Effects Of Public Sector Corruption On The Private Sector: Investigating The Market Value Of Political Connections Omotoke Paul-Lawal * Under the direction of Professor Arun Chandrasekhar Stanford University Department of Economics Honors Thesis May 30, 2016 Abstract This paper exploits public announcements of corruption allegations amongst Nigerian political figures to estimate the impact of political connections on firm value. Specifically, I draw results from market reactions to 27 events to determine whether politically connected firms suffer more from financial malpractice amongst government officials. This paper finds heterogeneous impacts of corruption announcements for firms’ returns. For some events involving larger sums of misallocated funds, politically connected firms suffer greater losses in their returns than unconnected firms. For other events involving smaller amounts of missing funds, I find that unconnected firms are actually punished more than connected firms. Still for other events, there appears to be no significantly differential impact of a government corruption scandal on connected and unconnected firms. Furthermore, surprisingly, some events actually show a positive return for all firms, and when this is the case, politically connected firms experience a larger increase in returns than their unconnected counterparts. To identify an overall impact, I run a pooled regression aggregating over all the events studied. I find that on average, there is a negative albeit insignificant impact on the returns of politically connected firms relative to unconnected firms in the aftermath of a corruption scandal. Keywords: political connections, corruption, firm value, Nigeria * Email address: [email protected]. I owe many thanks to my thesis advisor and Economics major advisor, Arun Chandrasekhar, whose advice, positivity, and endless support were invaluable in completing this work. My gratitude also goes to the Honors Director, Marcelo Clerici-Arias for organizing this program and always being available to answer questions. Special thanks go to Francisco Munoz for his comments and help in editing this paper. And last but not least, I am also greatly indebted to my family and friends without whose encouragement and confidence I could not have completed this project.
46
Embed
Effects Of Public Sector Corruption On The Private Sector ... Of Public Sector Corruption On The Private Sector: Investigating The Market Value Of ... Perceptions Index ... from the
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
Effects Of Public Sector Corruption On The Private Sector: Investigating The Market Value Of Political Connections
Omotoke Paul-Lawal*
Under the direction of Professor Arun Chandrasekhar
Stanford University Department of Economics
Honors Thesis
May 30, 2016
Abstract This paper exploits public announcements of corruption allegations amongst Nigerian political figures to estimate the impact of political connections on firm value. Specifically, I draw results from market reactions to 27 events to determine whether politically connected firms suffer more from financial malpractice amongst government officials. This paper finds heterogeneous impacts of corruption announcements for firms’ returns. For some events involving larger sums of misallocated funds, politically connected firms suffer greater losses in their returns than unconnected firms. For other events involving smaller amounts of missing funds, I find that unconnected firms are actually punished more than connected firms. Still for other events, there appears to be no significantly differential impact of a government corruption scandal on connected and unconnected firms. Furthermore, surprisingly, some events actually show a positive return for all firms, and when this is the case, politically connected firms experience a larger increase in returns than their unconnected counterparts. To identify an overall impact, I run a pooled regression aggregating over all the events studied. I find that on average, there is a negative albeit insignificant impact on the returns of politically connected firms relative to unconnected firms in the aftermath of a corruption scandal. Keywords: political connections, corruption, firm value, Nigeria
* Email address: [email protected]. I owe many thanks to my thesis advisor and Economics major advisor, Arun Chandrasekhar, whose advice, positivity, and endless support were invaluable in completing this work. My gratitude also goes to the Honors Director, Marcelo Clerici-Arias for organizing this program and always being available to answer questions. Special thanks go to Francisco Munoz for his comments and help in editing this paper. And last but not least, I am also greatly indebted to my family and friends without whose encouragement and confidence I could not have completed this project.
OmotokePaul-Lawal
2
1. Introduction
Historically, Nigeria has been infamous for its endemic fraud and corruption, and the
country’s economy has been fraught with the difficulties of overcoming the challenges posed by
these institutional problems. Nigeria ranked 136 out of 168 countries, with a score of only 26 out
of 100, on Transparency International’s widely quoted 2015 Corruptions Perceptions Index. As
a step towards combating these institutional challenges, in 2002, the Nigerian government
created an independent organization, the Economic and Financial Crimes Commission (EFCC), a
watchdog agency that acts as surveillance against money laundering, fraud, embezzlement, and
other forms of financial crime. Individuals and organizations report cases to the EFCC, who then
take up the case for research and investigation, often culminating in litigation. The importance of
the EFCC’s work is reflected in the widespread publicity of its cases – a fact that this paper
exploits.
When a case is reported to the EFCC, the organization usually takes the information to
the press, exposing details about the case, including names of the concerned individuals and their
associated organizations or government positions, as well as the amount alleged to be missing
under their control. This information often makes big news, and there has been much speculation
that the reports have large impacts on firms listed on the Nigerian Stock Exchange. This paper is
concerned with the extent to which the data supports this claim, and the extent to which investors
respond to such corruption announcements. The study focuses on high profile cases involving
public officials, with the aim of testing the impact of such notable cases of public sector
corruption on private firms. The fact that these cases involve well-known politicians, and that all
OmotokePaul-Lawal
3
the cases studied are reported in at least of one of the country’s major newspapers and/or online,
provides support for the belief that the information is readily available to the public and
investors. If the market response to the news of scandal is large, this informs us not only about
the importance of political connections, but also about the potential impact of corruption on the
financial economy, and the important role that fraud control agencies like the EFCC may play.
To investigate the value of political connections for firms, this paper studies 27 high-
profile corruption cases in which public officials are accused of some sort of financial
malpractice. The public officials include elected and appointed federal and state political figures.
Majority of the cases involve former governors of state, and the rest involve senators and
ministers of various government departments. This study takes a broad view of political
connections by defining a politically connected firm as a firm in which at least one member of its
board of directors had at any time in his or her past (before the event date) held a position such as
senator, member of the House of Representatives, or member of the administration, or had been a
director of an important government organization in the same party as the sitting party at the time
of the event. By taking a broad approach, I am able to examine whether the market punishes
firms with any type of link to the government, as defined above, in response to the actions of
one/a few government officials. If this is so, then there will be grave consequences of political
corruption on any scale for firms with potential access to political favors as suggested by their
connections.
OmotokePaul-Lawal
4
2. Literature Review
There is a growing body of empirical literature on the importance of political connections
to firm value. In different contexts, many researchers have found that political connections have
a significant positive impact on firm value. Fisman (2001) examines the impact of several news
announcements concerning the deteriorating health of President Suharto on firms in Indonesia
with political links to the Suharto family. He finds that well-connected firms suffer more than
less-connected firms in the event of a serious rumor, suggesting that a large share of a well-
connected firm’s value is derived from political connections. In her cross-country study, Faccio
(2006) finds that a company’s value increases when its large shareholders enter into politics, and
this result is most prominent in highly corrupt countries. Similarly, Do et al. (2015) find that
firms connected to elected governors experience a significant increase in their value compared to
those connected to the losing candidate. This study will enrich the existing international evidence
on the value of political connections (e.g. Faccio 2006; Fisman 2001; Prem 2015), whilst
complementing findings in the US (e.g. Do et al. 2015; Goldman, Rocholl, and So, 2009).
In addition to estimating the importance of political connections, a number of papers also
provide channels through which connections create value for politically connected firms. Several
studies have suggested that political connections are valuable to firms because they lead to
preferential access to bank finance (e.g. Gonzalez and Prem 2015; Li et al. 2008, Claessens et al.
2006; Kwhaja and Mian 2005). For example, Gonzalez and Prem (2015) argue that politically
connected firms are better able to make strategic investments when faced with a political
transition because of distortions in the credit market which favor connected firms, allowing them
to maintain their market dominance from the non-democratic regime into the new government.
OmotokePaul-Lawal
5
Kwhaja and Mian (2005) also find that political firms borrow more, despite having higher
default rates, and this preferential treatment occurs only in government banks and not in private
banks. Meanwhile, Faccio, Masulis, and McConnell (2006) suggest that connections create value
by showing that politically connected firms are more likely to be bailed out than their
unconnected counterparts.
This paper will also contribute to the literature that studies the value of political
connections through exploiting events that happened independently of the connections. Many
papers have exploited close election races, such as Do et al. (2015), which uses a regression
discontinuity design to examine the impact of a connected governor winning office on firm
value. Similarly, Goldman et al. (2008), Knight (2007), and Matozzi (2008) all exploit close
elections in US presidential races. Roberts (1990), Fisman et al. (2006), and Jayachandran (2006)
use events or news involving key political figures in the US, whilst Prem (2015), Fisman (2001),
and Ferguson and Voth (2008) exploit important political events in Chile, Indonesia, and Nazi
Germany, amongst other international studies of political connections (e.g. Johnson and Mitton
2003 in Malaysia; Imai and Shelton 2010 in Taiwan). This strategy of using independent or
unexpected events is favored in the literature because it provides a relatively clean solution to the
identification problem, whilst avoiding the reverse causation channel (Do et al. 2015).
Research on political connections in the Nigerian context is scarce. However, the few
existing results suggest that political connections do not play a huge role in the valuation of firms
in Nigeria. For example, Osamwonyi and Tafamel (2013) find that there is no significant
relationship between board composition, board political connections, and firm performance for
OmotokePaul-Lawal
6
their sample of thirty firms listed on the Nigerian Stock Exchange. Likewise, Aburime (2009)
finds that political affiliation has a positive but insignificant impact on bank profitability in
Nigeria. Some of the results of this paper seem to affirm these findings.
The current study is unique in the wealth of events it exploits to investigate the
comparative impact of bad news related to public officials on the value of politically connected
private firms and that of unconnected firms. The degree to which news of corruption affects the
health of the economy, as indicated by the value of firms that make up that economy, is
dependent on how much perceptions of government affect the willingness of investors to invest
in private Nigerian firms. This research will try to assess the magnitude of that dependence. If
the effect is significant, then this provides further impetus for government to clamp down on
instances of financial corruption, and informs corporate governance and institutional design.
3. Data and Descriptive Statistics
This paper uses two main bodies of data: it employs stock market data, and it also uses
information about the corruption events. The sample consists of all firms on the Nigerian Stock
Exchange (NSE) that remained listed between 2005 and 2015, the period of the events that this
paper studies. This leads to a total sample of 123 firms, of which I exclude those that lack
sufficient stock market and political connections data to observe returns for the corruption
announcements. This leads to a remaining base sample of 106 firms. All financial information
such as closing price, trade volumes, EPS, and market index values were retrieved from the
Nigerian Stock Exchange databases. Data about firms’ political connections was hand-collected
OmotokePaul-Lawal
7
by gathering the names of all members of the board in the relevant years from the firms’ annual
reports, and researching their career background in search of any political history. To aid this
process, I used online sources like Bloomberg and Reuters that detail biographies of reputable
business people, along with director introductions on the companies’ websites. I then verified the
political connections using Google searches that usually returned several articles mentioning the
political involvement of the director.
Data on the corruption events was acquired from the Economic and Financial Crimes
Commission (EFCC) in Nigeria. Relevant information about the corruption events include the
date that the case was first released to the public, the government personality involved, and the
amount alleged to be missing, stolen, or otherwise diverted. Where available, the date of EFCC’s
press release is used as day 1 of the event, or the event start date. When this was unavailable, I
used an online search to track down the first evident publication reporting the event in order to
determine the start date as this is key in identifying when the information first became publicly
accessible, allowing the stock market to react. One concern of this research is that particularly
connected investors or other insiders may have had access to the information before it became
public knowledge. If this is the case, then the public announcement may have no effect on the
firm, or the firm’s value may have adjusted earlier than the announcement date used. However, if
this is the case, then any findings here are lower-bounds on the true effects of the corruption
events. Furthermore, even if the market had anticipated the scandal, then the announcement
resolves any remaining uncertainty about whether the politician’s fraudulent actions were simply
rumors, or serious, actionable matters – a fact that should generate further market responses.
OmotokePaul-Lawal
8
As described above, this paper defines politically connected firms as those firms in which
at least one member of the board of directors had at any time in his or her past held a position
such as senator, member of the House of Representatives, or member of the administration, or
had been a director of an important government organization in the same party as the party in
government at the time of the event. These include both elected and appointed positions, but are
limited to federal and state government officials, who may be thought to be in a position to
provide significant access to political resources and favors. Furthermore, only one party has
been in government since Nigeria resumed elections in 1999 following the period of military
rule. This closed political structure, as well as the elitist club nature of politics in Nigeria, makes
this a valid definition of political connections.
Row 1 of Panel A in Table 1 reports descriptive statistics for the connections variable.
The firm with the greatest number of connections is United Bank of Africa (UBA) with a total of
7 connected board members over the studied years. In the first phase of the study, I assign a
value of 1 to all firms with any number of connected board members in the year of the event. On
average across the listed firms, the mean connections in the sample is 0.49. The average number
of connections conditional on a firm having at least one connection is 1.78.
Rows 2 to 4 in Panel A of Table 1 report basic financial information for the sample firms
from the Nigerian Stock Exchange database for 2014, supplemented by data from the financial
statements of the companies. Row 2 reports the firm’s size as the logarithm of total assets.
Return on equity is used for profitability, and leverage in Row 4 is the ratio of total debt to total
capital. Panel B reports differences in the means of these variables for connected and
OmotokePaul-Lawal
9
unconnected firms. As might be expected, connected firms are larger than unconnected ones (by
6.57%). Furthermore, connections appear to be significantly higher in firms in the finance and
agricultural industries. To ensure that these differences between connected and unconnected
firms are not driving the results, I conduct robustness checks that control for size and industry.
However, connected firms are not significantly more profitable, more leveraged, or older than
unconnected ones, so any differences we observe in market returns for the two types of firms
cannot be attributed to these characteristics.
4. Empirical Strategy
The econometric strategy exploits within firm variation and the corruption events as
exogenous announcements in a standard differences-in-differences analysis using panel data.
Since all firms are similar on most basic characteristics, as shown in the last section, observed
market responses reflect differences in political connections. A major concern of this research is
that the results may have been driven by other events in the macroeconomy. To control for the
fact that there were large market movements during the event windows, this paper calculates
abnormal returns for the event days. I follow Acemoglu 2013 and calculate abnormal returns
using the following market model:
Abnormal Returns or AR is calculated as
!"!" = !!" − [!!! + !!!!!"]
(1)
OmotokePaul-Lawal
10
where !"!" is the abnormal return for firm ! on event day t. !!" is the firm’s actual return, and
!!" is the market return, using the NSE’s All Shares Index (ASI). The parameters, !!! and !!!
are estimated from the pre-event relationship between market returns and firm returns:
!!" = !! + !!!!" + !!"
I follow the standard in the literature of using a pre-event period of 250 trading days ending 30
days before the event start date. The abnormal returns shows the actual returns to each firm
minus the predicted returns based on the firm’s performance relative to the market over the
estimation period.
After calculating abnormal returns in this way for all firms in each event window, the
main regression this research estimates is as follows:
TABLE5:ROBUSTNESSCHECKSNotes:Thetableshowstheresultsofrobustnesschecksforeventsfoundtohavesignificantresults.Thespecificationincolumnslabeled(a) is a replication of the standard regression of the extendedmodel in Table 3. Columns labeled (b)control for size, (c)controls forindustry,andcolumnslabeled(d)controlforbothsizeandindustry. (1a) (1b) (1c) (1d) (2a) (2b) (2c) (2d)
REFERENCES Aburime, Toni. (2009), Impact of Political Affiliation on Bank Profitability in Nigeria (July 16, 2009). African Journal of Accounting, Economics, Finance and Banking Research, Vol. 4, No. 4, July 2009 Claessens, S. Feijen, E. Laeven, L. (2008), Political connections and preferential access
to finance: the role of campaign contributions. Journal of Financial Economics 88(3), 554-80.
Do, Quoc-Anh, Yen Teik Lee, and Bang Dang Nguyen (2015), "Political Connections
and Firm Value: Evidence from the Regression Discontinuity Design of Close Gubernatorial Elections." CEPR Discussion Paper No. DP10526.
Faccio, M. (2006), “Politically Connected Firms”, American Economic Review, 96: 369-
386. Faccio, M., Masulis, R., McConnell. J. 2006. Political connections and corporate
bailouts. Journal of Finance 61(6), 2597-635. Ferguson, T., Voth, H. (2008), Betting on Hitler – the value of political connections in
Nazi Germany. Quarterly Journal of Economics 123(1), 101-37. Fisman, David, Raymond J. Fisman, Julia Galef, Rakesh Khurana, and Yongxiang
Wang (2012), "Estimating the Value of Connections to Vice-President Cheney." The B.E. Journal of Economic Analysis & Policy 12.3
Fisman, R. (2001), “Estimating the Value of Political Connections”, American Economic
Review, 91: 1095-1102 Goldman, E., Rocholl, J., So, J. (2009), Do politically connected boards affect firm
value?” Review of Financial Studies, 22(6), 2331-60. Gonzalez Felipé, Prem Mounu (2016), “Losing Your Dictator: Firms During Political
Transition”. Working Paper, Stanford University. Imai, Masami, and Cameron A. Shelton (2011), "Elections and Political Risk: New
Evidence from the 2008 Taiwanese Presidential Election." Journal of Public Economics 95(7-8), 837-49.
Jayachandran, S. (2006), The Jeffords Effect. Journal of Law and Economics 49(2), 397-425. Johnson, S. Mitton, T. 2003. Cronyism and capital controls: evidence from Malaysia.
OmotokePaul-Lawal
46
Journal of Financial Economics 67(2), 351-82. Khwaja, A. I., and Mian, A. (2005), Do lenders favor politically connected firms? Rent
provision in an emerging financial market. Quarterly Journal of Economics 120(4), 1371-411.
Knight, B. (2007), Are policy platforms capitalized into equity prices? Evidence from
the Bush/Gore 2000 Presidential Election. Journal of Public Economics 91(1-2) 389-409.
Li, H., Meng, L., Wang, Q., Zhou, L.-A. (2008), Political connections, financing and
firm performance: Evidence from Chinese private firms. Journal of Development Economics, 87(2), 283-99.
Mattozzi, A. (2008), Can we insure against political uncertainty? Evidence from the U.S.
Stock Market. Public Choice 137(1-2), 43-55. Roberts, B. E. (1990), A dead senator tells no lies: seniority and the distribution of
federal benefits. American Journal of Political Science 34(1), 31-58. Osamwonyi,I.O.,TafamelE.A.(2013),FirmPerformanceandBoardPolitical
Connection:EvidencefromNigeria.EuropeanJournalofBusinessandManagement5(26),83-96.Transparency International. "Corruption by Country / Territory."
https://www.transparency.org/country/#NGA, accessed 21 May 2016.