Economic subjects | Business economics » Does Product Market Competition or Firm Ownership Have Anything to do with the Payout Policy of Firms

Source: http://www.doksinet 1 SUMBISSION OF POSTDOCTORAL RESEARCH (FINAL DOCUMENT) “DOES PRODUCT MARKET COMPETITION OR FIRM OWNERSHIP HAVE ANYTHING TO DO WITH THE PAYOUT POLICY OF FIRMS?” Submitted to: Professor Kristine Hankins Submitted by: Shahid Ali INSTITUTE OF MANAGEMENT SCIENCES UK POSTDOC SCHOLAR 2015 UNDER PROGRAM CONSORTIUM OF FIVE KHYBER PUKHTUNKHWA UNIVERSITIRES WITH UNIVERSITY OF KENTUCKY SUBMISSION DATE: JUNE 17, 2016 Source: http://www.doksinet 2 Does product market competition or firm ownership have anything to do with the payout policy of firms? Shahid Ali Abstract Competition in product markets is posing fresh implications on payouts as researchers attempt to understand the theoretical relationships between the intensity of market competition and its role towards increasing or decreasing agency problems between owners and managers. Competition dynamically changes the market structure and firms have to strategize swiftly to remain in business. The

complex underpinnings of agency cost hypothesis and tax-based hypothesis burdened with assumptions of asymmetry demand investigation from the perspective of emerging markets. Studying the relationship between ownership structure and payouts should be intriguing as less developed markets can add more insights to theoretical finance in this area. This study is going to work both towards testing the effects of product market competition and ownership structures on payouts. The paper relies on a panel data of nineteen listed manufacturing industries of Pakistan from 2001 to 2015. Sufficient controls are exercised in Tobit and OLS models. There is evidence that agency cost hypothesis holds in the Pakistani context as with declining number of institutional investors overtime, industries seem to pay more dividends. No evidence is found for tax-based hypothesis More levered firms are paying less dividends whereas firms with high retained earnings are positively related to firms that pay more

dividends. Family ownership is evidenced as negatively related to paying dividends Product market competition is also found to have a negative relationship with payouts. There is evidence that when firms have lower representation of institutional owners on boards they pay more dividends. In the context of fifteen years of analysis it seems that for Pakistani listed manufacturing firms are paying lesser dividends over time. Source: http://www.doksinet 3 I. Introduction The payout of companies is a hot debate for many academics around the world. The countless dimensions of dividends are studied both in developed and developing markets and there is abundant information on the determining factors of dividends and the implying factors that shape the dividend policy of firms. This work is specifically aiming to explore and investigate the relationship of product market competition and payout policy of firms. It will also try to investigate that whether institutional or insider ownership

firms has anything to do with the payout of firms. This relates to looking in detail at the corporate governance implications on payout policy. The Asian crisis of 1997-99 is deemed responsible for the devastation of some fast growing economies of Asia which resultantly triggered a debate towards ensuring prudential regulations, building a wider and deeper understanding of product market competition, and thinking towards development of stricter codes of corporate governance. Over-investment fueled the Asian crisis due to disregard for corporate competition. This pertains to comparing the characteristics of markets in mature and developing economies. Emerging economies have industry structures where leading corporate organizations are family-owned. Reluctance to relinquish control and the degree of information asymmetry in emerging markets have interesting implications on the payout policies of firm which need to be studied. Product market discipline, capital markets liberalization, and

economic reforms by many emerging economies led to the growth of private sector as major public sector business were privatized. The interventions of stricter corporate governance rules as a result in emerging economies is a ripe area for research if one wishes to see its effects on the payout policy of firms. Asymmetric information under such assumptions should increase business costs due to severe contractual arrangements directing managers to serve best economic interests of the owners. Agency cost hypothesis predicts Source: http://www.doksinet 4 contractual costs would be less when managers become owners. What effects can we monitor on the payout policy in such scenarios should be interesting? The agency cost theory infers that payout policy should have an inverse relation with more institutional owners on boards while the tax-based hypothesis asserts otherwise. There should be more evidence that which one holds better in mature or emerging markets. The degree and intensity of

industry competition results in residual income that firms may want to distribute or retain keeping in view their business requirements. Literature shows that big mature firms are more likely to pay dividends whereas new and growing firms may take a while before they start sharing profits. This work builds a case for investigating the relationships of product market competition with dividend payouts and also works towards determining some important factors for firms that pay dividends. Governance mechanisms demand catering interests of all stakeholders and especially those of firm shareholders which include minority owners. Agency cost implications are different once insiders and institutions sit on boards. This paper is going to work towards finding evidence to such interesting and compelling questions by relying on data from Pakistani market. The paper relies on standard econometric methods and analyses data gathered from reliable governmental sources. Section 2 discusses the

relevant literature and lists the hypotheses, Section 3 outlines the methodology, Section 4 presents the data analysis and discussions, and finally Section 5 showcases the conclusions. II. Literature Review Literature on payouts is vast as researchers around the world have investigated firm-specific, industry-specific, and country-specific factors and components that may be responsible for shaping the dividend policy. The decision whether to distribute or retain the residual income to shareholders has been investigated from multiple dimensions. The dividend irrelevance theorem Source: http://www.doksinet 5 by (Miller & Modigliani, 1961) set the pace for a plethora of discussion and information on the relative importance of payouts. Generating a discussion on signaling hypothesis, this theorem asserts that managers may use dividends to show better economic performance compared to their competition currently and can also send signals of better anticipated future performance. This

argument leads to implications of information asymmetry on competitive players in an industry. Theorists relaxed the assumptions of MM dividend irrelevance theorem and showed that taxes do matter in opting for debt or in paying dividends (Miller M. H, 1986) Some authors supported relevance in the context of information asymmetry (Mayers, 1984: Ross, 1977) while others referred to the agency relevance of dividends (Jensen & Meckling, 1976) which presents that agents or managers can expropriate funds from debt-holders by issuing additional debts and distribute the cash proceeds as dividends or can invest in high risk business projects. Disclosure mechanism can reduce this kind of agency risk but dividends do take effect and affect the payout distributions. Corporate disclosures and their regulation makes a key difference in the level of trust by investors in emerging and mature economies. The economic rationale that may justify corporate disclosures is that investors have access to

private information of managers. Disclosure regulation aims to mitigate information asymmetry and therefore companies in US are required to comply to disclosure rules which are set by Securities and Exchange Commission. (Healy & Palepu, 2001) take of good stock of disclosure implications on investors, capital markets, and disclosure regulation. They theoretically attribute market imperfections and concerns other than market failures as the main reasons for building an economic rationale for corporate disclosures. They leave open many other researchable questions in corporate disclosures but conclude that regulated financial information does provide handy valuable information to investors but may not be superior to a free market disclosure approach. (Patal, Balic, & Bwakira, 2002) report after Source: http://www.doksinet 6 quantifying that Asian emerging markets have and ensure higher transparency and disclosure compared to Latin American, Middle Eastern, and Eastern

European emerging markets. The agency cost hypothesis infers that corporate managers may not serve the economic interests of shareholders and therefore if they distribute profits they will have little to play with. The MM dividend irrelevance theorem has recent critics and the notion that only investment policy determines value is discarded (DeAngelo & DeAngelo, The irrelevance of the MM dividend irrelevance theorem, 2006). (Glen, Karmokolias, Miller, & Shah, 1995) believe that the arguments given in the context of information asymmetry and agency cost hypothesis may have different implications for emerging markets as they represent concentrated ownership structures and have closely held corporations. (Miller & Rock, 1985) are of the view that if trading of shares is added to the standard finance model for investment/dividend/financing decisions it could restore the time consistency in investment policy. Building an argument for nonavailability of a full-information model

they infer that dividend announcement effects imply information asymmetries which may be tackled by keeping the assumption of information asymmetry and possibility of trading shares and then to seek consistent alternative investment decision rules. United States is one of the leading capital markets of the world. A vast literature covers the payouts topic for the US alone. After analyzing corporate firms in nearly two decades (Fama & French, 2001) assert that corporate landscape of America is leaving the conventional cash dividends. They conclude on the basis of firms studied between the period 1978 and 1999 that the primary reason for this change was the huge number of newly listed firms in 80’s and 90’s along with changes in other characteristics including book-to-market ratio, growth in assets, size, and profitability. But even controlling for these characteristics Fama and French report dividends Source: http://www.doksinet 7 are still declining in a phenomena called

“disappearing dividends”. (Baker & Wurgler, 2004) provide a behavioral explanation for this trend and call them temporary fads for dividends. When fascination is high for dividends the valuation of dividend payer increases and when this fascination drops the dividend payer valuation decline. This phenomenon is termed the Catering Hypothesis as such behavioral fads are suggested the first order determinants of fading dividends. This finding is further examined by (Hoberg & Prabhala, 2009) in the risk context of the firm and risk is found to be a significant determinant in roughly 40% of the disappearing dividends. The study further reports that Catering Hypothesis does not hold once risk is controlled and fads have no explanatory power. Another finding of this study is that firms are more likely to distribute dividends if they have higher profitability, low market-to-book ratio and asset growth. Contrasting this finding (Julio & Ikenberry, 2004) report that dividends are

reappearing and the notion set earlier that the propensity of US firms to pay cash dividends has declined is fading out. This view challenges the recently believed fundamental policy change for US firms that stock repurchases are replacing cash dividends is on trial again. Supporting his case of rebound in dividends the authors investigate five reasons for reappearing dividends including the (i) Maturity Hypothesis (ii) dividends as a signal of confidence (iii) a response to tax reforms (iv) Catering Hypothesis of (Baker & Wurgler, 2004) (v) short of ideas on NPV projects. The authors conclude in this work that dividends are reappearing in corporate America not because of fiscal adjustments and tax relief on dividends. Pointing to the marginal tax-cut by Bush Administration in May 2003 the authors discard the view that dividends are rebounding as a result. Their work finds support for the argument that firms that need to signal confidence use dividends to convey their message in

order to have more access to debt markets or to get better capital servicing commitments. They do not find compelling evidence for the argument that Source: http://www.doksinet 8 dividend payments are for catering investors’ whims as investors have an increased appetite for dividends. The work strongly supports the Maturity Hypothesis for the reappearance of dividends in corporate America. It carefully discards the notion that the shift may be due to temporal shifts in investment options. Their evidence suggests that asset growth, dividends, and cash takeovers are all increasing and the results are not consistent with a residual investment model of dividends. The shift in underlying economy is an additional factor on which authors could not find conclusive evidence. (Grullon & Michaely, Dividends, Share Repurchases and the Substitution Hypothesis, 2002) also assert that dividends and repurchases in terms of total cash payouts to shareholders have not declined. (Lintner, 1956)

report conservatism in setting of dividend policy by managers and in a recent survey research by (Brav A., Graham, Harvey, & Michaely, 2005) strangely this conservatism of managers still holds. (Allen & Michaely, 2003) while reviewing literature on payout policy consider taxes, information asymmetry, transaction costs, and incomplete contracting possibilities and report on the basis of accumulated evidence that payouts are motivated by reducing overinvestment by management. They further report that over 90% of dividends are paid by those firms that already pay dividends and the popularity of repurchases has actually increased the financial flexibility of firms. (DeAngelo, DeAngelo, & Skinner, Are dividends disappearing? Dividend concentration and the consolidation of earnings, 2004) believe that though the median firm shifted its payout policy away from paying cash dividends during 1990s yet the aggregate dollar dividends increased. There is higher concentration of paying

dividends among fewer dividend paying firms. The increase in cash holdings is strongly correlated to the fading dividends and new listings phenomena as documented by Fama & French (2001, 2004). (Fama and French, 2001) studied the patterns and the determinants of dividends over the period 1926-1999 for publicly traded US firms. The data Source: http://www.doksinet 9 covered Amex, NYSE and Nasdaq companies. The sample showed a large increase in the number of newly listed companies’ post 1978. They report that the decrease in the dividend payment propensity is attributed to an increase in the number of newly listed companies. In the authors’ view, the decline in the propensity of dividends for new listed companies implies that the companies have become aware of the tax disadvantage of dividends. Their findings further showed that firms that pay dividends are highly profitable and mature while fast-growing and young firms never paid dividends. Mature firms with high level of

profits have retained earnings that are adequate to meet their capital expenditures. (DeAngelo, DeAngelo, and Stulz, 2006) suggest that newly listed and young firms prefer not to pay dividends. In young firms, retention of cash dominates allocation as they face plentiful investment opportunities with inadequate resources. On the other hand, mature firms with high profitability and less attractive investments opportunities are considered better candidates to pay dividends. (Denis and Osobov, 2008) highlight similar findings in their study. The sample covered all companies in Japan, US, Canada, UK, France and Germany using Worldscope database over the period 1989–2002. For all the six countries, the key determinants for dividend policy are profitability, firm size, earned/contributed equity mix and growth opportunities. The findings showed high dividend payments for large and profitable firms. In another empirical study, (Bates, Kahle, & Stulz, 2009) proposed that firms with recent

listing cohorts generally hold large amount of cash. They added that more cash is held by an IPO firm and the cash ratio drops as the IPO gets more distant. The precautionary demand for cash theory provides a plausible explanation for the increase in cash holding for non-dividend payers. According to this theory, excess cash holdings act as a buffer for firms to protect them against undesirable cash flow shocks. It is also reported that established and older firms that pay dividends do not increase their cash holdings; however, Source: http://www.doksinet 10 the increase in cash holdings is much more considerable for recently listed and smaller firms. This discussion leads to look more deeply into questions like do large mature firms pay more dividends compared to their newly listed counterparts? Information asymmetry and agency problems determine the differences that exist in the payout policy of public firms and private firms. (Brav, Graham, Harvey & Michaely, 2005) provided

a distinctive viewpoint on corporate repurchase and dividend policies at the beginning of 21st century in US. The sample of their survey consists of responses from 384 financial executives, which include 128 private firms and 256 public companies. The findings showed a strong willingness for managers of private firms to cut down dividends. Transmission of information to stakeholders, monitoring of corporate managers and preventing them from excess spending is much easier in privately held firms. These firms view dividends to hold less information, therefore, the negative consequences of cutting dividends are less severe for privately-held firms than for public firms. (DeAngelo, DeAngelo, & Skinner, 2008) document that as private firms little exhibit; information asymmetry, therefore, these firms should cut or omit dividends more often than publicly traded companies. Similar view is presented in contracting model of dividends. According to this model, private firms with tight

ownership alliances have minimal agency problems. Therefore, the dividend payouts are unnecessary in these firms (Megginson & Smart, 2008 p.448) (Michaely & Robert, 2012) report that public firms with stronger corporate governance pay higher dividends compared to private firms having weaker corporate governance. They analyzed the difference in dividend behavior of private firms with that of public firms in United Kingdom. The study suggested that the differences in ownership structure and corporate governance mechanisms shape the dividend policy of public and private firms. Their findings showed that in comparison to privately dispersed and wholly owned firms, public Source: http://www.doksinet 11 firms distribute high percentage of their profits as dividends. If reliable data on private firms is available on emerging markets testing the hypotheses that ‘Private firms pay less dividends compared to publicly listed firms’ will be insightful. 2.1 Product market

competition and payouts Relating product market competition to payout policy is a relatively a new dimension. The degree of competition in a product market derives the market share in terms of business profitability. When competition is intense profitability may have an even distribution among competing players leaving low free cash-flows to disgorge. Firms compete for the size of market share on the basis of products, quality, price, and innovation. Business markets are volatile and instable as they change quickly. Payouts depend on the current business performance but they also depend on how markets or competition shapes-up in future. Firms may be compelled to halt payouts and prepare for competing the next round due to threats posed by competing firms who change their products swiftly. This happens in industries which are exposed to the shocks of technological advancement. We can take the example of Eastman Kodak company which was once a giant corporation serving the photographic

interests of its global customers. Failing to realize the changing dynamics of product markets it fell miserably and had to file for bankruptcy. (Hoberg, Phillips, & Prabhala, Product Market Threats, Payouts, and Financial Flexibility, 2014) are of the view that cash holdings or payout policy are influenced by product market threats. They have developed a new measure of the competitive threats in product market which are faced by firms from their competition. This fluidity is simply defined to be the difference in rival firms’ products relative to the firm. Their findings include that firms with higher product market fluidity are less likely to pay dividends. When product market fluidity is high and access to capital markets is low, such firms are more likely to keep cash rather than pay dividends. This Source: http://www.doksinet 12 work shows that the financial policies of firms depend on managing their product market threats. (Gasper & Massa, 2006) give and find

support for their hypothesis that the uncertainty of average profits increase for firms with increasing competition. They used the Lerner Index (Lerner, 1934) that is price-cost margin adjusted for industry to measure the market power of firms within industry. The market power between industries is measured by HerfindahlHirschman Index (Herfindahl, 1950: Hirschman, 1945) The uncertainty in average profits of firms may have implications on payouts. (Grullon & Michaely, Coporate Payout Policy and Product Market Competition, March 2007) assert the notion that product market competition may be a reason for firms to pay higher or lower dividends. They report that firms operating in more competitive industries are more likely to pay higher dividends compared to their counterparts in concentrated industries. This expectation needs a review as firms which are in tough competition and which are heading into more intense forms of competition will need cash to compete and sustain competition.

Market competition is believed to induce firms to payout dividends and the notion supplements the theoretical rationale that market competition offers incentives to efficient managers. Contemporary literature in economics seems to have addressed this debate further as there are numerous authors who believe that competition in markets is a more important governance mechanism than corporate control or institutional monitoring. This discussion has led to hypotheses including a ‘threat of liquidation hypothesis’ and a ‘yardstick competition hypothesis’. The ‘threat of liquidation hypothesis asserts that managers avoid negative NPV projects for their firms as they can potentially drive them out of the markets (Aghion, Dewatripont, & Rey, 1999). The ‘yardstick competition hypothesis’ argues that the monitoring and information asymmetry costs go down as product market competition offers outsiders opportunities to compare the performance of a firm with its competitors

(Shleifer, Source: http://www.doksinet 13 1985). Under both above stated hypotheses competition should serve as a governing mechanism reducing information asymmetry and business risk for firms. This discussion demands further investigation especially in emerging markets where the nature of competition, the number of market players, market regulation, rule of law, and market forces are going to be unique. (Chhaochharia, Grinstein, Grullon, & Michaely, 2009) report in their work that product market competition is a better governance mechanism for firms in competitive industries and is a substitute for corporate governance. Firms that do not compete actively are found to be inefficient. They assert that competition between firms reduces agency conflicts between owners and managers. For less competitive industries which are less efficient they report that the regulation titled Sarbanes Oxley improved efficiency. (He, 2012) asserts that companies working under weak legal regimes are

less likely to pay dividends and so is the case when companies suffer from the consequences of weak corporate governance. His research argues that under such conditions market mechanism serves as a governing arrangement that can compel companies to pay dividends to minority shareholders. Pakistani business is dominated by families and there is a visible presence of family groups and isolated businesses which are owned by them. Family owned publicly listed businesses may treat minority shareholders differently. This is evidenced by (Ghani & Ashraf, 2005) in their study about performance of firms operating in business groups and non-groups. They report that though the business performance of business groups is better than firms not operating in groups but investors view the group-mechanism as an expropriation mechanism to take advantage of minority shareholders. This finding implies that firms in groups may be able to pay dividends but may not choose to do so as the regulating or

governing mechanism is easy to be exploited. The reported literature prompts many interesting questions like when competition is high, do firms pay low or no dividends? Is product market Source: http://www.doksinet 14 competition really a substitute of corporate governance? Do firms in concentrated industries pay more dividends or does family controlled businesses really exploit minority shareholders and expropriate earnings by not paying any dividends? This paper is going to look in some of such hypothesis. 2.2 Corporate Governance and Payouts Corporate boards are considered to be the custodians of the interests of all stakeholders however the interests of the minority shareholders are often neglected by companies in developing markets. Institutional regimes usually work better toward ensuring their stakes They are also found responsible in discouraging firms paying higher dividends (Grinstein & Michaely, 2005). There are disagreements in literature over the debate that does

corporate governance affects payout policy? (La Porta, Lopez-de-Slanes, Shleifer, & Vishny, 2000) are of the view that firms in countries that ensure strong rights of minority shareholders pay higher dividends. They highlighted an “outcome model” and a “substitute model” as incentives to pay higher dividends by firms. The “outcome model” theorizes that dividends are paid by managers as minority shareholders exert pressure on them. In contrast the “substitute model” presents a view that managers pay dividends in future only to appraise minority shareholders and also aim at building a decent reputation of fair treatment of any such shareholders. Dividend policy should be designed to minimize the sum of agency, capital and taxation costs (Easterbrook, 1984). These costs are influenced by the percentages of shares owned by institutional investors. The degree of institutional ownership has an impact on firms’ agency costs and consequently on its dividend policy. A

large number of studies argued that monitoring of corporate managers by institutional investors help to resolve the agency conflicts e.g, (Shleifer & Vishny, 1986; Jarrell & Poulsen, 1987; Brickley, Lease, & Smith, 1988; Graves &Waddock,1990). (Han, Lee, & Suk, 1999) Source: http://www.doksinet 15 postulates that due to their monitoring capabilities, firms with high degree of institutional ownership would pay low dividends as they are relatively less concerned about the agency costs. An inverse relationship between institutional ownership and dividend payout is predicted under the agency-cost hypothesis. Institutional investors benefit from economies of scale in gathering and analysis of information. Managerial entrenchment would be more difficult in the presence of these investors (McConnell and Servaes, 1990). Therefore, corporate boards having institutional owners are less likely to use dividends as a means of reducing agency costs. (Giroud & Mueller,

2011) examine whether good governance matter more for firms in non-competitive industries. (Mitton, 2004) shows similar findings that firms who are better governed are more profitable and pay higher dividends whereas weak governance results in absence of dividend payments. They further elaborate several causes for firms in non-competitive industries that are due to weak governance including inefficiencies including minimal labor productivity, more input costs, and value-degrading acquisitions. Their work concludes that firms in competitive settings benefit more from good governance than those operating in non-competitive settings. (Michaely & Robert, 2012) report that public firms with stronger corporate governance pay higher dividends compared to private firms having weaker corporate governance. Their further findings include that publicly held firms distribute the highest % of profits as dividends whereas closely held or wholly owned firms pay a lesser percentage. Their

discussion that wholly owned firms are more sensitive to investment needs looks handy from the asymmetric information and corporate behavior point of views. (Januszewski, Koke, & Winter, 2002) in their study about testing the effects of corporate governance and product market competition on productivity growth conclude that competition in product market is positively associated with growth in productivity. They also find strong evidence for higher productivity that is due to tight control of ownership Source: http://www.doksinet 16 but they insert that the statement may not be true when the ownership is with a financial institution. (Allen & Michaely, 2003) took a clear stance contrasting the widely held stance of (Miller & Modigliani, 1961) that dividends are irrelevant for firm value. Survey evidence supporting this view is also reported by (Brav, Graham, Harvey, & Michaely, 2005). Dividends are also considered as substitutes for good corporate governance as firms

use dividends for reducing agency costs that may be incurred due to poor corporate governance (John, Kose, & Knyazeva, 2006). This view is further supplemented by some evidence that investors positively react when dividends increase while their reaction is negative when dividends decrease (Benartzi, Shlomo, Michaely, & Thaler, 1997). In emerging markets, dividend payout depends mostly on the availability of external finances. While in Pakistan, it is highly correlated with the type of governance. Ownership structure has a major impact on the dividend payout policy. Most firms in Pakistan have one key owner holding major percentage of shares in the affiliated firms. This creates an agency problem between the majority and the minority shareholders, thus, influencing the firms’ dividend policy. Firms are often family owned and liable to ignore the interests of minority shareholders with payouts (Mehar, 2005). This leads us to state hypothesis like do firms pay more dividends

when there are more institutions on boards? Or do firms pay more dividends when there is more presence of managers as owners on boards? These are open questions to be empirically examined as testing the relationship between institutional ownership and dividend payout policy may serve multiple academic and policy purposes. Source: http://www.doksinet 17 (Ahmed & Javid, 2009) report that dividend policy of Pakistani firms depends on their current profits and not much on the past dividend as the sensitivity of payments is more towards current earnings. (Afza & Mirza, 2010) are of the view that managerial ownership, size, and cash flows sensitivity negatively influence dividend payments whereas profitability and operating cashflows positively influence dividends in Pakistan. Similar findings are reported by (Imran, 2011) There is evidence that ownership concentration plays a key role in dividend payments and growing companies concentrate more on retaining income for business

needs than paying dividends (Mehar, 2005). This finding leads us to think that when more companies are incorporated in a sector then the average dividend payout of the sector should drop. (Rafique, 2012) report that corporate tax and firm size are significantly related to dividend payouts in Pakistan. Since dividends are taxed more than capital gains the tax preference theory may seem to hold. The tax environment in Pakistan is different if compared to developed markets The differences in tax system influence the dividend policy and also affect the magnitude of dividend smoothing. The payment of dividend is voluntary and major investors oppose paying dividends and consider appreciation in share prices as a major constituent of stock returns. The attitude of investors towards dividends is expected to have an impact on the way dividend policy is set by firms in Pakistan (Ahmed & Javid, 2009). The preferences of small investors towards dividends is adversely affected due to the

corporate taxation policy. Dividends are subject to double taxation as dividends are taxed at the rate of 10% and companies are taxed at the rate of 35% on their income. In comparison to dividends, capital gains are more attractive for investors as they are exempt from taxes (Afza & Mirza, 2010). In Pakistan, Modaraba companies or Islamic mutual funds came into existence after Modaraba Companies Ordinance 1980. These companies were tax-exempt only if they paid 90% or more of Source: http://www.doksinet 18 their profits as dividends. Post 1995 they started paying taxes like other companies due to regulatory changes and resultantly their dividend payments declined too. Another important dimension of this work considers the subsidiaries of multinational companies who are closely concentrated and since by law dividends can be paid to foreigners, the local subsidiaries pay higher dividends. But this may require further investigation in the indigenous context This present work is

looking for any evidence regarding the relationship of market competition and corporate governance with payout policies for firms in Pakistan. Government of Pakistan is a major stock holder in the capital markets with over more than half of the stock ownership. Dividend receipt is one major source of revenue for it. Government owned companies have a history of retaining their earnings resulting in revisions of dividend receipts by the State. Government of Pakistan tried to impose regulation in the past to fix the issue but had to step back due to severe criticism by market circles and threats of some major public companies that they will buy-back minority shareholders’ rights and get private. III. Methodology This section unfolds the methodological approach for the questions at hand. Pakistani financial databases are limited when it comes to finding information about variables of interest. Governmental sources have erroneous entries that may distort findings. For this research all

available data from listed Pakistani manufacturing businesses is gathered from different sources. Secondary Information is collected from Pakistan Stock Exchange, State Bank of Pakistan, and official statistics exhibited by Ministry of Finance, Government of Pakistan. Some information is gathered from annual reports of companies in the sampled period. Business Recorder is another reliable publisher of data on stock prices, which is used for some data. Initially an attempt was made to have a complete enumeration of companies for the variables of interest, however, Source: http://www.doksinet 19 information was found missing on several dimensions for the population of firms. Financial data for Pakistani listed business is scattered as it is provided by multiple suppliers. The data hubs that publish financial information have their own problems. The data used in this research is gathered, and painfully organized from multiple sources including sources mentioned above. Some figures

published by these sources may not be trusted therefore some effort is done in smoothing data by removing outliers. The application of conventional random sampling techniques was found limited, therefore, only available information for firms from listed nineteen industries was capitalized in the form of a panel that covers 5672 firm-year observations from 2001 to 2015. The resultant panel was random enough and perhaps caters enough for a scientific sample size. 2.1 Theoretical framework This study demands understanding the dividend policy and its implications from two different perspectives. The first is that whether dividend payout depends on institutional owners as being on boards they may like to have more dividends even though such earnings may be taxable. This would suggest that institutional ownership should have a positive association with dividend payouts under the tax-based hypothesis which is contrary to the argument of agency cost hypothesis that predicts a negative

relationship between payouts and the presence and the degree of institutional ownership on corporate boards. Companies may choose to pay or not their distributable income keeping in view their compelling business reasons. Dividends cannot be negative therefore the dataset includes zeros or greater than zero values, which suggests putting a Tobit model using maximum likelihood for estimating coefficients as the standard ordinary least squares method provides biased regression coefficients. Censoring the lower value for dividends, using a Type-I Tobit model will serve the purpose as the dependent variable will assume only Source: http://www.doksinet 20 positive numbers above zero. This warrants using dividend yield instead of using dividend payout ratio. The explanatory factors include the variables that need to be tested for the listed hypotheses and including controls in the model. The correlational design demands testing for finding evidence towards relationships of dividend

payouts with corporate governance, degree of product market competition, and institutional ownerships. The model also warrants controls for other intrinsic and extrinsic factors that may affect dividend yield. The main controls used are for profitability, leverage, income growth, earnings risk, investment, and target dividend payout ratio. 2.2 Econometric models and variables The data obtained is not in the form of a balanced panel as companies are not consistently showing observations on all important variables. This paper utilizes available observations by first showing results using ordinary least square estimates. Next, in agreement with (Han, Lee, & Suk, 1999) the following econometric model with some modifications is used in this research. Where denotes dividend payouts for firm j at time t and dividend payouts are measured as dividend yield, dividends scaled to sales, dividends scaled to earnings, and dividends scaled to total assets denotes the vector of explanatory

factors including X 1 : Dummy that shows a family owned or a family non-owned firm in the sample (Family ownership) X 2 : Price to book ratio (market worth) Source: http://www.doksinet 21 X 3 : Inverse of Herfindahl-Hirschman index for an industry in a given year X 4 : Past three years average sales growth for a company (Growth) X 5 : retained earnings as ratio of shareholders’ equity (earning potential) X 6 : Tobin’s Q (company worth as viewed by shareholders) X 7 : Debt to assets ratio (leverage) X 8 : standard deviation in return on assets of firm j in the past three-year period (business risk) X 9 : average of past three-year dividend payouts at time t (payouts targets) X 10 : % age of insider owners in the firm at time t (insiders on boards) X 11 : % age of institutional owners on boards at time t (institutions on boards) X 12 : ratio of operating income to total assets of firm j at time t (profitability potential) denotes regression coefficients and denotes error in the

model Keeping in view the earlier discussion that there may be important determining factors for firms that pay dividends the following logistic regression is going to be used = 1 if + + >0 = 0, otherwise denotes the standard logistic distribution error, one may want to use standard normal distribution instead which leads to estimating probit regression, however, this paper restricts itself to using logit regression. There are procedures like Herfindahl-Hirscman Index (HHI) using which product market competition can be measured for more or less concentrated industries. The HHI is a simple metric that can range from 0 to 10,000 as the share of a company in a market can be from 0% to 100%. The HHI for an industry in a year is computed as; HHI = Source: http://www.doksinet 22 This simply means that HHI = S 1 2+ S 2 2+ S 3 2+.+ S K 2 Where S 1 , S 2 , S 3 ,.S k represent market share of companies in a particular year for a particular industry. Number for firms for the index

calculation are usually 50 or less The range of HHI can be broken down like US Department of Justice puts markets having less than 1000 HHI to be competitive markets, 1000 to 1800 HHI to be moderately concentrated, and more than 1800 HHI to be highly concentrated. Ideally 0 HHI should denote perfect competition and 10000 HHI should denoted monopoly. This index faces the limitation of only including information about available firms, data about public firms is readily available in capital markets but lack of information about private firms may pose serious threats of reliance on the index scores as their competitive contribution remains un-measured. This paper investigates the nature and strength of relationship that may exist between the level of competition and the dividend payouts which may generate evidence towards one of our posed hypotheses. The literature has established that cash dividends have agency implications for both the controlling and minority shareholders. Following

(He, 2012) this research uses the following measure of dividend payouts. Dividend Payout = (cash dividend per share/earnings per share)* 100 The study also computes the following dependent variables from the sample data. Dividend payout scaled by net income (DPO1) = cash dividend/earnings Dividend payout scaled by sales (DPO2) = cash dividend/sales Dividend payout scaled by assets (DPO3) = cash dividend/assets Source: http://www.doksinet 23 Controlling shareholders will be forced to pay dividends in light of product market competition if we find positive association between dividend payout measures and HHI. (Denis & Osabov, 2008) and (Gul, 1999) report that more profitable and larger firms in Japan are more likely and driven to pay dividends as the newly listed firms fail to do so due to their compelling business reasons. Only a very small fraction of firms in an industry of Pakistan pay stock dividends therefore this paper excludes paying stock dividends in the dividend payout

ratios. The existing literature reports several factors that affects dividend payouts like severe competition leads to lower profitability which may result in lower payouts. The regression models used by researchers consider using firm size measured by market value of equity, book to market ratio, return on assets, industry competition measure by HHI, five-year growth rate in total sales. (Grullon & Michaely, Coporate Payout Policy and Product Market Competition, March 2007) in their research used volatility in stock returns (standard deviation of monthly stock returns in the previous year) which is a reasonable proxy for risk. Relating to literature on the topic we expect that dividend payouts should have a negative relationship with stock price volatility and sales growth, and should have a positive relationship with market value of equity and return on assets. This work is not considering this aspect here but other research attempts could unfold findings using such aspects. IV.

Findings and discussions There are several hypotheses for which data is analyzed using the sample of non-financial listed Pakistani firms which are listed on Pakistan Stock Exchange. Dividend payments by firms are subject to their relationships with agency hypothesis, signaling theory, information asymmetry etc. Listed Pakistani firms have a history of paying dividends but all firms don’t as they behave exactly like firms in other markets of the world. Pakistani economy has transformed from an Source: http://www.doksinet 24 agrarian status to an economy that is generating a huge revenue from the services sector. Large scale manufacturing or non-financial sector is making its contribution. According to Pakistan Economic Survey 2015-16, this sector contributes 10.9% to GDP This work relies on data for some of the major industrial sectors and companies of Pakistan. Data is taken from Pakistan Stock Exchange (formerly known as Karachi Stock Exchange) as it publishes information on

business and market performance on firms and industrial sectors regularly. Karachi Stock Exchange (KSE) is the biggest capital market in Pakistan as it houses all major businesses listed against its name. According to Pakistan Economic Survey 2014-15 there were 560 companies listed at KSE by the end of March 2015 with a market capitalization of 6760 Billion Rupees. KSE also publishes data on dividends paid by financial and nonfinancial sectors of Pakistan Table 41 shows new listings of firms annually along with other useful information. It is interesting to look at the historic data of dividend payments by large scale manufacturing firms in Pakistan. Pakistan Stock Exchange has been and is one favorite source of information on historic financial performance of firms. Their data publications on dividend payments are published in excel sheets however the change in display of information by shifting firms to different sectors for some years pose threats of consistency and confusion when

it comes to data organization. The flow of information in their publications is found uniform from the year 2001 to 2008, however, from 2008 to 2014 their publications exhibits data on firms not consistently placed in their sectors. Table 4.1 exhibits that even though the total number of firms have reduced over years, however the number of firms that pay dividends have increased and also the percentage of payments have increased too. This may reveal that as firms are getting mature they are perhaps distributing the earnings more. The table also exhibits that as the representation of institutional investors is Source: http://www.doksinet 25 decreasing on boards overtime the dividend payouts by firms are improving. This finding is evidencing the agency cost hypothesis and is not favoring the tax-based hypothesis. Insider ownership is not showing any sound relationship with increasing amount of dividends by Pakistani firms. Table 43 shows the dividend payout ratios for major listed

industrial sectors from 2001 to 2008 and reports that the biggest economic sector of textiles is paying less dividends over time. Textiles include firms in spinning, weaving, composite, woolen sectors There are high expectations from this sector as it is a major economic sector of Pakistan. Some sectors are paying more over time like food, cement, miscellaneous etc. The presence or absence of dividends may be attributed to a number of important factors that are shaping the manufacturing industries of Pakistan. Ownership concentration, expanding economy, new entrants, older firms gaining maturity, agency and tax-based assumptions, product market competition and numerous other factors may be attributed to increasing or decreasing dividends in the Pakistani context. In order to observe evidence based effects we are going to see how correlational design helps in identifying key factors that affect the dividend payouts. Table 46 shows the product market competition for the major Pakistani

industries. It can be seen that industrial sectors are prone to the changes that happen locally and globally. Pakistani consumerism is changing for the good as evidenced by the growth of services sector in the indigenous economy. The HHI index is computed for all major sectors for a period of 15 years and the industries were divided on the basis of a continuum of 0 to 10000. Where a score of 100 or below is defined to be a highly competitive sector, industries scoring 101 to 1000 are rated as un-concentrated, industries which get a score of 1001 to 1800 are defined to be moderately concentrated, and if a sector gets more than 1800 HHI score, it is defined to be highly concentrated. Table 46 exhibits that most of the listed Pakistani manufacturing industries range Source: http://www.doksinet 26 from un-concentrated to very highly concentrated. Family ownership may be a reason for high degree of concentration in some of these industries as business families in Pakistan have a high

degree of ownership concentration and have business groups. The table reflects that understanding the payout behavior of those few leading companies that hold major market share should be very interesting. Using pooled OLS and Tobit model analysis, the findings for family ownership show an inverse relationship with dividend payout. It implies that in Pakistan firms with family owners avoid dividends. This can be related to the fact that family owned firms have large amount of expenses as the members are compensated with high salaries. This may result in either very low or negative income to pay out any dividends. The findings for Tobin’s Q report a positive relation with dividends payout. This indicates that in Pakistan the payment of dividends sends out positive signals to the market about the company’s future earnings and performance while dividend cut is viewed negatively by the stakeholders as it reflects uncertainty about the future prospects of the company. The results for

insider ownership reveal that firms with high level of insider ownership are more likely to pay dividends in order to lessen the costs related with agency conflict. This is consistent with the findings of (Ahmad & Javid, 2009) (Ahmad & Javid, 2009) examined the determinants of dividend payout policy of 320 non-financial firms in Pakistan. The findings also suggest that firms’ profitability has a positive impact on both dividend payout and dividend yield ratio. Firms generating high operating cash flows from sales are more likely to pay dividends as they will earn higher net income. High profits imply stability in firms’ earnings over time and distribution of large amount of FCFs as dividend payments (Ahmad & Javid, 2009; Nazir, Nawaz, Anwar & Ahmed, 2010). Table 44 show an insignificant relationship between product market competition and dividend payout. However, in Table 45, Source: http://www.doksinet 27 findings from Tobit analysis reveal that in Pakistan,

dividends may act as a substitute for product market competition. Therefore, in absence of disciplinary force of product market competition, firms in less competitive industries want to establish good reputation so that they can reduce the agency costs and the costs of raising new finances. Leverage has a negative relationship with dividend payout as well as with dividend yield. Highly leveraged firms retain large amount of their earnings as they tend to avoid the cost of raising external financing. In order to maintain their cash flow and liquidity position, firms with high amount of debt pay low dividends to their shareholders. The findings for institutional ownership report mix results with dividend payout for both OLS and Tobit analyses. Table 44 shows that the percentage of shares owned by institutional investors is positively linked to dividend payout. Under tax-based hypothesis, institutional investors prefer dividends over capital gain. However, the findings for institutional

ownership in Table 4.5 show a negative relationship with dividends A negative relationship between institutional ownership and dividend payout is predicted under the agency-cost based hypothesis. Firms with high degree of institutional ownership are less concerned about the agency costs; therefore, these investors discourage dividend payments. V. Conclusions This work started with an inspiration to examine whether product market competition and institutional ownership are related with payouts in the Pakistani context. Pakistani industrial sectors have big family owned enterprises and disclosure mechanisms are exploitable. This poses serious threats of information asymmetry as owner-managers or bigger shareholders may disappropriate earnings and may get into risky business projects by retaining earnings. The work looked at major finance theories before setting theoretical and econometric models. There is fair degree of competition in Pakistani product markets and the recent emerging

market status is Source: http://www.doksinet 28 posing fresh implications on payouts as researchers may want to understand the theoretical relationships between the intensity of market competition and its role towards increasing or decreasing agency problems between owners and managers. This study looks at the complexity of agency cost hypothesis and tax-based hypothesis under the assumptions of asymmetry information from the perspective of emerging markets. Studying the relationship between ownership structure and payouts is engaging as less developed markets can add more insights to theoretical literature. This study works to test the effects of product market competition and ownership structures on payouts. The paper relies on a panel data of nineteen listed manufacturing industries of Pakistan from 2001 to 2015. There is evidence that agency cost hypothesis holds in the Pakistani context as with declining number of institutional investors overtime, industries seem to pay more

dividends. It is learnt that more levered firms are paying less dividends whereas firms having higher level of retained earnings are positively related to payment of more dividends. Family ownership is evidenced as negatively related to paying dividends. Product market competition is also found to have a negative relationship with payouts. There is evidence that firms having lower percentage of institutional owners on boards pay more dividends which goes in favor of agency cost hypothesis. Source: http://www.doksinet 29 References: Afza, T., & Mirza, H H (2010) Ownership Structure and Cash Flows As Determinants of Corporate Dividend Policy in Pakistan. International Business Research, 210-221 Aghion, P., Dewatripont, M, & Rey, P (1999) Financial Discipline and Growth The Review of Economic Studies, 825-852. Ahmed, H., & Javid, Y A (2009) The Determinants of Dividend Policy in Pakistan International Journao of Finance and Economics, 110-125. Allen, F., & Michaely, R

(2003) Payout Policy Handbook of the Economics of Finance, 337-429 Baker, M., & Wurgler, J (2004) A catering theory of dividends Journal of Finance, 1125-1165 Bates, T. W, Kahle, K M, & Stulz, R M (2009) Why do US firms hold so much more cash than they used to? The Journal of Finance, 64(5), 1985-2021. Benartzi, Shlomo, Michaely, R., & Thaler, R (1997) Do changes in dividends signal the future or the past? The Journal of Finance, 1007-1034. Brav, A., Graham, R J, Harvey, R C, & Michaely, R (2005) Payout policy in the 21st century Journal of financial economics, 483-527. Chhaochharia, V., Grinstein, Y, Grullon, G, & Michaely, R (2009) Product Market Competition and Agency Conflicts: Evidence from the Sarbanes Oxley Law. SSRN Working Paper Series DeAngelo, H., & DeAngelo, L (2006) The irrelevance of the MM dividend irrelevance theorem Journal of Financial Economics , 293-315. DeAngelo, H., DeAngelo, L, & Skinner, J D (2004) Are dividends disappearing? Dividend

concentration and the consolidation of earnings. Journal of Financial Economics, 425-456 DeAngelo, H., DeAngelo, L, & Skinner, D J (2008) Corporate payout policy Foundations and trends in finance, 3(2-3). DeAngelo, H., DeAngelo, L, & Stulz, R M (2006) Dividend policy and the earned/contributed capital mix: a test of the life-cycle theory. Journal of financial economics, 81(2), 227-254 Denis, J. D, & Osabov, I (2008) Why do firms pay dividends? International evidence on the determinants of dividend policy. Journal of Financial Economics, 62-82 Easterbrook, F. H (1984) Two agency-cost explanations of dividends The American Economic Review, 74(4), 650-659. Source: http://www.doksinet 30 Fama, E., & French, K (2001) Disappearing Dividends Changing Firm Characteristics or Lower Propensity to Pay? Jounal of Financial Economics, 3-43. Fama, E. F, & French, K R (2004) New lists: Fundamentals and survival rates Journal of financial economics, 73(2), 229-269. Gasper, J.-M,

& Massa, M (2006) Idiosyncratic volatility and Product Market Competition The Journal of Business, 3125-3152. Ghani, I. W, & Ashraf, J (2005) Corporate Governance, Business Group Affiliation, and Firm Performance: Descriptive Evidence from Pakistan. Wroking Paper No 05-35 Giroud, X., & Mueller, M (2011) Coporate Governance, Product Market Competition, and Equity Prices The Journal of Finance, 563-600. Glen, D., Karmokolias, Y, Miller, R R, & Shah, S (1995) Dividend Policy and behavior in emerging markets: To pay or not to pay. The World Bank Graves, S. B, & Waddock, S A (1990) Institutional ownership and control: Implications for long-term corporate strategy. The Executive, 4(1), 75-83 Grinstein, Y., & Michaely, R (2005) Institutional Holdings and Payout Policy The Journal of Finance, 1389-1426. Grullon, G., & Michaely, R (2002) Dividends, Share Repurchases and the Subsitution Hypothesis Journal of Finance, 57, 1649-1684. Grullon, G., & Michaely, R

(March 2007) Coporate Payout Policy and Product Market Competition AFA 2008 New Orleans meetings paper. Gul, A. F (1999) Growth opportunities, capital structure and diviend policies in Japan Journal of Corporate Finance, 141-168. Han, C. K, Lee, H S, & Suk, Y D (1999) Institutional shareholders and dividends Journal of Financial and Strategic Decisions, 53-62. He, W. (2012) Agency problems, product market competition and dividend policies in Japan Accounting & Finance, 873-901. Healy, M., & Palepu, G (2001) Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature. Journal of Accounting and Economics, 405-440 Herfindahl, O. (1950) Concentration in the US steel industry PhD diss Columbia University Source: http://www.doksinet 31 Hirschman, A. (1945) National power and the structure of foreign trade Berkely: University of California Press. Hoberg, G., & Prabhala, R N (2009) Disappearing dvidends, catering, and

risk Review of financial studies, 79-116. Hoberg, G., Phillips, G, & Prabhala, N (2014) Product Market Threats, Payouts, and Financial Flexibility The Journal of Finance, 293-324. Imran, K. (2011) Determinants of Dividend Payout Policy: A case of Pakistani Engineering Sector The Romanian Economic Journal, 47-60. Jarrell, G. A, & Poulsen, A B (1987) Shark repellents and stock prices: The effects of antitakeover amendments since 1980. Journal of financial economics, 19(1), 127-168 Januszewski, I., Koke, J, & Winter, K J (2002) Product market competition, corporate governance and firm performance: an empirical analysis for Germany. Research in Economics, 299-332 Jensen, M., & Meckling, W H (1976) Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 305-360 John, Kose, & Knyazeva, A. (2006) Payout policy, agency conflicts and corporate governance NYU Julio, B., & Ikenberry, L (2004) Reappearing Dividends

Journal of Applied Corporate Finance, 89-100 La Porta, R., Lopez-de-Slanes, F, Shleifer, A, & Vishny, W (2000) Agency Problems and Dividend Policies around the World. The Journal of Finance, 1-33 Lerner, A. (1934) The concept of monopoly and the measurement of monopoly power Review of Economic Studies, 157-175. Lintner, J. (1956) Distribution of incomes of corporations among dividends, retained earnings, and taxes. The American Economic Review, 97-113 Mayers, S. C (1984) The capital structure puzzle The Journal of Finance, 574-592 Mehar, A. (2005) Corporate Governance and Dividend Policy Pakistan Economic and Social Review, 93106 Michaely, R., & Robert, M (2012) Coporate dividend policies; Lessons from private firms Review of Financial Studies, 711-746. Miller, M. H (1986) Behavioral Rationality in Finance: The Case of Dividends Journal of Business, 451468 Source: http://www.doksinet 32 Miller, M. M, & Rock, K (1985) Dividend Policy under Asymmetric Information The

Journal of Finance, 1031-1051. Miller, M., & Modigliani, F (1961) Dividend Policy, Growth and Valuation of Shares Journal of Business, 411-433. Mitton, T. (2004) Corporate governance and dividend policy in emerging markets Emerging Markets Review, 409-426. Megginson, W., & Smart, S (2008) Introduction to corporate finance: Nelson Education Nazir, M. S, Nawaz, M M, Anwar, W, & Ahmed, F (2010) Determinants of stock price volatility in Karachi Stock Exchange: The mediating role of corporate dividend policy. International Research Journal of Finance and Economics, 55, 100–107. Patal, A., Balic, A, & Bwakira, L (2002) Measuring transparency and disclosure at firm-level in emerging markets. Emerging Markets Review, 325-337 Rafique, M. (2012) Factors Affecting Dividend Payout: Evidence From Listed Non-Financial Firms of Karachi Stock Exchange. Business Management Dynamics, 76-92 Ross, S. (1977) Teh Determination of Financial Structure: The Incentive Signalling Approach Bell

Journal of Economics, 23-40. Shleifer, A. (1985) A theory of yardstick competition The RAND Journal of Economics, 319-327 Source: http://www.doksinet 33 Table 4.1 SUMMARY OF LISTED INDUSTRIAL MARKET STRUCTURE LISTED CAPITAL (RS BILLION) 374.1 TURN OVER OF SHARES (IN BILLION) 97 AVERAGE DAILY TURN OVER OF SHARES (IN MILLION) 386.7 AGGREGATE MARKET CAPITALIZATION (RS BILLION) 1357.5 Year 2004 NUMBER OF LISTED COMPANIES 668 NEW COMPANIES LISTED 16 FUND MOBILIZED (RS BILLION) 70.7 2005 659 15 54 439 88.3 351.9 2013.2 2006 658 14 41.4 496 104.7 319.6 2801 2007 658 12 49.7 631.1 68.8 211 4019.4 2008 652 7 62.9 706.4 63.3 238.2 3777.7 2009 651 8 44.9 781.8 28.2 115.6 2143.2 2010 652 8 135.1 909.9 43 173.2 2732.4 2011 639 1 31.04 943.7 28 111.63 3288.7 2012 591 3 115.1 1069.8 38.1 150 3518.1 2013 569 4 29.5 1116 54.3 221 5154.7 2014 557 5 47.6 1100.3 56.6 229.1 6655.3 2015 560 6 29.1 1177.8 38.4

185.7 6760.8 Source: Karachi Stock Exchange Table 4.2 SUMMARY FINDINS ON PAYOUTS AND INSTITUTIONAL OWNERSHIP Year 2009 2010 2011 2012 2013 2014 2015 Total firms 476 452 450 401 399 398 416 No of firms that paid dividends 127 162 156 151 159 151 172 Average Dividend Payout Percentage 45.7% 34.6% 32.3% 27.5% 25.9% 24.5% 34.5% Institutional Ownership Percentage 46.10% 41.46% 29.14% 31% 34.60% 27.80% Not available Insider Ownership Percentage 2.72% 3.27% 2.76% 2.47% 2.33% 2.35% Not available Source: http://www.doksinet 34 Table 4.3 SECTOR-WISE SUMMARY OF PAYOUTS Years Industrial Sectors 2001 2002 2003 2004 2005 2006 2007 2008 Textile Spinning Total firms Paid dividends Average DPO 122 54 52.25% 115 43 55.48% 108 33 43.22 97 23 43.17 138 14 68.7 111 18 56.21 71 16 58.13 73 10 53.69 Textile Weaving Total firms Paid dividends Average DPO 15 6 97.80% 13 4 56% 10 4 29.65 9 4 91.52 13 1 65.74 20 2 58.54 7 1 64.49 7 1 33.27 Textile Composite Total firms

Paid dividends Average DPO 51 23 43% 49 21 47% 47 23 37.77 46 18 32.99 34 13 7.52 59 17 32.69 38 17 87.92 40 13 26.11 Woolen Total firms Paid dividends Average DPO 4 2 80% 4 2 81.75 4 3 57.6 3 0 0 3 0 0 5 0 0 1 0 0 2 0 0 Total firms Paid dividends Average DPO 20 8 79.52% 18 9 84.24 19 10 85.61 16 6 54.66 15 4 4.06 19 6 60.47 11 4 40.01 11 5 30.86 Jute Total firms Paid dividends Average DPO 6 80.855 81% 6 2 98.11 6 2 38.49 5 2 35.78 6 1 22.78 6 2 34.26 1 1 22.15 1 0 0 Sugar & Allied Total firms Paid dividends Average DPO 37 13 33.93% 37 14 30.32 37 10 64.66 34 10 36.9 38 12 25.23 37 15 42.64 31 6 66.44 31 11 37.29 Cement Total firms Paid dividends Average DPO 21 5 86.29% 22 6 86.55 22 8 68.75 21 9 64.89 21 5 60.66 21 11 34.09 18 4 32.26 19 1 24.88 Tobacco Total firms Paid dividends Average DPO 6 2 39% 5 2 49.29 5 2 66 5 3 67.6 5 3 15.09 5 3 81.96 2 2 75.1 2 2 65.68 Fuel & Energy Total firms Paid dividends

Average DPO 25 17 61.68% 25 16 61.12 24 18 63.64 25 18 56.12 29 12 58.53 28 14 66.86 27 15 64.44 26 16 46.96 Engineering Total firms Paid dividends Average DPO 12 7 46.90% 10 5 54.86 10 4 48.19 10 7 50.16 24 4 79.27 13 7 65.06 10 9 39.16 9 4 30.05 Auto & Allied Total firms Paid dividends Average DPO 23 10 53.63% 23 13 70.95 22 14 51.11 22 12 39.86 21 13 36.63 25 14 77.47 17 10 46.48 19 10 45.02 Synthetic & Rayon Source: http://www.doksinet 35 Cables & Electrical Total firms Paid dividends Average DPO 9 4 80.16% 8 4 65.08 7 3 64.63 6 4 51.36 6 3 86.49 9 2 41.91 6 2 29.08 6 2 22.31 Total firms Paid dividends Average DPO 35 23 70.89% 36 24 93.03 32 20 56.66 30 18 52.69 33 18 64.96 31 17 48.06 25 19 86.84 29 20 71.79 Total firms Paid dividends Average DPO 13 7 65.66% 12 7 65.91 11 7 54.75 11 7 54.06 9 7 42.77 11 5 44.1 7 3 35.56 8 3 30.72 Total firms Paid dividends Average DPO 13 0 0% 12 3 88.67 9 0 0 10 2 17.39

11 0 0 13 1 9.99 12 1 29.4 1 0 0 Total firms Paid dividends Average DPO 7 4 55.46% 7 4 53.51 5 3 46.18 5 4 29.56 5 2 57.06 5 3 45.17 4 2 25.19 4 2 19.56 Food & Allied Total firms Paid dividends Average DPO 18 12 3.04% 16 12 59.41 17 13 54.44 17 12 53.83 22 12 50.54 21 13 67.72 16 14 69.45 16 13 73.12 Glass & Ceramics Total firms Paid dividends Average DPO 7 5 63.75% 7 5 81.54 7 4 20.3 7 4 45.5 10 2 41.49 10 3 36.54 6 3 66.49 7 2 59.79 Fertilizer Total firms Paid dividends Average DPO 4 2 37.00% 4 3 65 4 3 70.8 4 4 77.5 2 1 67.67 4 4 73.41 4 4 59.64 4 4 58.71 Miscellaneous Total firms Paid dividends Average DPO 23 10 71.24% 23 7 58.67 21 5 46.56 20 11 24.2 27 3 25.17 27 11 81.78 18 8 78.53 19 8 97.42 Chemical & Pharmaceutical Paper & Board Vanaspati & Allied Leather & Tanneries Source: http://www.doksinet 36 TABLE 4.4 POOLED OLS MODELS VARIABLES FO PBR IHHI GS RETE TQ Lev BR PDPO IOW InOW ITA Constant

Observations R-squared Model 1 DY Model 2 DPO1 Model 3 DPO2 Model 4 DPO3 0.91178 (1.25171) -2.46495* (1.17589) -962.45784 (995.67062) -0.00005 (0.00006) 1.49091 (2.80796) 2.16133 (3.31576) 15.86772* (3.70005) 2.69884 (3.76105) 0.03980* (0.01987) 5.23582* (2.18980) 4.52500 (5.06117) 64.76699* (11.87100) -8.08927* (4.58277) -5.26504* (3.04432) 9.15006* (2.85991) -874.45850 (2,421.59712) -0.00014 (0.00015) -26.48821* (6.82932) 5.57483 (8.06436) 10.83672 (8.99900) 9.64807 (9.14735) 0.07546 (0.04832) 16.12678* (5.32586) 32.65765* (12.30942) 10.83780 (28.87179) 6.61179 (11.14587) -0.11923 (0.16388) -0.13448 (0.15396) -34.81424 (130.36156) -0.00000 (0.00001) 0.35023 (0.36764) 1.33482* (0.43413) -2.29053* (0.48444) 0.42405 (0.49243) 0.00403 (0.00260) 0.34236 (0.28671) 1.90850* (0.66265) 1.72964 (1.55425) 1.52327* (0.60001) -0.20985 (0.20471) -0.07504 (0.19231) -217.80479 (162.83704) -0.00000 (0.00001) -1.43221* (0.45923) 1.23286* (0.54228) 0.50498 (0.60513) 0.05770 (0.61510) 0.00375

(0.00325) 0.88572* (0.35813) 1.62761* (0.82773) 17.14830* (1.94144) -0.13058 (0.74949) 166 166 166 0.38025 0.36776 0.38140 Standard errors in parentheses * p<0.01, * p<0.05, * p<0.10 166 0.50962 Source: http://www.doksinet 1 Model 1 Table 4.5 Tobit Models FO PBR IHHI GS RETE TQ Lev BR PDPO IOW InOW ITA Constant Observations DY Model 2 Sigma 1.01747 (1.21115) -2.46314* (1.13444) -1,029.02213* (962.34879) -0.00005 (0.00006) 1.55636* (2.70959) 2.25930 (3.20003) -15.82232* (3.56984) 2.82607 (3.63019) 0.03672* (0.01936) 5.29505* (2.11324) -4.76547* (4.88732) 64.73085* (11.45257) -7.89669* 5.5418* (4.42445) (0.30549) 166 166 DPO1 Model 3 Sigma DPO2 -5.34988* (2.93826) 9.22261* (2.76013) -776.28198* (2,338.08256) -0.00014 (0.00014) -26.67967* (6.59143) 5.38361 (7.78271) -10.69702* (8.68354) 9.53327 (8.82639) 0.07607 (0.04663) 15.84180* (5.14481) -32.48139* (11.87773) 11.62057 (27.86546) 6.44098 (10.75513) 166 Model 4 Sigma DPO3 Sigma 13.4799* (0.74314) -0.10175

(0.15869) -0.13418 (0.14844) -45.82417* (126.02584) -0.00000 (0.00001) 0.36106 (0.35458) 1.35103* (0.41879) -2.29803* (0.46712) 0.44510 (0.47511) 0.00352 (0.00254) 0.35216* (0.27655) -1.94828* (0.63977) 1.72366 (1.49856) 1.55513* (0.57912) 0.7251* (0.03995) -0.18261 (0.19820) -0.07523 (0.18518) -225.93084* (156.91727) -0.00000 (0.00001) -1.40529* (0.44266) 1.25042* (0.52233) -0.48028* (0.58298) 0.08222 (0.59257) 0.00403 (0.00314) 0.89656* (0.34494) -1.65002* (0.79720) 17.16754* (1.86948) -0.13886 (0.72171) 0.90461* (0.04980) 166 166 166 166 166 Tablehttp://www.doksinet 4.6Herfindahl-Hirschman Source: Industry Auto Cables Cement Chemicals Engineering Fertilizers Foods & Personal Care Fuel & Energy Glass & Ceramics index (HHI) for listed manufacturing industries of Pakistan and the degree of competition 2001 2002 2003 2004 2006 2007 2008 2009 2010 2011 1395 2005 17511 HHI 1115 1180 1400 1509 1642 1555 1256 1894 1467 Level of

competition Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration High Concentration Moderate HHI 3003 3080 3352 3533 3942 4048 3897 4207 7581 4084 3460 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 744 709 726 726 722 782 1113 1077 1075 996 889 Level of competition Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Moderate Concentration Moderate Concentration Moderate Concentration Unconcentrated Unconcen HHI 916 993 1335 1463 1289 1426 1465 1269 1182 1308 1625 Level of competition Unconcentrated Unconcentrated Moderate Concentration

Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate HHI 2440 2904 2941 2462 1725 2479 3508 3116 2871 2187 2460 Level of competition High Concentration High Concentration High Concentration High Concentration Moderate Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 10000 10000 3661 3234 10000 3288 3206 2900 2833 5002 4030 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 3177 3038 2926 2655 1929 2425 2560 2398 2505 2525 2046 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High

Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 1547 1663 1803 1347 1243 1203 1359 1553 1440 1923 1689 Level of competition Moderate Concentration Moderate Concentration High Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration High Concentration Moderate HHI 1613 1662 1603 1695 1776 2048 2569 2179 2351 2255 2195 Source: http://www.doksinet 2 Level of competition Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration High Concentration High Concentration High Concentration High Concentration HHI 2724 3633 6115 7213 7405 7925 8815 8421 8074 10000 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration

High Concentration High Concentration High Concentra HHI 2531 3000 4457 4371 4332 4394 4398 4556 4770 4910 4957 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 2188 2312 2295 2362 2733 3294 5000 10000 10000 5289 7088 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 2707 2686 2859 2936 3090 3942 5847 6088 3911 2292 2111 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 2241 2157

2156 2378 2194 2225 2116 2225 2284 3600 3623 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra Sugar HHI 382 369 343 369 448 566 501 506 485 649 678 Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcen Textile Level of competition HHI 159 326 395 186 237 175 205 203 222 239 244 Level of competition Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcentrated Unconcen HHI 5005 5054 5143 5166 5064 5327 5421 5314 5278 5249 5299 Level of competition High Concentration High Concentration High Concentration High Concentration High

Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra HHI 3395 3417 3340 5004 10000 5014 10000 6292 6658 10000 10000 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentration High Concentra Jute Leather Oil & Ghee Paper & Board Rayon Tobacco Woolen High Concentration High Concentra Source: http://www.doksinet 3 Miscellaneous HHI 2548 2456 2436 2345 3005 1738 1692 1613 1481 8708 8637 Level of competition High Concentration High Concentration High Concentration High Concentration High Concentration Moderate Concentration Moderate Concentration Moderate Concentration Moderate Concentration High Concentration High Concentra Source: http://www.doksinet 1