Exploring the relationship between dividend policy, the COVID-19 crisis, and stock market reaction: empirical insights from Indonesian real estate and property firms

Abstract

The economic turbulence experienced during the COVID-19 crisis in Indonesia in 2020 provides a backdrop for this study, which aims to investigate the relationship between the crisis and dividend policies within the real estate and property sectors in Indonesia, along with the stock market’s response to corporate announcements. The examination involves multiple robustness tests, including the incorporation of various proxy measures for the primary variables augmented with control variables to fortify the model, as well as a sub-period robustness check to assess the relationship between the crisis and dividend policies, and market reactions. The findings of this research indicate that companies tended to adopt a negative dividend policy during times of crisis. The sub-period robustness check consistently reveals that dividend policies are positively correlated with macroeconomic conditions, a pattern also observed in the pre-crisis period. The statistical analysis tools used are dynamic panel data regression for testing the causality of crisis and dividend policy, and also one-sample T-test for testing stock market reaction. Furthermore, the results suggest a positive market response to dividend announcements during the 2020 crisis. A comparison with the pre-crisis year 2019 and the post-crisis year 2021 indicates that these comparative periods did not exhibit a similar positive reaction as observed during the crisis. Consequently, this study offers implications for real estate and property sector companies in Indonesia to consider adopting negative dividend policy to maintain the company’s survival amidst a crisis full of uncertainty.

Keywords:

• Dividend policy
• stock market reaction
• crisis
• covid-19
• real estate and property

JEL Classifications:

• C33
• G01
• G35

REVIEWING EDITOR:

• David McMillan, University of Stirling, United Kingdom of Great Britain and Northern Ireland

Introduction

The global economic landscape has been significantly disrupted by the COVID-19 pandemic crisis, manifesting in a forceful and precipitous decline in stock valuations and heightened levels of volatility within international markets, including Indonesia (Mazur et al., 2021). In response to the pandemic, the Indonesian government initially enacted measures to restrict human mobility, aiming to curtail viral transmission. Consequently, this imposed constraints on the movement of goods and disturbed the intricate cycles of trade and commerce. As a consequence, economic activity has been impeded, resulting in a state of crisis (Sholahuddin et al., 2021). In the Indonesian context, the ramifications of the COVID-19 pandemic materialized as an economic crisis in the year 2020. Indications of this crisis event became evident through the trajectory of Indonesia’s economic performance (Ssenyonga, 2021). In 2020, Indonesia experienced a stark decline in Gross Domestic Product (GDP) growth, plummeting to −2.07%. The assessment of GDP growth has proven to be a reliable metric for evaluating the impacts of the pandemic-induced crisis (Gunay & Can, 2022).

As documented by Tinungki, Hartono, et al. (2022), a nearly perfect Pearson correlation of 0.999 underscores the strong relationship between GDP growth and a binary dummy variable that represents crisis and non-crisis conditions, and they are robust to measure the crisis. However, given the recent resurgence in economic activities, a period of post-crisis recovery is anticipated by 2021, evident in the economy’s expansion, as indicated by a 3.69% upswing in GDP growth. Moreover, the inflation rate stood at 1.68%, signifying a decline in comparison to prior years. This underscores the subdued nature of the inflation phenomenon in the year 2020. This phenomenon can be ascribed to the prevailing belief that during that year, there was a conspicuous trend among individuals to deposit their funds in banks and curtail their expenditures, primarily due to the uncertainties surrounding the pandemic’s duration and resolution. Furthermore, the IDX composite index exhibited a sustained decline throughout 2020, reaching its nadir at 4194 on March 20, 2020. This underscores the pivotal notion that the descent of the capital market during the crisis induced by the pandemic is of paramount significance.

Multiple empirical investigations have underscored the adverse impacts of the COVID-19 pandemic on the performance of stock markets in various countries, as noted by Ashraf (2021); Cepoi (2020); Owusu and Bentum-ennin (2021); and Mazur et al. (2021). Moreover, analogous trends have emerged in different financial domains, including commodities, cryptocurrencies, and equity markets, as elaborated upon by Ahmed and Sarkodie (2021) and Conlon and McGee (2020), and also by Mazumder and Saha (2021). The Indonesian context reflects this pattern, where the adverse influence of the pandemic has reverberated across diverse sectors, encompassing the stock market, as documented by Utomo and Hanggraeni (2021); the domain of cryptocurrencies, as scrutinized by Gunawan and Anggono (2021); and within equity markets, as expounded by Kamaludin et al. (2021). Additionally, the question of dividend policy as a means of return on stock investments emerges as a significant inquiry, particularly during periods of crisis, when the capital market experiences a downturn.

Numerous empirical investigations have underscored the adverse effects of the COVID-19 pandemic on the stock market performance in Indonesia, as reported by Indrayono (2021); and Utomo and Hanggraeni (2021). Moreover, the crisis has impacted the financial market in Indonesia, as documented by Indrayono (2021). Beyond Indonesia, this impact has been evident in various countries, as demonstrated in stock markets worldwide, as reported by Chaudhary et al. (2020); and Engelhardt et al. (2021). Similar repercussions have been observed in commodity markets, as reported by Ferguson and Ubilava (2022), cryptocurrency markets, as reported by Akhtaruzzaman et al. (2022), and financial markets, as reported by Zhang et al. (2020). Additionally, questions regarding dividend policies as a means of stock investment return have emerged as significant concerns, particularly during crisis periods when the capital markets experience downturns.

Amid the COVID-19 pandemic crisis, several empirical investigations have explored its specific impact on dividend policies. Xu et al. (2023) reported that transportation and entertainment companies in China tended to reduce dividend payouts during the crisis. Kluzek and Schmidt-Jessa (2022) documented that companies in Poland curtailed dividend payments amidst the crisis, even though they received crisis assistance from the government. Similarly, Ali et al. (2022) observed diminished dividend payments in Pakistan. Furthermore, Boumlik et al., (2023) identified a negative trend in dividend policies among Moroccan firms. Conversely, some companies displayed a positive stance towards dividend policies. This was evidenced by Tinungki, Robiyanto, et al. (2022), who found that companies in Indonesia adhered to a positive dividend policy. Additionally, Tinungki, Hartono, et al. (2022) reinforced these findings among 24 green-listed companies in Indonesia. In parallel, Hartono and Raya (2022) affirmed that manufacturing companies in Indonesia adopted a positive dividend policy during the crisis. Moreover, a study conducted by Mazur et al. (2023) uncovered that companies entering the S&P1500 index tend to maintain or even increase their dividends during times of crisis. These findings for negative dividend policy in crisis compare to non-crisis condition align with the pecking order theory, which posits that companies prioritize stability and sustainability over dividend payouts during crisis periods. Otherwise, these findings for positive dividend policy in crisis align with the dividend signaling theory, which suggests that during crises, companies tend to maintain or increase their dividend levels to send positive signals to the market regarding their performance and sustainability amid turbulent times (Damodaran, 2015; Fumey & Doku, 2013; Jensen, 1986; Tinungki, Robiyanto, et al., 2022).

Moreover, in the context of such crises, it is imperative to scrutinize how the market responds to the distribution of dividends. This action serves as an indicator of the signals that companies convey to the market regarding their performance and sustainability during times of crisis. Additionally, it aligns with the perception held by investors, which tends to favor the ‘bird-in-the-hand’ theory, viewing dividends as a reliable return on their stock investments. Dividends are perceived as a guaranteed return, making them more responsive compared to capital gains, which experience high market volatility during crises (Ashraf, 2021; Cepoi, 2020; Hartono & Robiyanto, 2023). Research pertaining to stock market reactions to dividend announcements during crises was conducted by Anwar et al. (2017), who reported a positive market response to cash dividend announcements during the 2008–2009 crisis. Subsequently, Kumar (2017) found that dividend announcements conveyed positive signals, with increased dividends correlating with higher stock prices, whereas companies maintaining the previous year’s dividend levels exhibited no significant market response. Amid the COVID-19 crisis, Robiyanto and Yunitaria (2022) demonstrated a lack of positive market response among companies indexed in LQ-45 in Indonesia. Conversely, Pandey and Kumari (2022) identified a dearth of positive responses to corporate actions during the 2020 crisis among companies indexed in BSE 500. Furthermore, Tinungki, Robiyanto, et al. (2022) substantiated a positive market reaction to dividend policy during the COVID-19 crisis among non-financial companies in Indonesia. On the other hand, Tinungki, Hartono, et al. (2022), and Hartono and Raya (2022) confirmed that stock market reactions were muted concerning these corporate announcements during the COVID-19 crisis, respectively, for green-listed companies in Indonesia and manufacturing companies in Indonesia.

Relying on documented observations and empirical evidence, there arises an imperative need to investigate dividend policy in times of crises (Krieger et al., 2021; Tinungki, Robiyanto, et al., 2022). In the context of crisis-induced dividend policy, it is equally vital to explore how the market responds to dividend declarations within this critical timeframe, aligning with the established dividend policy. Can the distribution of dividends convey positive market signals? Particularly, if positive reactions are discernible, it is plausible that these dividends could play a constructive role in mitigating capital market distress stemming from prevailing market pessimism (Cejnek et al., 2021; Robiyanto & Yunitaria, 2022). Consequently, the exploration of dividend policy during the COVID-19 crisis, coupled with the subsequent analysis of market responses to such corporate actions, can effectively elucidate the impact of dividend policy in alleviating capital market distress during crises.

Therefore, this research endeavors to delve into dividend policy within the Indonesian real estate and property sector during the COVID-19 pandemic, concurrently scrutinizing market reactions to dividend announcements. The rationale for selecting this sector lies in its significant contribution to Indonesia’s economy. These companies are intricately involved in various other industries in their business operations, including banking, manufacturing, finance, transportation, and more. As reported by Grahadyarini (2023), this industrial sector has proven to support Indonesia’s economy by contributing approximately IDR 2.349 trillion to IDR 2.865 trillion per year, accounting for 14.63%–16.3% of the national GDP in total. Moreover, this industrial sector has demonstrated its interconnection with 185 other sub-industries in Indonesia. This underscores the pivotal role played by the real estate and property sector in the national economy of Indonesia, even during times of crisis (Hartono and Matusin, 2020; Hartono et al., 2021). On the other hand, some studies related to dividend policies in Indonesia have demonstrated that during a crisis, manufacturing companies, 212 non-financial companies, and companies included in the SRI-KEHATI index, named the Green Index, have established positive dividend policies, meaning they maintain or increase dividend payouts during the crisis. This is supported by Hartono and Raya (2022); Tinungki, Hartono, et al. (2022); and Tinungki, Robiyanto, et al. (2022). However, this positive dividend policy is not consistently met with a positive response from the stock market in Indonesia, and this is supported by Robiyanto and Yunitaria (2022) for LQ-45 companies.

Therefore, this study makes a substantial contribution to the existing body of literature. Firstly, it represents an evaluation of dividend policy within Indonesian real estate and property companies in the context of the COVID-19 crisis. This investigation conducts a comprehensive assessment, including rigorous examinations of variable measurements and sub-period consistency. To bolster the study’s robustness, empirical analysis employs a Dynamic Panel Data Regression approach, utilizing the System-Generalized Method of Moments estimation method with a two-step estimator technique. The utilization of this analytical tool is more pertinent compared to static panel data regression since it accommodates the dynamic nature of dividend policy (Tinungki, Hartono, et al., 2022; Tinungki, Robiyanto, et al., 2022). Secondly, this research provides a more extensive assessment of how the capital market responds to dividend declarations during the 2020 COVID-19 crisis. Furthermore, we conducted market reaction assessments for the years 2019 and 2021 for comparative purposes, thus yielding more robust implications. Henceforth, the research outcomes signify that real estate firms, amid the COVID-19 crisis, exhibited an unfavorable stance towards dividend policies, as indicated in the primary empirical models. Within the purview of sub-period robustness examinations, compelling evidence substantiates that the models for the 2014–2020 and 2014–2021 periods underscore the detrimental impact of dividend policies during the pandemic crisis. Furthermore, the findings disclose that, during the tumultuous year of 2020, there ensued a positive market response to dividend declarations, contrasting with the scenarios in the pre-crisis year of 2019 and the post-crisis year of 2021.

Literature review

Pecking order theory vs. dividend signaling theory: perspectives of dividend policy in the crisis

In moments of crisis, particularly during the COVID-19 pandemic, enterprises tend to strategize on how to anticipate the economic downturn. The pandemic-induced crisis, stemming from restrictions on the mobility of people and goods, has impeded the business and economic cycles. The situation in Indonesia was compounded by the absence of definite measures to combat the pandemic, leading to a crisis in 2020, evident in the economic growth represented by a year-on-year GDP growth rate of 2.07%. However, at the outset of 2021, the management of the pandemic began showing promise with the emergence of effective vaccines capable of suppressing virus transmission. Gradually but steadily, transmission rates were curbed, and public response to virus infections strengthened. This prompted the government to gradually ease movement and trade restrictions, subsequently restoring the flow of the business and economic cycles. As a result, the economy showed signs of improvement in 2021, reflected in a positive year-on-year GDP growth of 3.69% (Guedhami et al., 2022; Khoirunurrofik et al., 2022; Prasasti & Ekananda, 2023; Tinungki, Robiyanto, et al., 2022).

Amidst economic uncertainty, particularly in 2020, businesses tend to prioritize survival and sustainability. The decline in corporate performance during crisis scenarios is a natural outcome due to disruptions in the business cycle. Consequently, a more comprehensive examination of companies’ net earnings becomes imperative to determine whether they opt to distribute dividends or retain them as retained earnings. Following the pecking order theory, companies typically give precedence to internal funding sources over external ones due to their lower inherent risks. Furthermore, internal funding sources entail a lower cost of capital when compared to external sources like debt, bond issuance, and equity. Consequently, during the COVID-19 crisis, it is an ideal strategy for companies to limit or suspend dividend payouts to ensure their survival. Companies are inclined to preserve their earned profits, particularly in times of crises marked by downturns in the capital market, making external funding sources such as equities less attractive for the company (Ali et al. 2022; Fassas et al., 2021; Lim, 2016; Myers, 1984; Robiyanto & Yunitaria, 2022; Tinungki, Robiyanto, et al., 2022).

Despite the prevailing crisis, some companies adopt a positive dividend policy stance. Corporations employ this strategy to mitigate information asymmetry among shareholders and the market regarding their long-term growth prospects. The Dividend Signaling Theory asserts that the dissemination of dividend distribution information serves as a substantial signal of a company’s performance and growth trajectory. The Agency Theory also lends support to the notion that emphasizing or withholding dividends during periods of deteriorating company performance is linked to the personal interests of company management. The continuation or augmentation of dividend distributions is believed to maintain a positive signal to the market. Conversely, reduced dividend distributions can have adverse repercussions on market reactions (Ali, 2022; Baker et al., 2015; Hartono & Raya, 2022; Miller & Rock, 1985; Robiyanto & Yunitaria, 2022).

Dividend policy in the crisis due to COVID-19 pandemic

The challenges brought about by the COVID-19 pandemic, subsequently leading to a crisis, have profound implications for the financial strategies employed by corporations. The crisis has had a significant impact on the capital market, resulting in substantial fluctuations in stock prices and creating short-term uncertainties in returns, particularly in terms of capital gains. Consequently, companies are actively devising strategic policy concerning this external source of funding. Additionally, the decline in corporate performance due to the crisis also influences various corporate strategies. In light of these circumstances, companies need to formulate policy related to their profitability and equity, especially in the context of shares (Chowdhury et al., 2022; Ellul et al., 2020; Mohammad, 2022). When corporate profitability diminishes, leading to reduced net earnings during crises, companies must consider reducing or even forgoing the distribution of dividends to shareholders as returns on their stock investments (Ali et al., 2022). Instead of distributing dividends, retaining net earnings as retained earnings emerges as an appropriate policy to ensure corporate resilience amidst the uncertainties triggered by the crisis (Cejnek et al., 2021). Moreover, until the end of 2020, a year marked by the crisis, economic uncertainty has been exacerbated by uncertainties in pandemic management, primarily stemming from restrictions on the movement of people and goods (Krieger et al., 2021; Tinungki, Robiyanto, et al., 2022).

On the other hand, companies should carefully consider the signals they convey to the market regarding the dividends distributed to shareholders. This is crucial because companies must address the information asymmetry between shareholders and the market concerning the company’s performance and long-term outlook. Companies must set dividend levels that are at least equal to those of the previous period or even increase dividend levels to maintain a positive signal to the market. A reduction or elimination of dividends during certain periods sends a negative signal to investors about the company’s performance and sustainability. Furthermore, companies that distribute dividends can mitigate agency conflicts within the organization, as the management of net earnings is rife with conflicts of interest (Ali, 2022; Baker et al., 2015; Hartono & Raya, 2022; Tinungki, Hartono, et al., 2022).

One of the key macroeconomic indicators reflecting the economic condition is the growth of the gross domestic product (GDP). This metric serves as an illustration of the economic growth potential within a nation. This aspect is considered relevant as an indicator of the impact of the COVID-19 crisis on dividend policy, as demonstrated by several empirical investigations conducted by Hartono and Raya (2022); and Tinungki, Hartono, et al. (2022). Moreover, dividend policy can be assessed using two criteria: firstly, through the dividend per share, which measures the dividend received by shareholders per share held, and secondly, through the dividend payout ratio, which compares the dividend per share to earnings per share (Damodaran, 2015; Hartono, Sari, et al., 2021; Tinungki, Robiyanto, et al., 2022; Zutter & Smart, 2019). Previous research has reported a positive impact of the crisis on dividend policy, indicating a reduction or even elimination of dividends in response to economic downturns during the COVID-19 crisis, as evidenced by Ali et al. (2022); Kluzek and Schmidt-Jessa (2022); and Xu et al. (2023). In contrast, Ali (2022); Hartono and Raya (2022); and Tinungki, Robiyanto, et al. (2022) have proved that companies have adopted a positive dividend policy during the crisis. Therefore, based on logical reasoning, prior scholarship, and a rigorous examination of dividend policy using some proxies, the following hypotheses are formulated:

H1: Gross Domestic Product growth has an impact on Dividend per Share.

H2: Gross Domestic Product growth an impact on Dividend Payout Ratio.

Additionally, the study examined robustness of the influence of the crisis induced by the COVID-19 pandemic on dividend policy, and to enhance the reliability of the GDP proxy, binary dummy variables were utilized. The categorization of periods into crisis and non-crisis states followed the approach previously employed by Hartono and Raya (2022); Sari (2017); Tinungki, Hartono, et al. (2022); and Tinungki, Robiyanto, et al. (2022). These dummy variables were assigned a value of 1 to signify crisis conditions and 0 to denote non-crisis conditions. As a result, the hypotheses were formulated as follows:

H3: Crisis has an impact on Dividend per Share.

H4: Crisis has an impact on Dividend Payout Ratio.

Stock market reaction to dividend announcement the COVID-19 pandemic crisis

The act of companies distributing dividends to their shareholders as a return on their stock investments is believed to have an impact on the stock market. This is because the distribution of dividends is seen as a favorable signal to the market concerning the company’s performance and growth trajectory. The stock market’s response to dividend announcements can be gauged through the presence of positive abnormal returns and cumulative abnormal returns in the vicinity of the dividend announcement. During times of crisis, the stock market’s reaction tends to be more subdued compared to normal conditions. This phenomenon is attributed to investor tendencies to prioritize cash holdings, as the adage ‘cash is king’ holds true in times of crisis and uncertainty (Chang & Yang, 2022; Hartono & Raya, 2022; Khanal & Mishra, 2017; Mahata et al., 2021; Mirbagherijam, 2014; Robiyanto & Yunitaria, 2022).

Several previous studies reported the stock market reaction to dividend announcement in crisis era. Anwar et al. (2017) observed a positive reaction of market to dividend distributions during the 2008–2009 crisis. Specifically, with regard to the COVID-19 crisis, Tinungki, Robiyanto, et al. (2022) documented a positive response by stock market, characterized by strong abnormal returns and cumulative abnormal returns following dividend announcements by Indonesia firms during the COVID-19 crisis. Dividends are widely perceived as a dependable return on stock investments during economic downturns. Additionally, Khanal and Mishra (2017) documented a positive response of stock market to dividend announcements during the 2008–2009 crisis. However, Pandey and Kumari (2022) found that the market’s reaction was weaker during the 2020 crisis compared to the pre-crisis period in 2019. In alignment with these findings, Tinungki, Hartono, et al. (2022); and Robiyanto and Yunitaria (2022) have reported a similar trend for SRI-KEHATI indexed companies and LQ-45 companies in Indonesia. Therefore, we posit the following hypotheses:

H5: There exists a significant abnormal return in the vicinity of dividend announcements.

H6: There exists a significant cumulative abnormal return in the vicinity of dividend announcements.

Research methods

This study employs a quantitative methodology to empirically examine the formulated hypotheses. The assessment of how the COVID-19 crisis influenced dividend policy involves testing the causal relationships among variables within the established model. To investigate the stock market’s reactions to dividend announcements, an event study is conducted to analyze abnormal returns during the days surrounding dividend announcements. Secondary data is sourced from Indonesia Stock Exchange, the Bloomberg terminal, and finance.yahoo.com. The research period spans from 2014 to 2021. The sample selection process utilizes purposive sampling. For the analysis of the COVID-19 impact on dividend policy, specific criteria are applied: real estate and property sector companies listed on the Indonesia Stock Exchange that made at least one dividend payment during the observation period and had no instances of initial public offering or delisting. Comprehensive financial reports were mandatory for measuring variables. Additional criteria for testing stock market reactions involved the exclusion of companies with compounding events during the event window, avoidance of delays or substantial revisions, and the avoidance of actions such as stock splits or acquisitions that could influence abnormal returns (Anwar et al., 2017; Hartono & Raya, 2022; Robiyanto & Yunitaria, 2022; Sekaran & Bougie, 2016; Tinungki, 2019).

Hence, out of the 766 firms listed on the IDX in 2021, 79 were in the real estate and property sector. Therefore, a sample of 33 real estate and property firms was selected for the observation period. Over the 8-year duration, 264 observations were available for analyzing the impact of the crisis on dividend policy. Moreover, the event study encompassed 16 companies in 2020, 22 in 2019, and 9 in 2021. In addition to conducting thorough evaluations that compare various indicators for assessing the impact of the COVID-19 crisis on dividend policy, this study employs an additional robustness check approach known as sub-period robustness check (Lu & White, 2014). These sub-period robustness checks are divided into three intervals: the years 2014–2019, which serve as the pre-crisis benchmark; the years 2014–2020, encompassing both the pre-crisis period (2014–2019) and the crisis year (2020); and 2014–2021, spanning the pre-crisis (2014–2019), crisis (2020), and post-crisis (2021) periods. Subsequently, the analysis of the stock market’s response to dividend announcements during crises extends over an eleven-day event window, covering five days before the announcement and five days after. This assessment of stock market reactions extends over a three-year period for comparative purposes: 2019 as the pre-crisis year, 2020 as the year of the crisis, and 2021 as the post-crisis phase. The selection of this three-year era is justified to facilitate a comprehensive discussion of the findings (Lu & White, 2014).

In assessing the influence of the crisis on dividend policy, there is a primary independent variable and a primary dependent variable. As part of one of the robustness checks, the main predictor is the COVID-19 crisis, measured using year-on-year GDP growth and a binary dummy variable. Additionally, the proxy response variables utilized include Dividend per Share, and also Dividend Payout Ratio (Sari, 2017; Sharma, 2021). Subsequently, this analysis employs several control independent variables that are hypothesized to influence dividend policy, as supported by studies conducted by Ranajee et al. (2018); Hartono and Robiyanto (2023); Singla and Samanta (2018); Tinungki, Hartono, et al. (2022); Wahjudi (2020); and Tinungki, Robiyanto, et al. (2022). Recent research validating these hypotheses includes variables such as profitability, financial leverage, firm size, and firm age. Additionally, previous year’s dividend is employed as an instrumental variable to test the empirical model within the context of the dividend policy model.

Therefore, to measure the dividend policy variable, it is assessed using Dividend per Share (DPS), calculated as follows: $\mathit{\text{DPS}}=\frac{\mathit{\text{Total Dividend}}}{\mathit{\text{Outstanding Shares}}}$ (Hartono & Raya, 2022; Lestari, 2018). It is also measured using the Dividend Payout Ratio (DPR), which is calculated as $\mathit{\text{DPR}}=\frac{\mathit{\text{Dividend per Share}}}{\mathit{\text{Earning per Share}}}$ (Hartono, Sari, et al., 2021; Sharma, 2021). To measure the COVID-19 pandemic crisis variable, it is assessed using the year-on-year Gross Domestic Product (GDP) growth for each observation year (Hartono & Raya, 2022; Tinungki, Robiyanto, et al., 2022). It is also measured using a Binary Dummy Variable (BD) with a code of 1 for crisis conditions and 0 for non-crisis conditions (Sari, 2017; Tinungki, Robiyanto, et al., 2022). The first control variable, profitability, is measured using Earnings per Share (EPS), calculated as $\mathit{\text{EPS}}=\frac{\mathit{\text{Net Income}}}{\mathit{\text{Outstanding Shares}}}$ (Sharma, 2021; Sharma & Bakshi, 2019). The second control variable, financial leverage, is measured using the Debt to Equity Ratio (DER), calculated as $\mathit{\text{DER}}=\frac{\mathit{\text{Total Liability}}}{\mathit{\text{Total Equity}}}$ (Anggraeny, Robiyanto, & Sakti, 2020; Wahjudi, 2020). The third control variable, firm size, is measured using Total Assets (TA), calculated as $\mathit{\text{TA}}=\mathit{\text{ln}}\left(\mathit{\text{Total Asset}}\right)$ (Hartono, Wijaya, et al., 2023; Tinungki, Hartono, et al., 2022). The fourth control variable, company age, is measured as the square root transformation of company’s age since its establishment (AGE), calculated as the square root of firm age. The selection of the transformation of the firm size variable is based on the ladder of powers measurement, which is the most optimal and statistically significant above 5% (Badu, 2019; Sharma & Bakshi, 2019). Furthermore, instrumental variables in the dynamic panel data regression estimation using DPS and DPR measurements have been conducted in previous research by (Bostanci, Kadioglu, & Sayilgan, 2018; Hartono & Raya, 2022; Sharma, 2021).

Consequently, we proceed to assess the influence of dividend announcements as a predictor variable on stock prices, the dependent variable. The analysis of stock prices adopts a daily-based methodology. An event study is conducted to scrutinize the presence of significant abnormal returns (AR) and cumulative abnormal returns (CAR) during dividend announcement events. The event study analysis period encompasses five days preceding the dividend announcement (t − 5) up to one day before the announcement, including the announcement day itself, followed by one day after (t + 1) up to five days after the announcement (t + 5). AR and CAR are computed by evaluating realized returns, according to Ashraf (2021); Hartono and Raya (2022); Khanal and Mishra (2017); and Robiyanto and Yunitaria (2022).

The proxy formulation for calculating realized return is $R_{i,t}=\frac{P_{i,t}-P_{i,t-1}}{P_{i,t-1}}$, where, $R_{i,t}$: realized return on the $i$-th company and the $t$-th day; $R_{i,t}$: realized return on the $i$-th company and the $t$-th day; $P_{i,t}$: adjusted close price on the $i$-th company and the $\left(t\right)$-th; and $P_{i,t-1}$: adjusted close price on the $i$-th company and the $\left(t-1\right)$-th day. The formulation for calculating expected return is ${E\left(R\right)}_{i,t}=\frac{{\mathit{\text{IHSG}}}_t-{\mathit{\text{IHSG}}}_{t-1}}{{\mathit{\text{IHSG}}}_{t-1}}$, where ${E\left(R\right)}_{i,t}$: expected return on the $i$-th company and the $t$-th day; ${\mathit{\text{IHSG}}}_t$: IDX composite on the $\left(t\right)$-th day; and ${\mathit{\text{IHSG}}}_{t-1}$: IDX composite on the $\left(t-1\right)$-th day. Moreover, the formulations for calculating the abnormal return and cumulative abnormal return are ${\mathit{\text{AR}}}_{i,t}=R_{i,t}-{E\left(R\right)}_i$ and ${\mathit{\text{CAR}}}_{i,t}\left(t,K\right)={\sum }_{t=-5}^K{\mathit{\text{AR}}}_{i,t}$, where ${\mathit{\text{AR}}}_{i,t}$: abnormal return on the $i$-th company and the $t$-th day; and ${\mathit{\text{CAR}}}_{i,t}\left(t,K\right)$: cumulative abnormal return on the $i$-th company and the $t$-th day; $t=-5,-4,\dots ,+4,+5$.

The investigation into the influence of the crisis on dividend policies employed Dynamic Panel Data Regression. The method chosen for parameter estimation is the System-Generalized Method of Moments (SYS-GMM) with the Two-Step estimator technique. The examination commenced with a model specification test, encompassing an instrument validity assessment using the Sargan test, an autocorrelation evaluation using the Arellano Bond test, and an unbiased examination comparing the lagged-1 parameter coefficients of the SYS-GMM estimation method, the Ordinary Least Square Robust (OLSRob) method, and the Least Square Dummy Variable Robust (LSDVRob) method. Moreover, the test of parameter significance was conducted through both a simultaneous assessment using the Wald Chi-Square test and a partial examination employing the Z-test to scrutinize the research hypotheses (Baltagi, 2005; Biørn, 2017; Sharma, 2021). Consequently, this analysis is structured into four regression models, formulated based on the consistency of proxy robustness checks. Models 1 and 2, as presented in Equations (1)(1) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(1) and (2)(2) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(2) , assess the impact of the crisis measured by GDP growth (Equation (1)(1) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(1) ) and a binary dummy variable (Equation (2)(2) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(2) ) on dividend policies measured by DPS. Furthermore, models 3 and 4, as displayed in Equations (3)(3) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(3) and (4)(4) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(4) , gauge the effect of the crisis measured by GDP growth (Equation (3)(3) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(3) ) and a binary dummy variable (Equation (4)(4) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(4) ) on dividend policies measured by DPR. The empirical models are represented by the following equations: (1) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(1) (2) $${\mathit{\text{DPS}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPS}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(2) (3) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{GDP}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(3) (4) $${\mathit{\text{DPR}}}_{i,t}={\alpha }_{i,t}+\delta {\mathit{\text{DPR}}}_{i,t-1}+{\beta }_1{\mathit{\text{BD}}}_{i,t}+{\beta }_2{\mathit{\text{EPS}}}_{i,t}+{\beta }_3{\mathit{\text{DER}}}_{i,t}+{\beta }_4{\mathit{\text{TA}}}_{i,t}+{\beta }_5{\mathit{\text{AGE}}}_{i,t}+{\epsilon }_{i,t}$$(4)

In the provided equation: ${\mathit{\text{DPS}}}_{i,t}$: represents the dividend per share for the $i$-th entity and the $t$-th time; ${\mathit{\text{DPR}}}_{i,t}$: indicates the dividend payout ratio for $i$ -th entity and the $t$-th time;; ${\mathit{\text{DPS}}}_{i,t-1}$: signifies the dividend per share for the $i$-th entity and the $\left(t-1\right)$-th time; ${\mathit{\text{DPR}}}_{i,t-1}$: denotes the dividend payout ratio for the $i$-th entity and the $\left(t-1\right)$-th time; ${\mathit{\text{GDP}}}_{i,t}$: stands for the gross domestic product for the $i$-th entity and the $t$-th time; ${\mathit{\text{BD}}}_{i,t}$: represents the binary dummy variable for the COVID-19 crisis for the $i$-th entity and the $t$-th time; ${\mathit{\text{EPS}}}_{i,t}$: signifies the earnings per share for the $i$-th entity and the $t$-th time; ${\mathit{\text{DER}}}_{i,t}$: indicates the debt-to-equity ratio for $i$-th entity and the $t$-th time;${\mathit{\text{TA}}}_{i,t}$: represents the total assets for the $i$ -th entity and the $t$-th time; ${\mathit{\text{AGE}}}_{i,t}$: denotes the firm’s age for the $i$-th entity and the $t$-th time; ${\epsilon }_{i,t}$: represents the error for the $i$ -th entity and the $t$-th time;${\alpha }_{i,t}$: signifies the intercept for the $i$-th entity and the $t$-th time; ${\beta }_1,{\beta }_2,\dots ,{\beta }_5$:: represent the slopes of the predictors; $\delta$: indicates the slope of the instrumental variable.

Data processing for parameter estimation was carried out using STATA version 14, with Single-user 8-core perpetual license. Conversely, when addressing the regression model and testing the causal relationship, a challenge arises due to endogeneity, which may lead to biased and inconsistent parameter estimation. Previous studies have reported the impact of GDP on profitability (Ongore & Kusa, 2013) and the influence of financial size and firm leverage on profitability (Bangun et al., 2017). Additionally, Sunardi et al. (2020) provided evidence of firm size affecting financial leverage. These findings collectively highlight the inherent endogeneity in both financial leverage and profitability. However, Li (2016) demonstrated that the generalized method of moments represents the parameter estimation approach with the most substantial corrective impact when addressing endogeneity concerns among exogenous variables, outperforming alternative methods. Furthermore, this estimation technique has been shown to yield the most appropriate parameter coefficients. It effectively addresses endogeneity by utilizing lagged-1 endogenous variables as instruments for the endogenous variable. Hence, this challenge can be overcome with the generalized method of moments (Biørn, 2017; Chinoda & Kwenda, 2019; Li, 2016). Furthermore, to assess the market’s response to dividend announcements through an event study, the analysis employed a one-sample T-test (Hair et al., 2018; Robiyanto & Yunitaria, 2022).

Result and discussion

Descriptive statistics for each proxy measuring variables in dynamic panel data regression estimation are presented in Table 1. Extreme data distributions are evident in the DPS, DPR, BD, EPS, and DER proxies, as all five proxies exhibit overdispersion characteristics. In contrast, GDP, TA, and AGE display equidispersion characteristics. This indicates that when overdispersion is present, which means that the mean is smaller than the standard deviation, the data is more heterogeneous. On the other hand, when equidispersion is observed, meaning that the mean is greater than the standard deviation, the data is more homogeneous. Hence, DPS, DPR, BD, EPS, and DER exhibit higher heterogeneity compared to GDP, TA, and AGE (Hartono, Tinungki, et al., 2021). Moreover, negative DPR values indicate a corresponding negative EPS value under those conditions. Negative GDP values signify a year-on-year negative GDP growth, serving as an indicator of a crisis. Furthermore, the TA and AGE values are in decimal form for both maximum and minimum values due to transformation applied to each proxy measurement, as explained in their respective proxy measurements.

Table 1. Descriptive statistics in the dynamic panel data regression model.

Proxy	Min	Max	Mean	Stand. Dev
DPS	0.000	503.162	27.164	70.949
DPR	−0.529	8.326	0.179	0.544
GDP	−0.021	0.052	0.040	0.023
BD	0.000	1.000	0.125	0.331
EPS	−1616.927	3190.693	132.443	373.990
DER	0.084	35.466	1.399	2.394
TA	13.804	18.639	16.014	1.155
AGE	3.371	8.307	6.089	0.942

The Pearson correlation estimates between the tested variables are presented in Table 2. Moderate correlations are observed between DPS and DPR as proxies for dividend policy, indicating the suitability of both proxies in measuring consistency within the two variables representing a response. Furthermore, a strong correlation is evident between GDP and BD, illustrating GDP’s ability to represent crisis and non-crisis conditions as depicted by the dummy variable. The correlations among the predictors reveal values less than 0.7, thereby indicating the absence of multicollinearity.

Table 2. Correlation between two variable proxies.

Proxy	DPS	DPR	GDP	BD	EPS	DER	TA	AGE
DPS	1.000
DPR	0.539	1.000
GDP	0.076	0.092	1.000
BD	−0.082	−0.090	−0.982	1.000
EPS	0.395	0.017	0.217	−0.200	1.000
DER	−0.056	0.017	−0.030	0.032	−0.341	1.000
TA	−0.090	−0.107	−0.059	0.051	0.165	0.165	1.000
AGE	0.196	0.081	−0.102	0.082	0.112	0.112	0.245	1.000

The findings of dynamic panel data regression estimation aim to assess the impact of the COVID-19 crisis on dividend policy. Table 3 presents parameter estimates for Models 1 and 2 over the period from 2014 to 2021. The results of the model specification tests for both Model 1 and Model 2 using the SYSGMM method meet the required conditions, including the Sargan test, the Arellano-Bond test, as well as the unbiasedness test when compared against LSDVRob and OLSRob estimates. Furthermore, the model specification tests indicate that the estimated models are well-fitted, as demonstrated by the simultaneous Wald-test and partial Z-test. Subsequently, estimations of the main variables reveal that GDP, as a proxy of crisis, positively influences DPS, whereas the binary dummy variable (BD) exhibits a negative impact on DPS, as estimated using the SYSGMM method.

Table 3. Dynamic panel data regression estimation to assess the impact of the COVID-19 crisis on dividend policies using DPS as a proxy of response variable, covering the period of 2014–2021.

Variable	Model 1			Model 2
	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob
Lagged of DPS	−0.060 (0.243)	0.038*** (0.002)	0.517*** (0.155)	−0.057 (0.245)	−0.027*** (0.002)	0.519*** (0.156)
GDP	276.239** (105.149)	446.543*** (7.886)	91.667 (120.062)	------	------	------
BD	------	------	------	−18.384** (7.755)	−25.654*** (0.453)	−10.101 (8.031)
EPS	0.023 (0.143)	0.045*** (0.001)	0.044* (0.023)	0.023 (0.024)	0.047*** (0.001)	0.043* (0.023)
DER	0.193 (1.952)	4.378*** (0.066)	1.463 (1.351)	0.212 (1.967)	4.610*** (0.059)	1.441 (1.345)
TA	4.189 (16.752)	−103.808*** (1.236)	−4.681** (2.287)	5.097 (16.837)	−102.874*** (0.971)	−4.673** (0.014)
AGE	27.737 (26.881)	177.953*** (2.148)	7.723** (3.169)	22.013 (25.897)	170.402*** (1.904)	7.786** (3.224)
Constant	−221.801 (338.687)	577.168*** (16.661)	31.228 (36.452)	−188.203 (340.033)	628.290*** (14.305)	35.732 (37.570)
Estimation description
Number of Observations	231	231	231	231	231	231
Number of Groups	33	33	------	33	33	------
N umber of Instrument	------	33	------	------	33	------
Model specification test
Sargan → ${\chi }^2$	------	32.390*	------	------	32.290	------
Arellano Bond
Order-1	------	−1.396	------	------	−1.390	------
Order-2	------	1.499	------	------	1.493	------
Parameter significance test
Wald	------	4191.37***	------	------	8330.06***	------
F	3.00***	------	8.97***	3.30**	------	8.41***
Adj-$R^2$	0.023	------	0.352	0.022	------	0.353

The values within parentheses indicate the standard deviation for each estimation. The symbol (***) denotes significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

Furthermore, we present the estimated parameters for models 3 and 4 over the period from 2014 to 2021, as displayed in Table 4. The findings of the model specification assessments for both of these models, employing the SYSGMM technique, adhere to the requisite criteria. These criteria encompass the Sargan examination, the Arellano-Bond evaluation, and the assessment for unbiased when juxtaposed with the LSDVRob and OLSRob estimations. Additionally, the model specification evaluations signify the appropriateness of the estimated models, as affirmed through the concurrent Wald test and the partial Z-test. In subsequent analyses, estimations of the principal variables divulge that GDP, serving as a crisis proxy, exerts a positive influence on DPR. Conversely, the binary dummy variable (BD) demonstrates a negative effect on DPR, as appraised through the SYSGMM method.

Table 4. Dynamic panel data regression estimation to assess the impact of the COVID-19 crisis on dividend policies using DPR as a proxy of response variable, covering the period of 2014–2021.

Variable	Model 3			Model 4
	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob
Lagged of DPR	−0.157*** (0.010)	−0.016*** (0.002)	0.517** (0.155)	−0.154*** (0.011)	−0.011*** (0.002)	0.074** (0.029)
GDP	3.041* (1.782)	4.433*** (0.181)	91.667** (120.062)	------	------	------
BD	------	------	------	−0.178* (0.091)	−0.247*** (0.010)	−0.155*** (0.057)
EPS	−0.000 (0.000)	0.000*** (0.000)	−0.000 (0.000)	0.000 (0.000)	0.000*** (0.000)	−0.000 (0.000)
DER	0.012 (0.028)	0.056*** (0.004)	0.006 (0.014)	0.012 (0.028)	0.059*** (0.004)	0.006 (0.014)
TA	−0.307 (0.313)	−1.375*** (0.039)	−0.066*** (0.024)	−0.295 (0.309)	−1.390*** (0.041)	−0.066*** (0.024)
AGE	0.317 (0.384)	1.374*** (0.088)	0.068*** (0.013)	0.235 (0.331)	1.335*** (0.089)	0.066*** (0.012)
Constant	3.065 (2.742)	13.590*** (0.420)	0.715** (0.362)	3.520 (3.072)	14.278*** (0.443)	0.834** (0.403)
Estimation description
Number of Observations	231	231	231	231	231	231
Number of Groups	33	33	------	33	33	------
Number of Instrument	------	33	------	------	33	------
Model specification test
Sargan → ${\chi }^2$	------	31.129	------	------	30.761	------
Arellano Bond
Order-1	------	−1.039	------	------	−1.032	------
Order-2	------	−0.683	------	------	−0.664	------
Parameter significance test
Wald	------	2190.62***	------	------	2536.56***	------
F	361.02***	------	13.70***	331.19***	------	13.47***
Adj-$R^2$	−0.005	------	0.012	−0.006	------	0.012

The values within parentheses indicate the standard deviation for each estimation. The symbol (***) denotes significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

The results obtained from estimating Models 1, 2, 3, and 4 collectively guide us in making decisions regarding the acceptance of H1, H2, H3, and H4. Hence, it is robustly evidenced that GDP has a positive impact on dividend policy, whereas the Binary Dummy Variable exerts a negative influence on dividend policy. This indicates that during the COVID-19 pandemic crisis in 2020, real estate and property firms tended to curtail or even eliminate dividends distributed to shareholders. These findings align with reports by Ali et al. (2022); Kluzek and Schmidt-Jessa (2022); and Xu et al. (2023). Moreover, this comprehensive and pertinent outcome reinforces the research presented by Hartono, Tinungki, et al. (2023). The research findings strongly suggest that during times of crisis, real estate and property firms tend to curtail dividend levels as a strategy to preserve their financial health. This inclination arises due to reduced business activities and macroeconomic downturns during such periods, compounded by the uncertainty surrounding the duration of the crisis.

To bolster the findings in our primary model, we conducted a robustness check using not only alternative proxies for measuring the main variables but also a Sub-Sample Robustness Check. Estimations for the period 2014–2020 were structured under Models 1a, 2a, 3a, and 4a, while for the period 2014–2019, they were modeled under Models 1b and 3b. Models 2b and 4b were not available due to the binary dummy variable that represents crisis and non-crisis conditions, which could not be adequately captured with the pre-crisis (0) and during-crisis (1) coding differentiation.

Models 1a, 2a, 3a, and 4a are presented in Tables 5 and 6. The estimation results for these four models pass the model specification tests, namely, the instrument variables are valid, there is no second-order autocorrelation, and they pass the unbiased test against the OLSRob and LSDVROb estimation. The significance tests of the parameter estimates using simultaneous tests indicate that the estimated models are well-fitted. Furthermore, the estimation results for the main variables in all four models consistently align with the primary model, where GDP positively affects dividend policy, and BD negatively impacts dividend policy. This supports H1, H2, H3, and H4, suggesting that during crises, real estate and property sector firms in Indonesia tend to adopt a negative dividend policy, i.e. they reduce or even eliminate dividend payouts (Kluzek & Schmidt-Jessa, 2022; Xu et al., 2023). Furthermore, we conducted a sub-sample robustness check to investigate dividend policy behavior before the crisis, which is modeled in Models 1b and 3b.

Table 5. Dynamic panel data regression estimation to assess the impact of the COVID-19 crisis on dividend policies according to model 1, 2, and 3, covering the Period of 2014–2020.

Proxy	Model 1a			Model 2a			Model 3a
	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob
Lagged of DPS	−0.212 (0.169)	0.026*** (0.007)	0.501*** (0.153)	−0.213 (0.169)	0.025*** (0.007)	0.501*** (0.153)	------	------	------
Lagged of DPR	------	------	------	------	------	------	−0.216*** (0.013)	−0.025*** (0.003)	0.070** (0.030)
GDP	273.094* (145.575)	389.167*** (28.972)	122.204 (111.988)	------	------	------	3.791 (2.860)	4.093*** (0.400)	2.380** (1.006)
BD	------	------	------	−19.368* (10.267)	−27.151*** (2.042)	−8.334 (7.831)	------	------	------
EPS	0.025 (0.023)	0.038*** (0.004)	0.036* (0.020)	0.025 (0.023)	0.038*** (0.003)	0.036* (0.020)	−0.000 (0.000)	0.000 (0.000)	−0.000 (0.000)
DER	−0.274 (1.614)	4.020*** (0.168)	1.803 (1.196)	−0.282 (1.616)	4.018*** (0.166)	1.807 (1.199)	0.008 (0.023)	0.050*** (0.005)	0.008 (0.016)
TA	7.235 (10.601)	−124.608*** (3.279)	−3.955* (2.032)	7.526 (10.518)	−123.806*** (3.232)	−3.949* (2.030)	−0.323 (0.329)	−1.400*** (0.052)	−0.066** (0.027)
AGE	25.765 (35.003)	157.911*** (11.872)	5.937** (2.665)	26.339 (35.387)	157.912*** (11.826)	5.926** (2.665)	0.573 (0.712)	1.224*** (0.165)	0.066*** (0.013)
Constant	−254.383 (214.383)	1047.842*** (94.549)	27.337 (32.349)	−248.800 (214.539)	1054.54*** (91.500)	33.377 (34.004)	1.752 (1.227)	14.948*** (0.826)	0.727* (0.391)
Estimation description
Number of Observations	198	198	198	198	198	198	198	198	198
Number of Groups	33	33	------	33	33	------	33	33	------
Number of Instrument	------	26	------	------	26	------	------	26	------
Model specification test
Sargan → ${\chi }^2$	------	23.079	------	------	22.874	------	------	23.467	------
Arellano Bond
Order-1	------	−1.281	------	------	−1.281	------	------	−1.012	------
Order-2	------	1.076	------	------	−1.074	------	------	0.797	------
Parameter significance test
Wald	------	3071.73***	------	------	9690.61***	------	------	2924.40***	------
F	5.22***	------	11.10***	5.23***	------	11.11***	550.50***	------	15.39***
Adj-$R^2$	−0.031	------	0.393	−0.031	------	0.393	−0.015	------	0.006

The values within parentheses indicate the standard deviation for each estimation. The symbol (***) denotes significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

Table 6. Dynamic panel data regression estimation to assess the impact of the COVID-19 crisis on dividend policies according to model 4, covering the period of 2014–2020, and according to model 1 and 3, covering the period of 2014–2019.

Proxy	Model 4a			Model 1b			Model 3b
	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob	LSDVRob	SYSGMM	OLSRob
Lagged of DPS	------	------	------	−0.488*** (0.057)	−0.024 (0.025)	0.543*** (0.172)	------	------	------
Lagged of DPR	−0.217*** (0.013)	−0.026*** (0.003)	0.070** (0.030)	------	------	------	−0.216*** (0.013)	−0.092*** (0.028)	0.070** (0.029)
GDP	------	------	------	2378.147 (5792.414)	6961.508*** (1578.31)	4237.843 (5626.107)	3.791 (2.860)	157.589*** (26.396)	80.444 (82.227)
BD	−0.265 (0.201)	−0.278*** (0.028)	−0.163** (0.066)	------	------	------	------	------	------
EPS	0.000 (0.000)	0.000 (0.000)	0.000 (0.000)	0.027 (0.022)	0.033*** (0.010)	0.038 (0.025)	−0.000 (0.000)	0.000 (0.000)	−0.000 (0.000)
DER	0.008 (0.023)	0.050*** (0.005)	0.008 (0.016)	−1.611 (1.122)	1.983*** (0.456)	2.113 (1.484)	0.008 (0.023)	0.027** (0.011)	0.008 (0.018)
TA	−0.318 (0.326)	−1.392*** (0.053)	−0.066** (0.026)	36.718*** (11.430)	−129.157*** (10.812)	−3.667 (2.695)	−0.323 (0.329)	−1.305*** (0.246)	−0.077** (0.038)
AGE	0.575 (0.714)	1.212*** (0.166)	0.066*** (0.013)	−9.306 (42.132)	146.267*** (18.543)	7.056** (3.141)	0.573 (0.712)	0.600** (0.293)	0.067*** (0.014)
Constant	1.853 (1.274)	15.083*** (0.822)	0.845* (0.431)	−611.148** (299.405)	857.242*** (195.332)	−193.372 (251.35)	1.752 (1.227)	9.753*** (3.088)	−3.024 (3.566)
Estimation description
Number of Observations	198	198	198	165	165	165	165	165	165
Number of Groups	33	33	------	33	33	------	33	33	------
Number of Instrument	------	26	------	------	20	------	------	20	------
Model Specification Test
Sargan → ${\chi }^2$	------	23.447	------	------	20.493*	------	------	15.963	------
Arellano Bond
Order-1	------	−1.012	------	------	−1.112	------	------	−1.047	------
Order-2	------	0.795	------	------	1.208	------	------	1.180	------
Parameter Significance Test
Wald	------	2712.02***	------	------	1555.81***	------	------	274.51***	------
F	545.60***	------	15.24***	205.56***	------	11.25***	2006.31***	------	13.88***
Adj-$R^2$	−0.015	------	0.005	0.134	------	0.413	−0.038	------	0.003

The values within parentheses indicate the standard deviation for each estimation. The symbol (***) denotes significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

The estimation results for Models 1b and 3b are presented in Table 6. Model specification tests confirm that these models meet the required conditions, including the validity of instrument variables, the absence of second-order autocorrelation, and lack of bias in comparison to LSDVRob and OLSRob estimations. Simultaneous tests indicate that the estimated models are well-fitted. Furthermore, the estimation of the main variables consistently aligns with the primary model, showing that GDP has a positive impact on dividend policy, whether measured by DPS or DPR. This suggests that, during the pre-COVID-19 crisis period, dividend policies of real estate and property firms in Indonesia tend to be sensitive to the country’s economy, as evidenced by economic growth. Therefore, a similar trend is observed during crisis conditions, indicating that real estate and property companies in Indonesia consistently adopt dividend policies that follow the economic situation, which is positively correlated with its movement. These research findings are consistent with those reported by Ong et al. (2018).

This is supported descriptively by dividend per share (DPS) and dividend payout ratio (DPR) during the research period. The DPS values according to the observed sample averages during that period were IDR 23.37 in 2014, IDR 21.80 in 2015, IDR 31.51 in 2016, IDR 26.84 in 2017, IDR 40.33 in 2018, IDR 25.39 in 2019, IDR 11.78 in 2020, and IDR 36.30 in 2021. Furthermore, these results are corroborated by the sample’s observed average DPR values, which were 17.09% in 2014, 15.13% in 2015, 21.63% in 2016, 15.11% in 2017, 38.93% in 2018, 10.97% in 2019, 4.96% in 2020, and 19.01% in 2021. This indicates that in 2020, there was a significant difference in the dividend per share compared to other years, suggesting a decrease in the dividend amount for each outstanding share in 2020. Furthermore, the dividend payout ratio also exhibited a lower ratio in 2020 compared to other years, indicating a lower dividend payment relative to net income earned.

In addition, an analysis of the instrumental variable, namely the lagged-1 of the endogenous variable, was conducted. Models 1, 2, 3, 4, 1a, 2a, 3a, and 4a yielded inconsistent results. In models 1, 2, 3, and 4, which analyzed the period from 2014 to 2021, the coefficients were positive in model 1 and negative in models 2, 3, and 4. This suggests that the evidence from models 2, 3, and 4, which predominantly exhibit a negative impact, indicates that the previous year’s dividend has a negative effect on the current period’s dividend. This further supports that during the 2020 crisis, companies set lower dividend policies compared to 2019. Moreover, this is corroborated by the evidence in model 3b, showing that even before the crisis, dividend policies were negatively determined, and this behavior remained consistent throughout the crisis. Additionally, the evidence in models 1a, 2a, 3a, and 4a is balanced, where evidence with the DPS proxy tends to have a positive effect, while evidence with the DPR proxy tends to have a negative impact. These negative effects also support the evidence in models 2, 3, 4, and 3b. Therefore, these results suggest the irrelevance of the dividend signaling theory to dividend policies that can be considered signals to the market. Consequently, it is important to examine how the market responds to dividend policies during the crisis period and compare it to the pre-crisis and post-crisis periods to assess the relevance of this signaling theory. Furthermore, even though dividend policies are negatively adopted during crises, it is essential to examine the stock market’s reaction to these corporate actions.

Tabulated Descriptive Statistics for Evaluating Stock Market Responses to Dividend Announcements appear in Table 7. This table encompasses the range of Abnormal Return (AR) and Cumulative Abnormal Return (CAR) values, highlighting their maximum and minimum values across the years 2019, 2020, and 2021, observed on each announcement day. Additionally, mean and standard deviation statistics are included to characterize the average and variability within the data. The analysis reveals that the AR and CAR data for each observation day in 2019, 2020, and 2021 exhibit overdispersion (Sáez-Castillo & Conde-Sánchez, 2013), indicating significant heterogeneity. Moreover, there is a declining trend in the number of dividend announcement events, decreasing from 22 events in 2019 to 16 events in 2020 and further down to 9 events in 2021. These observations highlight a reduction in dividend distribution events in 2020 compared to the preceding year, followed by another decrease in 2021 relative to 2020.

Table 7. Calculation of descriptive statistics for abnormal returns and cumulative abnormal returns in testing stock market response to dividend announcements.

Year			Periods
			$D_{-5}$	$D_{-4}$	$D_{-3}$	$D_{-2}$	$D_{-1}$	$D$	$D_{+1}$	$D_{+2}$	$D_{+3}$	$D_{+4}$	$D_{+5}$
2019	$\mathit{\text{AR}}$	n	22	22	22	22	22	22	22	22	22	22	22
		Mean	0.013	−0.001	−0.019	−0.003	−0.000	−0.005	0.006	0.028	0.000	−0.008	−0.007
		St.Dev.	0.044	0.015	0.049	0.024	0.024	0.017	0.027	0.072	0.033	0.014	0.024
		Max.	0.147	0.033	0.043	0.050	0.047	0.025	0.053	0.250	0.117	0.017	0.023
		Min.	−0.047	−0.030	−0.207	−0.069	−0.049	−0.052	−0.079	−0.047	−0.038	−0.040	−0.083
	$\mathit{\text{CAR}}$	n	22	22	22	22	22	22	22	22	22	22	22
		Mean	0.013	0.012	−0.007	−0.010	−0.011	−0.016	−0.010	0.018	0.019	0.010	0.003
		St.Dev.	0.044	0.049	0.057	0.063	0.059	0.068	0.074	0.079	0.087	0.093	0.086
		Max.	0.147	0.159	0.154	0.150	0.146	0.159	0.167	0.177	0.247	0.238	0.189
		Min.	−0.047	−0.052	−0.100	−0.169	−0.135	−0.151	−0.154	−0.129	−0.132	−0.172	−0.158
2020	$\mathit{\text{AR}}$	n	16	16	16	16	16	16	16	16	16	16	16
		Mean	0.006	0.004	−0.006	0.014	0.005	0.022	0.008	0.002	0.012	−0.006	−0.011
		St.Dev.	0.026	0.024	0.026	0.032	0.026	0.066	0.049	0.037	0.056	0.029	0.027
		Max.	0.072	0.059	0.028	0.102	0.065	0.206	0.077	0.070	0.159	0.090	0.045
		Min.	−0.018	−0.032	−0.066	−0.018	−0.056	−0.069	−0.089	−0.050	−0.043	−0.047	−0.073
	$\mathit{\text{CAR}}$	n	16	16	16	16	16	16	16	16	16	16	16
		Mean	0.056	0.010	0.004	0.018	0.023	0.045	0.053	0.055	0.067	0.061	0.050
		St.Dev.	0.026	0.028	0.038	0.049	0.060	0.086	0.092	0.100	0.110	0.114	0.103
		Max.	0.072	0.066	0.083	0.111	0.141	0.181	0.201	0.251	0.258	0.236	0.219
		Min.	0.006	0.010	0.004	0.018	0.023	0.045	0.053	0.055	0.067	0.061	0.050
2021	$\mathit{\text{AR}}$	n	9	9	9	9	9	9	9	9	9	9	9
		Mean	0.001	−0.000	−0.009	−0.013	−0.009	−0.000	−0.004	0.015	0.003	−0.006	0.004
		St.Dev.	0.011	0.010	0.010	0.016	0.010	0.023	0.011	0.031	0.016	0.029	0.019
		Max.	0.025	0.021	0.002	−0.001	0.002	0.036	0.013	0.094	0.033	0.020	0.042
		Min.	−0.010	−0.011	−0.027	−0.052	−0.027	−0.041	−0.020	−0.008	−0.023	−0.057	−0.025
	$\mathit{\text{CAR}}$	n	9	9	9	9	9	9	9	9	9	9	9
		Mean	0.001	0.001	−0.008	−0.021	−0.030	−0.029	−0.033	−0.017	−0.014	−0.020	−0.016
		St.Dev.	0.011	0.013	0.008	0.017	0.026	0.028	0.031	0.029	0.034	0.042	0.048
		Max.	0.025	0.020	0.003	−0.007	0.005	0.018	0.007	0.028	0.030	0.050	0.060
		Min.	−0.010	−0.020	−0.024	−0.064	−0.091	−0.065	−0.075	−0.056	−0.079	−0.077	−0.102

The event study, which examines the stock market’s reaction to dividend announcements by real estate and property companies in Indonesia, is presented in Table 8 for abnormal return analysis. In the year 2020, during the crisis, significant abnormal returns were observed only on two days before the announcement. There was no positive reaction on the announcement day or up to five days after the announcement. Comparatively, in the pre-crisis year of 2019, there were no positive abnormal returns during the observation window. Instead, negative abnormal returns were observed on the third day before the announcement and the fourth day after the announcement. Furthermore, when comparing the post-crisis period, there were no significant positive abnormal returns during the observation period. In fact, there were significant negative abnormal returns observed from the third day before the dividend announcement to one day before the dividend announcement. Thus, we accept H5.

Table 8. Estimation of abnormal returns using one-sample T-test for the period spanning 5 days prior to dividend announcement to 5 days after dividend announcement.

Period	2019			2020			2021
	Mean	$d.f$	T-stat	Mean	$d.f$	T-stat	Mean	$d.f$	T-stat
$D_{-5}$	0.013	21	1.381	0.006	15	0.901	0.001	8	0.394
$D_{-4}$	−0.001	21	−0.284	0.004	15	0.681	−0.000	8	0.100
$D_{-3}$	−0.019	21	−1.771*	−0.006	15	−0.954	−0.009	8	−2.587**
$D_{-2}$	−0.003	21	−0.664	0.014	15	1.790*	−0.013	8	−2.545**
$D_{-1}$	−0.000	21	−0.086	0.005	15	0.739	−0.009	8	−2.701**
$D$	−0.005	21	−1.444	0.022	15	1.337	−0.000	8	0.026
$D_{+1}$	0.006	21	1.022	0.008	15	0.659	−0.004	8	−1.027
$D_{+2}$	0.028	21	1.847	0.002	15	0.267	0.015	8	1.550
$D_{+3}$	0.000	21	0.022	0.012	15	0.862	0.003	8	0.603
$D_{+4}$	−0.008	21	−2.826***	−0.006	15	−0.823	−0.006	8	−0.598
$D_{+5}$	−0.007	21	−1.347	−0.011	15	−1.631	0.004	8	0.632

Parameter estimation for assessing mean differences utilizes a two-tailed statistical method. Symbols (***) represent significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

Subsequently, market reaction tests were conducted on cumulative abnormal returns, as presented in Table 9. In the year 2020, it was evident that there were significant positive cumulative abnormal returns starting from the dividend announcement day up to five days after the dividend announcement. This indicates a positive stock market reaction to dividend announcements. Therefore, we accept H6. These findings align with the research conducted by Tinungki, Robiyanto, et al. (2022), which demonstrated a similar phenomenon for 212 non-financial companies in Indonesia. Comparing to the pre-crisis period in 2019, there were no significant cumulative abnormal returns during the observation period. Furthermore, the comparison with the post-crisis period in 2021 indicates no significant positive cumulative abnormal returns during the observation window. In fact, there are significant negative cumulative abnormal returns observed starting three days before the dividend announcement until the dividend announcement day.

Table 9. Estimation of cumulative abnormal returns using one-sample T-test for the period spanning 5 days prior to dividend announcement to 5 days after dividend announcement.

Period	2019			2020			2021
	Mean	$d.f$	T-stat	Mean	$d.f$	T-stat	Mean	$d.f$	T-stat
$D_{-5}$	0.013	21	1.381	0.056	15	0.901	0.001	8	0.394
$D_{-4}$	0.012	21	1.137	0.010	15	1.436	0.001	8	0.260
$D_{-3}$	−0.007	21	−0.544	0.004	15	0.378	−0.008	8	−2.698**
$D_{-2}$	−0.010	21	−0.745	0.018	15	1.466	−0.021	8	−3.572***
$D_{-1}$	−0.011	21	−0.828	0.023	15	1.527	−0.030	8	−3.482***
$D$	−0.016	21	−1.085	0.045	15	2.088*	−0.029	8	−3.175**
$D_{+1}$	−0.010	21	−0.628	0.053	15	2.299**	−0.033	8	−3.236
$D_{+2}$	0.018	21	1.097	0.055	15	2.229**	−0.017	8	−1.803
$D_{+3}$	0.019	21	1.001	0.067	15	2.457**	−0.014	8	−1.270
$D_{+4}$	0.010	21	0.519	0.061	15	2.163**	−0.020	8	−1.415
$D_{+5}$	0.003	21	0.188	0.050	15	1.961*	−0.016	8	−0.999

Parameter estimation for assessing mean differences utilizes a two-tailed statistical method. Symbols (***) represent significance at the 1% level, (**) at the 5% level, and (*) at the 10% level.

The results of market reaction tests during the crisis period, compared to the pre-crisis and post-crisis periods, indicate that during crisis conditions, the stock market’s sensitivity to dividend distribution announcements is significantly higher than in non-crisis conditions. This is attributed to the uncertainty in stock investment returns during crises, where dividends are seen as a certain form of return. Furthermore, companies that distribute dividends during a crisis are perceived as having strong performance, which attracts investors, thus resulting in a positive reaction to dividend announcements (Anwar et al., 2017; Damodaran, 2015; Tinungki, Robiyanto, et al., 2022).

Conclusion

Evidence reveals that during COVID-19 crisis, companies operating in the real estate and property sectors in Indonesia typically adopt a negative dividend policy, characterized by the reduction or elimination of dividend distributions. This observation holds true in our primary model, which assesses several proxy variables supported by control variables, and is also substantiated in our sub-period robustness checks. Furthermore, our findings are consistent with pre-crisis conditions, where it is observed that GDP exerts a positive influence on dividend policy, thus aligning with crisis conditions. This phenomenon suggests that dividend policies within this sector adhere to the pecking order theory, emphasizing the prioritization of less risky internal financing, particularly during times of crisis marked by heightened uncertainty. Furthermore, this research also reveals that during crises, the stock market’s response to dividend announcements in this sector tends to be more positively reactive compared to the periods before the 2019 and 2020 crises. This demonstrates heightened stock market sensitivity during crisis times compared to both pre-crisis and post-crisis periods regarding corporate actions, thus affirming that corporate actions provide positive signals to the market during crises.

In general, this study is consistent with the research conducted by Xu et al. (2023); Kluzek and Schmidt-Jessa (2022); Ali et al. (2022); and Boumlik et al. (2023), which demonstrate that transportation and entertainment companies in China, companies in Poland, companies in Pakistan, and Moroccan firms that adopt a negative dividend policy, meaning they suppress or eliminate dividend policies during crises. Consequently, this study carries significant implications for companies operating in the real estate and property sector in Indonesia. It underscores the necessity for such firms to consider adopting a negative dividend policy during times of crisis to maintain corporate sustainability during times of crisis and uncertainty. Although some studies report that during crises, there are companies implementing positive dividend policies. Despite the positive market reactions during crisis periods in this research, expected to provide a positive signal to the stock market, the overall positive reaction is not confirmed, as reported by Hartono and Raya (2022); Robiyanto and Yunitaria (2022); and Tinungki, Hartono, et al. (2022), thus, the justification for implementing positive dividend policies is slightly irrelevant in enhancing stock market stability during a crisis.

Furthermore, it’s essential to acknowledge that this research comes with specific limitations. Notably, it restricts its analysis to the pre-crisis period, the crisis period, and only one year of the post-crisis phase, which corresponds to 2021. Given the limitations in data availability, the examination of corporate dividend policies during the post-crisis era remains unexplored, primarily due to the need for a balanced dataset spanning the pre-crisis and post-crisis periods. As a result, this study suggests expanding research efforts to include an investigation of post-crisis dividend policy dynamics once data adequacy for the post-crisis period is met. Additionally, this study has not explored dividend policies categorized based on dividend levels or rankings, as suggested by prior studies (Abdulkadir et al., 2015; Anggraeny et al., 2020). On the other hand, Mazur et al. (2023) provides specific evidence regarding the COVID-19 crisis conditions. During this crisis, companies significantly reduced share repurchases, particularly those heavily affected by the crisis. Therefore, further research can investigate the relationship between share repurchases and dividend policy during the COVID-19 crisis in Indonesia, as this aspect has not been previously studied during this crisis period.

Authors contributions

Conceptualization, B.U., H.S.L.; methodology, B.U., H.S.L., S.S., L.E.; software, S.S., L.E.; validation, B.U., H.S.L.; format analysis, S.S., L.E.; investigation, B.U., H.S.L., S.S., L.E.; resources, B.U.; data curation, B.U., H.S.L.; writing—original draft preparation, B.U., H.S.L., S.S., L.E.; project administration, B.U., H.S.L., S.S., L.E.; funding acquisition, B.U., H.S.L.; writing—review and editing, B.U., H.S.L., S.S., L.E. All of the authors have thoroughly perused and assented to the final version of the manuscript that has been made available for publication.

Acknowledgment

The authors would like to express their gratitude to Professor Georgina Maria Tinungki, from the Department of Statistics at Hasanuddin University, Indonesia, for her guidance and invaluable insights into the research methodology, which significantly contributed to the successful completion of this study.

Disclosure statement

In accordance with the principles of academic integrity, it is important to ensure that the work presented is free from any potential conflicts of interest. Thus, it is with great confidence that we disclose that the authors have reported no such conflicts in relation to the subject matter discussed in this work.

Data availability statement

The corresponding author can be contacted to request access to the datasets used and/or analyzed during the current study. The availability of the datasets is subject to reasonable requests.

Additional information

Funding

In terms of financial support, it should be noted that this research did not receive any particular grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes on contributors

Bahtiar Usman

Bahtiar Usman is a Professor of Management at the Faculty of Economics and Business, Universitas Trisakti, Jakarta, Indonesia. He earned his Ph.D. in Financial Management from Universitas Indonesia in 2012. Dr. Usman holds a faculty position at the Faculty of Economics and Business, Universitas Trisakti, where he is deeply committed to teaching and advancing the field of management science. His academic interests encompass various aspects of management, ranging from operational to strategic, financial, and risk management. His expertise in managerial finance extends to areas such as banking, investment, portfolio management, encompassing both fundamental and technical analyses, as well as considerations of Islamic and ethical principles. Collaborating with fellow academics, he actively contributes to the publication of scholarly articles in national and international journals, specializing in the field of Finance.

Henny Setyo Lestari

Henny Setyo Lestari holds the position of Assistant Professor in Corporate Finance and Banking at the Department of Management, Faculty of Economics and Business, Trisakti University. She earned her Ph.D. in Financial Management from Universitas Padjadjaran, Indonesia, with a focus on Corporate Finance. Her research interests primarily revolve around the fields of Corporate Finance and Banking.

Syofriza Syofyan

Syofriza Syofyan holds the position of Associate Professor in Monetary Economics at the Department of Development Economics, Faculty of Economics and Business, Trisakti University. She obtained his Ph.D. in Economics from Universitas Indonesia, specializing in Economics. Her research interests primarily focus on Monetary Economics and Econometrics.

Lavlimatria Esya

Lavlimatria Esya serves as an Assistant Professor in Islamic Economics at the Department of Development Economics, Faculty of Economics and Business, Trisakti University. She earned her Ph.D. from Trisakti University, focusing on Islamic Economics. Her research interests predominantly lie in the field of Islamic Economics and Finance.