FREE ELECTRONIC LIBRARY - Dissertations, online materials

Pages:     | 1 |   ...   | 2 | 3 || 5 | 6 |   ...   | 12 |

«John R. Graham, Co-Supervisor David T. Robinson, Co-Supervisor Manuel Adelino Alon P. Brav Manju Puri Aaron K. Chatterji Dissertation submitted in ...»

-- [ Page 4 ] --

Earlier studies formalize this intuition and identify several mechanisms through which technological evolution affects firms’ ability to innovate. A negative shock to 26 the value of a firm’s accumulated knowledge space implies a longer distance to the knowledge frontier and a higher knowledge burden to identify valuable ideas and produce radical innovation (Jones, 2009). Firms working in a fading area benefit less from knowledge spillover (Bloom, Schankerman, and Van Reenen, 2013), which in turn dampens growth in innovation and productivity.2 To implement the idea and measure the influence of exogenous technological evolution on each firm’s capability to innovate, I build on the literature of bibliometrics and scientometrics, which measure the obsolescence and aging of a discipline or technology using the dynamics of citations that refer to the discipline or technology. The instrument, termed as Knowledge Obsolescence (Obsolescence hereafter), attempts to capture the τ -year (between t ´ τ and t) rate of obsolescence of the knowledge possessed by a firm. For each firm i in year t, this instrument is constructed in three steps (formally defined in formula (4.2)). First, firm i’s predetermined knowledge space in year t ´ τ is defined as all the patents cited by firm i (but not belonging to i) up to year t ´ τ. I then calculate the number of citations received by this KnowledgeSpacei,t´τ in t ´ τ and in t, respectively. Last, Obsolescenceτ is defined i,t as the change between the two, and a larger Obsolescence means a greater decline of the value and utility of a firm’s knowledge, Obsolescenceτ “ ´rlnpCitt pKnowledgeSpacei,t´τ qq´lnpCitt´τ pKnowledgeSpacei,t´τ qqs.

i,t (4.2) The validity of the exclusion restriction rests on the assumption that, controlling for industry-specific technological trends and firm-specific characteristics, the technological evolution regarding a firm’s knowledge space, which is predetermined and 2 One concern is that when a firm’s knowledge space becomes hotter, product market competition grows more severe, which in turn could disincentivize innovation and imply that emerging knowledge value could lead to lower innovation performance. This concern, however, is shown to be secondary by Bloom, Schankerman, and Van Reenen (2013) and is resolved by the first-stage regression in Table 4.3.

27 accumulated along its path, is orthogonal to its current decision on CVC other than through affecting innovation performance. One might worry that a firm’s knowledge space could be affected by the type and capability of its managers, but this concern should be minimized by using a predetermined knowledge space formed along the corporate history rather than the concurrent one. One might also worry that the firm itself could be the main driver of the technological evolution. This concern is addressed first by excluding patents owned by the firm from its own knowledge space and then by excluding all citations made by the firm itself in the variable construction.

It is mitigated further by a robustness check on a subsample of medium and small firms, which are less likely to endogenize technological evolution.

In Table 4.1, I report summary statistics for Obsolescence.

The number of citations received by a firm’s predetermined knowledge space decays by 8% in the control group, which can be interpreted as a mild three-year natural decay of knowledge. The knowledge space on average decays by 29% in the three years before a parent firm initiates its CVC arm, which demonstrate a much more severe hit by the technological evolution.

I exploit the instrument in a standard 2SLS framework. In the first stage, I

instrument the change in innovation with Obsolescenceτ using the following form:


–  –  –

4.2.2 2SLS Results Table 4.3 presents the estimation results of models (4.3) and (4.4). Column (1) reports a reduced-form regression in which Obsolescence is used to explain the decision to 28 launch a CVC program. The positive coefficient 0.001 indicates that firms experiencing larger technological decays are more likely to initiate CVC activities.

Columns (2) and (4) report first-stage regressions where ∆Innovation (Innovation measured by the quantity and quality of new patents) is predicted using Obsolescence and a larger Obsolescence (faster rate of technological decaying) is associated with poorer innovation performance. The estimate of -0.114 in column (2) translates a 10% increase in the rate of obsolescence of a firm’s knowledge space into a 1.14% decrease in its patent applications; this same change is associated with a 1.28% decrease of its patent quality as measured by lifetime citations. The F -statistics of these first-stage regressions are both well above the conventional threshold for weak instruments (Stock and Yogo, 2005).

Columns (3) and (5) show the second-stage estimation results. The key explanatory variables are now fitted innovation changes predicted from the first stage. The causal effect of innovation shocks on starting a CVC unit is both economically and statistically significant. The coefficient of -0.007 in column (2) translates a 2σ-change in ∆ lnpN ewP atentq to a 52% change in the probability of launching CVC investment.

The gaps between the OLS estimates (in Table 4.2) and the 2SLS estimates are small. This comparison suggests that endogeneity issues are not biasing the OLS estimation in any clear direction on net. This does not mean, however, that no endogeneity issues are involved—as discussed above, competing endogenous forces could drive the OLS bias in either direction, mitigating the net effect. The Appendix shows that the result is robust to several sampling criteria, such as excluding the IT and Pharmaceutical sectors, excluding California-based firms, and excluding very big or very small firms.

29 Table 4.3: This table documents the causal relationship between innovation deterioration and the initiation of Corporate Venture Capital. The analysis is performed using the following Two-Stage Least Squares (2SLS) specification: ∆Innovationi,t´1 “ { 1 1 1 π0,industryˆt ` π1 ˆ Obsolescencei,t´1 ` π2 ˆ Xi,t´1 ` ηi,t´1, IpCV Cqi,t “ αindustryˆt ` β ˆ ∆Innovationi,t´1 ` γ ˆ Xi,t´1 ` εi,t. The panel sample is described in Table 4.1.

{ Column (1) reports the reduced-form regression, which predicts the decision to initiate CVC using Obsolescence as defined in (4.2) in the paper. Columns (2) and (4) report the first-stage regression, which regress the three-year change in innovation quantity (the natural logarithm of the number of new patents in each firm-year plus one) and innovation quality (the natural logarithm of average citations per new patent in each firm-year plus one) on the three-year Obsolescence. Columns (3) and (5) report the second-stage regression, where IpCV Cqi,t is equal to one if firm i launches a Corporate Venture Capital unit in year t, and zero otherwise. ∆Innovationi,t´1 is { the fitted innovation change over the past three years (i.e., the innovation change from t ´ 4 to t ´ 1). In the 2SLS framework, firm-level controls Xi,t´1 include the ROA, size (logarithm of total assets), leverage, and R&D ratio (R&D expenditures scaled by total assets). Industry-by-year dummies are included in the model to absorb industry-specific time trends in CVC activities and innovation. T-statistics are shown in parentheses and standard errors are clustered by firm. *, **, *** denote statistical significance at the 10%, 5%, and 1% levels, respectively.

(1) (2) (3) (4) (5) Reduced Form First Stage 2SLS First Stage 2SLS

–  –  –

In previous analyses that focus on firm-level evidence, I control for industry-by-year fixed effects to absorb potential confounding factors. In this section, I look into this part of the variation that was controlled for by fixed effects in order to examine the industry-by-year pattern of CVC investment and how it speaks to the information acquisition view of CVC.

Existing CVC research documents that CVC investment clusters as waves and shows strong cyclicality (Gompers, 2002; Lerner, 2012). Figure 4.2 plots the time series of the instances of 381 CVCs studied in the sample. Both the number of launches of new CVC units and the number of deals invested are plotted. Similar to Gompers (2002) and Dushnitsky (2006), the graph highlights two waves—most CVC units were launched in either the early to mid-1980s or the later 1990s. More than 20 firms began CVC investments in each year from 1983 to 1986, and 71 firms started CVC units in 1999. CVC deals occurred in two similar waves: in the first wave, from 1983 to 1986, CVC units invested in about 60 deals each year; this number was reached again 10 years later, in 1996, at the beginning of the second CVC wave.

Existing explanations for these waves emphasize macro-level factors (tax change, market condition, etc.) that do not directly speak to one important aspect that attracts less attention—CVC waves do not happen uniformly in each industry. That is, some industries waved in only one of the two periods, with little activity in the other. In Figure 4.3, the sample is broken down to produce a by-industry CVC investment graph. Four industries are presented—machinery, printing and publishing, business services (including IT), and pharmaceuticals. Two observations can be gleaned from these figures. First, CVC investments cluster not only at the aggregate level (as in Figure 4.2) but also at the industry level, and this industry-level clustering is what can be termed an “industry CVC wave.” Second, and more important, 31 different industries waved at different times. Specifically, most CVC investments in the machinery industry were made in the 1980s, but the industry was not heavily involved in the second aggregate CVC wave in the 1990s. In contrast, printing and publishing firms were relatively silent during the 1980s CVC wave but rode the second wave in the later 1990s. Even IT firms, the overall most active group in the CVC field, were relatively uninvolved in the first aggregate wave but invested aggressively in the second wave. The pharmaceutical industry, another highly active industry in CVC investments, was almost equally active during the two aggregate waves, and this industry continued investing even outsider the waves (in contrast to most other industries).

–  –  –

Figure 4.3: This figure plots the by-industry time series (1980 to 2006) of CVC investments covered in the sample.

These are CVCs affiliated to US public non-financial firms that were started between 1980 and 2006. The CVC data are from the VentureXpert Venture Capital Firm Database, accessed through Thomson Reuters SDC Platinum. CVC investment is measured as the launch of new CVC units (left axis) and the number of deals invested in (right axis). CVC deals include only the first investment that a CVC invested in a portfolio company. Industries are classified by the Fama-French 48 Industry Classfications, based on the main SIC code of a firm reported in Compustat.

Table 4.4: This table studies the industry clustering in CVC investment on the sample of Corporate Venture Capital (CVC) which are affiliated to US public nonnancial firms between 1980 and 2006.

The CVC data are from VentureXpert Venture Capital Firm Database, accessed through Thomson Reuters SDC Platinum. Panel A compiles all of the industry CVC wave periods, jointly defined using the clustering of launches of CVC units and investment deals made by CVCs. Each wave period is limited to at most four years. Industry is defined using the Fama-French 48 Industry Classification. Panel B lists potential economic and technological changes affecting CVC investment for the important industry CVC wave periods, as discovered in Panel A. The explaining events column is partially motivated by Table 2 in Harford (2005).

–  –  –

37 fields Automobile 1984 to 1985, 2000 Technological innovation (FWD, brake system, etc.) and sharp decline of oil prices Machinery 1983 to 1985 Demand for microcomputers for numerical controls Healthcare 1983 to 1985 Industry consolidation and development of new drugs Measuring and 1998 to 2000 Strong structural changes in the semiconductor industry Control Equipment Figure 4.3 suggests that some industry-specific factor motivates firms in the industry to implement Corporate Venture Capital investment simultaneously. Table 4.4 Panel A compiles industry CVC wave periods, jointly defined using the clustering of CVC initiations and investment. I limit each wave period to at most four years. In general, most industries experience at least one wave period and more than 50% of the CVC investments were made during that short window. For example, printing and publishing firms initiated six CVC units and invested in 71 deals between 1997 and 1999; the total deals made by this industry between 1980 and 2006, however, number just 88. IT firms made most of their CVC investments during the dot-com boom. Pharmaceutical firms have a less clear wave pattern but still had two time windows when they were more active than usual.

4.4 Additional Economic Forces and Robustness

Table 4.3 controls the endogeneity problem when establishing the causality between innovation deterioration and the CVC initiation decision.

This is consistent with the information acquisition view, which predicts that firms in need of new knowledge are more active in reaching out to the innovative entrepreneurial sector. In this section I discuss additional tests that serve two main purposes— explore the informational motivation further and to study additional economic forces that could affect firms’ decision to take the CVC route.

Pages:     | 1 |   ...   | 2 | 3 || 5 | 6 |   ...   | 12 |

Similar works:

«Study of the Effect of a Flu Pandemic on Economic Values Using the Delphi Method Project Oversight Group Tom Edwalds, Chair Scott Cochran Robert Gleeson Max Rudolph Bill Sayre Jan Schuh, SOA staff Ronora Stryker, SOA staff, Principal Investigator May 2007 Please direct questions and comments to: Ronora Stryker SOA Research Actuary rstryker@soa.org ©2007 Society of Actuaries TABLE OF CONTENTS Introduction 1 Summary of Individual Responses 2 Question 1 3 Question 2 4 Question 3 6 Question 4 7...»


«Faculty of Economics Cambridge Working Paper Economics Cambridge Working Paper Economics: 1624 THE EMPIRICAL LANDSCAPE OF TRADE POLICY Chad P. Bown Meredith A. Crowley 1 April 2016 This chapter surveys empirically the broad features of trade policy in goods for 31 major economies that collectively represented 83 percent of the world's population and 91 percent of the world's GDP in 2013. We address five questions: Do some countries have more liberal trading regimes than others? Within...»

«C HA RL ES I VE S SEC ON D PI AN O S ON AT A “Co n cord, Mass., 184 0-1 86 0” EME RS ON HA WT HORNE THE AL C OTT S THORE AU GE ORG E P AP PA •ST A VR O U, pia no B ON NIE L IC HTER, flu te CHARLES IVES (1874-1954), as is well-known, made imp o r t a n t a n d l a s t i n g c o nt r i b u t i o ns t o t h e b u s i n es s of life insur ance. A long-time business partner a n d c l o s e friend has written that Ives’ “... creative mind, great breadth of culture, intensive sympathies...»

«Signposting Directory • Self-help and support groups • Education and Learning • Faith Communities • Health Care • Housing • Relationships • Family and Friends • Leisure and Recreation • Work and Money Promoting well-being through primary and community services July 2013 Contents Section 1 Signposting Directory Information on a wide range of local non-medical services that support well-being Section 2 Health Services Directory Information on local healthcare services and...»

«Curriculum Vitae CINDY VOLK Office Address Home Address University of Arizona 1271 W. Panorama Road College of Education Tucson, AZ 85704 Department of Special Education, Rehabilitation Tele: (520) 297-8873 and School Psychology P.O. Box 210069 Tucson, AZ 85721 (520) 621-5208 E-mail: cvolk@u.arizona.edu Education Ph.D. Higher Education Major: Finance Minor: Organization and Administration University of Arizona Tucson, Arizona May 1995 Dissertation: Assessing Competing Models of Resource...»

«13 MYANMAR’S COURTS AND THE SOUNDS MONEY MAKES Nick Cheesman In 2007, Deputy Township Judge U Sein Lwin lost his job and went to jail. His alleged crime was to have solicited bribes so that three women would not become co-accused in a case before him. A prosecutor charged the judge under the 1948 Prevention of Corruption Act, and a court in Taunggyi promptly sentenced him to seven years in prison. The Supreme Court dismissed him from office immediately (Supreme Court 2007a). If the facts of...»

«From Theodosius to Constans II: Church, Settlement and Economy in Late Roman and Byzantine Sicily (AD 378-668) Thesis submitted for the degree of Doctor of Philosophy at the University of Leicester by Denis Sami School of Archaeology and Ancient History University of Leicester August 2010 From Theodosius to Constans II: Church, Settlement and Economy of Late Roman and Byzantine Sicily (AD 379-668) Abstract This thesis explores the archaeology of late antique Sicily from the time of Theodosius I...»

«Land, Housing and Property Rights in zimbabwe REPORT MISSION COHRE Africa Programme LAND, HOUSING AND PROPERTY RIGHTS IN zimbabwe CO H R E A f r i ca P r o g r a m m e M i s s i o n R e p o r t, S e p t e m b e r 2 0 0 1 © Copyright 2001, Centre on Housing Rights and Evictions (COHRE) Land, Housing and Property Rights in Zimbabwe Centre on Housing Rights and Evictions, Geneva, Switzerland All rights reserved The Centre on Housing Rights and Evictions is registered in the Netherlands as a...»

«Richard Connolly Economic Structure and Social Order Type in Post-Communist Europe Diskurs 2009 – 4 Richard Connolly Economic Structure and Social Order Type in PostCommunist Europe Abstract Keywords: Richard Connolly Centre for Russian and East European Studies University of Birmingham Rxp297@bham.ac.uk This is work in progress: do not cite without the author’s permission © Richard Connolly, University of Birmingham Richard Connolly Economic Structure and Social Order Type in...»

«University of Pennsylvania ScholarlyCommons Publicly Accessible Penn Dissertations 1-1-2015 Essays in Corporate and Household Finance Christine Louise Dobridge University of Pennsylvania, christine.dobridge@gmail.com Follow this and additional works at: http://repository.upenn.edu/edissertations Part of the Finance and Financial Management Commons Recommended Citation Dobridge, Christine Louise, Essays in Corporate and Household Finance (2015). Publicly Accessible Penn Dissertations. Paper...»

«REVOLUTION IN PROGRESS: THIRD-PARTY FUNDING OF AMERICAN LITIGATION * Jason Lyon There is a growing phenomenon of for-profit investment in U.S. litigation. In a modern twist on the contingency fee, third-party lenders finance all or part of a plaintiff’s legal fees in exchange for a share of any judgment or settlement in the plaintiff’s favor. There are a number of international corporations, both public and private, that invest exclusively in this new “market.” Critics contend that...»

<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.