The Effects of Bilateral and Multilateral Official Development Assistance on Vietnam’s Economic Growth

Abstract

This study investigates the effects of bilateral and multilateral official development assistance on Vietnam’s economic growth from 1986 to 2022. Utilizing the autoregressive distributed lag (ARDL) bounds testing approach, our results show that in the shortrun, bilateral official development assistance has a significant positive influence on economic growth, whereas multilateral official development assistance has a significant negative influence on economic growth. However, the empirical findings reveal that both bilateral and multilateral official development assistance have no influence on economic growth in the longterm. Given that bilateral official development assistance has a significantly positive impact on economic growth in the shortrun, Vietnam should strengthen partnerships with donor countries. Tailoring projects to align with bilateral donors’ interests can lead to more effective interventions. In addition, multilateral official development assistance has been found to have a negative impact on economic growth in the shortrun, possibly due to complex approval and implementation processes. Therefore, the government should advocate for more flexible project requirements and reduce bureaucratic hurdles. Simplifying the approval process can help accelerate project implementation and enhance immediate economic benefits. Moreover, because official development assistance does not impact on economic growth in the longterm, Vietnam should focus on sustainable development strategies that reduce dependency on external aid. This includes investing in human capital, innovation, and technology to foster endogenous growth.

1. Introduction

Whenever a nation or territory cannot raise enough money from its citizens to support its economic expansion, it turns to outside lenders and investors. This external finance often consists of both state and private funding. In contrast to private financing, public financing has favorable terms. It is provided through grants or privileged lending with extended payback terms and lower interest charges than those on global personal capital markets (Amoa, 2020). This is made possible by official development assistance (ODA), also referred to as Foreign Development Aid, which is federal funding that primarily supports and fosters the prosperity and well-being of underdeveloped nations (Ono & Sekiyama, 2024). ODA differs from other monetary resource transfers from industrialized to underdeveloped nationsprimarily due to two factors. First, ODA is given to developing nations in two ways: directly and indirectly through multilateral institutions at the choice of the governing bodies of industrialized nations. These bodies are driven by their desires and objectives rather than market outcomes (Cassola et al., 2022). Secondly, ODA is offered on extremely lenient conditions; about 90% are donations, and if loans are issued, they are offered at modest interest rates for extended periods (Sengupta, 2002). Due to these qualities, ODA remains more appealing to less developed nations, making it superior to other forms of outside funding.

The relationship between ODA and EG is complicated. Theoretically, ODA could positively or negatively affect EG depending on governance quality, institutional capacity, and the broader economic environment in the recipient nation. According to the two-gap model proposed by Chenery and Strout (1966), developing countries face two primary gaps: the savings–investment gap and the foreign exchange gap. Therefore, ODA can stimulate EG by bridging these gaps. In addition, endogenous growth theory (Romer, 1994) suggests that internal elements, such as human capital, innovation, and knowledge, are crucial in promoting EG. ODA can boost EG by financing education, research and development, and technology transfer, which contribute to increasing returns to scale and sustained long-term growth. However, moral hazard theory posits that ODA can have adverse effects if governments of recipient countries change their behavior, as the aid they receive protects them from facing the full consequences of their actions. Moreover, public choice theory asserts that the influence of foreign aid depends on how political leaders in recipient countries distribute aid resources. ODA may be redirected to projects that serve political interests rather than those that enhance economic efficiency. Furthermore, Bauer (1972) argues that foreign aid often does more harm than good. He claims that foreign aid can create dependency, undermine local economies, and perpetuate poverty rather than alleviate it. Emphasizing the importance of market mechanisms, Bauer (1972) suggests that EG is best achieved through free markets and individual entrepreneurship rather than government intervention and aid. Easterly (2006) further critiques aid programs, arguing that Western efforts to aid developing countries often lead to unintended negative consequences. He emphasizes the importance of incentives in fostering economic growth, asserting that aid often disrupts local markets and creates dependency rather than encouraging self-sufficiency.

Over the past few decades, many empirical studies have analyzed the impact of ODA on economic growth (EG) in developing countries. Most studies in this research vane have focused on the effect of aggregating ODA on EG with mixed outcomes. Specifically, many empirical studies found a positive effect of ODA on EG (Burnside & Dollar, 2000; Gomanee et al., 2005; Refaei & Sameti, 2015; Das & Sethi, 2020; Azam & Feng, 2022; Wehncke et al., 2023; Fazlly, 2024). However, some studies reported a negative effect of ODA on EG (Mallik, 2008; Liew et al., 2012; Zardoub & Sboui, 2023) or no impact on EG (Ekanayake & Chatrna, 2010; Siraj, 2012; Stojanov et al., 2019; Awino & Kioko, 2022). A key distinction in the discussion of ODA is between bilateral and multilateral aid, each driven by different donor motives, characteristics and conditions of aid, and donor–recipient relationships. Bilateral aid is often aligned with the donor’s economic and strategic interests, as seen in the practices of countries like Japan, the United States, and former colonial powers. In contrast, multilateral aid, provided by institutions like the World Bank (WB) and the International Monetary Fund (IMF), typically includes stringent conditions, such as structural adjustments. The unique technical skills, historical ties, and specific field expertise that bilateral donors possess often create a more tailored and effective aid relationship. These differences underscore the necessity of distinguishing between bilateral and multilateral aid impacts on economic analyses, as their respective influences on growth can vary significantly.

Like other developing countries, Vietnam has depended on foreign capital to augment its resources and accelerate investment from abroad. ODA may complement local savings in a way that other types of foreign transfers cannot, owing to both attributes (Sengupta, 2002). In economically disadvantaged nations like Vietnam, where banking systems are fragile, physical infrastructure is lacking, resources are stationary, and market malfunctions are widespread, ODA turns into one of the most excellent and successful strategies to promote EG. However, Vietnam’s reliance on ODA can lead to vulnerabilities, particularly if donor priorities change or if there is a reduction in funding. In addition, complex procedures and regulations from donor agencies can delay project approval and implementation, reducing responsiveness to local needs. Although the effect of ODA on EG has been substantially discovered in developing countries, the different effects of bilateral and multilateral ODA on EG have not been sufficiently documented, particularly in the context of Vietnam, a transitional economy. Therefore, this study is devoted to filling this void in the literature by investigating the effects of bilateral and multilateral ODA on Vietnam’s EG separately.

The contributions of this study are to broaden the literature on the impact of ODA on EG in a transitional economy. Vietnam offers a compelling case for studying due to its rapid economic transformation with high EG and the substantial aid it has received for infrastructure, social services, and policy reforms over the period of 1986–2022. In addition, while most studies have examined the aggregate effect of ODA on EG, this study focuses on the effects of bilateral and multilateral ODA on Vietnam’s EG separately. This distinction is critical, as bilateral ODA often comes with specific conditions tied to the donor country’s interests, while multilateral ODA is generally managed by international organizations with broader development objectives. Studying the separate impacts of bilateral and multilateral ODA is essential to better understand the different effectiveness of these types of ODA and to provide tailored policy recommendations. Such insights are crucial for optimizing the use of ODA in Vietnam and contributing to the broader discussion on the role of ODA in the economic development of developing nations. As Vietnam progresses from a low-income to a lower-middle-income country, the strategic choice and management of ODA funding will significantly impact the nation’s sustainable development path.

The remainder of the paper is structured as follows. Section 2 reviews the literature and Section 3 describes the data along with the research methodology. Section 4 reports and discusses the empirical results. Lastly, Section 5 provides the conclusion.

2. Literature Review

Many empirical studies that investigate the impact of ODA on EG have been conducted in developing economies for the last few decades. Most empirical studies have examined the effects of aggregate ODA on EG. However, the empirical findings of these studies have been mixed. Many studies documented that ODA has positive effects on EG. Burnside and Dollar (2000) is an early and more influential study in this field. Using a dataset of 56 developing nations from 1970 to 1993, the researchers documented that foreign aid positively impacts EG, but this impact relies on the caliber of the recipient country’s economic policies. Foreign aid is more effective in stimulating EG in nations with good policies, but the influence of foreign aid on EG is negligible or even negative in countries with poor policy environments. Easterly et al. (2004) reassess the empirical findings of Burnside and Dollar (2000) using an expanded dataset. The authors found that the positive relationship between aid and growth in good policy environments does not hold when using the new data. The aid–policy interaction term becomes insignificant and even negative, raising doubts about the robustness of the original conclusions. Moreover, Gomanee et al. (2005) determined the impact of ODA on the EG of 25 sub-Saharan African countries from 1970 to 1997 and found that ODA has a significantly positive impact on EG. Similarly, Eregha and Oziegbe (2016) reported that ODA had a significantly positive effect on EG for Southern Africa, Central Africa, and oil-exporting countries. However, ODA had a significant positive influence on EG for West Africa only when macroeconomic policy environment variables were taken into account. Moreover, using a sample of 111 developing countries from 1970 to 2010, Rahnama et al. (2017) documented that ODA has positive effects on EG in high-income developing nations. Furthermore, Azam and Feng (2022) explored how foreign aid impacted the EG of 37 developing economies from 1985 to 2018. Generally, they found a positive effect of ODA on EG. However, the influence of ODA depends on some factors, such as the types of aid and the caliber of institutions within the recipient country. Additionally, Wehncke et al. (2023) estimated the relationships between ODA, foreign direct investment (FDI), and EG for 20 developing countries. The researchers found that ODA has a significantly positive influence on EG in both the shortterm and longterm, but the long-term impact of ODA on EG is more significant than those observed in the shortterm. Regarding specific countries, Feeny (2005) examined the effect of ODA on Papua New Guinea’s EG from 1965 to 1999. The empirical findings of the study confirmed that ODA positively influences EG. In addition, Refaei and Sameti (2015) explored the influence of ODA on Iran’s EG from 1980 to 2012. They reported that ODA positively impacted EG in the longterm. Moreover, Das and Sethi (2020) investigated how various external financial inflows, including FDI, remittances, and ODA, impacted the EG of India and Sri Lanka and found that ODA had a positive influence on EG for both countries. Nguyen et al. (2022) explored the influence of Japanese ODA on the EG of ASEAN countries for the period from 2008 to 2020. This study documents that the ODA has a positive effect on the EG of ASEAN countries. Similarly, Chansombuth (2023) investigated the relation between Japan’s ODA and EG in Laos for the period from 1990 to 2020 and concluded that ODA positively influences EG in the longrun, but it has no impact on EG in the shortrun. Additionally, Guo et al. (2024) examined the influence of ODA on the development of renewable energy in sub-Saharan African (SSA) countries. Using the double machine learning approach, they found that ODA has a positive influence on the development of renewable energy in these countries. Moreover, the effectiveness of aid in promoting renewable energy is significantly enhanced when recipient countries possess strong management systems and transparent policy environments. Recently, using the autoregressive distributed lag (ARDL) approach with the sample from 1980 to 2021, Fazlly (2024) also reported a long-term positive effect of ODA on EG for Afghanistan. In addition, Orchi and Ahmed (2024) investigated the symmetric and asymmetric effects of ODA and remittances on EG of Bangladesh. The findings derived from the ARDL and Non-Linear ARDL (NARDL) consistently confirm that ODA has a positive effect on Bangladesh’s EG.

In contrast, some studies have provided empirical evidence of the negative effect of ODA on EG. Mallik (2008) used a cointegration approach to ascertain the long-term relationship between ODA and the EG of the six poorest African countries. The findings of the study confirmed that ODA had a negative long-run effect on the EG of five countries, namely, Central African Republic, Malawi, Mali, Niger, and Sierra Leone. Similarly, Liew et al. (2012) also reported that ODA had a significantly negative impact on EG for five East African countries. In addition, Driffield and Jones (2013) investigated the effects of FDI, ODA, and remittances of migrants on EG in developing economies and found a negative effect of ODA on EG when the interaction terms are not considered. However, the impact of ODA on EG becomes positive when the bureaucratic quality factor is considered. In other words, the effectiveness of ODA on EG is conditional upon the quality of the bureaucracy in recipientcountries. Specifically, countries with stronger bureaucratic structures are better able to manage and distribute aid effectively. The findings suggest that improving bureaucratic quality should be a priority in recipient countries, as it enhances the positive effects of ODA on EG. Moreover, Rahnama et al. (2017) documented that ODA has a negative impact on EG in low-income developing nations. Additionally, Zardoub and Sboui (2023) analyzed the effects of capital inflows, including ODA on the EG of 41 developing countries from 1990 to 2016. Their findings also confirmed that ODA has a significantly negative effect on EG. In a recent study, Moloi (2024) investigated the relationship between ODA, FDI, and EG in 30 African countries. The findings from this study also revealed that ODA is negatively associated with EG. Several studies, in particular, have found no significant effects of ODA on EG (Ekanayake & Chatrna, 2010; Siraj, 2012; Stojanov et al., 2019; Awino & Kioko, 2022).

Some empirical studies have investigated the effects of bilateral and multilateral aid on EG for developing countries. Specifically, Ram (2003) examined how bilateral and multilateral ODA impacted the EG of 56 developing countries. The empirical findings confirmed that multilateral ODA has a negative effect on EG, whereas bilateral ODA greatly boosts it. Additionally, Amoa (2020) determined the effects of bilateral and multilateral ODA and EG of six African countries from 1978 to 2010. Amoa (2020) reported that both types of ODA have negative effects on EG. In addition, this study revealed that the interaction between ODA and FDI significantly boosted EG in the studied countries. Moreover, Edo et al. (2023) investigated how different types of ODA, including bilateral and multilateral ODA, impacted the EG of sub-Saharan African economies. The researchers documented that both bilateral and multilateral ODA had significantly positive effects on the countries’ EG, but the effects depended on how they were administered and utilized. Specifically, the impact of bilateral ODA is more significant than multilateral. Similarly, Wambaka (2023) estimated the effects of bilateral and multilateral aid on EG for 28 middle- and low-income sub-Saharan African nations from 1999 to 2015. The empirical findings indicated that only multilateral aid had a significantly positive impact on EG and that this impact depended on the presence of high-quality institutions.

In summary, theoretical frameworks posit the potential for ODA to promote EG but also caution against potential negative effects. Moreover, analysis of the influence of ODA on EG has not reached a consensus. Specifically, most studies reported a positive effect of ODA on EG, whereas others found a negative or no influence of foreign aid on EG. The diverse effects of ODA on EG often depend on the types of ODA, institutional capacity, and quality of governance in the recipient country.

3. Data and Research Methodology

3.1. Data

This study utilizes the annual series of economic growth, bilateral ODA, multilateral ODA, FDI, and gross savings on the GDP of Vietnam from 1986 to 2022. This period was chosen for this study because Vietnam initiated its comprehensive economic reforms (Doi Moi) in 1986, marking the beginning of its economic opening and the end of a period of central planning. The data for this analysis were acquired from the World Bank. Specifically, the data source and measurements are presented in

Table 1. Data source and measurements.

Data	Measurements	Data Source
Vietnam’s economic growth	Percent (%)	World Bank (WB)
Vietnam’s bilateral ODA inflows	USD	World Bank (WB)
Vietnam’s multilateral ODA inflows	USD	World Bank (WB)
Vietnam’s FDI inflows	USD	World Bank (WB)
Vietnam’s gross savings on GDP	Percent (%)	World Bank (WB)

.

3.2. Research Methodology

Theoretically, the two-gap model, presented by Chenery and Strout (1966), asserts that developing countries face two primary gaps: the savings–investment gap and the foreign exchange gap. Therefore, ODA can stimulate EG by bridging these gaps. In addition, endogenous growth theory suggests that internal elements, such as human capital, innovation, and knowledge, are crucial in promoting EG. ODA can boost EG by financing education, research and development, and technology transfer, which contribute to increasing returns to scale and sustained long-term growth. Empirically, several empirical studies confirmed that ODA has positive effects on EG (Burnside & Dollar, 2000; Gomanee et al., 2005; Refaei & Sameti, 2015; Das & Sethi, 2020; Azam & Feng, 2022; Wehncke et al., 2023; Fazlly, 2024). As a result, it is hypothesized that ODA has a significant positive influence on Vietnam’s EG. In addition, according to the endogenous growth theory, higher national savings can lead to more investment in research and development, which boosts productivity and growth. Therefore, it is expected that national savings have a positive effect on EG. Moreover, several studies document that foreign direct investment (FDI) has a significant influence on EG (Driffield & Jones, 2013; Amoa, 2020; Wehncke et al., 2023; Zardoub & Sboui, 2023; Moloi, 2024). Drawing from prior research, it is expected that FDI has a positive influence on EG. Therefore, this study investigates the effects of ODA on the EG of Vietnam by employing the baseline regression model as follows:

$${\mathrm{LNGROWTH}}_t={\beta }_0+{\beta }_1{\mathrm{LNBODA}}_t+{\beta }_2{\mathrm{LNMODA}}_t+{\beta }_3{\mathrm{LNSAV}}_t+{\beta }_4{\mathrm{LNFDI}}_t+{\epsilon }_t$$

(1)

where

-

LNGROWTH is thenatural logarithm of Vietnam’s EG;

-

LNBODA is thenatural logarithm of Vietnam’s bilateral ODA inflows;

-

LNMODA is thenatural logarithm of Vietnam’s multilateral ODA inflows;

-

LNSAV is thenatural logarithm of Vietnam’s gross savings on GDP;

-

LNFDI is thenatural logarithm of Vietnam’s FDIinflows.

To capture the short-run and long-run effects of bilateral and multilateral ODA on the EG of Vietnam, the ARDL bounds testing approach developed by Pesaran et al. (2001) is applied in this study. This study employs the ARDL approach because this technique has some advantages compared to other cointegration methods (Truong et al., 2024). Firstly, the ARDL model does not require all variables to have the same integration order. Instead, this approach just requires that the integration order of all studied variables does not exceed 1. Secondly, the ARDL approach is generally more reliable and robust than other approaches for studies with a limited number of observations, like this study. Finally, an error correction model (ECM) can be estimated from the ARDL model, and, hence, the short-run and the long-run effects of explanatory variables on the dependent variable can be simultaneously estimated. The ARDL representation for Equation (1) is specified as follows:

$$\begin{array}{l}{\Delta \mathrm{LNGROWTH}}_t={\beta }_0+{\displaystyle \sum _{i=1}^{q_1}{\beta }_{1i}}{\Delta \mathrm{LNGROWTH}}_{t-i}+{\displaystyle \sum _{i=0}^{q_2}{\beta }_{2i}}\Delta {\mathrm{LNBODA}}_{t-i}+{\displaystyle \sum _{i=0}^{q_3}{\beta }_{3i}}\Delta {\mathrm{LNMODA}}_{t-i}+\\ {\displaystyle \sum _{i=0}^{q_4}{\beta }_{4i}\Delta {\mathrm{LNSAV}}_{t-i}}+{\displaystyle \sum _{i=0}^{q_5}{\beta }_{5i}}\Delta {\mathrm{LNFDI}}_{t-i}+{\lambda }_1{\mathrm{LNGROWTH}}_{t-1}+{\lambda }_2{\mathrm{LNBODA}}_{t-1}+{\lambda }_3{\mathrm{LNMODA}}_{t-1}+\\ {\lambda }_5{\mathrm{LNSAV}}_{t-1}+{\lambda }_5{\mathrm{LNFDI}}_{t-1}+{\epsilon }_t\end{array}$$

(2)

where

-

$\Delta$ is the first difference of the variables;

-

q₁, q₂, q₃, q₄, and q₅ are the optimal number of lags for the variables. In this study, the optimal number of lags for variables in the model (the optimal model) is determined using the Akaike information criterion (AIC) technique.

It is noted that the ARDL model requires that the integration orders of all variables do not exceed 1 (Truong et al., 2024). Therefore, unit root tests need to be conducted to ensure that all studied variables fulfill the model requirements. This study employs the widely used augmented Dickey–Fuller (ADF) test to determine the integration orders of all variables. Before estimating the long-term effects, it is crucial to know whether the variables in the model are cointegrated. If they are not, then the long-term estimation would provide misleading results. The ARDL bounds test helps to confirm whether the variables are suitable for long-term analysis. The null hypothesis (H₀) for the ARDL bound test is that no cointegration exists across the variables. If the F-statistic computed from the bounds test exceeds the critical value at the given level, the null hypothesis is rejected, indicating the existence of cointegration across the variables. Once cointegration is confirmed, the short-term and long-term effects of bilateral and multilateral ODA on the EG are simultaneously estimated by the ARDL error correction model. Finally, to check the robustness of the model, the Breach–Godfrey test for autocorrelation and the ARCH test for heteroscedasticity are employed in this study.

4. Empirical Results

4.1. Descriptive Statistics of the Sample

Based on the data obtained from the WB, some descriptive statistics for the studied variables over the period from 1986 to 2022 are calculated and presented in

Table 2. Summary statistics of the sample.

Variables	Minimum	Mean	Maximum	Standard Deviation
GROWTH (%)	2.55	6.45	9.54	1.70
BODA (million USD)	124.28	2107.96	5379.60	1531.00
MODA (million USD)	10.99	40.56	72.06	15.75
Gross savings on GDP (%)	−2.65	24.52	35.77	10.97
FDI (million USD)	0.04	5677.55	17,900.00	5798.15

. Table 2 shows that Vietnam’s EG over the sample period was rather high and stable. Specifically, the average EG over the studied period is 6.45%, while the standard deviation is only 1.70%. In addition, Figure 1 illustrates Vietnam’s EG from 1986 to 2022. It is observed that EG appears to accelerate notably during the period from 1986 to 1997, suggesting a period of robust economic reforms and integration into the global economy. The growth rate seems to stabilize from 1998 to 2019, but it decreased significantly during the period of 2020–2021 due to the COVID-19 pandemic. Moreover, statistics presented in Table 2 show that the value of bilateral ODA is significantly higher than multilateral ODA inflows to Vietnam, meaning that Vietnam’s ODA is mainly bilateral. In fact, the mean of bilateral ODA during the sample period was USD 2107.96 million, whereas the average of multilateral ODA was only USD 40.56 million. Specifically, it is observed in Figure 2 that the bilateral ODA was highly volatile over the studied period, with a range from USD 124.28 million to USD 5379.60 million. Thebilateral ODAincreased continuously during the period from 1986 to 2012, but it dropped dramatically from 2013 to 2022. Additionally, Figure 3 shows that Vietnam’s multilateral ODA inflows started modestly in the late 1980s but began to show significant growth in the mid-1990s. However, it dropped remarkably during the period from 1994 to 2003. Then, Vietnam’s multilateral ODA inflows exhibited a consistent upward trend during the period from 2005 to 2021, reflecting enhanced engagement with multilateral organizations. Additionally, Table 1 indicates that the average of Vietnam’s gross savings on GDP for the period from 1986 to 2022 is 24.52%, ranging from −2.65% to 35.77%. Finally, statistics presented in Table 1 reveal that Vietnam’s FDI highly fluctuated from 1986 to 2022. In fact, the mean of FDI inflows was USD 5677.55 million, with a standard deviation of 5798.15.

4.2. Results of the ADF Test

The analysis for the ADF test provided in

Table 3. Results of the ADF test.

Variables	Without Trend	With Trend
LNGROWTH
Level	−5.01 ***	−5.05 ***
LNBODA
Level	−2.53	−0.36
First difference	−7.24 ***	−9.40 ***
LNLNMODA
Level	−0.57	−1.93
First difference	−5.77 ***	−5.88 ***
LNSAV
Level	−2.32	−1.93
First difference	−5.92 ***	−6.17 ***
LNFDI
Level	−5.81 ***	−7.16 ***

*** represents significance at the 1% level.

indicates that the LNGROWTH and LNFDI series are I(0), while the LNBODA, LNMODA, and LNSAV series are I(1). These findings suggest that all the variables satisfy the requirements of the ARDL bound test.

4.3. ARDL Bound Test for Cointegration

This research utilizes the bounds test introduced by Pesaran et al. (2001) to examine the long-term relation across the variables. According to the Akaike information criterion, the optimal model to use for the bounds test is ARDL(4,4,4,4). The analysis for the bounds test contained in

Table 4. Results of the bounds test.

Model	k	F-Statistic	Significance Level	Critical Value
				I(0)	I(1)
ARDL(4,4,4,4)	4	5.55 **	5%	3.23	4.35
			1%	4.29	5.61

k represents the number of explanatory variables. ** indicates significance at the 5% level.

provides evidence that the null hypothesis of no cointegration among the variables is rejected at the five percent level. This evidence indicates a long-term equilibrium relationship across the model variables.

4.4. Short-Term and Long-Term Effects of Bilateral and Multilateral ODA on EG

With the evidence of cointegration among the variables in the model, the short-run and long-run coefficients are estimated using the ARDL error correction model. The empirical findings presented in

Table 5. Estimated results of the short-term and long-term coefficients.

Variables	Coefficients	t-Statistics
Panel A: Short-term estimates
$\Delta \mathrm{LNGROWTH}(-1)$	1.2007	2.72 **
$\Delta \mathrm{LNGROWTH}(-2)$	0.2323	0.48
$\Delta \mathrm{LNGROWTH}(-3)$	1.2227	2.80 **
$\Delta \mathrm{LNBODA}$	0.5235	1.98 *
$\Delta \mathrm{LNBODA}(-1)$	−0.4778	−1.72
$\Delta \mathrm{LNBODA}(-2)$	0.6001	2.01 *
$\Delta \mathrm{LNBODA}(-3)$	0.7918	3.43 ***
$\Delta \mathrm{LNMODA}$	−0.3198	−1.09
$\Delta \mathrm{LNMODA}(-1)$	0.1326	0.50
$\Delta \mathrm{LNMODA}(-2)$	0.0854	0.31
$\Delta \mathrm{LNMODA}(-3)$	−1.5157	−3.96 ***
$\Delta \mathrm{LNSAV}$	1.5327	2.32 ***
$\Delta \mathrm{LNSAV}(-1)$	1.4675	2.24 *
$\Delta \mathrm{LNSAV}(-2)$	0.9505	1.55
$\Delta \mathrm{LNSAV}(-3)$	1.0136	2.55 **
$\Delta \mathrm{LNFDI}$	0.9070	3.47 ***
$\Delta \mathrm{LNFDI}(-1)$	−0.2606	−1.40
$\Delta \mathrm{LNFDI}(-2)$	0.2160	1.22
$\Delta \mathrm{LNFDI}(-3)$	−0.1618	−0.89
ECM(−1)	0.8653	−3.90 ***
Panel B: Long-term estimates
Constant	−4.3432	−1.86
LNBODA	−0.1174	−0.81
LNMODA	0.4891	1.76
LNSAV	0.6043	1.20
LNFDI	0.1682	2.68 **

***, **, and * represent significance at the 1%, 5%, and 10% levels, respectively.

indicate that in the shortrun, bilateral ODA has a significant positive influence on EG for different time lags. Specifically, a 1% increase in bilateral ODA results in a 0.60% and 0.79% increase in EG for the 2-year and 3-year lags, respectively. However, it is found in Table 5 that in the shortterm, multilateral ODA has a negative effect on EG at a significance level of 1%. Specifically, a 1% increase in multilateral ODA leads to a 1.52% decrease in EG for the 3-year lag. In addition, Table 5 reveals that in the shortrun, the national gross savings (SAV) has a significantly positive effect on EG. In fact, a 1% increase in SAV is contemporaneously associated with a 1.53% increase in EG at asignificance level of 5%. Moreover, the results reported in Table 5 indicate that in the shortrun, FDI positively impacts EG at a 1% level of significance. Specifically, a 1% increase in FDI is contemporaneously associated with a 0.91% increase in EG. Furthermore, the coefficient of error correction terms is −0.8653, indicating that 86.53% of the divergence from the long-run equilibrium due to a shock in the current year will be adjusted back toward equilibrium in the next year.

As well as approximating short-term effects, the ARDL methodology allows for the estimation of the long-term ODA on EG. The empirical results reported in Table 5 confirm that both bilateral and multilateral ODA have no statistically significant impact on EG. Especially, in the longrun, the empirical findings confirm that FDI has a significant positive influence on EG at the 5% level. Specifically, a 1% increase in FDI leads to a 0.17% increase in the longrun. However, we found no evidence of a long-term effect of SAV on EG.

Overall, this study finds a considerable difference between the effects of the two ODA components. Specifically, bilateral ODA has a significantly positive impact on EG whereas multilateral ODA has a negative effect on EG. However, our findings reveal that both bilateral and multilateral ODA have no significant influence on EG in the longterm. These findings are totally consistent with previous empirical findings of Ram (2003) and partially align with the findings of Edo et al. (2023). However, this evidence is contrary to the findings documented by Amoa (2020) and Wambaka (2023). There are some reasons to explain these different effects. First, bilateral donors could have closer ties and a better understanding of Vietnam’s context, challenges, and development needs. Therefore, bilateral ODA is financed by more targeted and effective interventions that address the specific needs of the country. Second, bilateral ODA could have deeper expertise and experience in specific sectors or development challenges, allowing it to provide more tailored and effective assistance. Moreover, the negative impact of multilateral ODA on the EG of Vietnam could be explained by several reasons. First, multilateral ODA projects generally require strict compliance with international environmental, social, and financial management standards. While these standards are crucial for sustainable development, they can impose significant administrative burdens, slow project progress, and reduce the ability to achieve immediate economic impacts. Second, some multilateral ODA projects could not fully consider the environmental and social context of Vietnam in the transitional period. Therefore, projects that overlook the environmental sustainability or social impacts might lead to long-term costs that offset any short-term economic gains. Third, multilateral ODA projects in Vietnam often involve multiple stakeholders, including international organizations, donor countries, and domestic management agencies. This complexity results in prolonged approval and implementation processes. Delays in implementation not only increase project costs but also reduce the immediate economic impact, negatively affecting short-run economic growth.

4.5. Diagnostic and Structural Stability Tests

The results of the Breusch–Godfrey test reported in

Table 6. Results of the Breusch–Godfrey and ARCH tests.

Diagnostic Tests	Statistics	p-Value
Breusch–Godfrey test H0: No autocorrelation	1.08	0.398
ARCH test H0: No ARCH effects	0.49	0.488

reveal that autocorrelation among the residuals does not exist in the model. Moreover, the results of the ARCH test indicate that the residuals exhibit homoscedasticity. The results derived from these tests confirm the reliability and validity of the estimated results.

Furthermore, this study employs the cumulative sum (CUSUM) and cumulative sum of squared (CUSUMSQ) methodologies to examine the long-term stability of the model. The findings illustrated in Figure 4 and Figure 5 confirm model stability across the study period.

5. Conclusions

This study empirically examined the short-term and long-term influence of bilateral and multilateral ODA on the EG of Vietnam over the transitional period from 1986 to 2022. The results obtained from the ARDL model confirmed that there is a significant difference between the effects of the two ODA components on EG in the shortrun. Specifically, bilateral ODA has a significant positive influence on EG, whereas multilateral ODA reveals a significant negative influence on EG. However, the empirical findings indicated that both bilateral and multilateral ODA have no impact on EG in the longrun. In addition, the findings confirmed that FDI has a significantly positive effect on EG in the shortrun but has no impact on EG in the longrun. Finally, the findings obtained from the ECM indicated that 86.53% of the divergence from the long-term equilibrium results from a shock in the current year will be adjusted back toward equilibrium in the next year.

Based on the findings, some policy implications can be drawn for Vietnamese policymakers to enhance the effectiveness of ODA and drive sustainable economic growth. First, given that bilateral ODA has a significantly positive impact on EG in the shortrun, Vietnam should strengthen partnerships with donor countries. Tailoring projects to align with bilateral donors’ interests can lead to more effective interventions. Second, multilateral ODA has been found to have a negative impact on EG in the shortrun, possibly due to complex approval and implementation processes. Therefore, the government should advocate for more flexible project requirements and reduce bureaucratic hurdles. Simplifying the approval process can help accelerate project implementation and enhance immediate economic benefits. Third, while ODA does not impact EG in the longterm, Vietnam should focus on sustainable development strategies that reduce dependency on external aid. This includes investing in human capital, innovation, and technology to foster endogenous growth. Finally, Vietnam should improve governance and institutional frameworks to amplify the effectiveness of both bilateral and multilateral ODA. Policies should aim to enhance transparency, accountability, and capacity in managing ODA projects.

The contributions of this study are to enrich the literature on the different effects of bilateral and multilateral ODA on EG in the context of Vietnam, a transitional economy. Studying the separate impacts of bilateral and multilateral ODA is essential to better understand the different effectiveness of these types of ODA and to provide tailored policy recommendations. Such insights are crucial for optimizing the use of ODA in Vietnam and contributing to the broader discussion on the role of ODA in the economic development of developing nations. While this study makes several contributions to the literature, it still has some limitations that future research has the potential to explore. Due to the limitations of the data, this analysis explores the influence of bilateral and multilateral ODA on EG without taking into account the moderating effects of the caliber of governance and institutional capacity variables on the effects of bilateral and multilateral ODA. Therefore, future studies should investigate how the caliber of governance and institutional capacity moderate the impact of bilateral and multilateral ODA on Vietnam’s EG. Another limitation of this analysis is the COVID-19 pandemic’s possible influence on Vietnam’s ODA inflows and economic growth, but it is not examined when analyzing the influence of ODA on the EG of Vietnam. This limit is a topic of possible interest for future research.