Diversification and Corporate Performance

Sample Selection

The sample covers the period running from the first quarter of 2009 to the third quarter of 2015. The sample includes the listed firms in the "coal", "oil and refinery", "wind power" and "solar power" sectors according to the CITIC Industrial Classification Standard developed by CITIC Group Corporation (The paper uses listed companies in four energy sectors due to China's energy structure, sectorial market levels and disclosed company information.). Of these firms, the paper first eliminates special-treated firms (Special-treated firms have suffered losses for two consecutive fiscal years.) marked as "ST" and "*ST", "other refinery" firms listed under the "oil and refinery" sector, and firms with no connections to the energy industry. Firms in the "wind power" and "solar power" sectors are counted as the renewable energy industry, and firms in the "coal" and "oil and refinery" sectors are counted as the conventional energy industry. As a result, the paper compiled a sample of 102 firms. Accounting and product information are available from the WIND database, which is a financial terminal developed by Wind Info (www.wind.com.cn), and firm financial reports. The accounting database (WIND) is commercially available from the Chinese Academy of Sciences.

Firms are considered as industrially diversified (multi-segment) if they operate in more than one ISIC (International Standard Industrial Classification, Revision 4) three-digit code industry. Single-segment and specialized firms are those that operate within a single ISIC three-digit code industry. Moreover, diversified firms operating within a single ISIC two-digit code are considered industrially diversified, and diversified firms are considered unrelated to industrial diversification if they operate within more than one ISIC two-digit code industry. Firms are considered internationally diversified when they operate in overseas and domestic markets.

In addition, the paper eliminates and corrects outliers with rates of change that exceed 50%.

Measures of Industrial Diversification, International Diversification, and Corporate Performance

In measuring industrial diversification, the paper constructs four alternative proxies. In the industrial diversification proxies, the first three measures are sales-based entropy (EI), the adjusted Herfindahl index (HHI), and the specialization ratio (Rs) (see Table 1). Another measure (Segment) measures the total number of segments in which a certain firm engages. In addition, Related is a dummy variable that takes a value of 1 if a firm engages in related industrial diversification and a value of 0 otherwise.

Table 1. Indicators of different business strategies.

Notes: $P_{i}$ is the proportion of the sales categorized under a single ISIC three-digit code industry in a certain firm. $Q_{i}$ is the proportion of sales of a firm's overseas business.

In measuring international diversification, the paper uses three alternative proxies. Among these proxies, Roverseas is the proportion of the overseas sales of a certain firm. The other two measures are the entropy ( MarketE) and Herfindahl index (MarketH) based on overseas sales.

Higher values of EII HHI, Segment, MarketH, MarketE and Roverseas denote higher levels of diversification. Higher values of Rs denote lower levels of industrial diversification. Calculation methods and explanations of these proxies are shown in Table 1.

Within the scope of corporate performance and value management, the corporate evaluation system is a core issue. Generally speaking, the evaluation of corporate performance considers many indicators that reflect corporate profitability, solvency and growth, and so on. According to corporate performance and value management principles, companies should optimize their operations to maintain good performance in all respects. However, the paper only used returns on assets (ROA), chief returns on assets (CROA), market-to-book (M2B), and Tobin's Q (TQ) as proxies of corporate performance, as these four proxies reflect corporate profitability, which is the basis of a company's survival and growth in competitive markets.

ROA and CROA measure a firm's level of investment efficiency. ROA is the ratio of earnings before interest and tax (EBIT) to total assets for a certain period. CROA is the ratio of the sum of operating revenue and interest expenses to total assets for a certain period. ROA can alleviate the effects of corporate capital structures and tax policies on net returns, and CROA can also eliminate profit manipulation in non-core businesses for net returns. M2B and TQ represent a firm's market value. M2B is the ratio of the market value to the book value of shareholder equity. TQ is the ratio of the sum of the market value of shareholder equity and the book value of liabilities to total assets.

Market value proxies reflect capital market expectations of future firm profit and growth.

As many studies show, company market value is not usually consistent with corporate fundamentals. The paper uses this indicator to verify the empirical results based on the perspectives of corporate fundamentals and capital markets. In addition, all corporate performance proxies are adjusted by subtracting industrial medians.

Model

The paper constructs the regression models as follows. Equation (1) examines the correlation between business strategies and corporate performance, and Equation (2) analyzes differences in the correlation above among conventional and renewable energy firms.

Performance = α+β₁·Ln (asset)+β₂·EBIT_Sales +β₃·Exp_Sales

(1) +β₄·Diversification

Performance = α+β₁·Ln (asset)+β₂·EBIT_Sales +β₃·Exp_Sales

(2) +β₄·Diversification +β₅·Industry· +β₆·Diversification·Industry

We use three variables to control for the effects of firm characteristics. The first variable is the natural logarithm of the total asset (Ln(asset)), which measures a firm's size. The second variable is EBIT, which measures a firm's profitability (EBIT_Sales). The third variable is the capital expenditure, which measures a firm's investments (Exp_Sales). For example, larger companies can adjust or improve their operations with access to more resources, such as cash flows, potentially rendering operations more efficient. Therefore, EBIT_Sales and Exp_Sales are scaled by firm's sales to alleviate the fixed effects of firm size.

Summary Statistics

As Table 2 shows, the investment efficiency and profitability of China's conventional energy companies exceed that of China's renewable energy companies. However, the market value of China's renewable energy companies is higher than that of China's conventional energy companies, showing that the capital market holds appreciated expectations of renewable energy companies. China's conventional energy companies are larger than renewable energy companies.

Table 2. Descriptive statistics on listed Chinese energy firm characteristics.

In an industrial structure, conventional energy companies tend to diversify more than renewable energy companies regardless of EI, HHI, Segment, and Rs. For industrially-diversified energy companies, the business strategies of renewable energy companies are more closely related than those of conventional energy companies. From a business distribution perspective, China's renewable energy companies are more internationalized than China's conventional energy firms. China's large overseas market for renewable energy companies may explain their position as original equipment manufacturers (OEMs) for European and American countries.