An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh

Studies Management and Finance Economics, of Journal 0504-2644 (online): ISSN 0490,-2644 (print): ISSN 5202 December 12 Issue 80 Volume 8.317 Factor: Impact ,15-i12-10.47191/jefms/v8 DOI: Article 3277-7722 No: Page JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7722 An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh Mithila Tanzil1, Shahed Ahmed2* 1 Assistant Professor, Department of Economics, Islamic University, Kushtia-7003, Bangladesh 2 Professor, Department of Economics, Islamic University, Kushtia-7003, Bangladesh *Corresponding author’s Name: Shahed Ahmed ABSTRACT: The study examines the economic growth of Bangladesh by analyzing the interactions of GDP with key macroeconomic factors such as Remittances (REM), Foreign Direct Investment (FDI), Gross capital Formation (GCF), and Annual Net National Income (ANNI) covering the time period from 1980 to 2024 using the Vector Error Correction Model (VECM) framework. The Johansen Cointrgration test confirms the existence of long-run cointegration relationship and highlights short-run stability, with particular relevance to the post pandemic context. The findings reveal that GDP acts as the sole endogenous stabilizer, adjusting 19.13% to restore equilibrium annually, while FDI, GCF, and ANNI remain weakly exogenous and showing minimal responsiveness to domestic shocks. On the other hand, REM destabilizes the system in the short run. The result suggests that the stability of Bangladesh's economy cannot rely on external or weakly exogenous variables. Policy implications underscore the need to strengthen GDP's internal corrective capacity through targeted domestic investment while mitigating volatility from external remittance inflows. KEYWORDS: GDP, Remittances, FDI, GCF, ANNI, VECM. 1 INTRODUCTION Bangladesh’s economic growth depends on how its main macroeconomic factors work together, and understanding this relationship is important for making good policies. Over the last four decades, the country has changed a lot, moving from a lowincome economy to a lower-middle-income one. Real GDP growth averaged about 6.3% between 2010 and 2023, driven by structural change, rising investment, export expansion, and improvements in human development. The poverty rate fell from 14.7% in 2010 to 5.9% in 2022, reflecting broad-based progress, yet the economy has also shown growing vulnerability to shortterm shocks. However, the economy has also faced ups and downs. For example, growth dropped to 3.45% in 2020 during the COVID-19 pandemic, then rose to 6.94% in 2021 and 7.10% in 2022, before slowing to around 5.78% in 2023 and about 5.82% in 2023–24. These changes show that the economy can grow strongly but is also sensitive to shocks. Therefore, it is important to study how factors like Remittances (REM), Foreign Direct Investment (FDI), Gross capital Formation (GCF) and Annual Net National Income (ANNI) affect GDP growth both in the long run and the short run. A clearer empirical understanding of these interactions is essential for designing policies that promote stable, resilient, and sustainable economic growth in Bangladesh. 1.1 Background of the Study: The journey of Bangladesh’s economic growth has been shaped by several important domestic and external macroeconomic factors. Remittances remain one of the strongest pillars of the economy, with inflows reaching around USD 21–24 billion in recent years, helping stabilize foreign currency reserves and supporting household income and consumption. Foreign Direct Investment (FDI), although still modest compared to peer economies, has generally ranged between USD 1.5–3 billion annually, contributing to the transfer of technology, managerial skills, and capital needed for industrial growth. Gross Capital Formation (GCF), which reflects domestic investment, has consistently remained high at around 30% of GDP, showing strong efforts to build productive capacity. Meanwhile, Annual Net National Income (ANNI) has kept rising steadily along with economic growth, reflecting the country’s expanding ability to generate income. Despite this progress, the long-term stability and structural consistency of these growth drivers remain uncertain, especially given fluctuations in remittances, investment, and external economic conditions. An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7723 Therefore, this study aims to go beyond simple correlations and examine how these key macroeconomic factors are structurally connected to Bangladesh’s sustainable GDP growth. 1.2 Statement of the Problem: Despite maintaining an average GDP growth rate above 6 percent for more than a decade, Bangladesh’s macroeconomic structure continues to exhibit signs of fragility, reflected in recurring external imbalances, fluctuating investment inflows, and periodic pressure on foreign exchange reserves. For example, while annual remittance inflows have remained strong—often exceeding USD 20 billion—FDI inflows have remained comparatively low, generally fluctuating between USD 1.5 and 3 billion per year, raising questions about the actual contribution of foreign investment to long-term growth. Similarly, although Gross Capital Formation has consistently hovered around 28–31 percent of GDP, the efficiency and stability of this investment in sustaining future growth remain uncertain. These inconsistencies highlight a deeper structural issue: the long-run equilibrium among key macroeconomic variables remains unclear, making it difficult to determine whether components such as FDI, remittances, and domestic investment act as stabilizing forces or potential sources of volatility. In the absence of clear empirical evidence on which variables anchor long-run stability, which respond passively to shocks, and which may destabilize the system, policymakers face significant challenges in designing policies that ensure sustainable economic growth. To address this gap, the present study employs a Vector Error Correction Model (VECM) to examine both the long-term structural relationships and the short-run dynamic adjustments among the selected macroeconomic indicators. This approach is expected to offer a more rigorous understanding of the mechanisms through which these variables influence sustainable GDP growth, thereby providing a stronger foundation for evidence-based economic planning and policy formulation. 1.3 Rationale of the study: This study is essential because it provides a comprehensive, up-to-date, and structurally grounded assessment of Bangladesh’s macroeconomic environment. First, it incorporates a broad set of domestic and external variables—such as remittances, FDI, Gross Capital Formation, and national income—into a unified analytical framework to capture the interconnected nature of the economy. Second, it applies rigorous econometric techniques, including unit root tests, the Johansen Cointegration approach, and the Vector Error Correction Model (VECM), enabling a clear distinction between short-run adjustments and long-run equilibrium relationships. Third, the study employs the most recent data available up to 2024, covering both the pre-pandemic period and the post-pandemic recovery phase, a period marked by notable volatility in GDP growth, remittance inflows, external balances, and investment patterns. By integrating these methodological strengths and contemporary data trends, the study provides timely, evidence-based insights that can guide policymakers in strengthening macroeconomic stability, managing external vulnerabilities, and promoting sustainable economic growth. 1.4 Objectives of the study: • To analyze the historical trends and structural volatility of GDP and selected macroeconomic indicators in Bangladesh. • To examine the short-run dynamics and long-run equilibrium relationships between GDP and key macroeconomic variables. • To identify the weakly exogenous macroeconomic variables that influence the growth system. 1.5 Research Questions: 1. How have GDP and key macroeconomic variables (REM, FDI, GCF, ANNI) historically evolved? 2. What are the short-run dynamic and long-run equilibrium relationship between GDP and these macroeconomic variables? 3. Which macroeconomic variables operate as weakly exogenous component within Bangladesh’s growth system? 1.6 Hypotheses: H1: A long-run cointegrating relationship exists among the variables H2: At least one of the macroeconomic variables significantly influence GDP in the short-run H3: At least one of the macroeconomic variables is weakly exogenous within the VECM framework, meaning it does not respond to deviation from long-run equilibrium 2.0 LITERATURE REVIEW 2.1 Theoretical Background: Economic growth is a fundamental indicator of a country’s economic performance and is shaped by a range of domestic and external macroeconomic factors. Identifying the most effective determinants of sustainable growth is particularly important for An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7724 developing economies such as Bangladesh, where structural transformation and external vulnerability coexist. This study draws on several classical and modern economic growth theories to establish its analytical foundation. Gross Capital Formation, a key variable in this study, aligns with the Solow–Swan neoclassical growth model, which emphasizes capital accumulation as a driver of productive capacity and long-run output. Foreign Direct Investment (FDI), which introduces capital, advanced technology, and managerial skills, is supported by endogenous growth theory, where external knowledge spillovers and innovation contribute to sustained economic expansion. Insights from open-economy growth theory further clarify how trade openness enhances efficiency, broadens market access, and encourages technological progress in developing economies. According to Keynesian theory, remittances play a countercyclical role by raising household income, consumption, and investment, thereby stimulating aggregate demand. Additionally, the inclusion of Annual Net National Income (ANNI) is grounded in aggregate income theory, which links overall economic activity and national income generation to GDP growth. By integrating these theoretical perspectives, the study develops a comprehensive framework to examine the dynamic nexus between domestic and external macroeconomic determinants and their contribution to sustainable economic growth in Bangladesh. 2.2 Review of Literature The literature review has been systematically organized in a table format, summarizing key studies, their objectives, methodologies, variables, and main findings to provide a clear and comparative overview of existing research on macroeconomic determinants of economic growth. Author(s) & Year Country/Period Variables Examined Methodology Key Findings Anuno et al. (2025) Timor-Leste, 2006–2023 FDI, Remittance, GCF, GDP ARDL FDI positive in short run, insignificant in long run; GCF positive in long run; remittance insignificant; FDI Granger causes GDP Bikorimana et al. (2021) East African Community, 1970–2017 FDI, Capital Formation, Inflation, Population Growth, Consumption, Trade, GDP VECM, Granger causality FDI & capital formation positive; inflation, population growth, consumption negative; trade negative; GDP & population growth bidirectional causality Poudel & Giri (2025) Nepal, 1975– 2023 Export, Import, Remittance, GFCF, GDP VAR, Johansen Cointegration Remittance positive; export & GFCF insignificant; import, investment, remittance affected by GDP shocks; export unaffected Akhter (2022) Bangladesh, last 20 years FDI, Remittance, External Debt, GDP Granger causality FDI insignificant on GDP; external debt unidirectional relation with FDI & remittance; FDI Granger causes ED, ED causes remittance Nath & Begum (2021) Bangladesh, 1996–2018 Exchange rate, Inflation, FDI, Remittance, Trade, GDP ADF, Cointegration, VECM, Granger causality Remittance positive; FDI negative impact on GDP Jacob & Raphael (2021) Bangladesh, 1990–2020 Inflation, Exchange rate, Trade openness, FDI, GDP ARDL bounds testing FDI negative; inflation, exchange rate, trade openness positive & significant Khan et al. (2024) Bangladesh Capital, Domestic Credit, GCF, Government Expenditure, Net Trade, GDP ARDL Short run: capital & credit negative, government expenditure & net trade positive; long run: GCF & government expenditure strongly positive An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7725 Haque et al. (2023) Bangladesh, 1980–2022 Inflation, Real Interest Rate, Exchange Rate, GDP VAR, Johansen Cointegration, VECM Inflation insignificant; interest rate negative; exchange rate positive; long-run causality present; short-run effects weak Hasan et al. (2022) Bangladesh Import, Export, Inflation, GDP Multiple regression Import & export positive; inflation negative Chowdhury & Rahman (2020) Bangladesh, 1972–2018 Official Exchange Rate, External Debt, Terms of Trade, GDP VECM Exchange rate & external debt positive long-run effect; terms of trade negative; unidirectional causality to GDP Hao Wu (2020) Bangladesh, 1980–2018 Investment, Labor, Technology, GDP Neoclassical growth framework Investment strongest determinant; suggested diversification beyond garment sector Chowdhury et al. (2019) Bangladesh, 1987–2015 Inflation, Exchange Rate, Interest Rate, Household Consumption, GDP Correlation, Multiple Regression Inflation, exchange rate, consumption positive; interest rate negative; significant relationships overall In conclusion, the reviewed literature highlights that macroeconomic factors such as investment, trade, and government expenditure are key drivers of economic growth, while the impact of FDI and remittances is mixed and context-dependent. This highlights the need for a comprehensive analysis to understand their dynamic interactions and inform policies for sustainable economic growth in Bangladesh. 2.3 Research Gap: Although numerous studies have explored macroeconomic influences on economic growth, notable gap persists in the concerning literature of Bangladesh. Bangladesh specific literature such as Khan et al., 2024; Haque et al., 2023; Nath and Begum, 2021; Chowdhury and Rahman, 2020) used data before 2020, consequently failing to capture post covid phase. Even the studies with recent time coverage (e.g., Anuono et al., 2025; Poudel and Giri, 2025) mostly situated within global and regional framework and often restrict its scope to few selected variables such as FDI, remittance, or capital formation. Thus the literature remains incomplete, as to overlooks the combined influence external and domestic factors on growth, which this study systematically incorporates a unified framework including remittance, FDI, trade in the external sides and Annual Net National Income along with gross capital formation on the domestic side. Moreover, while most of the existing studied applied simple VAR and ARDL methods, this study utilized advanced and complete econometric techniques such as unit root, Johansen cointegration and VECM. This underscores a clear gap that necessitates a more robust and methodologically rigorous empirical assessment. 3.1 METHODOLOGY This chapter describes about the type and sources of data, mention set of variables, and the methodological approaches that have been adopted to estimate the objectives of this research. 3.2 Sources of Data: This study uses time series data covering the time period from 1980 to 2024. Most of the data of selected macroeconomic variables for this study have been collected from World Development Indicator (WDI) and World Bank (WB). For cross checking and confirming the reliability some other sources like Bangladesh Bank, Bangladesh Bureau of Statistics (BBS) and other online journals and articles have been used. 3.3 Key Variables: The study aims to investigate the short-run dynamic and long run equilibrium between selected macroeconomic variables and economic growth of Bangladesh. The main key variables are Remittances (REM), Foreign Direct Investment (FDI), Gross Capital Formation (GCF) and Annual Net National Income (ANNI), which have significant impact on Gross Domestic Product (GDP). Here GDP is considered as dependent and others as independent variables. An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7726 3.4 Model Specification: Since this study deals with time series data of 44 years long, this can be often nonstationary in nature, which cause their mean and variance to change over time. Therefore, the variables must be tested for stationarity to avoid the spurious results. 3.4.1 Model Specification and Estimation The empirical analysis utilized the Vector Error Correction Model (VECM) to investigate the dynamic relationships between fivetime series variables: Gross Domestic Product (GDP), Remittances (REM), Foreign Direct Investment (FDI), Gross Capital Formation (GCF), and Annual Net National Income (ANNI). 3.4.2 Unit Root and Cointegration Tests Before estimating VECM estimation, the time series properties of the variables were examined, assuming all variables are integrated of order one, I(1) (non-stationary). The following steps were followed: 1. Unit root testing ADF 2. Lag selection through VAR 3. Johansen Cointegration Test 4. VECM estimation 3.4.3 Augmented Dickey–Fuller (ADF) Test Specification The stationarity of the variables was confirmed using the Augmented Dickey-Fuller (ADF) test. The test generally employed the specification including both a constant (α) and a linear trend (βt): ∆Yt= α+βt+γYt−1 +∑δi∆Yt−1 +εt p i=1 Where, ∆Yt = The first difference of the time series variables at time t . Yt-1 = The lagged level of time series variables γ = The key coefficient. The null hypothesis is H0 : γ = 0 (Unit root) ∑δi∆Y p i=1 t-1 = A sum of lagged difference term included to correct autocorrelation. ϵt = The white-noise error term. H0: γ = 0 (Unit root); Result confirms all variables are I(1). 3.4.4 Optimal Lag Selection The analysis began with an unrestricted Vector Autoregression (VAR) model to determine the optimal lag structure. Based on the majority of criteria (LR, FPE, and AIC), an optimal lag length of p = 2 was selected. This choice determines the number of lagged difference terms (p-1=1) used in the subsequent VECM. 3.4.5 Johansen Cointegration Test Specification and Results The Johansen Cointegration Test was performed to confirm the existence of a long-run relationship The model is specified as: ∆Xt= ПXt +∑Гi ∆Xt−1+εt П= αβʹ contains long-run equilibrium (β) and adjustment coefficients (α) The cointegration rank (r) is tested using two likelihood ratio statistics: The Trace Test and the Max-eigenvalue Test. Two cointegrating vectors found (r=2). 3.4.6 Long-Run Cointegration Equation GDPt = β0 + β1REMt + β2FDIt + β3CGFt + β4ANNIt + εt 3.4.7 VECM Equation The estimated VECM model utilizes a rank of r=2 and p-1 = 1 lagged difference, given the optimal lag selection. The general form of the VECM is: ∆Xt= αβʹXt−1 +∑Гi∆Xt-1+εt Where, αβʹXt-1 = The Error Correction Term (ECT). Represents the system's deviation from the longrun equilibrium in the previous period. βʹ = The Cointegrating Vector (Long-Run Relationship). An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7727 α = The Adjustment Vector. Г1∆Xt-1 = The Short-Run Dynamics Component. ∆Xt = Vector of the first differences of the variables. 3.4.8 Error Correction Term: Normalization on GDP sets GDP coefficient=1 Thus, ECT appears with reverse signs: ECTt-1 = GDPt-1 - (β₀ + β₁ REMt-1 + β₂ FDIt-1 + β₃ GCFt-1 + β₄ ANNIt-1) 3.4.9 Diagnostic Checking The estimated VECM model (rank 2, 1 lagged difference) was validated using residual tests. • Model Stability: The underlying VAR(2) model is stable , as the Inverse Roots of the AR Characteristic Polynomial all lie inside the unit circle. This is essential for valid analysis. • Residual Serial Correlation (LM Test): This test confirms the model is correctly specified by checking if residuals are white noise. o Null Hypothesis (H0): No serial correlation. 3.4.10 Identification of Weakly Exogenous Variables To identify the weakly exogenous variables in the system, follows the standard VECM approach through Error Correction Term (ECT), the two variable form of VECM formulated as, ∆Yt= αYECTt−1 +∑Гi∆Zt−1 +εYt ∆Xt= αXECTt−1 +∑Гi∆Zt−1 +εXt ECTt-1= captures the deviation from long-run equilibrium αy, αx measure how strongly each variable adjust to recover equilibrium Гi∆Zt-1 represents short-run effects from lagged differences Applying this structure in this study Yt = GDPt X t = REMt, FDIt, GCFt, ANNIt 4.0 RESULT AND DISCUSSION 4.1 Objective 01: Analyzing the historical trends and structural volatility of GDP and selected macroeconomic factors (REM, FDI, GCF, ANNI) in Bangladesh. This objective tends to investigate the historical trend and structural volatility of GDP and selected macroeconomic variables such as remittances, foreign direct investment, gross capital formation and annual net national income in the context of Bangladesh. A sample of 44 observations covering the time period 1980 to 2024 have been selected to analyze how variables behave over the long period of time. By utilizing descriptive statistic which includes Mean, Median, Maximum, Minimum and Standard Deviation to provide essential insights in understanding the traditional pattern and instability of economic growth and other selected key macroeconomic variables priory mentioned. Table: 1 Descriptive Statistics for key Macroeconomic Indicators Statistic GDP (Gross Domestic Product) REM (Remittances) FDI (Foreign Direct Investment) GCF (Gross Capital Formation) ANNI Mean 803.4882 4.980352 0.458681 23.47668 20.55375 Median 396.6707 4.596677 0.350304 24.34142 16.92402 Maximum 2716.486 10.59151 1.735320 32.21373 39.17621 Minimum 184.7068 1.867162 -0.029894 14.43939 10.96872 Std. Dev. 780.6227 2.534164 0.492685 5.793651 7.620551 Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. This descriptive statistic shows trend as well as volatility between economic growth and selected macroeconomic variables. Over the sample period from 1980 to 2024, the trend of GDP is strongly raising, indicated by the high Maximum value (2716.486) compared to Minimum (184.7068). The upward trend of GDP is also confirmed by the huge gap between Mean (803.48) and An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7728 Median (396.67). Despite that, high volatility is also observed in this trend, reflecting by its very large standard deviation (780.6227), indicating notable instability in the overall economic growth. On the other hand, remittances found to remain consistent over the past years since the Mean (4.98) and Median (4.60) provides very close values with a moderate Standard deviation (2.53), which reflects comparatively stable trend. Although FDI maintains a low Mean (0.46) and minimal dispersion (0.49), the negative Minimum value (-0.028894) reveals a period of net disinvestment, suggesting even if the fluctuations are limited, yet it caused instability. While gross capital formation exhibits a relatively stable investment trend which shows manageably moderate volatility in domestic investment, ANNI presents a upward trend moving between Maximum of 39.18 and Minimum of 10.97, showing income growth with periodic fluctuation. Overall, the result from this analysis suggest that, although the economic growth of Bangladesh has a sharply upward trend, it also faces volatility. It is also found that, in compare to FDI inflows which exhibits inconsistent nature, domestic investment and remittances display better stability. Table: 2 Descriptive Statistics of Distributional Properties for Macroeconomic Variables Statistic GDP REM FDI GCF ANNI Skewness 1.348480 (RightSkewed) 0.820119 (RightSkewed) 0.929778 (RightSkewed) -0.043727 (Symmetric) 1.176980 (RightSkewed) Kurtosis 3.361037 (Slightly Leptokurtic) 2.460079 (Platykurtic) 2.829890 (Platykurtic) 1.574328 (Platykurtic) 3.092477 (Slightly Leptokurtic) Jarque-Bera 13.88239 5.591060 6.537910 3.825354 10.40564 Probability 0.000967 0.061083 0.038046 0.147685 0.005501 Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. This above distributional table shows GDP, REM and ANNI are all right skewed, indicating long right tails, whereas, CGF is nearly symmetric. On the other hand, the values of Kurtosis suggest that, GDP and ANNI are slightly platykurtic, indicating a little heavier tail, whereas, REM, FDI, GCF are playkurtic, indicating flatter distribution. The Jarque-Bera test confirms that GDP, FDI and ANNI significantly driven from normality, but REM and CGF do not reject normality at 5% level. 4.2 Objective 2: Examining short run dynamic and long run equilibrium between key macroeconomic variables and economic growth. This objective estimates both short run adjustment and long run equilibrium between economic growth as Gross Domestic Product (GDP) and the major macroeconomic variables such as remittance (REM), Foreign Direct Investment (FDI), Gross Capital Formation (GCF) and Annual Net National Income (ANNI) applying Vector Error Correction Model (VECM). 4.2.1 Long-run equilibrium: to analyze long run equilibrium between GDP and key macroeconomic variables let’s start with generating long-run regression equation considering GDP as dependent variable and REM, FDI, GCF and ANNI as independent variables; Long-rum regression equation: GDP = β0 + β1REM + β2FDI + β3CGF + β4ANNI………… (1) Long-run econometric equation with error term, GDPt = β0 + β1REMt + β2FDIt + β3CGFt + β4ANNIt + εt………… (2) Where, €t = Error term; capturing all missing factors Subscript t = Time dimension Long-run estimated equation with error term: GDPt = -2737.1 + 5.29REMt – 14.56FDIt + 999.24CGFt + 1480.99ANNIt + εt……….(3) Table: 3 Johansen Cointegration Test (Summary) Test Result Notes An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7729 Rank (r) r = 2 Maximum Eigenvalue test suggested two cointegrating vectors; supports long run linkage Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. The Johansen Cointegration Test identifies the number of long-run equilibrium relationships among the variables. In Table 3, the test results indicate a cointegration rank of r = 2, meaning there are two distinct long-run relationships among GDP, REM, FDI, GCF, and ANNI. The Maximum Eigenvalue test confirms the presence of these two cointegrating vectors, which supports the existence of stable, long-term linkages between the selected macroeconomic variables in Bangladesh. This result justifies the use of the VECM framework to analyze both short-run dynamics and long-run equilibrium. Table: 4 Long-Run Cointegration Coefficients: Variable Coefficient t-statistic REM 5.29 6.504 FDI -14.56 -2.859 GCF 999.24 __ ANNI 1480.99 __ Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. From above table, it is observed that, most of the estimated long run coefficients among the selected, have positive and significant impact on economic growth except FDI. Starting with remittance, the value of this coefficient is 5.29, which means 1% increase in remittance is associate with 5.29% increase in GDP growth in the long run. For gross capital formation the coefficient is 999.24 which indicates a very strong positive contribution to long term GDP growth. Moreover, the coefficient for annual net national income 1480.99 which also confirms a strong and positive contribution. However, the coefficient of FDI is negative with -14.56, which thus suggests, in long run FDI has a limited or inverse effect on GDP growth. 4.2.2 Short-Run Dynamics: To evaluate the short-run dynamics the Vector Error Correction (VECM) has been utilized. This model includes Error Correction Term (ECT), General form of ETC (Lagged), ECTt-1 = GDPt-1 - (β₀ + β₁ REMt-1 + β₂ FDIt-1 + β₃ GCFt-1 + β₄ ANNIt-1)…….(4) The estimated form of ECT, ECTt-1 = GDPt-1 + 2737.1 - 5.29 REMt-1 + 14.56 FDIt-1 - 999.24 GCFt-1 - 1480.99 ANNIt-1…(5) VECM specification of GDP using ECT, ΔGDPt = αGDP × ECTt-1 + Σγi ΔXt-i + εt………………………………………(6) Estimated VECM equation of GDP, ΔGDP_t = -0.19133 × ECTt-1 + (short-run lagged Δ terms) + εt……….(7) Interpretation: -0.19133 indicates negative but significant value which means GDP corrects 19.13% disequilibrium each year previously occurred. This quantifies speed of adjustment. Table: 5 Error Correction Term (ECT) coefficients Variable ECT Coefficient (α) t-Statistic GDP -0.19133 -3.19133 REM 0.002930 2.96409 FDI __ 1.44584 GCF __ __ ANNI __ __ Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. From the above table it is observed that, the ECT coefficient for GDP is negative and statistically significant with -0.19133, which implies approximately 19.13% disequilibrium has been corrected each year, reflecting adjustment happen in a moderate speed. Remittance showing a significant but positive coefficient which suggest that remittance appears to push the system further from equilibrium rather than guiding it back towards stability. Since the adjustment coefficients of remaining variables such as FDI, GCF An Empirical Analysis of the Dynamic Nexus Between Key Macroeconomic Factors and Sustainable GDP Growth in Bangladesh JEFMS, Volume 08 Issue 12 December 2025 www.ijefm.co.in Page 7730 and ANNI are not statistically significant, they can be considered as weakly exogenous. Because they fail to adjust long-run disequilibrium, shocks related to GDP do not meaningfully influence their short run dynamics. Since GDP is the only variable exhibiting a significant adjustment coefficient, the short run correction towards equilibrium is predominantly driven by change in GDP, rather than change in other variables. 4.3 Objective 3: Identifying weakly exogenous variables in the growth system The third and last objective is to find out the weakly exogenous variables within the GDP growth system. Weak exogeneity refers to the variables that do not adjust in the short-run and to deviations from the long-run variables. When evaluating through VECM model, a weakly exogenous variable has no error correction term variable. Mathematically, The general VECM format for two integrated variables can be presented as, ∆Yt= αYECTt−1 +∑Гi∆Zt−1 +εYt……..(8) ∆Xt= αXECTt−1 +∑Гi∆Zt−1 +εXt…….(9) The estimated long run ECT derived from the cointegrating vector is: ECTt−1 =GDPt−1 −(273701−5.29REMt−1 +14.56FDIt−1 −999.24CGFt−1 − 1480.99ANNIt−1)…………………………(10) The significance of the coefficient of the ECT model determines whether a variable adjust to long-run disequilibrium. Identification of weakly exogenous variables: 1. Gross Domestic Product (GDP): GDP has a negative and significant adjustment coefficient, which means GDP is the endogenous variable in the adjustment process. 2. Remittances (REM): Remittance has a significant and positive adjustment coefficient, which means remittance respond to disequilibrium but in a disruptive direction. Remittance is neither weakly exogenous nor a stabilizer. 3. FDI, CGF and ANNI: These variables have statistically significant adjustment coefficient, which means they do not respond to disequilibrium. Therefore, they considered as weakly exogenous variables. Table: 6 Error Correction Term (ECT) Coefficients (α vector) Variable ECT Coefficient (α) t-Statistic Interpretation GDP -0.19133 -3.19133 Significant; adjusts to restore equilibrium (19.13%) REM 0.002930 2.96409 Significant positive; tends to move system away FDI Not significant 1.44584 Weakly exogenous GCF Not significant — Weakly exogenous ANNI Not significant — Weakly exogenous Source: Author’s calculation based on data from the World Development Indicators (WDI), World Bank. The finding showed in the table indicates a clear definite hierarchy for how the economic variables work together to stabilize the economy, and it is found that GDP is the single stabilizer mechanism which is statistically significant adjustment coefficient with a negative sign. This finding establishes GDP as an endogenous adjustment factor, which automatically adjust to correct the imbalance and restore the stability to ensure sustainable economic growth. On the contrary, the remaining factors such as Foreign Direct Investment and capital formation found to be weakly exogenous since they show no significant coefficient, which means they are not responsive to the error and also not concerned in fixing long term imbalance. This result is consistent with Bangladesh economy, suggesting that, the movement of FDI and Capital formation are governed by external forces rather than adjusting to domestic adjustment issues. Remittances exhibit different outcome. Positive value of the coefficient indicates they push the economy away from equilibrium and possibly amplify short-run instability. However, remittances are neither stabilizer nor weakly exogenous, they introduce short-run volatility because their flow is highly sensitive to external labor market and domestic households. The structural analysis of the VECM exposes a clear overview of long-run economic sustainability in Bangladesh. GDP is the key internal force that fixes the economy. On contrary, FDI, Gross Capital Formation, and National Income are weakly exogenous and externally driven; they fail to adjust to the long-run error. Furthermore, Remittances remain insignificant in correcting error in long-run rather boosts volatility in short run. Hence, government policies should focus directly stimulate GDP to control the fluctuation arising from remittance inflows. 4.4 Diagnosis Tests: