Evidence of Dividend Catering Theory in Malaysia: Implications for Investor Sentiment

This study investigates the key determinants of corporate performance in Malaysia. Using panel data (2002-2007) of 361 companies listed in Malaysia, the study finds dividend per share, use of debt, number of board members, and last year’s performance to be the most significant determinants of corporate performance across four selected industries: trading or services, property, consumer products, and industrial products. This study also finds that dividend per share is influenced by market performance and is followed by last year’s dividend and size of the dividend. These findings exhibit the presence of dividend catering incentives. As such, market demand for dividends drives corporate dividends. The study concludes that investor sentiment influences corporate decisions in Malaysia.


Introduction
Dividend policy is a major financial decision. Despite theories suggesting that dividend policy has no significant impact on the changes in corporate value (Miller & Modigliani, 1961), extant studies find that dividend works as a signal and influences asset valu-expected that dividends are perceived as tangible benefits to investors when valuing any company. Investors desire more dividends. However, Fama and French (2001) presented the disappearing dividends effect among American investors claiming that dividends are no longer an important vehicle for attracting investors. In response to this argument, 2004) found that the propensity to pay dividends is driven by the catering incentive. The catering theory of dividends purports that corporations will pay dividends only if they perceive a demand for the same from the market (Baker & Wurgler, 2004).
Thus, there will be higher dividend payouts if the market provides a premium for the stock price. It is this premium that creates the catering incentive and thus explains the corporate tendency to pay dividends. Baker and Wurgler (2004) argued that if dividend payment is influenced by stock market performance (or vice versa), investor sentiment would be a major reason behind this causal relationship. Consequently, disequilibrium in the market reveals the tendency to relate dividends to the market value of corporations.
This study examines the presence of dividend catering theory and the influential power of dividends in corporate valuation among the listed firms in Malaysia. Similar to Baker and Wurgler (2004), if corporate performance influences corporate dividend payment, the study may conclude that some performance-related motivational force is driving the propensity to pay dividends. The study also investigates the influence of different industries (such as construction and trading) on the determinants of corporate value and dividend catering incentives.

Dividend and Other Determinants of Corporate Valuation
A number of studies in the West and recently in emerging markets have identified a list of determinants for corporate valuation. Theoretically, better quality investments should positively influence the value of corporations in the market (Morgado & Pindado, 2003).
The term quality investment, in most of the studies, refers to investments having a positive net present value. Myers (1977) found that the effect of debt financing in investment decisions was negative. Jensen (1986) argued that investment interacts with availability of free cash flow, agency conflict and corporate financing policies when determining the value of corporations.
Companies with higher debt have the opportunity to offer external stakeholders control, providing the corporations with transparency and effective checks and balances. However, involving external stakeholders may result in agency conflict, which could be flagged by investors as being a negative signal. On average, however, an increase in positive NPV projects would positively influence corporate value. Baker, Stein and Wurgler (2003) presented similar results by reporting that investment performance depends on how financing decisions are made.
The vehicle for corporate financing -debt or equity -has a significant impact on the value of corporations. Fama and French (1999) found that U.S. corporations mostly use long-term debt for expansion. These firms rely on equities only during merger and acquisition activities. Investment performance and dividend policy simultaneously influence the effect of external financing on firm performance. Although firms gain external control through debt financing (Berger & Di Patti, 2006), higher dependency on debt financing may result in poor performance given that the investment decisions are below average quality (Abor, 2005;2007;Lang, Ofek, & Stulz, 1996). On the other hand, the use of debt positively influences the performance of reputed firms (Campello, 2006;Harris & Raviv, 1991).
The existing literature displays the effect of dividends on corporate value. Dividends work as a signal and carry both positive and negative impacts. Grinblatt, Masulis and Titman (1984) found that dividends positively influence corporate value. Baker and Wurgler (2007) discovered that dividend premiums are a significant proxy of investor sentiment in the market. They concluded, similarly to Brown and Cliff (2005), that investor sentiment works as a contrarian predictor of future stock returns, thus proving that the global presence of dividend premiums is a deter- other studies found a significant negative influence of large board size on firm performance (Eisenberg, Sundgren, & Wells, 1998;Haniffa & Hudaib, 2006).
Various proxies measure corporate performance.
Many studies used firm performance and firm value (especially market performance and market value) interchangeably. Return on Asset (ROA) is used as a measure for financial performance (Haniffa & Hudaib, 2006), whereas Tobin's Q is reported as a proxy for financial and market performance in various studies (Chua, Eun, & Lai, 2007). Chua et al. (2007) reported that Tobin's Q is the proxy for perceived corporate value by the investors. Tobin (1969) explained the Q ratio as being the determinant of how investors reward and penalize the firms' financial decisions. Thus, Tobin's Q can work as a proxy for financial, market and investor perception. Other than Tobin's Q, various studies rely on stock price as a measure for market performance. However, due to the frequent volatility of the stock price, which requires additional analysis, making a valid proxy from stock returns to represent firm performance is somewhat questionable.

Empirical Models
The major objectives of this study include determining key factors behind corporate valuation among listed firms in Malaysia to determine the influence of significant factors in different industries and to determine the presence of the dividend catering incentive in overall corporate valuation and in selected industries.
Equation 1 lists a number of determinants along with the proxy for corporate value. Table 2 gives the descriptions of the variables. Haniff and Hudaib (2006) found a significant influence of different industry groups in linking corporate value and corporate governance.
Their study uses data from six industries, including the consumer, trading, property, construction, plantation and industrial sectors. This study incorporates the analysis of four significantly large industries that include industrial production (IP), consumer products (CP), property (PR), and trading and service (TS). Table 1 provides descriptive statistics on industry groups.
Because the number of companies under each of the four selected industries is suitable for conducting multiple regression analysis, equation (1) will be examined for four industries to compare the beta coefficients. year. Thus, more than an average dividend in the previous year would create a positive demand for dividends in the current year. To examine these three conditions, the study uses the following three equations.

Data and Method
Due to the structural differences of listing require- The study uses panel data, which has become increasingly important in developing countries due to the paucity of time series data (Gujarati, 2003). In the panel data method, the study can control for cross section fixed effects (Baltagi, 2005). To provide a simple understanding, (2), (3) and (4) (Table 3).

Duality is insignificant in almost all sectors except for
property. It was interesting to observe a conflict between duality and the lag value of the Q ratio, which may lead to challenging future research on governance and firm performance.     (2006) found a negative relationship between the number of board members and value. Conflict of interest is the primary reason behind such relationships. The previous year's performance (Q t-1 ) also influences the current year's performance and boosts the R 2 of the estimates. The variable is robust across all of the sectors and is consistent with the suggestions of Haniffa and Hudaib (2006). Table 5 shows that the R 2 of the estimates are significantly above conventional norms. Additionally, the Durbin Watson (DW) statistics are under control. Higher standardized beta coefficients of DPS and Q (t-1) lead to further inquiry on the dividend catering incentive.

Dividend Catering
Dividend catering theory argues that corporations offer dividends if there is market demand for dividend pay-ment. Thus, to examine the existence of dividend catering incentives, equations (2), (3) and (4) should be significant and robust across industries. Table 6 highlights the tests for these three equations for the total sample and for four industry groups. One of the major arguments behind dividend catering theory is that market value drives the propensity to pay dividends. Table 6 shows that the Q ratio (proxy for market performance) Additionally, investors may expect that the companies with higher dividends may continue to pay higher dividends. Thus, they will expect higher dividends and by the grace of catering incentives, managers should look for sources of income to provide higher dividends. The proxy for higher dividends, DPOUT, significantly influences DPS in all sectors as well as the total sample. The beta coefficients are also high.
Thus, market forces drive corporations and investor sentiment while paying higher dividends. Three of our proxies, through equations 2, 3 and 4, establish that corporate managers time the market for their dividend announcement activity. 2004) theoretically support the performance proxy (DPS and Q in equation 2) and size proxy (DPS and DPOUT).

A Comprehensive Model
After analyzing the dividend catering incentive, the study revises the preliminary estimates of the key determinants. Table 7 exhibits robust results for DPS, DEBT, BOARD and Q (t-1) . The study finds a new variable, DPOUT, significant while explaining the changes in corporate value in Malaysia. Additionally, the R 2 and DW statistics for the estimates are satisfactory. Among these variables, Q (t-1) is the most influential variable, followed by DPS, DEBT, DPOUT and BOARD. Notes: Beta Coefficients are standardized *** = Significant at 1%, ** = at 5% and * = at 10%. Dependent Variable: Tobin' s Q.