The Relationship Between Foreign Aid and Economic Growth: Empirical Evidence from Somalia

The Keynesian paradigm of economic expansion [8, 16] studies the role that foreign aid plays in bridging the savings gap and spurring domestic investment, which helps developing countries grow their economies more quickly. According to Rosenstein-Rodan [17], the well-known big-push principle, an underdeveloped economy to address growth hurdles, a minimum amount of high investment is required. In this case, foreign aid the following studies [13, 18, 19] concluded that foreign aid enhances a direct investment in non-industrialized nations by indirectly raising capital through increased domestic income and savings. Additionally, assistance in the form of technical assistants improves technical knowledge and human talent, thereby contributing to the economies of developing countries. Numerous theoretical and empirical studies on the effectiveness of international aid have been conducted in different parts of the world. As many economists attempted to experimentally investigate the relationship between the foreign aid and economic growth, the discussions on the impact of foreign aid on economic growth became a topic of interest. The study examines the pertinent empirical literature that used various modeling techniques to determine the scope.

According to Geetilaxmi et al. [20] this study used annual data from 1970 to 2014 to evaluate how foreign aid affects the expansion of the Indian economy. The co-integration test confirms a long-term link between India's real GDP per capita and foreign aid. The analysis concludes that both in the long run and the short run, foreign aid has a positive and considerable impact on India's economic development. Since 1960, countries in sub-Saharan Africa have received a significant amount of international aid. Throughout the 1960s, 1970s, and 1980s, assistance payments to the region grew steadily, but a significant drop in both multilateral and bilateral payments began in the mid-1990s [4]. It is concerning that this decrease in foreign aid was not entirely attributable to a drop in global aid donations but partly to donors' propensity to divert funds away from Africa and toward East Europe and East Asia, regions with significantly higher living standards [14].

Numerous studies on the relationship between foreign aid and economic growth have been conducted over the past three decades from a variety of ideological and methodological perspectives [15]. Empirical studies between foreign aid and economic growth have not been able to provide conclusive results. Some studies reported a positive significant effect between foreign aid and economic growth, other studies reported a negative significant effect, and a few others claimed that foreign aid had no significant impact on economic development. According to this study Karki [21] evaluated the effect of foreign aid on Tanzania's economic development. For instance, the study made use of yearly data spanning the years 1992 to 2014. Error Correction Model (ECM) was also used in the study to analyze the short- and long-term effects of international trade on Tanzania's economic development. The result shows that the model's long-term relationship between economic development and overseas aid is present. Although this study looked at how international trade affects economic growth, it did not directly address how foreign aid affects economic growth. This is supported Mustafa et al. [22], the study used autoregressive distrusted lag (ARDL) bounds tests to look into the relationship between foreign aid and Sudan's economic development. using information from time series that ranged from 1980 to 2015. The study's conclusions show that official development assistance (ODA), a type of foreign aid, has a sustained, significant, and positive relationship with Sudan's growth in the economy. Despite the fact that the study explicitly stated a relationship between foreign aid and economic growth, it exclusively focused on Sudan's economic development. As a result, it is impossible to generalize Somalia's economic growth to other African nations with diverse economic circumstances, even though it has made some improvement in recent years. Somalia has seen poor economic growth over the past two decades. Lastly, this study is also supported by Hongxing et al. [1] Foreign aid and economic growth: Do energy use, trade openness, and CO₂ emissions matter? is a well-researched paper? African trading blocs' heterogeneous evidence from DSUR. The findings of this study indicated a significant correlation between foreign aid and economic growth in African nations. Although though the study focused on the relationship between economic development and foreign aid in African nations, the economic conditions and foreign aid in various African nations vary. For instance, foreign aid to Somalia has increased recently while foreign direct investment has decreased [11]. Therefore, there is little research regarding the exact relationship between increasing foreign aid and economic growth in Somalia.

On the other hand, other studies indicate a negative correlation between economic growth and foreign aid. For example, Yiew and Lau [23] This study used 95 developing countries as their sample to examine the function of foreign aid (ODA) and its effects on GDP. According to the study's findings, foreign aid initially has a detrimental impact on a country's growth but eventually helps that country's economy thrives. Hence, if foreign aid discourages the labor force from going to work since they are receiving it without working, it could hinder the country's economic expansion. Hence, staying at home instead of working is preferable. For instance, the Somali folks assume that foreign aid only comes to the nation when the farmers produce their own products, which deters farms from abstaining to farm. Therefore, that could cause the foreign to have negative impact on economic growth. Similarly a study conducted in Bangladesh by Zobair and Uddin [18] examined historical data on economic growth indicators and foreign investment in Bangladesh to learn more about the effects of these inflows. In this regard, data from the World Bank's database have been gathered for the years 1976 to 2017. The ARDL technique is used to do the analysis. The study came to the conclusion that foreign aid had a negative impact on economic growth. In other study in conducted in South Asian and South-East Asian by Rao et al. [5] the interactions between foreign aid, foreign direct investment (FDI), and South Asian and South-East Asian economic growth between 1980 and 2016. The study's conclusions showed that there was little correlation between economic growth and foreign aid. Lastly, a study conducted in Ghana by Appiah-Konadu et al. [24], this study used time series data from 1972 to 2012 to investigate how foreign aid affected Ghana's economic growth. The long-run and short-run links between aid and economic growth were investigated using the ARDL approach to co-integration (bounds test). The study concludes that foreign aid has negative significant impact on economic growth.

Apart from a positive and negative impact of foreign aid on economic growth, there are other studies indicating there is no a significant relationship between foreign aid and economic growth. For instance, Babalola and Shittu [25] studied the relationship between foreign aid and economic growth in West Africa: Examining Institutional Roles. The study analyzes panel data from the World Bank's global development and governance indices, spanning the years 1996 to 2017, using the autoregressive distributed lag technique. The results show that foreign aid has a neutral effect on economic growth. In Somalia it is conducted a study titled: effects of foreign debt and foreign aid on economic growth in Somalia [4]. However there are three issues with the study. First, the authors employed ordinary least squares (OLS) approach as fundamental model without exploratory analysis or causal inferences. second, the work was also published in un rebuttable journal. Third, the OLS method is only used. Therefore, J & J co-integrating method was hardly ever used in studies on Somalia. Co-integrating vectors are easily used in conjunction with FMOLS and DOLS. Contrary to the Somali-related study mentioned above, this one tackles the shortcomings by analyzing the effect of foreign aid on Somalia's economic growth. Thus, this study uses a recently developed econometric technique based on the co-integrating method over the years 1989 to 2017.

Researcher provides the findings of this empirical study in this section. The descriptive statistics of the data being considered are presented first (see Table 2). The summary statistics of the pertinent factors taken into account for this empirical study of Somalia from 1989 to 2017 are shown in Table 2. We observe that the dependent variable LRGDP has a mean value of 20.849 and a standard deviation of 0.174, and that it fluctuates regularly from 20.578 to 21.140. Table 2 displays a matrix of correlation coefficients for several factors. We note that there is a positive correlation between our main variables—the per capita GDP and foreign aid—and that this correlation is statistically significant. Further econometric research is motivated by this positive connection between the variables.

Table 2. Descriptive statistics

The paper does an empirical investigation using a wide range of econometric methodologies. Since the study focuses on 28 years of annual time-series data, it is crucial to solve the non-stationary problem, which is why multiple panel unit root tests are conducted. The panel unit root test results for different tests at both levels and the initial differences are shown in Table 2. For the level and first differenced series of these variables for the two categories of individual intercept and individual intercept and trend, unit root statistics are presented. All of the variables are level non-stationary and first difference stationary, according to the unit root tests. To do that, author must first determine the ideal lag time. The right lag must be selected in order to use advanced econometric techniques like co-integration tests, hence the best lag selection criteria are also used. The right lag must be selected in order to use advanced econometric techniques like the co-integration test, hence the best lag selection criteria are also used.

4.1 Unit root

Testing the stationary of time series data is necessary because if author estimates a non-stationary series, it may generate a spurious regression that provides misleading statistical evidence of a linear relationship between variables. The researcher used both Phillips Perron (PP) and Augmented Dickey-Fuller (ADF) techniques to examine the unit root problem. The alternative hypothesis demonstrates the absence of a unit root problem while the null hypothesis of ADF and PP demonstrates the series' non-stationarity, which indicates the presence of a unit root problem. Therefore, the t-statistic of the variable must be higher than the variable's given critical t-value in order to reject the null hypothesis that the data is non-stationary and fail to reject the null hypothesis that it is stationary. However, if we fail to reject the null hypothesis, the variables are non-stationary. The results of the unit root tests are displayed in Table 3 together with the integration order at level (I (0)) and the first difference level (I (1). As shown in Table 3, all the variables are non-stationary in the levels in both ADF and PP results. Therefore, at a level, all of the variables have a unit root, but they are stationary at the first difference level (I (1)) at conventional significance levels. Because all the variables become stationary at the first difference (I (1)) and are free of unit root issues at the first difference, this suggests that a J & J co-integrating regression can be used. Given that ADF and PP show that our data is free of unit root issues, this suggests that a co-integrating regression can be done comprehensively. The researchers chose three as the highest lag order because the study's sample is only 29 observations, and we subsequently decreased it to two and one until the ideal lag order is free of diagnostic errors and demonstrates model stability.

Table 3. Unit root tests

***, **, and * indicate significance level at 1%, 5%, and 10%, respectively ADF augmented Dickey-Fuller, PP Phillips–Perron

Table 4. Long run cointegration

Author used the J & J co-integrating to regress for the long-run estimation (Table 4). Table 4 reflects the results after generating variables using J & J co-integrating method. The result shows that foreign aid in Sweden, US aid, capital, and labor are statistically substantial and have a positive association with economic growth and UK aid is statistically insignificant. This result is consistent with [1, 6, 21]. However, UK aid is not statistically significant which is consistent with this study [25] studied the relationship between foreign aid and economic development in West Africa: Examining Institutional Roles. The study uses the autoregressive distributed lag technique to analyze panel data from the World Bank's global development and governance indices, covering the years 1996 to 2017.

In time series analysis, a co-integration test can be performed once it has been established that the series have achieved the first-order stationary and the number of lags. The Johansen co-integration test does not require the separation of the dependent and independent variables in the model. The data level results from the Johansen co-integration test are used for the analysis in Table 5 and Table 6.

4.2 Johansen co-integration test

After determining the ideal lags, we may proceed to determine the number of equilibrium relations that exist between the four variables (1 lag). For the Johansen co-integration test, the system under test has eight variables (LRGDPt, LGFKt, LLt, LSWEDEN AIDt, LUK AID, and LUS AIDt). The Maximum Eigenvalue and Trace tests, as summarized in the following table, must be run in order to conduct the test:

Table 5. Johansen cointegration test results

Trace test indicates a cointegrated relationship of 5%

Table 6. Unconstrained co-integration Eigenvalue Rank Test (Eigenvalue)

The maximum eigenvalue test indicates a co-integrated relationship at the level of 5%.

The results of the trace and Maximum Eigenvalue tests show that there is co-integration across variables since the null hypothesis of the absence of co-integration is rejected at the 5% level. Table 3 shows that the Max-Eigen test finds 3 co-integration equations at the 5% threshold, but the trace test finds 5 co-integration equations. Since there is just one co-integration connection.

It is essential for policy development to improve and revitalize the economic shocks to conduct a thorough empirical study that examines the influence of foreign aid, capital, and labor on economic development in Somalia from 1989 to 2017. This study used the J & J co-integrating framework, which simulates long-run co-integration. This study's weakness is that it did not account for other significant macroeconomic variables that have an impact on economic growth. The factors that our study specifically left out of the model include exchange rate export and import, economic stability index, and foreign direct investment (FDI). However, due to the time limit and data availability concerns, and Somalia's over-decadal data constraint, these variables were not included.

According to the study's findings, the effects of Sweden, US, and UK aid, capital, and labor on economic growth are all long-run co-integrated. First, the relationship between Sweden's aid and long-term economic growth is noted. economic growth rises by 0.030% over time for every 1% increase in Sweden's aid. Long-term US aid affects economic growth as well. Long-term, an increase in US aid of 1% raises economic growth by 0.028%. Additionally, capital is seen to contribute to long-term economic growth. Long-term economic growth progress of 0.794 percent is induced by a 1% increase in capital, suggesting that capital growth is more responsive to economic growth. First, the sensitivity might be attributed to growth being more vulnerable to capital shocks as a result of having the greatest and newest technologies. Second, economic development is more susceptible to the increasing capital that has been used which results in economic development and prosperity for the country.

On the contrary hand, it appears that UK aid does not have a significant effect on economic growth as long the UK aid is mostly related to military aid rather than development aid. However, the bulk of past studies found that foreign influences economic growth favorably. This negative effect is primarily due to the underutilization of skilled workers in the local economy. In addition, in order to make sure that our results are accurate and free of bias and incorrect inferences, we performed a number of diagnostic and model stability checks in the estimated model. All diagnostic and stability tests verified the estimated model's dependability and consistency over the course of the investigation.

The study's conclusions have various policy ramifications. To improve the economy of the country by encouraging factors that contribute to economic development, the first step for the government should be to create a strong economic growth strategy that permits full participation of the commercial, public, and civil society sectors, as well as development partners, NGOs, farmers, and specialists during the planning and implementation phases. Also the government should prioritize where the foreign aid will be putting because if the aid concentrates more project developments that could contribute the economic development of the country. To use foreign aid for development rather than other things like securities, adaptation and intervention policies in this regard should design measures particularly to improve the capital or equipment used. The government needs to implement a plan to increase the labor force's abilities to achieve long-term economic growth. Finally, new economic growth strategies must be created while taking into account all the variables that affect the country's economic growth, such as Sweden, US aid, capital, labor, and so forth.