Value relevance of government ownership of equity (evidence from Russian listed companies)

Размещено на http://www.allbest.ru/

Размещено на http://www.allbest.ru/

Value relevance of government ownership of equity (evidence from Russian listed companies)

Anastasia Maga

Elizaveta Baranova

Zhargal Tumunbayarova

The impact of government ownership of a firm's equity has generated much discussion in finance, economics, and politics. Extant literature provides evidence of both positive and negative effect of government ownership on the market value of firms with multiple reasons in favor of both effects. There has also been research on how such effects may differ in different markets. This paper aims to explore value-relevance in Russian financial markets. We are exploring the relationship between government ownership of a firm's equity on its market value in Russia with the view of answering the research question of how investors in Russia perceive the fact and degree of government ownership on equity measured through the market value of equity. We are using a sample of 159 Russian listed companies to identify relationships between market value and government ownership of equity. Previous studies support a positive relationship between such variables, however evidence from Russian listed firms proves otherwise. We find little statistically significant evidence of such association in the market in general, however the industry effect proved to be significant; we find that government control is severely penalized in manufacturing, while rewarded in service sectors; this effect also depends on the concentration of government ownership. This research connects to the efficient market hypothesis (EMH).

The structure of ownership of firm equity has been widely discussed as contributing to both positive and negative influences on various aspects of performance. Two main streams of literature exist in this relation, the most prominent one being the one exploring various effects government ownership of a firm's equity exerts on its financial performance, and the less significant in volume being the stream on the direct market value relevance of government control. Both literatures have in common the view that governmental ownership can have both positive and negative effects on either one of the performance indicators. Our research contributes to the latter and dwells upon the efficient market hypothesis (EMH) [Fama, 1998] supporting current market value of the firm's equity contains the publicly available data on the equity structure as well as other information, meaning that investors making decisions to purchase a particular share of stock will consider this information, thus knowing about the fact and degree of state-control over a firm and expressing their attitude towards it.

We are exploring the relationship between government ownership of a firm's equity on its market value in Russia with the view of answering the research question of how investors in Russia perceive the fact and degree of government ownership on equity measured through the market value of equity. Carney and Child (2013) report a growing role of government in equity ownership in East Asian countries between 1996 and 2008. Boubakri, Ghoul, Guedhami and Meggisnson (2017) report that in emerging markets SOEs (State owned enterprises) account for 28% of the largest companies. However, despite active privatization in both developed (mostly Europe) and developing markets (Asian countries, East Europe, Latin America) the share of state-controlled companies continues to grow, but a more active phase of this growth occurred after the financial crisis, when government ownership started to be associated with a bailout strategy and certain references in receiving support [Guedhami, 2012; Nash, 2017].

Further observation with regard to government ownership of firms' equity, found some companies were formed due to privatization of SOE's capital and the government retained a proportion of equity (usually around 50% as reported by Tran, Nonneman & Jorrisen (2014)). In other cases, ownership was obtained by purchasing a share of equity in the market, the result being seen in the form of low percentage of government ownership of the firm's equity. Such acquisition was reported by Eckel and Vermaelen (1986) for North American firms. Thus, the trend reported by the extant research shows the proliferation of the latter form of acquisition after the financial crisis in Europe and Asia, presumedly in response to the crisis effects, with some evidence in favor of a more positive perception of such a type by investors, but again the findings differ depending on economy [Huang, Xiao, 2012] and firm size [Truong et al., 2006].

Our research explores the relationship between government ownership and market value for Russian listed firms. We are using evidence from a sample of 159 publicly traded firms listed in the stock exchange of Russia found in Bloomberg database, the study is cross-sectional and relies on data obtained for the year 2017.

Literature review

The extant literature abounds in research on value relevance of various corporate engagements, including financial and non-financial information with the main theoretical concept being the efficient market hypothesis (EMH) [Fama, 1970; 1998; Kendall, 1958] stating that share prices contain all publicly available information, including information on the public ownership of the firm's equity.

The literary background of market value research started in 1953, when Maurice Kendall published his seminal paper elaborating on the «random walk» proposition [Kendall, 1953] having proved that stock prices fluctuate independently of one another, that they have the same probability distribution, hence probability theory can be applied, and that stock and commodity prices in the long term tend to maintain an upward trend. The stock price movements in Kendall's model are martingales. The same concept was later formalized by Samuelson (1965) in his seminal paper «Proof that properly anticipated prices fluctuate randomly» and the Random Walk Hypothesis (RWH) thus took its shape, and later came to be known as weak-form efficiency.

Subsequently the concept was developed by Gene Fama with his Efficient Market Hypothesis (EMH), which stipulates that stock prices contain all relevant information, but there are three different forms of efficiency under the hypothesis [Fama, 1970; 1991]:

- The weak form, under which the current prices reflect all information contained in the record of past prices.

- The semi-strong form suggests that current prices contain not only historical prices but also all other publicly available information.

- The strong form states that all information, both public and private, is instantly priced into stocks.

The evidence in support of the EMH has been substantial over the years, mainly relying on the failure of fund managers to outperform the market [Fenton O'Creevy et al., 2005; Fama, 1991].

Various effects on market value were being explored at that same time, for instance the Modigliani-Miller hypothesis (M&M) stated that, under the conditions of a perfect market, in the absence of taxes, the total market value of a firm is not affected by its capital structure [Modigliani, Miller, 1958]. In the light of the M&M paradox a question logically arose as to what affects the market value and how firms capitalize on it. The help came from Jensen and Meckling's Agency Cost theory (1976). The agency approach stated that the capital structure matters as debt is a good disciplinarian for managers, and the ownership structure matters as well because managerial behavior is affected by their own interest, that assumption will work in case of managerial ownership. Under more realistic conditions, however, with inclusion of taxes and bankruptcy costs, the M&M paradox did not hold [Modigliani, Miller, 1963; Morck, Shleifer, Vishny, 1988] and two streams of literature developed in relation to the effect of ownership structure on firm value: management ownership and concentration of ownership, both discuss the ways to reduce agency costs.

The managerial ownership, also referred to as insider ownership, dwells on the convergence of interest hypothesis to explain the positive impact of managerial ownership on market value [Morck, Shleifer, Vishny, 1988]. However, results of studies attempting to find the link between insider ownership and market value have been mixed. The concentration of ownership effect is based on the efficient monitoring hypothesis [Pound, 1988], stating that firms with more concentrated ownership structures have better financial and market performance, because of more efficient management input coming from the single investor, shareholders' activism has less embedded conflicts, the efforts of managers are better coordinated with those of the shareholder, hence agency cost is reduced.

Another view at ownership structure is government ownership of a firm's equity, However, research on the impact of government ownership on market value lacks an extensive history. The object of such a consideration is a mixed enterprise, defined by Spenser (1959) as joint ownership by public and private management. Such ownership is described as the result of either state-owned enterprise (SOEs) being privatized but government retaining its stake, or state acquisition of shares of a private jointly owned enterprise.

The first intensive research into the effect of state ownership on the market value of a firm was made by Eckel and Vermaelen (1986), who reported both negative and positive impact, the positive including expectation of lower risk and access to subsidy, synergistic effects; the negative ones being lower expected profitability capitalizing into share prices. The authors connected research to a conceptual framework of agency approach under the perception of alignment of agents' interests with principal ones, but dependent upon the degree of regulation.

Since then there has been two parallel streams of literature exploring the effects of public ownership: market value relevance effects and the financial performance effects, the results of both streams can be linked by the overarching concept of EMH and agency approach under it, dwelling on possible agency costs from misalignment of owner's interests with the agent's ones, and possible agency gains from the same misalignment.

It has to be noted that significantly less studies found purely positive effects (agency gains) of state involvement [Jiang et al., 2008; Liao, Young, 2012], and more studies identified various either negative [Sappington, Stiglitz, 1987; Shleifer, Vishny, 1994; Boycko et al., 1996] or mixed [Wei et al., 2005; Hess et al., 2010; Alfaraih, Alanezi, Almujamed, 2012] effects. The main reasons for negative effects listed as agency conflict, and positive impact attributed to lower risk perception. It has also been found that transition economies may show different effects [Huang, Xiao, 2012]. The most interesting explanation provided by the current literatures is the one contrasting the «helping hand» with the «grabbing hand» of the government, which might take advantage of its controlling position and abuse the earnings [Sappington, Stiglitz, 1987]. Thus, the main assumption in exploring the value relevance of government ownership is the cost effect.

We will connect to the stream of literature of purely market effects of governmental ownership, examining the latter as an independent variable in market valuation [Alfaraih, Alanezi, Almujamed, 2012; Razak, Saidi, Mahat, 2013] in the widely used valuation model by Ohlson (1995), which serves to measure certain value effects [Ohlson, 1995; Hassel, 2005; Loh et al., 2017]. We will also connect to literature on the concentration of ownership effect.

Our goal is to explore the effect of government ownership of firm equity in Russia. In doing so, we have developed three hypotheses.

Hypothesis 1 (H1): Firms with any degree of government ownership (> 0) have lower total market values than purely privately owned firms.

H1 explores the market value relevance in association with government control, connecting to the EMH and previous research supporting the expectation of agency conflict [Eckel, Vermaelen, 1986] and the «grabbing hand» effect by the investor [Huang, Xiao, 2012].

Hypothesis 2 (H2): Firms with higher concentration of government ownership have lower market value values than firms with lesser degree of government control.

H2 connects to the concept of ultimate control [Wang, Xiao, 2009] where the degree of government control is ranging between 0 and 100% assuming that less state control translates into better performance of firms in terms of profitability and productivity [Tran et al., 2014].

Hypothesis 3a (H3a): Firms with higher concentration of government ownership (over 25%) have higher market values than firms with the concentration of government ownership less than 25%.

Hypothesis 3b (H3b): Firms with higher concentration of government ownership (over 50%) have higher market values than firms with the concentration of government ownership less than 50%.

H3 tests the concentration of ownership effect to see if it will overpower the effect of ultimate control, with the cutoff values of 25% and 50% of the equity ownership [Husain, Hazoor, Sabir, 2014].

The empirical analysis relies on a sample of 159 firms out of the target population of 260 corporations listed in the Russian financial market. We used data derived from the Bloomberg database including those on government ownership, market value, and other variables. Data analysis is done with the help of statistical software IBM SPSS and R.

In this study, we are using a derivative of Ohlson's (1995) model of a firm's market value relation to accounting data and other information. In finance and accounting research Ohlson's valuation model is viewed as a conventional tool in determining value-relevance of various data [Hassel, 2005; Loh et al., 2017; Lourenco, Eugenio, 2011].

The model is using a regression tool to determine the nature of the relationship between variables and the impact of independent variables on the intercept. As long as the data on market value is heterogeneous, we are using the weighted least squares regression method. Our model derivation is based on the method used by Loh and Thomas (2017) with modifications in terms of ownership variable. We will test several specifications of the model with the baseline having the following shape:

In the model MV_{i t}+₄ is the market value four months after the financial year-end of company i; BV_{i t} is the book value of common equity at the year-end of company i; EARN_{i t} is earnings before extraordinary items at the year-end of company i; NEG_{t t} is a dummy variable equal to 1 if at the t year end the firm had losses and 0 if otherwise, and ei, t is the error term. We are including book value and earnings, because in previous research [Loh et al., 2017; Hassel, 2005; Ohlson, 1995] book value shows a positive relationship with market value, earnings, in contrast, it can show a negative relationship with market value, because profit is usually rewarded by the market and loss is usually penalized [Brecht et al., 2018]. First we run it as Model (1) and test its specifications.

As the next step we will include the ownership variable first as a dummy variable, with 0 - if the firm had no state ownership and 1 - if there was some degree of it (dummy variable OWNj _t), to find the link between the ownership and the market value to accept/reject the hypothesis H1 (Model 2). To test the hypothesis 2 (H2) we explore the relationship between the degree of government ownership and market value replacing the dummy variable O WN_{t t} with a continuous variable, the share of equity owned by the government (between 0 and 100%) and produce model (3). After that we will generate the model (4), containing a control variable; for the control variable we chose to use the firms belonging to service or manufacturing sector, the INDj _t variable, which we assign the value - «0» if its service and «1» if the company works in the manufacturing sector.

Assessing the model specification and predicting power we will examine the R² for the three models and the p-values for the significance of the relationships, we are expecting to see a value of determination coefficient of more than 0,5, which would mean that the model can predict 50% of the variance of market value by the change in book value, earnings and profit/loss (model 1), we expect the value of R² to remain relatively constant for subsequent model specifications, at the significance level of p below 0,05. Therefore if the association proves to be statistically significant (p-value below 0,05) we will accept that the association did not happen by chance; if otherwise we will accept the null hypothesis (no significant relationship).

1019 firms in total are listed in Russian financial markets, however out of those firms only 260 are active and report financials. And only 159 of the active 260 firms have enough values reported for analysis. The selected 159 firms are characterized by high market values, have varied earnings and mostly larger book values of equity. The distribution of firms by main variables is positively skewed with a number of large outliers.

Table 1 summarizes the summary statistics of all variables used in this research describing the central tendency and the dispersion of variables used in the model.

Table 1. Summary statistics of variables