Coles Research Magazine Fifth Issue | 2019
Welcome to the fifth edition of the Coles Research Magazine.
Coles Research Magazine highlights outstanding business research conducted by College faculty and students. The abstracts and executive summaries cover a broad range of topics and include both applied and discipline-based research. In that way, the magazine embodies the College’s research mission to develop thought leaders who advance business practice and theory. This edition features abstracts of Doctoral student and faculty collaborations, papers that won College research awards, and papers presented at the first Coles College Research Symposium. Held in fall 2018, the symposium was a showcase for research on improving all aspects of global security. Our faculty research builds on the solid foundation of business education to create ideas, elucidate and challenge accepted concepts, and develop leaders who change their communities and the world. I remain exceedingly proud of the work of our faculty members and guarantee that you will enjoy this edition of the Coles Research Magazine.
Kathy S. Schwaig Provost and Vice President for Academic Affairs Michael J. Coles College of Business Kennesaw State University
Table of Contents
Journal Publications - Financial Times Top 50 Journals Informed traders’ performance and the information environment: Evidence from experimental asset markets By Lucy F. Ackert, Bryan K. Church, and Ping Zhang Do debt covenants constrain borrowings prior to violation? Evidence from SFAS 160 By Moshe Cohen, Sharon Katz, Sunay Mutlu, and Gil Sadka 4 6
Journal Publication - Outstanding Impact
8
Corporate governance research in accounting and auditing: Insights, practice implications, and future research directions By Joseph V. Carcello, Dana R. Hermanson, and Zhongxia (Shelly) Ye
Journal Publication - Distinguished Journal
10
Espionage and the optimal standard of the Customs-Trade Partnership against Terrorism (C-TPAT) program in maritime security By Aniruddha Bagchi and Jomon A. Paul
Doctoral Research Summaries
12
An exploratory study of organizational security risk management for improved effectiveness By Angela Jackson-Summers, Humayun Zafar, Traci Carte, and Adriane Randolph Intuition in employee selection: Examining the conditions for accurate intuitive hiring decisions By Vinod Vincent, Rebecca Guidice, Neal Mero, and Stacy Campbell
14
Applied Research
16
Logistics jobs in Georgia By Michael Maloni and Stacy Campbell
Coles Research Symposium
18
Optimal standard of human rights in countering an insurgency By Aniruddha Bagchi and Jomon A. Paul Socially optimal IT investment for cybersecurity By Jomon A. Paul and Xinfang (Jocelyn) Wang Conflict and aid dependency – An explorative study motivated by the case of Palestine By Marcus Marktanner and Almuth D. Merkel Natural disasters and their impact on cooperation against a common enemy By Timothy Mathews and Jomon A. Paul
20 22
24
Summer Research Fellowship
26
Competitive awareness of emerging and developed economy multinationals By Canan C. Mutlu Do foreign lenders face different enforcement in the U.S.? By Sunay Mutlu
28
Working Papers
30
Decision-support model for cybersecurity risk planning: A two-stage stochastic programming framework featuring firms, government, and attacker By Jomon A. Paul and Minjiao Zhang
32
The dark side of crowdfunding platforms By Susan L. Young and Birton Cowden
* Coles College of Business faculty highlighted in bold.
Informed traders’ performance and the information environment: Evidence from experimental asset markets
Lucy F. Ackert, Bryan K. Church, and Ping Zhang
Accounting, Organizations and Society, Vol. 70 (October 2018), pp. 1-15
Overview We designed 18 experimental markets to investigate the effect of the information environment on informed traders’ performance. We vary the number of informed traders across experimental markets, representing different information environments and posit that the cost of information affects traders’ ability to prosper. Traders bid to acquire costly but imperfect information on asset value, then participate in a double-auction asset market. Using the inverse relationship between cost of information and number of informed traders, we assessed whether traders can properly determine the value of the information under environments defined as enriched or impoverished based on the number of informed traders. We find that traders in an impoverished environment pay too much for information and, once informed, do not transact enough to recover its cost. Surprisingly, traders who compete for information that confers a strong advantage do worse than those who compete in an environment where information is more widely available.
4 | Journal Publications - Financial Times Top 50 Journals
Executive Takeaways
■ When market information is very costly, traders overestimate their ability to generate profit. ■ Self-focus leads traders to pay too much for information and risk too little to protect this advantage. ■ To hide their advantage, informed traders do not trade enough to recover the cost of information. ■ When information is less costly, many traders become informed, speeding dissemination.
Lucy F. Ackert, Professor of Finance
Do debt covenants constrain borrowings prior to violation? Evidence from SFAS 160
Moshe Cohen, Sharon Katz, Sunay Mutlu, and Gil Sadka
The Accounting Review (forthcoming)
Overview Prior evidence shows that leverage is reduced after covenant violations, but we do not know whether leverage is affected before covenants are violated. In December 2007, the US Financial Accounting Standards Board (FASB) issued SFAS 160, which mechanically relaxed certain covenant types. This study uses SFAS 160 as an exogenous, accounting-based shock to debt covenants that relaxed their tightness to examine whether covenants constrain leverage for borrowers that are close to violation, even when they are financially healthy. We find that SFAS 160 increased debt levels in firms that were close to violation and that financially healthy firms drove this increase. We conclude that the likelihood of future covenant violations could impede borrowing. We also find an increase in investment sensitivity to Q after SFAS 160 in firms close to violation, suggesting that the additional debt was used to make legitimate investments. Because SFAS 160 was passed during the financial crisis, generalizing our findings to normal financial periods is difficult.
6 | Journal Publications - Financial Times Top 50 Journals
Executive Takeaways
■ A new accounting standard in 2007, SFAS 160, mechanically relaxed certain covenant types. ■ Financially healthy firms with relaxed covenants borrowed more after SFAS 160. ■ These firms used their additional borrowings to fund legitimate investments. ■ This evidence shows that tight covenants restrict borrowing even before they are violated.
Sunay Mutlu, Assistant Professor of Accounting
Corporate governance research in accounting and auditing: Insights, practice implications, and future research directions
Joseph V. Carcello, Dana R. Hermanson, and Zhongxia (Shelly) Ye
Auditing: A Journal of Practice & Theory, Vol. 30, No. 3 (August 2011), pp. 1-31
Since the mid-1990s, academic accounting researchers have examined the relation between board of directors and audit committee characteristics and a variety of accounting and auditing outcomes, including fraud, earnings management, earnings quality, auditor quality, and internal control effectiveness. Researchers consistently have found that good governance “on paper” (e.g., boards and audit committees with independent, expert members) consistently is associated with better accounting and auditing outcomes. However, reforms, such as the Sarbanes-Oxley Act, to improve the independence and expertise of corporate directors and the adoption of governance best practices now are causing many boards and audit committees to look similar. As a result, the ability to link good governance on paper with good accounting now is fading. We discuss a number of practice implications and avenues for research in the post-Sarbanes-Oxley environment. Overview
8 | Journal Publication - Outstanding Impact
Executive Takeaways
■ Good board / audit committee traits have been associated with good accounting outcomes. ■ Regulation and best practices are causing boards and audit committees to look similar. ■ The ability to link good governance on paper with good accounting now is fading. ■ We offer many avenues for future research in the new governance landscape.
Dana R. Hermanson, Professor, Dinos Eminent Scholar Chair of Private Enterprise
Espionage and the optimal standard of the Customs-Trade Partnership against Terrorism (C-TPAT) program in maritime security
Aniruddha Bagchi and Jomon A. Paul
European Journal of Operational Research, Vol. 262, No. 1 (October 2017), pp. 89-107
Overview
We examine the effect of espionage on optimal design of the C-TPAT (Customs Trade Partnership Against Terrorism), a trusted-trader program administered by US Customs and Border Protection (CBP) on behalf of the Department of Homeland Security. A major CBP concern is that one of the many containers passing through ports can be used to import a “dirty” bomb. Inspecting every container would be expensive and slow down trade. C-TPAT was introduced to manage this problem. Firms that sign up make some expenditures to secure their supply chain in exchange for priority processing of their goods at the port of entry. We examine how other factors, such as the quality of information about terrorist groups, affects C-TPAT efficacy. First, we predict that terrorist groups are likely to smuggle weapons infrequently but to focus on highly lethal weapons. Second, if the CBP requires a higher standard of supply-chain security from trusted traders, we identify circumstances that will decrease or increase firms’ propensity to sign up. Third, we demonstrate that, although counterintuitive, social welfare may increase in response to an increase in the opportunity cost of time.
10 | Journal Publication - Distinguished Journal
Executive Takeaways
■ Espionage and role of standards in the implementation of trusted trader program considered. ■ Government underspends on espionage relative to the congestion-minimizing level. ■ Membership size depends non-monotonically on each of these policies. ■ Highly motivated terrorists will smuggle deadlier weapons less often. ■ An increase in the opportunity cost of time can enhance social welfare in certain situations.
Jomon A. Paul, Professor of Quantitative Analysis Aniruddha Bagchi, Professor of Economics
An exploratory study of organizational security risk management for improved effectiveness Angela Jackson-Summers (PhD Graduate) Humayun Zafar (Dissertation Chair) Traci Carte (Second Supervisor) Adriane Randolph (Reader)
Information technology (IT) executives are increasingly concerned about security threats and the total costs of data breaches. While organizations have made large investments (i.e. hardware, software) to solve these problems, IT executives believe they are not enough. Few studies have considered the organizational security risk management (SRM) process itself. This study assesses the SRM process and proposes three ways to strengthen it. First, it develops a customized SRM dictionary and a modified capability maturity model to assess security risk management-related textual content. Second, it uses validation feedback from information SRM experts to clarify differences in how leaders view SRM capability and maturity levels. Finally, it creates a conceptual model for future research testing of internal and external SRM capabilities based on the textual content assessment. Overview
12 | Doctoral Research Summaries
Executive Takeaways
■ In assessing an organization’s security risk management (SRM) process, written artifacts may be used: ■ to convey the specific capability and maturity levels in place; ■ to identify possible drivers and challenges of its maturity and effectiveness; and ■ to define opportunities to strengthen the overall process.
Humayun Zafar, Associate Professor of Information Systems Angela Jackson-Summers, PhD Graduate Adriane Randolph, Associate Professor of Information Systems
Intuition in employee selection: Examining the conditions for accurate intuitive hiring decisions Vinod Vincent (DBA Graduate) Rebecca Guidice (Dissertation Co-Chair) Neal Mero (Dissertation Co-Chair) Stacy Campbell (Reader)
Overview
In complex organizational environments, managers often rely on intuition to make decisions. In contrast to analysis, which is a conscious and rational cognitive process, intuition is unconscious, automatic, and rapid. Although research has found intuition to be useful under certain conditions, little empirical evidence indicates when it will be effective in such organizational contexts as employee selection. Using an experimental design based on expert and non-expert interviewer samples, this study finds that when recruiting for a complex job, interviewer expertise has a positive impact on intuition. The intuitive decisions of expert interviewers are as good as their analytical decisions, and giving some weight to their intuitive judgment when hiring for complex jobs may be prudent. On the other hand, non-expert interviewers—for example, a new manager with no prior hiring experience—perform significantly better when using an analytical decision-making process rather than intuition. They should be provided with the training, tools, and procedures necessary to make analytical hiring decisions.
14 | Doctoral Research Summaries
Executive Takeaways
■ Experts make better intuitive decisions than non-experts. ■ Non-experts perform better when they use analytical rather than intuitive decision processes. ■ Expert intuition is effective in hiring for complex jobs.
Vinod Vincent, DBA Graduate Stacy Campbell, Professor of Management
Logistics jobs in Georgia
Michael Maloni and Stacy Campbell
Logistics is a $15 billion industry in Georgia, involving more than 11,000 companies and one million jobs. Yet, the global logistics industry faces a severe talent shortage. In response, an interdisciplinary team of experts in logistics and human resources at Kennesaw State University, including business and engineering faculty, conducted a multiyear study in coordination with logistics practitioners and the Georgia Department of Economic Development’s Center of Innovation for Logistics. The team focused on understanding how to increase the number of students entering the field and how to retain the existing Georgia workforce. The study collected data from over 400 logistics practitioners and over 1,900 college students and 250 faculty at 17 Georgia universities and technical colleges. Outputs include a series of academic publications, white papers, and presentations helping (1) students to understand the strengths of logistics jobs; (2) the logistics industry to better market its strengths to students; (3) individual logistics companies to become employers of choice; and (4) faculty to better connect with students and their career expectations. Overview
16 | Applied Research
Executive Takeaways
■ College graduates seek intrinsic skills that lead to extrinsic outcomes and stability. ■ Logistics offers intrinsic skills and social and altruistic work outcomes. ■ Logistics employers must improve opportunities for advancement and career stability. ■ Supervisors and mentors are crucial to retaining logistics professionals, especially women. ■ Logistics organizations can customize their retention programs to become employers of choice.
Stacy Campbell, Professor of Management Michael Maloni, Professor of Management
Optimal standard of human rights in countering an insurgency
Aniruddha Bagchi and Jomon A. Paul
Coles Research Symposium on Homeland Security Special Issue, SIFALL18-05, October 2018
We explore why human rights violations take place during a rebellion. Authoritarian governments may not care about human rights, but surprisingly, several democratic governments have also condoned violations. We show that the primary cause is faulty intelligence; we consider both types, missed alarms and false alarms. We also identify incentives that drive government armed forces to violate human rights as well as the optimal human rights standard in such circumstances. We then examine the effect of lowering the human rights standard on the probability of quelling the rebellion. In our theoretical model, this effect is indeterminate; that is, it can be positive or negative. We quantify it empirically using the case of the Armed Forces Special Powers Act in India. Since the probability of quelling a rebellion is not directly observable, we use the magnitude of violence as its indicator and posit that the magnitude of violence should be negatively related to the probability of government success. We find that lowering the human rights standard increases violence and reduces the chance of quelling the rebellion, and this effect is statistically significant. Overview
18 | Coles Research Symposium
Executive Takeaways
■ We study why countries faced with insurgencies violate human rights. ■ We find that imperfect intelligence is a primary cause of such violations. ■ We also identify the conditions under which a tough policy does not work. ■ Empirical evidence from India suggests that a tough policy worsens the insurgencies.
Aniruddha Bagchi, Professor of Economics Jomon A. Paul, Professor of Quantitative Analysis
Socially optimal IT investment for cybersecurity
Jomon A. Paul and Xinfang (Jocelyn) Wang
Coles Research Symposium on Homeland Security Special Issue, SIFALL18-01, October 2018
This paper uses the concept of social cost, comprised of private and externality costs, to examine the optimal balance among prevention, detection, and containment measures to safeguard against cybersecurity breaches under three sources of uncertainty. We propose a robust optimization model in tandem with distribution-free ellipsoidal uncertainty sets to ease the challenge of providing exact estimates for uncertain input. Validated on a case study, results from 25 deterministic scenarios reveal, first, a strong preference for allocating resources toward prevention. When budget constraints are relaxed, this preference shifts to containment and detection safeguards. Results from 54 robust test instances indicate that, among the three sources of uncertainty, adjusted effectiveness of prevention safeguards has the greatest impact on both the social cost and optimal balance of the safeguards. Our analysis points to some serious flaws in the existing cybersecurity framework, which mainly relies on prevention, and provides urgently needed guidelines on cybersecurity to decisionmakers. Overview
20 | Coles Research Symposium
Executive Takeaways
■ Examine the optimal balance between prevention and detection & containment safeguards. ■ Resource allocation preference shifts from prevention to detection and containment with relaxing budget constraints. ■ Among all the uncertainties tested, the effectiveness of the prevention safeguard had the greatest impact. ■ The study quantified the flaws in current practice, which relies on prevention to the exclusion of the other safeguards.
Jomon A. Paul, Professor of Quantitative Analysis
Conflict and aid dependency – An explorative study motivated by the case of Palestine
Marcus Marktanner and Almuth D. Merkel
Coles Research Symposium on Homeland Security Special Issue, SIFALL18-07, October 2018
Confronted by external and internal conflict, foreign aid has become Palestine’s lifeline. However, the interaction between aid and conflict is unclear from two perspectives. First, prevalent aid-dependency concepts ignore the dynamic effects of accumulation. Second, while aid is given under the assumption that both donors and recipients want to build peace, shirking for the purpose of generating income from aid cannot be ruled out. Motivated by Palestine’s case, we explore cumulative aid dependency as a function of cumulative conflict from a cross-sectional perspective. Our results suggest both long- and short-term relationships between aid dependency and conflict. We find that aid dependency and conflict cause each other, in Granger’s sense, and conclude that aid may be less effective in reducing conflict than conflict is in securing income. Overview
22 | Coles Research Symposium
Executive Takeaways
■ NGOs’ delivering aid may not reduce conflict. ■ Massive aid inflows often mask the strength of an economy. ■ Aid reduction might reduce conflict.
Marcus Marktanner, Professor of Economics Almuth Merkel, Student, School of Conflict Management, Peacebuilding and Development
Natural disasters and their impact on cooperation against a common enemy
Timothy Mathews and Jomon A. Paul
Coles Research Symposium on Homeland Security Special Issue, SIFALL18-02, October 2018
Overview We develop a simple game theoretic model to study the impact of a natural disaster on the coordination of defensive efforts by a target state (G) and an ally (A) in relation to the choice to stage an attack by a terrorist (T) , to examine how the realization of a natural disaster can impact strategic choices in such a setting. We focus on “long term impacts” in which a natural disaster increases costs of defensive efforts for G . When costs for G are higher in this way, the optimal choice by A to provide assistance or not and the ultimate optimal choice by T to attack or not could be altered. For all possible parameter values, a unique Subgame Perfect Nash Equilibrium for the specified sequential move game is identified. A comparative statics analysis is conducted to determine how the natural disaster could alter the equilibrium outcomes. We show that for a natural disaster of sufficiently small magnitude, we could realize counter-intuitive outcomes: (i) A not providing assistance which otherwise would have been provided and (ii) T not staging an attack which otherwise would have been staged. That is, after a natural disaster occurs, it could be that (i) the ally of a potential terrorist attack is less likely to provide defensive support and (ii) a terrorist attack is less likely to be launched.
24 | Coles Research Symposium
Executive Takeaways
■ Our game-theoretic model assesses the long-term impacts of a natural disaster on allied nations’ defense coordination against terrorism. ■ An ally might withhold assistance from a targeted nation suffering a small-scale natural disaster. ■ Similarly, terrorists might hold back on attacks during such events.
Jomon A. Paul, Professor of Quantitative Analysis Timothy Mathews, Professor of Economics
Competitive awareness of emerging and developed economy multinationals
Canan C. Mutlu
Overview How does the competitive awareness of multinationals from developed economies differ from that of multinationals from emerging economies? A firm’s strategy depends on how accurately it perceives its competition. We adopt the classical Johari window to firm-level interactions, categorizing the sources of competitive awareness into four quadrants based on the information each type of firm either knows or does not know. Each quadrant represents different challenges. Leveraging the awareness-motivation-capability perspective, we elucidate awareness in the competition between emerging and developed economies’ multinationals.
26 | Summer Research Fellowship
Executive Takeaways
■ Multinational corporations in developed and emerging economies have different levels of competitive awareness. ■ Competitive awareness is critical when data are not fully available, and cause-effect links are ambiguous. ■ Multinationals in developed economies know more about internal assets and market uncertainty. ■ Multinationals in emerging economies know more about nonmarket dynamics and institutional uncertainty.
Canan C. Mutlu, Assistant Professor of Strategic Management
Do foreign lenders face different enforcement in the U.S.?
Sunay Mutlu
Overview Prior studies have shown that legal enforcement informs the features of debt contracts. We examine whether, in the United States, the debt contractual terms of foreign lenders differ from those of domestic lenders. We argue that foreign- led debt contracts include looser covenants, lower likelihood of performance pricing provisions, and fewer negative covenants. We control for variation in legal environments by identifying debt contracts governed by the contract law of the state of New York. We attribute the differences in enforcement within this setting to the liability of foreignness faced by foreign lenders, rather than legal differences that lead to heterogeneous contract enforcement across the United States.
28 | Summer Research Fellowship
Executive Takeaways
■ We provide evidence that foreign lenders face different debt contractual features in the United States. ■ We show that foreign lenders face weaker enforcement in their contracts. ■ The results partly explain the increasing trend of foreign borrowing by US firms.
Sunay Mutlu, Assistant Professor of Accounting
Decision-support model for cybersecurity risk planning: a two-stage stochastic programming framework featuring firms, government, and attacker
Jomon A. Paul and Minjiao Zhang
Coles Working Paper Series, SPRING19-02, March 2019
Overview We study the decisionmaking problem in cybersecurity risk planning concerning resource allocation strategies by the government towards intelligence and firms’ investments in detection and containment safeguards. Aiming to minimize the social costs incurred due to cyberattacks, we consider both the initial monetary expenditure and the deprivation costs due to delayed detection and containment. We also consider the effect of positive externalities of the overall cybersecurity investment on an individual firm’s resource allocation attitude. The optimal decision derived will guide firms in determining their countermeasure portfolio mix (detection vs. prevention vs. containment) and government intelligence investments while accounting for budgetary limitations and the possible actions of a strategic attacker. In our two-stage stochastic programming model, first, firms decide on prevention and detection investments, including investments in government intelligence that improve detection effectiveness. In the second stage, once attacker actions are realized, firms evaluate their impact and decide on containment investments. We use a case study to demonstrate the applicability of our model. We find that externality can reduce government’s intelligence investment and that the firms’ detection investment receives priority over containment. We also note that while prevention effectiveness has a decreasing impact on intelligence, intelligence effectiveness has an increasing impact on intelligence.
30 | Working Papers
■ We focused on government and firm resource allocation strategies involved in cybersecurity risk planning. ■ Firms focus on prevention, detection, and containment safeguards, while government focuses on intelligence. ■ The countermeasure portfolio accounts for a strategic attacker and firm budgetary constraints. ■ Externality may reduce government intelligence. ■ Firms give preference to detection investment over containment safeguards. Executive Takeaways
Minjiao Zhang, Associate Professor of Quantitative Analysis Jomon A. Paul, Professor of Quantitative Analysis
The dark side of crowdfunding platforms
Susan Young and Birton Cowden
Coles Working Paper Series, SPRING19-03, March 2019
Overview Crowdfunding platforms, such as Kickstarter, provide a way for new entrepreneurs to fund their projects. By 2025, such platforms are predicted to account for over $300 billion in cumulative transactions. Researchers and practitioners alike generally find these platforms to be beneficial for new ventures. However, anyone with an internet connection can view these early projects, particularly third parties in countries with different laws and different levels of regulatory enforcement, which can result in idea stealing. We argue that the threat of opportunistic copycats means the startup must rethink its financing, marketing, and especially its research and design practices. We provide an in-depth case of a U.S. startup that used Kickstarter to raise nearly $1.9 million in 2018 for a reusable drinking straw. The venture, FinalStraw, demonstrates the downside of a successful campaign on an open platform where economic actors in other institutional environments can capitalize on an idea with few consequences. Using institutional economics theory, we examine the costs copycats can impose on the venture, despite its apparent success.
32 | Working Papers
■ Startups should be more aware of copycats’ access to crowdfunding sites. ■ Startups using crowdfunding sites should consider barriers to copycats, such as patent protection. ■ Ventures with ideas that can be easily imitated may be better off using traditional financing routes. ■ Startups using crowdfunding should have their logistics ready to go as soon as the campaign ends. Executive Takeaways
Birton Cowden, Assistant Professor of Management Susan L. Young, Assistant Professor of Management & Entrepreneurship
Special thanks to the following faculty and committees for their significant contributions to the Coles Research Magazine .
Coles Working Paper Series
Editor: Jomon Paul
Editorial Board: Aniruddha Bagchi ■ Canan Mutlu ■ Humayun Zafar ■ James Meurs Jennifer Hutchins ■ Jesse Schwartz ■ Leo MacDonald ■ Marcus Caylor Maria Kalamas ■ Mark Hiatt ■ Rongbing Huang ■ Stacy Campbell Sweta Sneha ■ Timothy Mathews ■ Vineeta Sharma ■ Xuepeng Liu
Research and Development Committee Canan Mutlu ■ Jennifer Hutchins ■ Justin Cochran Leo MacDonald ■ Mona Siha ■ Muhammad Obeidat Saurabh Gupta ■ Sunay Mutlu ■ Vineeta Sharma ■ Xuepeng Liu
Summer Research Fellowship Committee Jomon Paul ■ Marcus Caylor Maria Kalamas ■ Michael Maloni ■ Saurabh Gupta
PhD Program Academic Director: Brian Rutherford
Aspire to More ™ | ColesCollege.com
Page 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16 Page 17 Page 18 Page 19 Page 20 Page 21 Page 22 Page 23 Page 24 Page 25 Page 26 Page 27 Page 28 Page 29 Page 30 Page 31 Page 32 Page 33 Page 34 Page 35 Page 36Made with FlippingBook - Online catalogs