To effectively reduce poverty, it is important to understand its determinants. The factors affecting poverty vary significantly across different regions, predominantly shaped by distinct cultural, political, and socio-economic contexts [5] [10] . Despite these regional variations, numerous studies highlight several key factors believed to significantly contribute to poverty alleviation [5] . Economic growth, for instance, is recognized as a major catalyst for reducing poverty and enhancing the quality of life in developing countries [11] . Extensive research finds compelling evidence that rapid and sustained economic growth is the most effective way to alleviate poverty [11] . Furthermore, extensive studies reveal a strong correlation, showing that a 10% increase in a country’s average income is associated with a considerable decrease in the poverty rate [11] .

Furthermore, economic growth is linked to another key determinant of poverty―inflation. The relationship between economic growth and inflation represents one of the most important macroeconomic concerns [12] [13] . The research by Erbaykal and Okuyan (2008) [13] showed a significant and negative short-term correlation between inflation and economic growth in Turkey. Akinsola and Odhiambo (2017) [14] further studied and concluded there is strong evidence, particularly in developed economies, that inflation and growth are negatively correlated. The paper mentioned a great deal of discrepancy in the precise inflation threshold that promotes economic growth. Their findings indicated no causal relationship between economic growth and inflation, but they identified the causality from inflation to economic growth. These results are consistent with earlier studies by Karaca (2003) [15] and Berber and Artan (2004) [16] in Turkey, suggesting that higher economic growth could help reduce inflation.

Another consideration is the frequent association between household consumption and income levels. Income is predominantly utilized to measure economic deprivation due to its ease of reporting and availability across much larger samples [17] . The income effect states that when consumers experience an increase in income, their consumption tends to rise [18] . This effect highlights a direct correlation between income and spending behaviors, where higher household consumption levels are associated with an improved standard of living. This relationship has broader economic implications, particularly regarding reducing poverty [18] . Furthermore, numerous statistical studies have identified a strong correlation between a national per capita income and its poverty indicators, using both income and non-income dimensions of poverty [19] .

The final factor to be examined is trade openness, which plays a crucial role in determining the level of extreme poverty. There has been a notable increase in the involvement of developing countries in global trade, which has corresponded with a reduction in extreme poverty worldwide [4] . Specifically, the share of developing countries in world trade has risen from 33 percent in 2000 to 48 percent [4] . This growth in trade participation has decreased the number of people living in extreme poverty and has also been crucial in creating jobs, spurring economic growth, and boosting productivity in developing nations. The study by Agusalim (2017) [20] examines the long-term effects of international trade openness on poverty in Indonesia. Past studies indicate varying impacts across different countries. The dataset includes export-import values, gross domestic product, income per capita, open unemployment rate (OUR), and poverty rate (POVR) from 1978 to 2015. The results reveal that while trade openness shows no significant effect on poverty in the short term, it does significantly reduce poverty over the long term. While trade openness shows no significant effect on poverty in the short term, it does have a significant impact on reducing poverty over the long term.

Numerous studies have been conducted across various regions to investigate the impact of fintech on poverty reduction, showing a general decrease in poverty rates. Emara and Mohieldin (2021) [5] reveal that fintech has a statistically significant effect on reducing extreme poverty within the entire sample, including both the MENA and SSA regions. The results confirm that a ten percent increase in the fintech index resulted in a fall in the poverty headcount ratio by about 0.06% for the full sample and with larger impacts of 0.37% and 0.39% in the MENA and SSA regions [5] . Similar research in China has found that fintech and third-party payment and credit services significantly boost household per capita consumption. More precisely, a 1% increase in fintech, third-party payment, and credit leads to increases in household per capita consumption of 1.337%, 0.720%, and 0.368%, respectively, according to the IV-GMM model. These results also suggest that economic growth, facilitated by advancements in fintech, effectively reduces poverty by increasing household consumption [21] .

Fintech has proven to be a powerful instrument in combating poverty by expanding access to financial services, supporting microfinance, facilitating digital payments, delivering financial education, and generating employment opportunities. In a nutshell, as fintech promotes financial development and inclusion, it also has the capacity to reduce poverty. Therefore, the hypothesis suggests that since fintech influences these key factors, it can play a significant role in poverty alleviation.

The assumptions of this study are as follows.

H0: Fintech has no significant impact on poverty rates in Southeast Asian countries.

Ha: Fintech has a significant impact on poverty rates in Southeast Asian countries.

The dataset comprises a panel of observations from seven Southeast Asian countries: Thailand, Indonesia, Malaysia, Myanmar, the Philippines, Cambodia, and Vietnam, covering the period from 2018 to 2021. These countries are categorized into two groups based on their income levels: middle-income and lower-income, as classified by the International Monetary Fund (IMF). Thailand, Indonesia, and Malaysia are middle-income groups, while Myanmar, the Philippines, Cambodia, and Vietnam are lower-income groups. Given that each variable has a unique scale, it is crucial to standardize all variables to a consistent, quantitative scale prior to analysis. Table 1 provides a detailed list of the macroeconomic and fintech variables used, including their definitions, units of measurement, abbreviations, and data sources.

The information input for this study included various types of data such as units, percentages, currencies, and numerical quantities. A standardization process is applied to maintain consistency across diverse data types. In this system, each data point is adjusted relative to the highest value observed in the dataset, transforming all data into a scoring-based system. This conversion involves dividing each value by the highest value and multiplying the result by 10 to create a 10-point scoring system. A score of 10 represents the highest performance in a category, while lower scores reflect relative performance within the established scoring framework.

For some variables, such as the poverty rate, direct scoring would mean that a higher rate (negative in reality) would misleadingly result in a higher score (intended to be positive). To correct this, such variables are calculated inversely. That means, instead of scoring the poverty rate directly, it is scored in such a way that lower poverty rates (which are positive) receive higher scores. This inversion ensures that all variables maintain the intended meaning, where higher scores consistently reflect better outcomes. Therefore, using relative ratios ensures a uniform ratio system across all data types and their respective definitions. This standardization technique allows a deeper analysis of the data.

This research utilizes two analytical tools: linear regression and Pearson correlation coefficient analysis. Linear regression analysis explains how independent variables affect dependent variables. The poverty headcount ratio of $2.15 a day as a percentage of the population is used as a proxy for poverty. The poverty headcount ratio at $2.15 per day represents the proportion of the population living below the threshold of $2.15 a day, considering purchasing power adjustments at 2017 prices (World Bank, 2021) [9] . The set of dependent variables contains common determinants of poverty, including GDP per capita, inflation rate, trade openness as a percentage of GDP, household consumption, and fintech index indicators. The measures of the fintech index include the number of access points per 100,000 adults at a national level, the percentage of people in the population who use the internet, the percentage of people in the population who report having an account at a commercial bank, the percentage of adults who report having an account with formal financial institutions or mobile providers, and the volume of electronic money transactions constructed to create the fintech index.

Pearson correlation coefficient analysis explains how each indicator correlates with another indicator as a pair, providing in-depth details on the impact and direction of the coefficient. Following the initial hypothesis, the collected data are calculated to explore the relationship between fintech and its effect on poverty in seven Southeast Asian countries.

The regression equation represents the direct effect of fintech on poverty reduction.

Y = b₀ + b₁X₁ + b₂X₂ + b₃X₃ + b₄X₄ + b₅X₅ + ε

where:

Y: The dependent variable Denotes pov, the poverty headcount ratio, defined as the number of individuals living on less than $2.15 per day.

b₀: The intercept term.

b₁: The slope coefficient, each associated with a paired independent variable i.

X₁: Denotes the independent variables, where fin represents fintech indicators,

X₂: Denotes to represents trade openness.

X₃: Denotes hc represents household consumption.

X₄: Denotes gdp represents GDP.

X₅: Denotes inf represents inflation.

ε: The random error component in the model.

Regression analysis concludes the analysis of poverty using fintech and economic factors. The equation below indicates that the weight coefficients of independent variables affect the poverty of two groups: middle-income and lower-income countries.

In linear regression, a coefficient represents the numerical value multiplied by an independent variable in the regression equation. This coefficient quantifies the relationship between the independent variable and the dependent variable.

In Table 2, a variance test was performed on the result of the regression analysis in middle-income countries. In this context, the variable fin is the only one that shows a statistically significant positive with the poverty rates at 95% confidence levels, suggesting a positive relationship with poverty reduction. Other variables such as to, gdp, and inf also exhibit positive relationships with poverty rates, but these factors are not statistically significant. On the other hand, the variable hc demonstrates a negative relationship with poverty rates, but this correlation is also statistically insignificant.

In Table 3, a variance test has been conducted on the result of regression analysis in lower-income countries. In this context, the variable fin shows a statistically significant positive with the poverty rates at 95% confidence levels, suggesting a positive relationship with poverty reduction. The P-value is slightly above the typical significance threshold, indicating marginal statistical significance. Furthermore, the variable inf also shows a statistically significant negative with the poverty rates at 95% confidence levels, implying that higher inflation is associated with worsening poverty conditions. Other variables, such as to and hc, also exhibit positive relationships with poverty rates, but these factors are not statistically significant. On the other hand, the variable gdp demonstrates a negative relationship with poverty rates, but this correlation is also statistically insignificant.

The Pearson correlation coefficient (Table 4) based on an analysis of real-world data, shows relationships between key variables and poverty improvement metrics in middle- and lower-income countries. Financial technology (fin) demonstrates a strong positive correlation with poverty rates in middle-income countries (0.734) and lower-income countries (0.619). This indicates that advancements in financial technology correspond with improvements in poverty conditions, highlighting the importance of technological innovations in promoting economic development.

The Pearson correlation coefficient (Table 4) shows that trade openness (to) has a very strong positive correlation of 0.945 with poverty rates in middle-income countries, indicating a significant link between increased international trade and improvements in poverty conditions. However, the correlation of trade openness in lower-income countries is considerably weaker at 0.246; while there is still a positive impact of trade on poverty reduction, it is significantly less pronounced in lower-income countries, possibly due to varying trade capacities or economic structures. Household consumption (hc) demonstrates a strong negative correlation with both middle-income countries (−0.954), implying that higher household consumption levels are closely linked to a reduction in poverty. On the other hand, household consumption in lower-income countries reflected the opposite result.

Furthermore, GDP growth (gdp) reveals a negative correlation with both middle-income (−0.374), higher GDP may help reduce poverty, though the relationship is not as strong as other factors. The analysis in lower-income countries reflected different results. As lower-income countries’ GDP increases, the poverty rate increases (0.552). The inflation rate (inf) positively correlates with poverty rates in middle-income countries (0.704), suggesting that higher inflation tends to correlate with increased poverty, possibly due to higher living costs not being matched by income growth. The inflation rates in lower-income countries show a weak negative correlation (−0.128), implying that inflation has a slightly different and less pronounced impact on poverty.

Table 4 reveals significant differences in how economic variables interact with poverty in middle- and lower-income countries. Financial technology and trade openness correlate positively with poverty reduction in both groups, though the effects are stronger in middle-income countries. Household consumption and GDP data suggest positive impacts on reducing poverty. Inflation shows varying effects associated with increasing poverty in middle-income countries but less in lower-income countries. Policymakers can use these correlations to develop targeted strategies such as promoting the adoption of financial technology, increasing international trade openness, and regulating household consumption patterns. It is important to note that the effectiveness of these measures may differ across various dimensions of poverty reduction.