The increase in per capita income is one of the main indicators used to measure the success of a country's economic development, including Indonesia. Per capita income not only reflects the level of public welfare but also illustrates the economic capacity to create sustainable value added. In recent years, Indonesia has experienced an upward trend in per capita income, supported by various factors from both domestic and external sources.

One of the important factors affecting per capita income is investment, originating from both Foreign Direct Investment (FDI) and Domestic Direct Investment (DDI). Investment plays a strategic role in promoting economic growth through increased production capacity, technology transfer, and job creation. Empirical studies indicate that FDI and DDI, both simultaneously and partially, have a positive and significant effect on economic growth in Indonesia (Antari et al., 2025). In addition, the unequal distribution of investment across regions remains a challenge in achieving inclusive economic growth (Martin, 2025).

Apart from investment, remittances from Indonesian migrant workers also constitute an important source of external financing. Remittances contribute to increased household consumption, educational investment, and family economic stability. Research shows that remittances have a positive effect on per capita income in Indonesia, as they enhance purchasing power and stimulate economic activity (Fahruddin & Aji, 2021). Furthermore, remittances have been proven to contribute to poverty reduction, although their impact is not always statistically significant (Abdi & Hakim, 2016).

However, the relationship between FDI, DDI, remittances, and per capita income is not static. In many cases, the effects of these variables are dynamic and interrelated over time. Per capita income in a given period is likely influenced by conditions in previous periods, giving rise to endogeneity issues and long-term dynamic effects. Therefore, the use of a dynamic panel data approach becomes essential to capture lag effects and provide a more accurate understanding of causal relationships in regional economic analysis.

Previous studies have generally employed static regression or time-series approaches, which have not fully captured the long-term dynamics among variables. Moreover, studies that simultaneously incorporate FDI, DDI, and remittances within a dynamic panel framework remain relatively limited, particularly in the context of Indonesian provinces. Yet, such an approach is crucial for understanding how these factors interact in influencing public welfare sustainably.

Based on this background, this study aims to analyze the effects of FDI, DDI, and remittances on per capita income in Indonesia using a dynamic panel data approach. This research is expected to provide an empirical contribution to the development economics literature and serve as a basis for formulating more effective investment and remittance management policies to improve public welfare.

The Solow Growth Theory explains that long-term economic growth is influenced by capital accumulation, labor force growth, and technological progress. The model assumes diminishing returns, implying that increases in capital or labor alone are insufficient to sustain long-term growth. In the long run, the economy reaches a steady-state condition in which growth in output per capita can only be achieved through technological advancement. Therefore, technology becomes the primary driver of long-term economic growth (Solow, 1956).

The Harrod–Domar Growth Theory was developed by Harrod and Domar to explain the conditions necessary for an economy to achieve long-term growth. The theory is based on the assumptions that the economy operates at full capacity, savings are directly proportional to income, and the capital-output ratio remains constant (Sukirno, 2006). In this theory, investment plays a crucial role because it not only increases aggregate demand but also expands production capacity through the addition of capital stock. As long as investment continues, income and output will increase. However, to maintain equilibrium, income growth must be consistent with the growth of production capacity (Jhingan, 2003). Furthermore, capital formation is regarded as the primary factor driving economic growth because it not only enhances productive capacity but also expands effective demand. Therefore, new investment is required as an addition to the existing capital stock (Todaro, 2006).

Per capita income refers to the average income earned by each individual within a region during a specific period. It is calculated by dividing national income or Gross Domestic Product (GDP) by the total population. This indicator is widely used to measure welfare levels and compare prosperity across regions (Todaro & Smith, 2010). Per capita income reflects the purchasing power of a population; however, it does not provide information about income distribution. Therefore, although it is frequently used as an indicator of development success, additional indicators are required to assess equity and social justice more comprehensively.

The variables used in this study consist of dependent and independent variables. The dependent variable is per capita income (PCI), while the independent variables include Foreign Direct Investment (FDI), Domestic Direct Investment (DDI), and remittances. In addition, the study incorporates the lagged value of per capita income, LOG(PCI(-1)), as a dynamic variable in the dynamic panel data model.

The estimation method employed in this study is the dynamic Generalized Method of Moments (GMM), specifically the Difference GMM estimator developed by Arellano and Bond (1991). This method is chosen because the model contains a lagged dependent variable, LOG(PCI(-1)), as a regressor. Estimation using conventional Fixed Effects or Random Effects methods would produce biased and inconsistent estimators due to the correlation between the lagged variable and the error term (Nickell, 1981).

This study employs the Difference Generalized Method of Moments (GMM) approach to analyze the effects of Foreign Direct Investment (FDI), Domestic Direct Investment (DDI), and remittances on per capita income across 34 provinces in Indonesia during the 2013–2023 period. A dynamic panel approach was selected to accommodate the persistent nature of per capita income and to address potential endogeneity issues commonly encountered in panel data estimation. The estimation results indicate that all independent variables—namely FDI, DDI, and remittances—have a positive and statistically significant effect on provincial per capita income in Indonesia at the 99% confidence level (p = 0.0000). Among these variables, remittances exhibit the largest coefficient, amounting to 0.2314, implying that a 1% increase in remittances leads to a 0.23% increase in per capita income. FDI ranks second with a coefficient of 0.1128, followed by DDI with a coefficient of 0.0654. These differences in coefficient magnitudes reflect variations in the transmission mechanisms and productivity levels associated with each source of financing in regional economies.

In addition to the three main explanatory variables, the estimation results reveal a negative and statistically significant coefficient for the lagged per capita income variable, LOG(PCI(-1)), amounting to −0.1076. This negative coefficient indicates the presence of conditional convergence among Indonesian provinces, suggesting that provinces with lower levels of per capita income tend to grow faster than more developed provinces. The estimated speed of adjustment toward the long-run equilibrium is approximately 10.76% per year, with a convergence half-life estimated at around 6–7 years. The validity of the model is confirmed through a series of diagnostic tests. The Sargan test produces a J-statistic value of 33.334 with a probability of 0.3082, indicating that all instruments used in the model are valid and uncorrelated with the error term. The number of instruments (34) remains below the number of cross-sectional units (N = 38), suggesting that the model is free from the problem of instrument proliferation.

This study aimed to analyze the effects of Foreign Direct Investment (FDI), Domestic Direct Investment (DDI), and remittances on per capita income across 38 provinces in Indonesia during the 2013-2023 period using the Arellano-Bond Difference GMM approach. Based on the estimation results and discussion, several conclusions can be drawn.

First, FDI has a positive and significant effect on per capita income, with a coefficient of 0.1128, indicating that a 1% increase in FDI raises per capita income by approximately 0.11%. Second, DDI also exerts a positive and significant influence, with a coefficient of 0.0654, implying that a 1% increase in DDI leads to an increase in per capita income of approximately 0.07%. Third, remittances emerge as the most dominant variable, with a coefficient of 0.2314, demonstrating that a 1% increase in remittance inflows raises per capita income by approximately 0.23%. Fourth, evidence of income convergence across provinces is found through the negative coefficient of LOG(PCI(-1)) amounting to -0.1076, indicating that provinces with lower income levels tend to grow more rapidly toward their long-run equilibrium.

Based on these findings, this study recommends that the government continue promoting a favorable investment climate for both FDI and DDI through regulatory simplification and improvements in regional infrastructure. In addition, remittance management policies should be directed toward productive economic activities through empowerment programs for the families of Indonesian migrant workers. Such measures would enable remittances to generate a more optimal and sustainable contribution to per capita income growth in the long run.