Analysing the dynamic interplay of remittance inflows, manufacturing output and foreign direct investment (FDI) on Gross Domestic Product (GDP) of Zimbabwe: VAR and VECM Approach spanning 1980-2022

Introduction

Background of the problem

Zimbabwe's economic decline since 1997 has seen country's macro-economic deteriorating in to a hyperinflationary economy. This has resulted in an unsustainable socio-economic condition that forced 37.8% of the population to migrate internationally (International migration organization 2008). According to C.L, Mnangagwa (2009) past researchers have concluded that the economic decline was caused by poor monetary policies, failure of fiscal policies to control budget deficit though some officials put blame on other factors such as corruption, sanctions, droughts. The average GDP in Zimbabwe averaged 8.44USD billion from 1960-2022, Agriculture and mining largely contribute to the government's exports with a large margin of contribution in GDP. The government officials have been left fingering different sectors, introducing policy after policy of the economy as they continue to fight the economic recession some of the policies include ESAP 1990, ZIMPREST 1996-2000, etc. Other factors ie natural disasters such as Cyclone Idai, COVID-19, Elnino droughts, floods continue to draw back efforts being made to improve economic performance of Zimbabwe.

The research problem

Zimbabwe, a developing country located in Southern Africa, has faced significant economic challenges since gaining independence in 1980. The country's economy has experienced periods of high inflation, currency fluctuations, and declining economic growth.

In recent years, Zimbabwe has become increasingly dependent on remittance inflows, manufacturing output, and foreign direct investment (FDI) to drive economic growth and development. Remittances inflows have become a vital source of foreign exchange earnings for Zimbabwe, with millions of Zimbabweans living and working abroad sending money back home to support their families as way to alleviate poverty back home. Manufacturing output has also been identified as a key sector for economic growth, with the government implementing policies to revitalize the industry. Foreign Direct Investment (FDI) has also played a crucial role in Zimbabwe's economic development, with foreign investors attracted to the country's natural resources and strategic location. Despite the importance of these factors, there is limited research on the dynamic interplay between remittance inflows, manufacturing output, FDI, and GDP in Zimbabwe. This study aims to address this knowledge gap by statistically analyzing the short-run and long-run relationships between these variables using Vector-Autoregression (VAR) and Vector Error Correction (VECM) models.

Aim

This study aims to conduct a comprehensive analysis of the relationships and impact of remittance inflows, manufacturing output, foreign direct investments on real GDP of Zimbabwe using VAR and VECM models with the ultimate goal of informing evidence-based decision. \subsection {Economic growth of Zimbabwe from independence:}

Real GDP of Zimbabwe 1980-2022

From 1980 since the British flag was ceremonially pulled down it is thus an appropriate moment to reflect country's performance and to see how far hopes of many Zimbabweans and other nations who wished her independence before 1980 have been fulfilled. According to Rodger C Rodinell in his analysis of the performance of the economy 4 years after independence he showed that Zimbabwe economy performed so much better than its neighbours. However, depressed foreign demand for the country's mineral exports and the onset of drought cut sharply into the growth rate in 1982, 1983, and 1984. In 1985, the GDP rebounded strongly due to a 30% jump in agricultural production. However, it falls in 1986 to a zero-growth rate and registered a negative of about 3% in 1987, primarily because of drought and the foreign exchange crisis faced by the country. In 1997, Zimbabwe's economic decline began to visibly take place. It began with the crash of the stock market on November 14, 1997. This was mainly due to private sector that was suffering from decreasing output and increasing competition from cheap imported products. The average growth of the economy from 1980 -1990 was about 4.5%. According to Mandivamba (2006), Between 2000 and 2007, GDP slowly declined with inflation vaulted to over 66,000%, and there were persistent shortages of hard currency, fuel, medicine, and food. GDP per capita dropped by 40\%, agricultural output dropped by 51% and industrial production dropped by 47%, the government however blamed sanctions. According to world bank indicators. The country experienced hyperinflationary conditions in 2008-09 however the government of national unity introduced the de-dollarization which resulted in the economy appreciating foreign currency transactions. In 2009, Zimbabwe recorded a period of economic growth for the first time in a decade. The GDP experienced a sharp increase up-to 2016 when the bond notes were introduced. Staff Reporter (2019-02-20). In August 2014, Zimbabwe began selling treasury bills and bonds to pay public sector salaries that have been delayed as GDP growth weakens while the economy experiences deflation. In June 2019 the use of foreign currencies in local transactions was prohibited as part of the prospective plan for a new national currency and thus ended the dollarization period, Samaita, K (15 July 2019) [21] . In 2019 midyear, After the adoption of a new Zimbabwe dollar, inflation had increased to 175\% and banning the use of foreign currency thereby sparking fresh concerns that the country was entering a new period of hyperinflation. The GDP experienced a low increase up-to 2016,however from 2019 the economy experience an increase in economic growth this can be attributed to the economy appreciating the foreign currency transactions and re-industrialisation through foreign investment.

Manufacturing output (% to GDP) continue to show large percentage during the entire period this implies that the sector's growth is more significant part of the economic growth. Comparing the plots, it is clear that percentage contributions of FDI and remittance inflows are below production output. However, the slight decrease from 1990 to year 2000 this can be attributed to lack of investment during that period, investors lacking confidence in investing in Zimbabwe, lack of infrastructure such as roads and electricity. During 2008 the decline was caused by hyperinflation that was experienced by nation, many industries closed. However during and after the government of national unity the manufacturing output recorded an increase this was due to effective policies introduced during that period. The plot of remittance inflows, shows a low to negative personal remittance inflows % of economic growth from 1980 to 1993 with a steady weak increase from 1993 to 2008. During that period the economy was performing normal to an extend that majority of Zimbabweans would find no reason to migrate. Factors such as the economic crisis (2000-8) which was characterised by high unemployment levels and hyperinflation in 2008 (ZimStat, 2009) [32] has led to many Zimbabweans living the country in search for greener pastures, the economy recorded remittance inflow's sharp increase in 2008 for the first time since attaining independence. The steady increase in worker's remittances over the years in specific 2008 going forward, can be attributed to factors such as increase Zimbabweans migrating to other countries and lowering of transactions costs. Low inflows might be due to money is being send through unofficial channels because people lacks confidence to send money through the formal channels. The graph makes it abundantly evident that between 1980 and 1990, Zimbabwe saw very little net foreign direct investment (FDI) inflows. This was because of a policy environment that was unfavorable to international investors at the time. Gwenhamo (2011) [10] claims that the late 1980s saw a shift in Zimbabwe's government's policies toward foreign investors, which resulted in a marginal improvement in FdI. The Zimbabwe Investment Centre (ZIC), which was set up as a one-stop shop for approvals, was the reason for the spike in FDI that was observed between 1992 and 1997. FDI decreased from a positive to a negative percentage of GDP between 1985 and 1990. FDI net inflows increased from 1990, when they were negative, to 1995, when they were positive. The net inflow of foreign direct investment (FDI) into Zimbabwe increased between 2000 and 2005, while its contribution to the Zimbabwean economy decreased between 2005 and 2010.The country saw a positive improvement from 2020 to 2022, although the FDI contribution to GDP had a negative rise from 2018 to 2020. This can be linked to multinational corporations closing because of a lack of confidence in the local currency at that time.

Literature review

The impact of remittances inflows on GDP Optimistic view examines remittance inflows as with a positive implication to both the sending and receiving households and their nations and was pioneered under the Laws of migration by Ravenstein in 1889. Migrant remittances are viewed as an instrument that stimulate savings and investment and also as an instrument for restoring balance of payment at the country of origin. Ratha (2007) [19] supported the neo-classicalist, by emphasising that remittances assume an important role in raising the capital market sector activities and help in giving profitable foundation and also in raising the viable interest for products and administration. By applying the Auto-Regressive Distributed Lagged model (ARDL), Odionye and Emerole (2015) identified that remittances positively impact the economic development of the receiving country. The study postulated that proper utilisation of international remittances flow enhances development of the receiving economies and this can be achieved through diversification of proceeds into productive purposes. A three staged least squares estimation (OLS) technique was used by Tambama to see the impact of remittances on poverty alleviation in Zimbabwe from 1980 to 2007. The study provided evidence that a unit increase in the share of remittances on GDP reduces poverty by 52\% and increases human capital accumulation by 11.5\% in Zimbabwe. She concluded that remittances contribute significantly to development objectives such as Millennium Development Goals (MDGs). Akonji and Wakoli (2013) [3] conducted research on impact of net migrant remittances on economic growth of Nigeria, which used the Seemingly Unrelated Regression (SUR) analysis and Error Correction Model (ECM). Their finds established a significant relationship between net migrant remittances on Nigerian economic growth. However, they also concluded that impact of migrant remittances can be more meaningful and contribute to economic growth only if financial institutions are well organised and be made more competitive to provide remittances services at low cost so that remittances flow into the country through official channels.

Production output impact on GDP

According to Su and Yao, who looked at the role of manufacturing during the middle-income development phase, manufacturing continues to be the primary driver of economic growth for middle-income economies during this phase, pulling alongside all other sectors, including the service sector. The Attiah research, which looked at how the manufacturing and service sectors contribute to economic growth in emerging nations, produced a similar finding. The findings indicate a favorable correlation between economic growth and the manufacturing sector's GDP share. Within the African context, Moyo and Jeke investigated the connection between economic growth and manufacturing in 37 African nations. The findings demonstrate that industrial value contributes to economic growth in African nations. To boost production output, policymakers are advised to take action.

Impact of FDI on economic growth

Numerous researches have attempted to establish the link between FDI and macroeconomic performances including GDP, however, the results are rather mixed. Balasubramanyam, Salisu and Sapsford (1996) [5] argued that FDI can speed up growth of the receiving countries through improving foreign trade and ensuring stability of macroeconomic variables.\\ Further, they concluded that FDI inflows can effectively boost economic growth than local investments in developing economies which implement export promotion policies. For nations with high institutional competence, FDI has a significant beneficial influence on their growth. (Olofsdotter, 1998) [16] . Akiri, Vehe and Ijuo (2016) [2] used VECM and determined positive impact of inward FDI to Nigeria's GDP growth during from 1981 to 2014. Stoneman (1975) [24] investigated the effect on growth caused by foreign direct investment in the growing economies from 1945 to 1970. He came to the conclusion that FDI had a detrimental effect on economic growth. However the FDI can be have a negative impact on economic growth, The effect of FDI on Nigeria's GDP was negative and negligible when ECM was used (Akinlo, 2004) [1] .

Methodology and Data analysis Research Design

The research design encompasses the aims,approach, type of design, experimental research design. Quantitative research will be used in this study, According to (Kisely and Kendal 2011) quantitative research entails using numerical data from a group of people or using the secondary data and statistical techniques to achieve research objectives.

Data collection and Data source

The study employs secondary annual time series data set to examine the influence of the selected variables on GDP of Zimbabwe from 1980 to 2022.The data was obtained from the following websites which are official boards for disseminating national data such as ZIMSTAT (Zimbabwe Statistics Agency) [32] , IMF (International Monetary Fund) and World Bank. Direct links are www.macrotrends.co.zw, https://www.fred.stlouisted.org (Federal Reserve Economic Data).

Model specifications.

The researcher adopts the 'principle of parsimony' which states that the model used should require the smallest possible number of parameters that will actually represent the data. Models that capture all of the signal and none of the noise provide the best possible predictions to unseen data from the same source. More simplistic models, may, fail to capture part of the signal and these models under-fit the data and provide poor predictions however overly complex models,over-fit the data and again provide poor predictions. Thus, parsimony is essential because it helps discriminate the signal from the noise, allowing better prediction and generalization to new data. In addition, most people feel naturally attracted to models and explanations that are easy to understand and interpret.

1.Vector Autoregressive model (VAR model).

Sims (1980) [22] established the VAR model as a method that macroeconomists might use to analyze the joint dynamic behavior of a group of variables without the requirement for stringent constraints, such as those required to uncover underlying structural factors. In reality, the VAR model is a synthesis of multiple autoregressive (AR) models, which create a vector between the variables that influence one another. A quantitative forecasting method that is typically used with multivariate time series data is the VAR model. The VAR model is a kind of stochastic model that compares the current observations of a variable to its historical data as well as those of other variables in the system. It records the evolution and interdependencies among many time series. One of the most popular, adaptable, and userfriendly models for analyzing multivariate time series is the vector auto regression model. The univariate autoregressive model is a logical progression to dynamic multivariate time series. The Vector Autoregressive (VAR) model is a potent tool in econometrics and time series analysis for comprehending and predicting the dynamic interrelationships among several variables. Because they can capture intricate relationships between variables, VAR models have become increasingly popular. This has allowed them to shed light on financial trends, economic system behavior, and other dynamic processes. Let m be variables that are described as a linear function of p-lags of itself and of all the other m-1 variables, plus an error term, and consider the VAR(p) model of x and y variables:

In time series analysis it is important to check stationarity on the variables, non stationary variables can lead to false correlations and relationships, biased and inconsistent estimates of a model parameters. Stationarity ensures that the results of time series analysis are meaningful and influenced by trends or seasonality. From the ADF and PP test results above we fail to reject null hypothesis and conclude that manufacturing output, remittance inflows were all non-stationary at level I(0) at 5% level of significance, only FDIs was stationary at 0.05, Since the test has concluded that other variables are nonstationary at their level therefore we proceed to difference these variables if they became stationary at their first difference I(1) then the variables can fit VECM once there exists a cointergration among them.

Differencing non-stationary series:

Differencing makes a non-stationary series stationary, which is a requirement for time series models such as VEC and VAR. Differencing also stabilize the variance making it constant overtime. Based on the results above the VAR model will be fitted with a lag of 2 according 'AIC'. The AIC (Akaike Information Criteria) strikes a balance between model fit and complexity avoiding over fitting and under fitting therefore it estimates better models,this means the VAR model (Short run model) will be lagged with the previous two years.

Granger Causality test

Granger causality, as described by Clive Granger, can be used to clarify the direction of potential causality between pairs of variables under certain contentious assumptions. It is crucial for statisticians and econometricians to understand that "correlation does not imply causality." Since correlation or covariance is a symmetric, bivariate connection, we cannot simply determine the presence or direction of causality between x and y by looking at non-zero covariance; cov(x, y) = Cov(y, x). Granger causality defines whether past values of x help anticipate y t , provided that the effects of past values of y on y t have already been taken into consideration. They are said to "Granger cause" y if they do.

Assuming that x and y are variables, the first equation represents y t as a linear function of both x's and its own historical values. Some or all of the lagged x values have nonzero impacts if x Granger causes y (which we write as x implies y): lagged x influences y t conditional on the effects of lagged y. In the VAR model of variables x and y, testing for Granger causation entails checking to verify if the joint blocks of coefficients beta xy and beta xy are zero. H_0 is the null hypothesis in the VAR, which states that x does not Granger cause y. The formula is \beta_yx =\beta_yx =……. =\beta_yxp =0.

A typical Wald F or \chi^2 test can be used to test that. In the same way, H_0 for the null hypothesis y implies x is \beta_xy =\beta_xy = …=\beta_xyp = 0. There are four possible results from these two tests. When analyzing VAR models, the Granger Causality test is crucial because it facilitates the identification of cause-and-effect linkages between variables.

The economics community has made extensive use of the test.

Alternative: REM, MAN and FDIs does granger cause GDP (they are independent variables of GDP).

From the results above, p-value is less 5\% significant level for all variables, therefore we reject the null hypothesis with sufficient evidence to conclude that REM, MAN and FDIs, they do cause GDP meaning changes, shocks or innovations on these variables do affect the real GDP of Zimbabwe.

Interpretation

These are short run impacts estimated after first difference except for FDIs that was stationary at level. The dependent variable is a function of its lagged values (AR) and the lagged values of other variables in the model. The positive coefficients in remittance, manufacturing and foreign direct investment indicates the impact of the lagged values of these variables on the current value of GDP. The positive coefficient suggests that a positive relationship meaning the past value of positively influence the current value of GDP.

Vector Error Correction model (VECM model)

For non-stationary series that are known to be co-integrated, a vector error correction (VEC) model is a constrained VAR model with cointegration restrictions incorporated into the specification. Multivariate time series modeling includes the Vector Error Correction Model (VECM). ECM (Error Correction Modelling), which examines the long-term relationship between a few non-stationary variables in the original data, is the most straightforward type of univariate modelling. 24 With a combination of linear variables, the formation of this cointegration is like a new hope for the longterm establishment of a stationary situation. Therefore, if the point eventually reaches the equilibrium point, it is said to be cointegrated.

ECM can be employed if cointegration analysis is feasible.

Testing will encounter the Error Correction Term (ECT) if it reaches the ECM analysis (short-term relationship). This is employed because the equilibrium state's adjustment speed is anticipated to be negative (convergent). While permitting a broad range of short-term dynamics, the VEC specification limits the endogenous variables' long-term behavior to converge to their cointegrating relationships. Since the departure from long-run equilibrium is gradually corrected through a succession of partial short-run adjustments, the cointegration component is also referred to as the error correction term.

Since there is a long-term relationship between variables if there is co-integration of variables in a series, we use VECM to assess the short-term characteristics of the co-integrated series; if co-integration is absent, we use the Granger Casuality test to look for casual relationships between variables. The cointegration rank in VECM indicates how many co-integrating vectors there are. Any short-term variations between independent and dependent variables will result in a stable long-term relationship between them, according to an ECM coefficient that is both negative and significant.

As a result, we may estimate short-term behavior, long-term cointegrating relationships, and short-term adjustment coefficients using the VECM model. The adjustment coefficients indicate the rate at which the short-term departures from long-term equilibrium are adjusted. The VECM model is provided in the following form after taking into account the short-and long-term relationships between y, the dependent variable, and x 1 and x 2 , the independent variables, together with their lagged values: Interpretation: the best lag is selected on the basis of "AIC" which is 4 with a p value less 0.05 according the results in the above table, Too many lags lead to loss of degrees of freedom, statistically insignificant coefficient and multicollinearity while too few lag lead to under parametrisation and specification errors.

Test for co-intergration with p-1 which 3

In VECM cointegration is an important aspect, variables are cointegrated when there is a long run relationship between multiple non-stationary time series variables. The variables may individually follow random walks(non -stationary), but there exists a combination of them that is stationary. Two variables m produce m-1 co-intergration equation so with three variables the number of cointergrating equations is m-1 which is 2, as selected by the trace statistic at 0.05 critical value.

According to Green, WH(2003), If the variables are cointegrated, construct both short-run (VAR) and the long-run model(VEC).

ECT term

After the shortrun VEC model was estimated the adjustment term was (-0.2848578) statistically significant at 5% level of significance, suggesting that the previous year's errors (or deviations from the long run equilibrium) are corrected for within the current year at the convergence of 28.48%. Note: The signs of the coefficients are reversed in the longrun hence interpreted as positive if negative or vice-versa. GDP variable is always positioned first as the dependent variable.

In the long-run Personal remittances, Manufacturing output, have a positive impact on GDP of Zimbabwe, all the coefficients are statistically significant at 5% level of significant.

In conclusion: Personal remittances, Manufacturing output have asymmetric effects on Zimbabwe's Real GDP in the long run on average ceteris paribus.

Model selection

After fitting the VAR and the VEC model, it is good to decide which model performs well than the other. The best model was selected on the basis of AIC and model coefficients significance. Based on the results above VAR has AIC" of 18.38497 while VECM has AIC" of 14.56084 the best model is the one with the smaller AIC, in addition VEC coefficients are statistically significant in the long run, indicating the VEC model as the preferred choice that has a better fit to the data and the parameters are more efficient in capturing the relationship between variables hence we proceed to diagnostics and model predictions with the VEC model.

Model Diagnostics Arch Effects

The Arch effects test checks for the presence of Autoregressive Conditional Heteroscadasticity (ARCH) effects (it tests whether the variance of the residuals changes over time) in the residuals of the VEC model. Based on the results provided above, fail to reject null hypothesis and conclude that there is clear indication that there are no significant arch effects in model residuals, it suggests that the variance of the residuals is relatively constant over time and the model is not missing any significant time varying volatility patterns. In conclusion the model is capturing the relationship between variables reasonably well.

Test serial correlation(autocorrelation)

It is important to check autocorrelation of the model in order to avoid inaccurate estimation because serial correlation can lead to in accurate forecasting and violation of the assumption of independent errors. The results above are from Breusch-Godfrey test for autocorrelation in the residuals of the VEC model with moving average (MA).

At lag order 1 we fail to reject the null hypothesis and conclude that there is no significant autocorrelation in the residuals at lag order 1, indicating that MA(1) is sufficient to capture any short-term dynamics. At lag order 2 we fail to reject the null hypothesis and conclude that there is no significant autocorrelation in the residuals at lag order 2, indicating that MA (2) is sufficient to capture any short-term dynamics. In conclusion the test indicate that there is no autocorrelation assuring that the assumption of independent errors is not violated.

Normality of residuals

Testing of residuals is a very crucial step in model diagnostics test as it checks if the assumption of normality of residuals is not violated, also normality of residuals ensures reliable inferences and hypothesis testing.

Alternative: Residuals are normally distributed Based on the above Jarque-Bera test for residuals normality, all variables have significant values at 1\% except for Manufacturing output, however the test further conclude that ALL (0.0000) residuals are normality distributed, we reject the null hypothesis and conclude that the residuals are normally distributed ALL at 1\% level of significance meaning the model assumption is satisfied allowing valid inferences and forecasting. The results show that an increase in remittance inflows and production output, leads to an increase in economic activity (GDP), the fact that the black line is above and between the dotted lines suggests that the positive response is statistically significant. In conclusion the effect of remittance inflows and production output, on GDP is positive and significant hence remittances and production output, are an important driver of economic growth.

Conclusions Recommendations

Strategies to improve the Zimbabwe Remittance inflows towards economic growth. Based on the above findings, remittances are a source of capital that can stimulate economic growth of Zimbabwe. The government of Zimbabwe should strengthen efforts to attract more remittances inflows into the country. The objectives of the diaspora policy should include developing mechanisms for dialogue and partnership with Zimbabweans abroad and should develop trust between community diaspora and government and then focus on availing incentives to those who wish to invest back home however uncertain macroeconomic conditions which include inflation, unfavourable exchange rates, high black-market premiums which might implicate on remittance flow must be corrected first before the maximisation of policies. Finally the government of Zimbabwe must adopt a policy that was used in Philippine, the government of Philippine through the Executive Order Number 857 which was issued in 1982. This required all workers who work outside the country of Philippians to remit back seventy percent of their monthly salary through banks, the compliance of the order was compulsory before the worker's passport, VISA could be renewed. That's making remitting mandatory for those workers who are abroad this will increase inflow volumes of remittances. It is also important that mere collection of remittances through banks and other outlays cannot bring forth a desirable progress in the economy but those funds must be utilised properly in the economy so as to sustainably develop the economy of Zimbabwe and for poverty alleviation purpose. According to Thabani Nyoni, the government of Zimbabwe need to revisit the investment legislation in the country, the consistent and transparent application of the Indigenisation Act in the country is very critical in attracting both foreign direct investment and domestic private investment as it limits the scope of discretion and reduces uncertainty with regards to the regulatory framework. The government should implement macroeconomic policies that improve the investment climate in the country such as reviewing the indigenization policy to make it attractive and to quicken the introduction of Special Economic Zones as well as industrial parks. Such policies are imperative as they will attract foreign investors.

Strategies to improve the

The government of Zimbabwe need to enhance good governance and corruption free environment by ensuring there is transparency in paperwork and sound accountability.

The respect of property rights should also be prioritised, especially concerning agricultural land foreigners are inadequately protected, with instances of land expropriation without compensation, it shouldn't be a risk to invest in Zimbabwe.

Conclusions

The Zimbabwe manufacturing sector is facing serious challenges but however the sector can still recover in the long term. Although many challenges can be cited, it is generally the hyperinflation period that has placed the manufacturing companies at a disadvantage compared to their counterparts from other countries. All sectors of the economy including the manufacturing sector must redefine the key economic drivers which will be necessary to take Zimbabwe to the next level of economic growth. The government should turn the conducive environment to attract foreign direct investment provided that operational costs are favourable. Remittance inflows amounts should also reflect the large number of people abroad therefore policies, proper formal channels must be put in place to realize the economic development brought by remittances.

Recommendations for

References

1. Akinlo AE. Foreign direct investment and growth in Nigeria. J Policy Model. 2004;26(5):627-639.
2. Akiri SE, Vehe BM, Ijuo AO. Foreign Direct Investment and Economic Growth in Nigeria: An Empirical Investigation. IIARD Int J Econ Bus Manag. 2016;2(5):112.
3. Akonji DR, Olateju, Abba MW. Nexus between public expenditure and economic growth by testing Wagner's law time series: Evidence from Nigeria. Int J Dev Sustain. 2013;2(4):2383-2395.
4. Alistar T, Murapiro et al. The Impact of Remittances as a Source of Economic Growth in LDCs: Case of Zimbabwe Since Independence. Int J Arts Soc Sci. 2023;6(11). Available from: www.ijassjournal.com
5. Balasubramanyam VN, Salisu MA, Sapsford D. Foreign Direct Investment and Growth in EP and IS Countries. Econ J. 1996;106:92-105.
6. Chidochashe L, Mnangangwa. The Economic Decline of Zimbabwe. Gettysburg Econ Rev. 2009;3:Article 9.
7. Confederation of Zimbabwe Industries (CZI). Back to Basics: Post-Dollarisation Manufacturing Recovery, Manufacturing Sector Survey. 2010.
8. Dickey D, Fuller W. Distribution of the estimators for autoregressive time series with a Unit Root. J Am Stat Assoc. 1979;74(366):427-431.
9. Engle RF, Granger CWJ. Co-integration and errorcorrection: Representation, estimation and testing. Econometrica. 1987;55(2):251-276.
10. Gwenhamo F. Foreign direct investment in Zimbabwe: the role of institutional and macroeconomic factors. S Afr J Econ. 2011;79(3):211-223.
11. International Migration Organization (IMO). 2008.
12. Johanssen S. Statistical Analysis of Cointegration Vectors. J Econ Dyn Control. 2012;12(2):231-254. 14. Macrotrends Data. Available from: www.macrotrends.co.zw.
13. Moyo C, Jeke L. Manufacturing sector and economic growth: a panel study of selected African countries. J Bus Econ Rev. 2019;4(3):114-130.
14. Olofsdotter K. Foreign Direct Investment, Country Capabilities and Economic Growth. Weltwirtschaftliches Archive. 1998;134(3):534-547.
15. Pasura DM. A Fractured Diaspora: Strategies and Identities among Zimbabweans in Britain. Unpublished PhD thesis, University of Warwick; 2008.
16. Phillips PC, Perron P. Testing for a unit root in time series regression. Biometrika. 1988;75(2):335-346.
17. Ratha D, Mohapatra S. Increasing the macroeconomic impact of remittances on development. World Bank. 2007;3(1):178-192.
18. Samaita K. Zimbabwe's inflation doubles up to 175%. 15 July 2019.
19. Sims CA. Macroeconomics and Reality. Econometrica. 1980;48(1):1-48.
20. Staff Reporter. RBZ introduces "RTGS Dollars". 20 Feb 2019.
21. Stoneman C. Foreign capital and economic growth. World Dev. 1975;3(1):11-26.
22. Tanzania Invest Report. Foreign Direct Investment. Available from: https://www.tanzaniainvest.com/fdi. 2018.
23. Su D, Yao Y. Manufacturing as the key engine of economic growth for middle-income economies. J Asia Pac Econ. 2017;22(1):47-70.
24. Tambama J. The impact of remittances on Zimbabwean economic development. Doctoral dissertation, University of Zimbabwe; 2011.
25. United Nations Conference on Trade and Development (UNCTAD). Foreign direct investments in Africa. New York and Geneva: United Nations; 2014.