Still Potent, Dividends Remain the Strongest Signal to Predict Blue-Chip Stock Prices in Indonesia’s LQ45 Index

Image Illustration by AI

FORMOSA NEWS - Jakarta - Amidst the dynamic landscape of Indonesia's post-pandemic capital market, investors are frequently faced with uncertainty when evaluating the intrinsic value of market leaders. Addressing this challenge, a new study released in May 2026 evaluates the effectiveness of stock valuation models in Indonesia. This collaborative research was spearheaded by Icha Mustamin from Politeknik Indonesia, alongside Amiruddin and Darmawati from Universitas Hasanuddin, and Endang Sriningsih from Universitas Teknologi Akba.

They discovered that cash dividend distributions remain the primary psychological anchor for the market in determining the stock price movements of highly liquid companies within the LQ45 index. These findings are crucial for market participants looking to filter out the best investment strategies to minimize risk amid market fluctuations.

The Dilemma of Dividends Versus Retained Earnings in Emerging Markets

Capital markets in emerging economies like Indonesia possess unique and highly dynamic characteristics. Many large enterprises listed on the LQ45 index have begun altering their corporate profit management policies, opting to retain earnings or execute share buybacks instead of distributing regular dividends. This phenomenon has sparked intense debate among financial analysts: is the traditional dividend-based formula still relevant, or is the modern earnings-based model far more accurate?

Theoretically, two popular valuation models are often compared. The first is the Dividend Discount Model (DDM), a classical approach asserting that a stock's real value stems purely from its future dividend cash flows. The second is the Residual Income Valuation (RIV) model, which focuses heavily on a company's ability to generate economic profit after accounting for the cost of capital—making it ideal for firms with inconsistent dividend patterns. Through the file "759-770 Icha Mustamin.pdf", the researchers put these two models to the test on LQ45 stocks.

Sifting Through Blue-Chip Financial Data with Statistics

To prove which model reigns supreme, the research team employed a comparative quantitative research design. They collected secondary data consisting of annual financial reports and stock price movements from emiten that consistently maintained their positions within the LQ45 index on the Indonesia Stock Exchange.

The sample was selected using a purposive sampling method, filtering out companies that possessed comprehensive financial data throughout the observation period—including net income, book value of equity, dividends per share, and closing market prices. This numerical data was then processed computationally using SPSS statistical software via a multiple linear regression approach. The predictive accuracy of both models was measured and compared using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) metrics.

Key Findings: The Dominant Power of Dividends Per Share

The data analysis revealed several fascinating insights for the national investment landscape:

• Strong Combined Explanatory Power: Simultaneously, net income, book value of equity, and dividends demonstrate high value relevance, explaining 65.8% of the variance in LQ45 stock price movements.
• Dividends Are King: When tested partially (individually), dividends per share emerged as the most dominant and statistically significant variable influencing stock prices.
• Net Income Blindsided by the Market: Conversely, net income and book value of equity independently showed no significant impact on stock prices at the time official reports were published.

The statistical insignificance of standalone net income is driven by a market trend where stock prices generally move ahead of time to price in earnings information long before official statements are released (prices lead earnings). Based on prediction accuracy, the Dividend Discount Model (DDM) proved to be much more relevant and possessed stronger predictive power compared to the RIV model in the context of LQ45 companies.

Strategic Implications for Investors and Public Policy

"The results of this study indicate that investors in the Indonesian capital market still heavily rely on actual dividends as the primary signal of a company's stability, management commitment, and fundamental quality," wrote Icha Mustamin and her team in the scientific paper.

For the general public and retail investors, this research provides valuable financial literacy, showing that one should not merely be enticed by high paper profits but should look closely at an emiten's consistency in paying out cash dividends. For corporate managers, these findings serve as an alarm that dividend policies are highly sensitive in shaping market perceptions and stock valuations. Meanwhile, for stock market regulators, this empirical data can be utilized to strengthen investor protection policies, particularly regarding the transparency of accounting information and profit distribution mandates.

Author Profiles

• Icha Mustamin, S.E., M.Si. (Lead & Corresponding Author) – Academic and researcher at Politeknik Indonesia, focusing on Financial Accounting, Capital Markets, and Stock Valuation.
• Prof. Dr. Amiruddin, S.E., M.Si., Ak., CA. – Senior lecturer and researcher at Universitas Hasanuddin, an expert in Financial Accounting and Financial Statement Analysis.
• Dr. Darmawati, S.E., M.Si., Ak. – Academic at Universitas Hasanuddin, specializing in Behavioral Accounting, Value Relevance, and Capital Market Theory.
• Endang Sriningsih, S.E., M.M. – Lecturer and researcher at Universitas Teknologi Akba, focusing on Financial Management and Investment Strategy.

Research Publication Source

Journal Article Title: Comparative Analysis of the Residual Income Valuation Model and the Dividend Discounting Model in Predicting Stock Prices of LQ45 Companies

Journal Name: Indonesian Journal of Advanced Research (IJAR), Vol. 5, No. 5, Pages 759-770
Publication Year: 2026
Official DOI: https://doi.org/10.55927/ijar.v5i5.16520