. Our results indicate that not all LBO firms eventually will reverse, i.e. the net benefits of private status for some firms appear to be permanent. For those LBOs that will reverse, reversal probabilities are found to increase over the first seven or eight years following a typical LBO, then to decline thereafter.},
number = {4},
urldate = {2016-06-17},
journal = {Journal of Empirical Finance},
author = {Van de Gucht, Linda M. and Moore, William T.},
month = oct,
year = {1998},
pages = {299--315},
file = {Van de Gucht and Moore (1998) - Predicting duration and reversal prob of LBOs.pdf}
}
Van de Gucht and Moore find that hazard rate of LBO reversals peaks at about 7 to 8 years after the buyout.
Data:
*343 LBO transactions $100 million+ (1980-1992) Securities Data Corporation
*Restricted to those with stock returns in CRSP in year before LBO, post-LBO status identifiable in Newspaper Abstracts or Ward's Business Directory (1994)
Variables:
*Full LBO
*Size
*Industry Q
*R&D/Sales
=Innovation Factors/Phenomena Papers=
===Artz et al 2010===
@article{artz_longitudinal_2010,
title = {A {Longitudinal} {Study} of the {Impact} of {R}\&{D}, {Patents}, and {Product} {Innovation} on {Firm} {Performance}},
abstract = {Because of increasing levels of competition and decreasing product life cycles, a firm's ability to generate a continuous stream of innovations may be more important than ever in allowing a firm to improve profitability and maintain competitive advantage This paper investigates several issues that are central to an examination of the innovation productivity in a firm. First, the relationship between a firm's commitment to research and development and its innovative outcomes is examined. Two innovative outcomes are analyzed: (1) invention, which focuses on the development of new ideas; and (2) innovation, the development of commercially viable products or services from creative ideas. Invention is measured by the number of patents granted, and innovation is assessed by the number of new product announcements. Second, because many inventions ultimately result in marketable innovations and because patents may provide protection for new products, the relationship between patents and product announcements is also investigated. Finally, the ability of a firm to benefit from its inventions and innovations is studied by examining their separate effects on firm performance, measured as return on assets (ROA) and sales growth. Drawing from a sample of 272 firms in 35 industries over 19 years, the results from a model of simultaneous equations provided support for some of the hypotheses, but several other surprising findings were found. As expected, R\&D spending was positively related to patents. This finding is consistent with others who argue that internal research capabilities, particularly those with a strong basic research component, is key to enabling a firm to generate creative outputs. More surprising was the finding of increasing returns to scale to R\&D spending. While this contradicts much of the existing research, it is consistent with economic arguments for the advantages of scale in innovation. Also interesting is the finding that, while a significant curvilinear relationship exists between R\&D spending and product announcements, it is not the predicted inverse-U but instead a U-shaped relationship. Consistent with previous work, product announcements were found to be positively related to both performance measures. A negative relationship was found between patents and both ROA and sales growth. While these findings were unexpected, they are intriguing and call into question the value of patents as protection mechanisms. In addition, these results may be resulting from the rise of strategic patenting, where an increasing number of firms are using patents as strategic weapons. As expected, a positive relationship was found between patents and new product announcements.},
language = {en},
number = {5},
urldate = {2016-06-17},
journal = {Journal of Product Innovation Management},
author = {Artz, Kendall W. and Norman, Patricia M. and Hatfield, Donald E. and Cardinal, Laura B.},
month = sep,
year = {2010},
pages = {725--740},
file = {Artz et al (2010) - Impact of RD Patents and Innovation on Firm Performances.pdf}
}
Analyzes relationship between firm performance and R&D/innovative activity.
*Argues for increasing returns to scale in R&D spending.
*Also claims patents are negatively related to growth in this sample.
Data:
*272 firms for 19 one-year periods
*COMPUSTAT 1984 to 2004, NBER Patent Citation Data Files (2001), Patents BIB, new product announcements from Factivia (cross-checked with F&S Predicast)
Variables:
*Measures innovative activity with patents and new product announcements.
*Measures firm performance with after tax ROA, sales growth 3-year moving average.
*Controls for year and industry (two digit SIC code)
*Does not control for firm size, since it is correlated with R&D spending.
===Hitt 1991===
@article{hitt_effects_1991,
title = {Effects of {Acquisitions} on {R}\&{D} {Inputs} and {Outputs}},
volume = {34},
issn = {0001-4273},
url = {http://www.jstor.org/stable/256412},
doi = {10.2307/256412},
abstract = {Making acquisitions, although a popular strategy, may not always lead to positive firm performance. Researchers have offered several explanations for this relationship. One is that acquisitions lead to lower investments in R\&D and curtail the championing process whereby organization members internally promote new products and processes in firms. The current research found that acquisitions had negative effects on "R\&D intensity" and "patent intensity."},
number = {3},
urldate = {2016-06-17},
journal = {The Academy of Management Journal},
author = {Hitt, Michael A. and Hoskisson, Robert E. and Ireland, R. Duane and Harrison, Jeffrey S.},
year = {1991},
pages = {693--706},
file = {Hitt et al (1991) - Effects of Acquisitions on RD.pdf}
}
Hitt et al argue that acquisitions are substitutes for R&D/patents and other investments. We should keep this in mind when using these as regressors in the same equation.
===Hagedoorn 2003===
@article{hagedoorn_measuring_2003,
title = {Measuring innovative performance: is there an advantage in using multiple indicators?},
volume = {32},
issn = {0048-7333},
shorttitle = {Measuring innovative performance},
doi = {10.1016/S0048-7333(02)00137-3},
abstract = {The innovative performance of companies has been studied quite extensively and for a long period of time. However, the results of many studies have not yet led to a generally accepted indicator of innovative performance or a common set of indicators. So far the variety in terms of constructs, measurements, samples, industries and countries has been substantial. This paper studies the innovative performance of a large international sample of nearly 1200 companies in four high-tech industries, using a variety of indicators. These indicators range from R\&D inputs, patent counts and patent citations to new product announcements. The study establishes that a composite construct based on these four indicators clearly catches a latent variable 'innovative performance'. However, our findings also suggest that the statistical overlap between these indicators is that strong that future research might also consider using any of these indicators to measure the innovative performance of companies in high-tech industries. (C) 2002 Elsevier Science B.V. All rights reserved.},
Seeks to establish commonality in the measurement of innovative performance. Its indicators include R&D inputs(expenditures), patent counts, patent citations, and new product announcements. Results of study are that any of these four indicators could be taken as a measure of innovative performance in the broad sense.
=Unsorted=
====Hagedoorn 2003====
@article{hagedoorn_measuring_2003,
title = {Measuring innovative performance: is there an advantage in using multiple indicators?},
volume = {32},
issn = {0048-7333},
shorttitle = {Measuring innovative performance},
doi = {10.1016/S0048-7333(02)00137-3},
abstract = {The innovative performance of companies has been studied quite extensively and for a long period of time. However, the results of many studies have not yet led to a generally accepted indicator of innovative performance or a common set of indicators. So far the variety in terms of constructs, measurements, samples, industries and countries has been substantial. This paper studies the innovative performance of a large international sample of nearly 1200 companies in four high-tech industries, using a variety of indicators. These indicators range from R\&D inputs, patent counts and patent citations to new product announcements. The study establishes that a composite construct based on these four indicators clearly catches a latent variable 'innovative performance'. However, our findings also suggest that the statistical overlap between these indicators is that strong that future research might also consider using any of these indicators to measure the innovative performance of companies in high-tech industries. (C) 2002 Elsevier Science B.V. All rights reserved.},
file = {ScienceDirect Full Text PDF:C\:\\Users\\James Chen\\AppData\\Roaming\\Zotero\\Zotero\\Profiles\\g2eepc1b.default\\zotero\\storage\\SEU8CIRN\\Cloodt et al. - 2006 - Mergers and acquisitions Their effect on the inno.pdf:application/pdf;ScienceDirect Snapshot:C\:\\Users\\James Chen\\AppData\\Roaming\\Zotero\\Zotero\\Profiles\\g2eepc1b.default\\zotero\\storage\\RDJIG292\\S004873330600045X.html:text/html}
}
====Van de Gucht 1998====
@article{van_de_gucht_predicting_1998,
title = {Predicting the duration and reversal probability of leveraged buyouts},
abstract = {We examine the probability that a firm will return to public status following a leveraged buyout (LBO) transaction and for those LBOs that will eventually reverse, we examine the factors that impact the timing of the reversal. These two dimensions of the reversal decision are studied by estimating standard and split population hazard models for a sample of 343 LBO transactions. Our results indicate that not all LBO firms eventually will reverse, i.e. the net benefits of private status for some firms appear to be permanent. For those LBOs that will reverse, reversal probabilities are found to increase over the first seven or eight years following a typical LBO, then to decline thereafter.},
number = {4},
urldate = {2016-06-17},
journal = {Journal of Empirical Finance},
author = {Van de Gucht, Linda M. and Moore, William T.},
month = oct,
year = {1998},
pages = {299--315},
file = {ScienceDirect Full Text PDF:C\:\\Users\\James Chen\\AppData\\Roaming\\Zotero\\Zotero\\Profiles\\g2eepc1b.default\\zotero\\storage\\CCNSKDXX\\Van de Gucht and Moore - 1998 - Predicting the duration and reversal probability o.pdf:application/pdf;ScienceDirect Snapshot:C\:\\Users\\James Chen\\AppData\\Roaming\\Zotero\\Zotero\\Profiles\\g2eepc1b.default\\zotero\\storage\\2IA32N4Q\\S0927539897000236.html:text/html}