CEO Characteristics, Firm Performance, and Corporate Political Contributions: A Firm Level Analysis Manohar Singh, The Pennsylvania State University-Abington Email: [email protected]Vijaya Subrahmanyam, Mercer University-Atlanta Email: [email protected]Anita Pennathur, Florida Atlantic University Email: [email protected]
46
Embed
CEO Characteristics, Firm Performance, and Corporate ...€¦ · CEO Characteristics, Firm Performance, and Corporate Political Contributions: A Firm Level Analysis Abstract: We investigate
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CEO Characteristics, Firm Performance, and Corporate Political Contributions:
where, contemporaneous stock return, ROA, and ROE, are the firm performance measures. The
predicted contributions are obtained from the first step of the estimations, and the control variables
are firm size, two-year sales growth rate, debt ratio, and industry effects.
Participating Firms Sample:
While one of the motivations of the paper is to analyze differences between
participating firms and non-participating firms, a more discerning aim is to examine whether firms
are able to benefit from such participation. We therefore repeat the estimations above for our
sample of participating firms alone. We develop proxies which focus on the identity of politicians
receiving political contributions from firms. We define a ‘value-relevant’ (good) contribution as
one where a firm donates to a politician who is deemed to be in a position to positively impact the
firm.
In order to develop variables that represent proxies for ‘value-relevant’ (good)
contributions by firms’ PACs, we hand-collected data from the FEC database using the following
steps. We first search in the full sample database for a PAC associated with each firm name in the
FEC database. If we find a PAC associated with the firm, we include the firm in a new
‘participating firm’ subsample to gather detailed data for each of these firms. For each of these
participating firms, we create three separate subsample datasets for each year to gather information
from the FEC database. Next, under the ‘PAC’ information from the FEC Database, we use the
‘Summary’ tab to find the ‘location’ and ‘Industry’ for each year for the participating firm in
question and add these fields to the participating firms’ subsample database. Following this, from
the ‘Recipients’ tab within the PAC Data (within the FEC Database), we collect the entire list of
names of House members and Senate recipients for each year along with the amount that each
politician received in ‘contributions’ for the year from the firm’s PAC for each participating firm.
We sort, by year, each list of Senators and House members, by the ‘amount of contribution’. We
then highlight the top five contributions for each firm for each year4 separately for the House and
Senate. Next, we select the tab ‘Politicians & Elections’ within the FEC database and do a search
within this section for each politician’s location and committee assignments for our list of
recipients for each firm for every year. Using the data from this section, we create corresponding
fields in the participating firms’ subsample database on each politician’s committee assignments
and location for each year.
In summary, we first examine if a firm is a participating firm and include it in a
‘participating’ firm database and collect information on its location and industry for the years 2002,
2004 and 2006. Following that, we gather information on a firm’s PAC contribution to the House
members and Senators (using the FEC database) and rank these contributions in descending order
of the dollar value of the contributions and also gather information on the location of the
politicians’ in question for each year for each firm. We then focus on the top 5 contributions of the
PAC to the House (and the Senate) members and define value-relevant contributions in two ways:
1. A contribution is considered value-relevant (good) if the recipient’s committee
assignment in the year in question (2002, 2004 or 2006) matches the industry in which
the contributing firm operates. If not we consider it a ‘non-value relevant’ (bad)
contribution.5 For each year, we define the value relevance ratio based on ‘industry
match’ as follows:
Industry Match = $ of value relevant contributions to the House (Senate)
members/Total $ value of all contributions to House (Senate) members.
4 The top five contributions may include several or all the contributions made. For instance if 30 house members received $5000 each, all 30 would be included. 5 The Congressional committee assignment was matched to the industry of the firm. The industry and location for each firm was listed under the FEC database under ‘Summary Information for each firm’ and was collected in creating the database. Besides matching the industry, in all cases, if the Congressional committee assignment for the House member or Senator was in ‘Appropriations’, ‘Budget’ or ‘Ways and Means’, we included it as value-relevant as well.
2. A contribution is considered value-relevant (good) if the Location of the House member
(or Senator) in the year in question (2002, 2004 or 2006) matches the contributing firm’s
headquarter location. If not, we consider it a ‘non-value relevant (or bad) contribution.
For each year, we define the value relevance ratio based on ‘location match’ as follows:
Location Match= $ value relevant contributions to the House (Senate) (based on
location)/Total $ value all contributions to the House (Senate).
4. EMPIRICAL RESULTS
Our first set of estimations test the CEO dominance hypothesis. We model the propensity
to make PAC contributions as a function of CEO dominance, as proxied by the ratio of inside
directors, CEO duality, CEO age, rCEOage (CEO age relative to the board) and CEO tenure. We
check for the VIF statistics and our results do not find any strong indication of multicollinearity
among our explanatory variables. Table 2 provides the results of the probit estimations for the
CEO dominance hypothesis.
<insert Table 2 about here>
Contrary to our null hypothesis that firms with a higher ratio of inside directors would be
more likely to make a PAC contribution; we find that such firms are significantly less likely to be
politically engaged via a PAC involvement. This result is both statistically and economically
significant. A 1% increase in the ratio of inside directors leads to a 2.25% decrease in the
probability of a PAC contribution. While the above results do not support the agency view of
political engagements, our results on CEO duality (Model 2 of Table 2) strongly support the
hypothesis that dominant CEOs who hold the dual title of CEO/Chair would be more likely to be
politically engaged.
We hypothesized that firms with older, and therefore more entrenched CEOs, would be
more likely to engage in a PAC contribution; however our results do not support such a link
between CEO age and PAC contributions. However, it is possible that the dynamics of the board
decision-making are influenced by the relative age of CEO with respect to that of the rest of the
board. For example, if there are several older board members, the CEO may not be as dominant as
in a firm where the board, on average, is younger. To measure the CEO age relative to the board
we construct a new variable, rCEOAGE. We define this variable as CEO age divided by the
number of directors on the board who are more than 70 years of age. This ratio captures relative
experience of the CEO compared to the board. The expectation is that for firms where the board
is composed of relatively younger directors, the CEO is more influential or dominant. This
variable combines the plurality of older directors as well as the relative age of the CEO. Indeed,
we find that firms with relatively older CEOs have a stronger positive propensity to make PAC
contributions (Model 4 of Table 2). Therefore, while CEO age in absolute terms may not influence
corporate political decisions, CEO age relative to the board impacts such decisions. Using tenure
as a proxy for CEO dominance, we had predicted that firms with longer serving CEOs would be
more likely to contribute to a PAC. Our results show the opposite; CEOs with longer tenures are
significantly less likely to be involved in PAC contributions.6
Overall, our results show mixed evidence that firms with dominant CEOs are more likely
to make firm level political contributions via PACs. However, for firms with entrenched CEOs,
6 The impact of CEO age and tenure may vary over firm performance; older CEOs or those with longer tenures in poorly performing firms may
be more likely to engage in PAC contributions. In order to test the impact of performance and CEO age (tenure), we interact return with CEO age (tenure). We find no significant relation between the interaction variables and the likelihood of making a PAC contribution.
the likelihood of PAC contributions decreases, except in instances where the CEO holds a dual
title, and when the CEO is older relative to the board of directors.
Examining the control variables, we find that firms that are larger, older, and those with a
greater number of directors are significantly more likely to be involved in a PAC contribution,
while firms with lower levels of leverage (total debt ratios) are significantly less likely to make
PAC contributions. We do not find any link between political contributions and lagged return or
market to book ratios. As we had anticipated that the likelihood of PAC contributions could
potentially vary by industry, our estimations include dummies for one-digit industry SIC codes.
We find that firms in SIC code 4, which encompasses transportation, communications, electric,
gas, and sanitary services are significantly more likely to be politically involved, perhaps due to
the regulatory nature of these industries. Thus, it appears that some industries see a significant
advantage to being politically active.
<insert Table 3 about here>
While the results above pertain to the complete sample of participating and non-
participating firms, we next turn to the sample of participating firms alone. As mentioned earlier,
this smaller set comprises of a hand-collected dataset where we match the politicians to firm-level
PAC donations by House (Senate) Committee assignments and by geographical location of the
firm headquarters and the home-state of the recipient politician. Table 3 provides the results for
CEO dominance estimations for value-relevant contributions made by committee assignment and
by location.7 Unlike the results for the full sample, we document a positive relation between the
ratio of inside directors and the percentage of value-relevant contributions made by the firm,
7 We report the results for contributions made to the U.S. House of Representatives members. Results for Senate contributions are available upon request.
measured by the Location variable. While the results for the full sample show that firms with more
inside directors are less likely to participate in a PAC, our results here show that in instances when
such firms do make PAC donations, inside directors appear to see the advantages of making
donations to ‘local’ politicians who may be in the position to impact the firm. However, when
value-relevance is measured by the committee assignment of the recipient politician, we are unable
to find a significant link between CEO dominance, as measured by the ratio of inside directors and
PAC contributions.
We find mixed results for our second model; CEO duality. While we find that for the full
sample, dual CEOs are more likely to contribute, our analysis of the value-relevance of
contributions shows that CEOs holding a dual title of CEO/Chair are significantly less likely to
make contributions to politicians in based on the location proxy. On the other hand, dual CEOs are
marginally more likely to make contributions to politicians holding committee assignments in the
industries in which their firms operate. We do not find any significant impact of CEO age, relative
CEO age (rCEOAGE), or CEO tenure on the value-relevant PAC contribution made by the firms,
measured by either geographical location or by committee assignment. So, in summary, while
overall the relative age of the CEO to the board (rCEOAGE) and CEO tenure does impact firm
decisions regarding whether to contribute to a PAC or not, it does not seem to stem from additional
value perceived to arise from the politician’s geographical location or committee assignments in
the given year. Examining the control variables, it appears that larger firms are significantly less
likely to make value-relevant contributions, regardless of the dominance proxy used in the models.
We also find some evidence that firms with higher debt ratios are less likely to participate in value-
relevant PAC donations. Overall, our results for CEO dominance follow our results for the full
sample, while there are some linkages between value-relevant contributions and our proxies for
CEO dominance, there is no strong evidence pointing to entrenched or dominant CEOs making
higher contributions to politicians who may be in positions to help their firm.
Our next set of estimations presents the impact of interest alignment characteristics on the
likelihood of PAC contributions by the firm. Based on the literature (Farrell, Hersch, and Netter,
2001; Ozer, 2010; Aslan and Grinstein, 2012), we hypothesized that firms having CEOs who are
more aligned with their shareholders are more active through PACs. Table 4 provides the results
of the interest alignment probit estimations, where we model the likelihood of setting up a PAC as
determined by the ratio of CEO holdings, institutional ownership, CEO total compensation, and
CEO variable compensation.
<insert Table 4 here>
Our results suggest that interest alignment, as proxied by CEO holdings, is negatively
related to the likelihood of making political contributions. Thus, our first proxy for interest
alignment refutes our hypothesis that CEOs who are more aligned with shareholders are more
likely to be politically active. Our compensation-based measures of interest alignment, total
compensation and incentive-based variable compensation, are positively related to the probability
of participating in firm’s PAC. Our results are consistent with past studies (Farrell, Hersch, and
Netter, 2001; Ozer, 2010; and Aslan and Grinstein, 2012); we also find that CEOs with higher
amounts of total compensation and variable compensation are significantly more likely to make
political contributions (at the 1% and the 5% levels of significance, respectively). Our findings of
positive link between total pay and the likelihood of having a PAC may reflect both, CEO
dominance and/or better interest alignment. In either case, it suggests that CEOs view political
activities as positive value propositions. We interpret the variable compensation results as
indicating that political engagements are viewed by CEOs as value-adding.
Given the argument that institutional ownership serves as an effective governance
mechanism, we expect that firms with greater institutional ownership are more likely to be
politically active via PACs. However, our results show that institutional ownership is not a
significant determinant of PAC contributions. This may appear to be in contrast with Ozer,
Oneonta, and Ahsan (2010) who note that institutional investors favor firms’ investment decisions
in corporate political strategies. However, these authors also state that the roles institutional owners
play in corporate political strategies differ based on the horizon of these institutional firms making
these decisions.
Our analysis of the control variables shows that larger firms and firms in the SIC code 4
(transportation, communications, electric, gas, and sanitary services) are significantly more likely
to be politically active. We also find that older firms and firms with larger boards are more likely
to make PAC contributions. These results are robust to choice of interest alignment proxy. There
is also evidence that larger firms are more likely to contribute to a PAC.
<insert Table 5 here>
Again, we examine the impact of value-relevant contributions for participating firms and
we next provide the results for the interest alignment variables. We had predicted a positive
relation between the interest alignment proxies and the percentage of value-relevant contributions
made by the firm. A perusal of Table 5 shows that all our CEO interest alignment variables are
highly significant (at the 1% level), as measured by the Location match variable. While our full
sample results show that CEOS with larger holdings are less likely to participate in PACs.
However, for those firms that do participate, we find that the CEO holdings is positively and
significantly related to the percent of value-relevant donations made by the firm. CEOs with larger
equity stakes donate more to local politicians whom they perceive to be in positions to help the
firm. On the other hand, we find the percentage of institutional ownership is negatively related to
the percentage of value-relevant contributions, as measured by location match of the politician
receiving such contributions. It may be that institutional owners are hesitant to allow the firm to
openly court politicians and may serve as a neutralizing agent when it comes to making blatant
political spending decisions. Ozer et al. (2010) find that institutional investors are mostly
concerned about the long-term value of the firm, and to this end, it appears that such owners do
not perceive contributing to local politicians as being of a long-term strategic advantage.
The other variables of total and variable compensation are also significantly negatively
related to the percentage of value-relevant contributions made by the firms. Contrasting these
findings with those for the full sample of participating and non-participating firms, our results
indicate that while CEOs earning higher total and variable compensation are more likely to
participate in PACs, it appears that this does not incentivize them to focus on local (value relevant)
contributions alone. The results are consistent with literature that indicates that political
contributions may be more a management perquisite consumption rather than being value-adding
for the firm. Interestingly, we are unable to document any significant linkages between our other
value-relevant proxy, the committee assignment of the recipient politician, and any of our interest
alignment proxies.
Furthermore, to judge if decisions by the CEO to engage in PAC is agency driven, we must
test if CEO characteristics relate to the decision to engage in a PAC, and to the extent that they do
relate, how PAC participation contributes to performance. Coates (2011) finds that outside of the
regulated industries where there is extensive political lobbying, political activity is associated with
lower firm value. Aslan and Grinstein (2012) note that compensation explained by ‘political
capital’ of executives is associated with an average increase of 0.3% in firms’ operating
performance. Our next set of estimations test for this relationship as defined in our firm
performance hypothesis, and we provide these results in Table 6. In the first step of our
estimations, we quantify the extent to which CEO dominance and interest alignment predict
political contributions. In the second step we relate the predicted contributions to performance.
Thus, we ask if CEOs make value-maximizing decisions for the firm when engaging in political
contributions. Our proxies for performance are contemporaneous firm return, firm ROA, and firm
ROE.
<insert Table 6 about here>
We present the results of our analysis for our first performance variable, contemporaneous return
in Columns 1 and 2 of Table 6, wherein we model total contribution as a function of the CEO
dominance and interest alignment characteristics detailed in the previous estimations. In the first
step, we ascertain whether dominant CEOs and those CEOs whose interests are aligned with those
of the firm are politically active (Column 1, Table 6). We find that the amount of total contribution
is significantly and negatively related to CEO holdings and increases when there are more inside
directors in the firm. Further, we document that firms with higher market value, firms with larger
boards, and more leveraged firms also make larger political contributions.
Having established that strong and less aligned CEOs are politically active, we next
investigate if such CEOs are able to make value-maximizing decisions for the firm. In the second
step of our estimations, we relate the predicted contributions to the measures of corporate
performance (measured by contemporaneous return). Column 2 of Table 6 presents these results.
Our results show that total PAC contributions, to the extent explained by CEO characteristics, are
strongly and positively related to firm performance, as measured by stock returns. Thus, the results
suggest that political engagement via PACs are performance enhancing and are not necessarily
agency driven and detrimental to shareholder interests.
Our second proxy for firm performance is contemporaneous ROA, and we provide the
results of the estimations in Column 3 and 4 of Table 6. We find that the amount of total
contribution is significantly and positively related to CEO total compensation and marginally
increases when there are more inside directors in the firm. Turning to the control variables, we
document that firms with higher market value, firms with larger boards, and more leveraged firms
also make larger political contributions. Once again, we find that firms with higher levels of
political contributions are better performing firms, as measured by their ROAs (Column 4). It
appears that the dominant and/or interest aligned CEOs use political engagement through PAC
contributions to enhance firm performance. Therefore, not only are PAC contributions value-
adding, they may not actually be agency conflict driven. Thus, our results support Cooper, Gulen,
and Ovtchinnikov (2010) who note that public firms that contribute to political campaigns are
significantly correlated with a high rate of return, but do not support those of Coates (2011) who
finds that more politically engaged firms tend to be poorly performing firms. In our third set of
performance estimations, we examine the impact of contributions on firm ROE. We note in
Column 5 that the amount of total contribution is positively and significantly related to both CEO
total compensation and the ratio of inside directors. However, unlike our other performance
variables, we are unable to document a significant relation between ROE and PAC contributions.
With respect to control variables, we document that firm size is negatively related to both firm
return and firm ROA, while sales growth is positively related to firm performance, measured by
both return and ROA.
In our last set of estimations, we examine the same performance variables for our sub-
sample of contributing firms and report these results in Table 7.
<insert Table 7 about here>
Panel A of Table 7 reports the results for the location match. Columns 1 and 2 examine
performance as measured by return. We show that total value-relevant contributions are positively
related to CEO holdings and negatively related to total compensation and firm size in the first stage
estimations. Examining the impact on contemporaneous return (Column 2), we note contributions
to local politicians are very significantly and negatively related to contemporaneous return. Thus,
value-relevant contributions, as defined by our measure in terms of contributions to local
politicians, may not be much value adding.
Turning to our next proxy of performance, ROA, we show that similar to
contemporaneous return, total value-relevant contributions are negatively related to CEO total
compensation and positively related to CEO holdings (Column 3). However, contributions based
on location appear to have no significant impact on firm ROA. One explanation for this could be
that ROA is typically used as a managerial performance measure that may have more to do with
asset efficiency and therefore may be influenced minimally by value relevant contributions based
on location
We turn to our last performance proxy, ROE, in Columns 5 and 6 of Table 7. Once again
we note a positive relation between CEO holdings and the firm’s value-relevant contribution
noting that as CEOs interests are more aligned to the shareholders, value-relevant contributions
increase. These value relevant contributions to local politicians appear to pay off in terms of
accounting returns, as measured by ROE, and we document a very significant and positive relation
between value-relevant contributions and firm ROE in Column 6 in Table 7, Panel A.
Panel B of Table 7 repeats the analysis for our other value-relevant measure, industry
match, based on committee assignments of the politicians. We do not find any significant relation
between value-relevant contributions and performance as measured by contemporaneous return
(Column 2). Thus, while the full-sample results show a positive relation between PAC
contributions and market return, there appears to be no additional value added by donating to
politicians holding industry relevant Committee Assignments. However, when examining
accounting measures, we see a significantly negative relation between firm performance
(measured by ROA and ROE) and value-relevant contributions (Columns 4 and 6). Thus, PAC
contributions based on industry match may in fact be detrimental to value. We also note that
smaller firms made larger value-relevant contributions. As political engagement is more costly for
smaller firms overall, we see that while they made lower contributions overall (Table 6), those
that contributed made significantly more value-relevant contributions based on industry match as
they perhaps more judiciously chose to contribute to those politicians that they perceived as value-
adding. Overall, the performance estimations provide evidence demonstrating a negative relation
between firm performance and value-relevant PAC contributions implying value relevant
contributions did not lead to better performance. Our estimations in Table 7 allow us to examine
the linkages among CEO characteristics, political engagement, and firm performance. Our results
show that value-relevant contributions are not agency driven.
5. CONCLUSIONS
Motivated by studies on corporate governance and agency theory, our study extends the
literature on CEO dominance and corporate strategy by examining CEO characteristics and firm
performance to investigate their impact on strategic choices regarding political engagement by
corporations. Academic research also explores whether corporate political activity, as a non-
market strategy, is value relevant. Therefore, our study uses a unique, hand-collected dataset to
investigate whether firms make value relevant political contributions.
The purpose of our study is two-fold. First we want to investigate the linkages between
CEO characteristics and the likelihood of PAC participation. For the full sample of firms, our
results on CEO characteristics and the likelihood of PAC contributions shows mixed results. We
find that firms with dominant CEOs, as measured by CEO duality, have a greater propensity to
make PAC contributions, while longer tenured CEOs are less likely to contribute. Furthermore,
while CEO age in absolute terms does not influence corporate political decisions, CEO age relative
to the board impacts such decisions. Firms with relatively older CEOs have a stronger propensity
to contribute to PACs. A unique contribution of our paper is that we examine the value-relevance
of firm contributions by matching firm location and industry with the home-state and committee
assignments of the politicians receiving contributions. Overall, our results for CEO dominance
follows that of the full sample results; while there are some linkages between CEO dominance and
value-relevant contributions, there is no strong evidence documenting that dominant CEOs
contribute more to politicians who are in the position to influence the environment in which the
firm operates.
Examining the interest alignment variables for the full sample, we find that our
compensation-based measures of interest alignment are positively linked to the firm participation
in a PAC. Looking at the value-relevance for contributing firms, we find that CEOs with a greater
equity stake in the firm align themselves by making higher contributions to politicians from their
home-state. However, firms with higher institutional ownership tend to donate less to politicians
from the home, perhaps reflecting a reluctance of large investors to openly court politicians.
It is plausible that PAC contributions impact performance. Therefore, the second objective
of our paper is to investigate the relationships between CEO characteristics, political participation,
and firm performance The results for the full sample of firms show that strong CEOs who are
aligned with the firm make value-maximizing decisions for the firm, as proxied by firm return and
firm ROA. We also investigate how firm performance is related to the value-relevance of the
contributions. Overall, our results demonstrate that contributions to local politicians are not value
adding and donations made to politicians who hold Committee assignments in the industry in
which the firm operates do not perform better.
Our results have implications in the CEO dominance and agency-theoretic literature since
we focus on CEO influence on decision making and strategy. Although not unequivocal, our
results - that dominant CEOs and CEOs with better interest alignment are more likely to be
politically active, and that PAC contribution associate positively with firm performance - suggest
that strategic political decisions may not be agency driven and may in fact be value enhancing.
REFERENCES
Adams, R.B., H. Almeida and D. Ferreira (2005), ‘Powerful CEOs and Their Impact on
Corporate Performance’, Review of Financial Studies, Vol. 18, pp. 1403-1432.
Aggarwal, R. K., Meschke, F., and Wang, T. Y. (2012). Corporate political donations: investment or agency?. Business and Politics, 14(1).
Agrawal, A. and C. Knoeber (2001), ‘Do Some Outside Directors Play a Political Role?’, Journal
of Law and Economics, Vol. XLIV, (April), pp. 179-198.
Aslan, H. and Y. Grinstein (2012), ‘Political Contributions and CEO Pay,’ Working paper,
University of Houston.
Barclift, J.Z. (2011), ‘Corporate Governance and CEO Dominance,’ Washburn Law Journal,
Vol. 50, pp. 611-634.
Bebchuk, L.A. and A. Cohen (2005), ‘The Costs of Entrenched Boards’, Journal of Financial
Economics, Vol. 78, pp. 409-33.
Bebchuk, L.A., M. Cremers and U. Peyer (2008), ‘CEO Centrality’, The Harvard John M. Olin
Discussion Paper Series, Discussion Paper No. 601, ISSN 1936-5349 (print), ISSN 1936-5357
(online).
Bebchuk, L.A., M. Cremers and U. Peyer (2011), ‘CEO Pay Slice and Firm Performance’,
Journal of Financial Economics, Vol. 102, Issue 1, (October), pp. 199–221.
Bertrand, M., F. Kramaraz, A. Schoar, and D. Thesmar (2007), ‘Politicians, Firms and the Political Business Cycle: Evidence from France’, Working Paper, University of Chicago.
Bennett, R.W. and C. Loucks (2008) ‘PAC Contributions from Sectors of the Financial Services
Industry - 1998-2002’, Atlantic Economic Journal, Vol. 36, pp. 395-405.
Boubkari, N., J.C. Cosset, and W. Saffar (2012), ‘The Impact of Political Connections on Firms’
Operating Performance and Financing Decisions’, The Journal of Financial Research, Vol .
34(3), pp. 397-423.
Brown, J. R. (2012), ‘Neutralizing the Board of Directors and the Impact on Diversity’,
Available at SSRN: http://ssrn.com/abstract=1938427 or http://dx.doi.org/10.2139/ssrn.1938427
Claessens, S., E. Feijen and L. Laeven (2008), ‘Political Connections and Preferential Access to
Finance: The Role of Campaign Contributions’, Journal of Financial Economics, Vol. 88, Issue
3, (June), pp. 554–580.
Coates, I. V., and John, C. (2012). Corporate politics, governance, and value before and after
Citizens United. Journal of Empirical Legal Studies, 9(4), 657-696.
Coles, J.L., N.D. Daniel and L. Naveen (2008), ‘Boards: Does One Size Fit All?’, Journal
of Financial Economics, Vol. 87, No. 2, pp. 329-356.
Hillman, A.J., G.D. Keim and D. Schuler (2004), ‘Corporate Political Activity: A Review and
Research Agenda’, Journal of Management, Vol. 30, pp. 837
Jensen, M.C. (1993), ‘The Modern Industrial Revolution, Exit, and the Failure of Internal
Control systems’, Journal of Finance, Vol. 48, pp. 831-880.
Johnson, S. and T. Mitton (2003), ‘Cronyism and Capital Controls: Evidence from Malaysia’,
Journal of Financial Economics, Vol. 67, pp. 351–382.
Khurana, R. (2002), ‘Searching for a Corporate Savior: The Irrational Quest for Charismatic
CEOs’, Published by Princeton University Press, (pages 20-21).
Khwaja, A.I. and A. Mian (2005), ‘Do Lenders Favor Politically Connected Firms? Rent
Provision in an Emerging Financial Market’, Quarterly Journal of Economics, Vol. 120,
pp.1391–1411.
Klein, A. (1998), ‘Firm Performance and Board Committee Structure,’ Journal of Law and
Economics, Vol. 41, No. 1, pp. 275-304.
Liebman, B.H. and K.M. Reynolds (2009), ‘The Returns from Rent Seeking: Campaign
Contributions, Firm Subsidies and The Byrd Amendment’, Canadian Journal of Economics,
Revue Canadienne d’Economique, Vol. 39, pp. 1345-69.
Lasfer, M. A. (2006), 'The Interrelationship Between Managerial Ownership and Board
Structure', Journal of Business Finance & Accounting, Vol. 33, (7) & (8) (September/October),
pp. 1006-103
Lord, R. A., and Saito, Y. (2012). Does Compensation Structure Alleviate Personal CEO Risks?.
Journal of Business Finance & Accounting, 39(9‐10), 1272-1297.
Macher, J.T., J.W. Mayo and M. Schiffer (2011), ‘The Influence of Firms on Government’, The
B.E. Journal of Economic Analysis & Policy, Vol 11, Issue 1, pp. 1-25.
Mande, V. and M. Son (2012), ‘CEO Centrality and Meeting or Beating Analysts’ Forecasts’,
Journal of Business Finance and Accounting, Vol. 39, pp 82-112.
Mathur, I. and M. Singh (2011), ‘Corporate Political Strategies’, Accounting and Finance, Vol.
51, pp. 252-277.
Mathur, I., Singh, M., Thompson, F., and Nejadmalayeri, A. (2013). Corporate governance and
lobbying strategies. Journal of Business Research, 66(4), 547-553.
Mobarak, A.M., and D.P. Purbasari, (2006), ‘Corrupt Protection for Sale to Firms: Evidence from
Indonesia’, Unpublished working paper, University of Colorado at Boulder.
Morse, A., V. Nanda and A. Seru (2011), ‘Are Incentive Contracts Rigged by Powerful CEOs?’
The Journal of Finance, Vol. 66, No.5, (October), pp. 1461–1867.
Ozer, M. and S.H. Lee (2009), ‘When Do Firms Prefer Individual Action to Collective Action in
The Pursuit of Corporate Political Strategy? A New Perspective on Industry Concentration’,
Business and Politics, Vol. 11, Issue 1, article 4.
Ozer, M. (2010), ‘Top Management Teams and Corporate Political Activity: Do Top
Management Teams Have Influence on Corporate Political Activity?’, Journal of Business
Research, Vol. 63, Issue 11, pp. 1196-1201.
Ozer, M., E. Alakent and M. Ahsan (2010), ‘Institutional Ownership and Corporate Political
Strategies: Does Heterogeneity of Institutional Owners Matter?’, Strategic Management Review,
Vol. 4(1), pp. 18-29.
Prabhat, S. (2012), ‘Do Political Contributions Enhance, Shareholder Equity Value?’ Indian
School of Business, Working Paper.
Schuler, D., K. Rehbein and R. Cramer (2002), ‘Pursuing Strategic Advantage Through Political
Means: A Multivariate Approach’, Academy of Management Journal, Vol. 45, pp. 659-672.
Yermack, D. (1996), ‘Higher market Valuation of Companies with a Small Board of Directors’,
Journal of Financial Economics, Vol. 40, No. 2, pp. 185-211.
Table1: Descriptive Statistics
This table provides the univariate tests for difference of means for firms making contributions to political action
committees (PACs) and for firms with no PAC contributions.
Variable PAC (mean)
(std. error)
NON PAC (mean)
(std. error)
t stat
(p-value)
Ratio of Inside Directors .137447
(.0023707)
.1715893
(.0036216)
8.1154
(.0000)
CEO Age 55.59153
(.2246658)
54.76706
(.2826332)
-2.3119
(0.0209)
CEO Tenure 6.44
(.2358615)
7.85119
(.3049674)
3.7143
(0.0002)
Total Compensation 3109066
(98113.93)
2302123
(109871.7)
-5.4902
(0.0000)
Variable Compensation % .5715474
(.0109258)
.5066234
(.0129564)
-3.8585
(0.0001)
CEO Holdings 1457.762
(483.3209)
676.6232
(853.4716)
-0.8382
(0.4021)
Institutional Ownership % .7018847
(.0052983)
.758979
(.0058918)
7.2101
(0.0000)
Total # Directors 11.52547
(.0887923)
10.1363
(.0999996)
-10.4159
(0.0000)
Firm Age 58.45515
(1.761582)
47.6865
(1.61909)
-4.4297
(0.0000)
Market Value 30583.56
(2105.859)
11925.25
(762.3762)
-7.5156
(0.0000)
Return 0 8.235622
(1.393896)
8.226134
(1.944239)
-0.0041
(0.9968)
ROA 0 .0451787
(.0033774)
.0501913
(.0071527)
0.6795
(0.4969)
Debt Ratio .2697358
(.0060789)
.2208658
(.008266)
-4.8713
(0.0000)
Market/Book Ratio 5.003562
(1.274594)
4.471595
(.4338186)
-0.3558
(0.7221)
Number of observations 682 522
Table 2: Full Sample CEO Dominance Probit Estimations
This table reports the coefficients (std. errors) for the probit estimations for the CEO dominance Hypothesis, proxied
by the number of inside directors, whether the CEO has a dual title as the Chair of the Board, CEO age, CEO age
relative to the board (rCEOAGE), and the length of CEO tenure. All variables are calculated at the end of the previous
year.
CEO Dominance Ratio Inside
Directors (Model 1)
CEO Duality
(Model 2)
CEO Age
(Model 3)
rCEOAGE
(Model 4)
CEO Tenure
(Model 5)
sic1 .2724492
(.4400334)
-.232116
(.4996397)
.2716348
(.4352322)
.070438
(.5812199)
.2518409
(.4442467)
sic2 .2415012
(.4221549)
-.1271451
(.4851193 )
.2363694
(.4181347)
.233308
(.565334)
.1845773
(.4268812)
sic3 .1072101
(.4165076)
-.2885936
(.4776364)
.1090767
(.4126817)
-.2373496
(.5560905)
.0634694
(.4216539)
sic4 1.329513***
(.4359185)
.7848245
(.5018661)
1.357128***
(.4315208)
.9812011*
(.5824477)
1.311347***
(.4401104)
sic5 .2078236
(.4267784)
-.3304208
(.4880364)
.1823735
(.4226368)
.0253779
(.5686572)
.1290337
(.4312793)
sic6 .0190804
(.426561)
-.5376937
(.4896169)
.0060405
(.422303)
-.2289545
(.5656053)
.0086847
(.4307792)
sic7 .2635717
(.4275815)
-.0398671
(.4901924)
.2431479
(.423712)
-.1096676
(.5767403)
.2083619
(.4322126)
Ln Market Value .308081***
(.0402442)
.2651298***
(.0451656)
.3134045***
(.0405298)
.2522236***
(.0611903)
.315614 ***
(.0405958)
Market/Book
Ratio
.0006684
(.0014367)
-.0004285
(.0038517)
.0005471
(.0014368)
-.0069998
(.0058167)
.0006318
(.0014148)
Debt Ratio -1.28e-06**
(6.45e-07)
2.65e-06*
(1.55e-06)
-1.21e-06**
(6.44e-07)
-5.61e-07
(8.92e-07)
-1.35e-06**
(6.49e-07)
Return -.0014497
(.0010274)
-.001332
(.0011494)
-.0015811
(.0010234)
-.0012662
(.0016107)
-.0014018
(.0010221)
Firm Age .0025754***
(.0009261)
.0022639**
(.0010804)
.0028941***
(.0009302)
.0043517***
(.0016008)
.0027108***
(.0009294)
Total # of
Directors
.0963153***
(.0180895)
.0891413***
(.0201602)
.1042348***
(.0179102)
.1373882***
(.0254834)
.1014996***
(.0180277)
Ratio of Inside
Directors
-2.250875***
(.5266472)
Duality
.4227159***
(.0912257)
CEO Age .0006131
(.0060358)
rCEOAGE .0074441**
(.003393)
CEO Tenure -.0170916***
(.0055503)
Constant -3.793653***
(.5440881)
-3.408217***
(6148857)
-
4.315296***
(.6239018)
-
4.351267***
(.782413)
-4.102112***
(.5409512)
Number of Obs.
1197 969 1194 553 1190
Psuedo R2
.1588 .1549 .1497 .1666 .1536
***, **, * represents significance at the 1%, 5%, and 10% levels respectively
Table 3: Value-Relevant CEO Dominance Tobit Estimations This table reports the coefficients (std. errors) for the tobit estimations for the CEO dominance Hypothesis, proxied by the number of inside directors, CEO Duality, CEO age,
rCEOAge and the length of CEO tenure. All variables are calculated at the end of the previous year.
Ratio inside directors
(Model 1)
CEO Duality
(Model 2)
CEO Age
(Model 3)
rCEOAge
(Model 4)
CEO Tenure
(Model 5)
Location
match Industry match
Location match
Industry match
Location match Industry match
Location match Industry match Location match Industry match
sic1 .0800499
(.1395015)
-.2122155**
(.1082116 )
.1109114
(.1393431)
-.2392486**
(.1070961)
.0625371
(.1400335)
-.2139453**
(.108241)
-.02689
(.1024339)
-.4350935**
(.1800085)
.0612782
(.1407323)
-.213107*
(.1088758)
sic2 .2220131*
(.1331246)
-.1694254*
(.1023402)
.2699449**
(.1325165)
-.1971599*
(.1008072)
.2169994
(.1336127)
-.1732522*
(.1024067)
.208743***
(.0806271)
-.3116757*
(.1730292)
.221188*
(.1338836)
-.1679486
(.1025783)
sic3 .0756399
(.1324228) -.1725086* (.1016582)
.1230636 (.1319078)
-.2053093** (.100391)
.0726133 (.1328998)
-.1722673* (.1017178)
.0233285 (.0806784)
-.2345786 (.172921)
.0738211 (.1331413)
-.1710207* (.1018633)
sic4 .1007082
(.1336653)
-.1440668
(.1025958)
.1377206
(.1330155)
-.1575922*
(.1009825)
.0865913
(.1343069)
-.1486849
(.1027655)
.1017274
(.0899022)
-.2376519
(.176019)
.0923009
(.1343681)
-.141486
(.1027888)
sic5 .2875483**
(.1351916)
-.1716375
(.104387)
.3227413**
(.1344614)
-.1766264
(.1026474)
.285863**
(.1355723)
-.1706971
(.1043819)
.265681***
(.0902677)
-.2751599
(.1761258)
.2810341**
(.1360861)
-.1650424
(.1047238)
sic6 .1987382
(.1349213) -.1051829 (.1038049)
.2376684* (.1341018)
-.1270211 (.1020486)
.1904297 (.135428)
-.1073918 (.1038614)
-2.074651 .
-.2412094 (.1761215)
.1883079 (.1358638)
-.1040484 (.1041964)
sic7 .0101689
(.1378295)
-.144308
(.1057349)
.0430815
(.1363864)
-.1637005
(.1034058)
.0075184
(.1383616)
-.152997
(.1058453)
-.0607032
(.1127683)
-.2180805
(.1788966)
.0055094
(.1388273)
-.1471024
(.1062503)
Ln Market
Value
-.09315***
(.0110254)
-.027743***
(.0089597)
-.098425***
(.011079)
-.026763***
(.0088528)
-.092451***
(.0111301)
-.028755***
(.0090099)
-.098499***
(.0217532)
-.032134**
(.0159837)
-.093252***
(.0111416)
-.028670***
(.0090817)
Market/Boo
k Ratio
.0011014
(.0009592)
-.0013206
(.0009761)
.0011631
(.0009443)
-.0013611
(.0009477)
.0011464
(.0009637)
-.0013641
(.0009739)
.0023232
(.0020161)
-.0000886
(.0014859)
.0011031
(.0010385)
-.0016308
(.0011539)
Debt Ratio -5.67e-07*
(3.44e-07)
-3.78e-07
(2.35e-07)
-5.92e-07*
(3.39e-07)
-3.37e-07
(2.30e-07)
-5.83e-07*
(3.47e-07)
-3.69e-07
(2.36e-07)
-4.51e-07
(6.46e-07)
1.45e-07
(3.37e-07)
-5.78e-07*
(3.49e-07)
-3.69e-07
(2.37e-07)
Return .0005164
(.0003533)
.0003451
(.0002831)
.0004933
(.0003683)
.000305
(.0002912)
.0005416
(.000356)
.0003795
(.0002841)
.0011252*
(.0006505)
.0006101
(.0004683)
.0005031
(.0003556)
.0003571
(.0002842)
Ratio of Inside
Directors
.3265228*
(.1824166)
-.1564889
(.1484811)
CEO Duality -.0576845** (.0264947)
.0353349* (.0214088)
CEO Age .0019899
(.0021378)
.0020433
(.0017056)
rCEO Age .0009991
(.00137)
-.0002074
(.0009999)
CEO Tenure .0019008
(.0020817) .0002049
(.0016553)
Total # of
Directors
.0020398
(.0055736)
-.0019886
(.0045391)
.0023646
(.0055658)
-.0023168
(.0044657)
.0006163
(.0056259)
-.0025514
(.0045656)
.0048803
(.0108755)
-.0177831**
(.0080698)
.001258
(.0055919)
-.0017246
(.0045401)
Constant .821437***
(.1639589)
.6153104***
(.1285483)
.9139671***
(.1604144)
.5873986***
(.1240614)
.7724822***
(.1910032)
.4992958***
(.1501632)
.8731509***
(.2289847)
.9485118***
(.2283868)
.8705177***
(.1630825)
.5980518***
(.1275357)
# of Obs. 548 548 532 532 546 546 232 232 545 545
Psuedo R2 0.2738 0.1778 0.3005 0.2398 0.266 0.1812 0.2138 0.25 0.2643 0.1746 ***, **, * represents significance at the 1%, 5%, and 10% levels respectively
Table 4. Full Sample Interest Alignment Probit Estimations
This table reports the coefficients (std. errors) for the probit results for the Interest Alignment Hypothesis, proxied by
the natural log of CEO holdings, the percentage of institutional ownership, the natural log of CEO total compensation,
and the percentage of CEO variable compensation. All variables are calculated at the end of the previous year.
Compensation/Ownership Ln CEO
Holdings
(Model 1)
Institutional
Ownership
(Model 2)
Ln Total
Compensation
(Model 3)
Variable
Compensation %
(Model 4)
sic1 .0734766
(.4541163)
.3229737
(.4387778)
.199795
(.4362087)
.2646499
(.4372379)
sic2 . 0920088
(.4374143)
.2507958
(.4208645)
.2200498
(.4186916)
.2527113
(.4207103)
sic3 -.0337732
(.4317228)
.1334377
(.4154506)
.1540256
(.4130811)
.1510309
(.4151474)
sic4 1.171834***
(.4504957)
1.395333***
(.4355296)
1.363466***
(.4319658)
1.397182***
(.4343331)
sic5 .0340309
(.4418448)
.2360167
(.4264649)
.185081
(.4232739)
.2111039
(.4250827)
sic6 -.1197039
(.4409047)
.0389146
(.4252148)
-.0191071
(.4232317)
.0143457
(.4247056)
sic7 .189098
(.4428801)
.2749732
(.4274126)
.24377
(.4245339)
.286345
(.4264315)
Ln Market Value .2889431
(.0415943)
.3059919***
(.0416397)
.2861705***
(.0409176)
.3077733***
(.0406442)
Market/Book Ratio .0019686***
(.0018471)
.0006156
(.0014304)
.0004558
(.0014685)
.0005474
(.0014574)
Debt Ratio -1.09e-06*
(6.59e-07)
-9.62e-07
(7.04e-07)
-1.13e-06*
(6.76e-07)
-1.05e-06
(6.63e-07)
Return -.0011806
(.0010435)
-.001629
(.0010354)
-.0019041*
(.00105)
-.001521
(.0010257)
Firm Age .0029603***
(.000946)
.0028402***
(.0009339)
.0026425***
(.0009378)
.0028963***
(.000934)
Total # of Directors .1016598***
(.0183709)
.0974317***
(.0182235)
.1023293***
(.0182022)
.1027044***
(.0180578)
Ln CEO Holdings -.0663442***
(.0205638)
Institutional Ownership % -.3427152
(.2990284)
Ln Total Compensation .156605***
(.0438465)
Variable Compensation % .2950879**
(.1252167)
Constant -3.590641***
(.5775117)
-3.903024***
(.6235146)
-6.269178***
(.7923597)
-4.406727***
(.539842)
Number of Obs.
1149 1182 1185 1188
Pseudo R2
.1555 .1507 .1557 .1588
***, **, * represents significance at the 1%, 5%, and 10% levels respectively