The Impact of Government Capital Expenditures on Economic Growth and Sustainability in Nigeria (1981 to 2023): A Comparative Analysis Using Multiple Regression Models
OYEBAMIJI, Timothy Adeola1
OYEBAMIJI, Moses Bolaji2
2024-11-23
Introduction
Capital expenditure is a key factor in any national economic growth including Nigeria. Studies varied in their findings on the impact of capital expenditure on the economic growth of the country (Nwamba et al., 2023; Okonkwo et al., 2023). Major aspects of the capital expenses of Nigeria as reported by the Central Bank of Nigeria and the National Bureau of statistics are Capital expenditure on administration, Capital expenditure on economic services, Capital expenditure on social-community services, capital expenditure on economic services, lastly, Transfers. (Statistical Abstract, 2023). Capital expenditure can stimulate economic growth by improving public services, improve the quality of life for citizens, increasing labor productivity (Okonkwo et al., 2023). Therefore, this study investigates the extent at which various aspects of capital expenditures of Nigeria has significantly impacted her economic growth.
Literature Review
The relationship between capital expenditure and economic development has been extensively studied, with diverse methodologies and findings offering nuanced insights. Chandana et al. (2020) analyzed the impact of Nigerian government expenditures, disaggregated into capital and recurrent components, on economic growth from 1970 to 2019. Using the ARDL model and accounting for structural breaks, they found that capital expenditure positively and significantly influenced economic growth in both the short and long run, while recurrent expenditure showed no significant impact. This research emphasized the importance of allocating funds to capital projects that directly benefit citizens’ welfare. These findings provided a baseline for subsequent research exploring disaggregated impacts of expenditure across sectors.
Building on this foundation, Iriabije et al. (2023) explored capital expenditure’s sectoral effects on economic growth as a planning tool. Using the ARDL bounds technique, the study analyzed data from 1990 to 2020, revealing that manufacturing, services, and agriculture were key growth-enhancing sectors. Both short- and long-run analyses affirmed the significant impact of capital expenditure on economic growth, particularly when funds were directed to productive sectors. Unlike Chandana et al., who provided a broader evaluation of expenditure categories, Iriabije et al. offered a more granular analysis, focusing on sectoral priorities that could guide government planning.
Okonkwo et al. (2023) introduced a post-pandemic dimension to the discourse, examining the performance of disaggregated government expenditures, including administration, social and community services, and economic services, on economic growth from 1981 to 2021. Employing the ARDL model, the study found strong positive relationships between administrative and economic services expenditures and economic growth, aligning with Chandana et al. and Iriabije et al. However, this study uniquely contextualized capital expenditure within the pandemic recovery phase, highlighting the sustained relevance of administrative and economic investments in bolstering growth during periods of economic uncertainty.
In contrast, Mohammed et al. (2024) focused on the balance of payments as a developmental outcome, analyzing government capital expenditure’s effects from 1990 to 2022. Using ARDL and ECM methods, the study revealed mixed impacts. In the short term, administrative and transfer expenditures had positive but insignificant effects, while economic services and social/community services expenditures negatively influenced the balance of payments. Over the long term, administrative and economic services expenditures showed positive but insignificant effects, while social/community services expenditures negatively impacted the balance of payments significantly. These findings diverged from the largely positive impacts identified in other studies, reflecting the complex and sometimes conflicting influences of expenditure across economic indicators.
Further diversifying the discourse, Nwaba et al. (2023) examined the relationship between sector-specific expenditure costs and economic development, measured using the Human Development Index (HDI), from 2003 to 2022. They found a significant positive relationship between health sector expenditure and economic development but an insignificant negative relationship with education expenditure. Unlike other studies that focused on GDP, this research adopted a broader measure of development, emphasizing the socio-economic implications of capital spending. The results also showed no sustained long-run relationship between the variables, contrasting with the findings of Chandana et al. and Iriabije et al., who identified consistent long-run impacts of capital expenditure on growth.
Ukoh and Nwaoha (2024) offered a focused analysis of capital expenditure’s effects on administration and economic services from 2007 to 2022. Utilizing a log-linear econometric model, they demonstrated robust positive correlations between these expenditures and GDP. Administrative spending, in particular, was found to significantly contribute to economic growth, underscoring the importance of investments in infrastructure and governance. While their findings echoed those of Okonkwo et al. regarding the positive role of administrative expenditure, they contrasted with Mohammed et al.’s results, which suggested only insignificant long-run impacts in similar areas.
Research Gap
While previous studies have enriched understanding of capital expenditure’s impacts, gaps remain in addressing its sustainability and long-term implications. Most studies focus on traditional metrics like GDP or the balance of payments without adequately incorporating broader sustainability indicators, such as environmental and socio-economic equity considerations. Furthermore, the interaction between sectoral expenditures and their combined effects on holistic development remains underexplored. The present study aims to fill these gaps by integrating sustainability-focused metrics into the analysis and employing advanced econometric techniques to assess capital expenditure’s comprehensive and enduring role in Nigeria’s economic development.
Methodology
The methodology of this research utilizes R and R Markdown for data analysis and reporting. Data was collected from relevant sources and analyzed using various econometric techniques, including time series analysis, regression models, and Granger causality tests. The analysis was performed using R, with data processing, statistical modeling, and visualization being conducted through R scripts integrated into R Markdown. This allowed for a seamless combination of code, results, and narrative, ensuring transparency and reproducibility of the findings. The R Markdown document facilitated dynamic reporting, enabling the automatic generation of tables, figures, and statistical summaries, which were essential for interpreting the relationship between government expenditures and economic growth in Nigeria.
# Calculate descriptive statistics
descriptive_table <- reframe(
data,
Variable = c("Administration", "Economic_Services", "Social_Community_Services", "Transfers", "GDP"),
Mean = c(mean(Administration, na.rm = TRUE),
mean(Economic_Services, na.rm = TRUE),
mean(Social_Community_Services, na.rm = TRUE),
mean(Transfers, na.rm = TRUE),
mean(GDP, na.rm = TRUE)),
Median = c(median(Administration, na.rm = TRUE),
median(Economic_Services, na.rm = TRUE),
median(Social_Community_Services, na.rm = TRUE),
median(Transfers, na.rm = TRUE),
median(GDP, na.rm = TRUE)),
SD = c(sd(Administration, na.rm = TRUE),
sd(Economic_Services, na.rm = TRUE),
sd(Social_Community_Services, na.rm = TRUE),
sd(Transfers, na.rm = TRUE),
sd(GDP, na.rm = TRUE)),
Min = c(min(Administration, na.rm = TRUE),
min(Economic_Services, na.rm = TRUE),
min(Social_Community_Services, na.rm = TRUE),
min(Transfers, na.rm = TRUE),
min(GDP, na.rm = TRUE)),
Max = c(max(Administration, na.rm = TRUE),
max(Economic_Services, na.rm = TRUE),
max(Social_Community_Services, na.rm = TRUE),
max(Transfers, na.rm = TRUE),
max(GDP, na.rm = TRUE)),
SE = c(sd(Administration, na.rm = TRUE) / sqrt(length(na.omit(Administration))),
sd(Economic_Services, na.rm = TRUE) / sqrt(length(na.omit(Economic_Services))),
sd(Social_Community_Services, na.rm = TRUE) / sqrt(length(na.omit(Social_Community_Services))),
sd(Transfers, na.rm = TRUE) / sqrt(length(na.omit(Transfers))),
sd(GDP, na.rm = TRUE) / sqrt(length(na.omit(GDP))))
)
# Print the descriptive summary table
print(descriptive_table)## # A tibble: 5 × 7
## Variable Mean Median SD Min Max SE
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Administration 178. 73.6 242. 0.263 1131. 36.9
## 2 Economic_Services 320. 215. 410. 0.656 1961. 62.5
## 3 Social_Community_Services 85.8 32.5 116. 0.238 540. 17.6
## 4 Transfers 119. 43.6 182. 0 855. 27.8
## 5 GDP 45962. 11501. 62608. 139. 234426. 9548.The GDP and various capital expenditure constructs were plotted over time to visually examine their trends and movements, allowing for a clear understanding of their fluctuations and potential relationships. These plots provided insight into how changes in capital expenditures, such as administration and economic services, correspond to shifts in economic growth, helping to identify patterns or anomalies in the data.
Due to the similarities in the movement of independent variables, a multicolinearity problem is suspected. Hence, a simple way of addressing this is to standardize the variables. The correlation matrix is then developed to test if the high correlation has been addressed.
The log-transformed correlation matrix reveals strong positive linear relationships among most variables, particularly between GDP and Economic Services (0.96), Administration and Economic Services (0.97), and Social Community Services and GDP (0.96), indicating these variables move closely together after transformation. Transfers show moderate correlations with other variables (e.g., 0.73 with GDP and Administration), suggesting weaker but still positive associations. The high correlations among variables indicate a likelihood of multicollinearity, which could impact regression analysis, while the log transformation likely mitigates heteroskedasticity by stabilizing variances and improving linearity.
# Data Transformation
# Standardize the data
data_scaled <- data %>%
mutate(across(-Year, scale))
# Log-transform data to handle high variances
data_log <- data %>%
mutate(across(-Year, log1p))
# Check correlations post-transformation
corr_matrix_log <- cor(data_log[-1])
ggcorrplot(corr_matrix_log, method = "circle", lab = TRUE, title = "Log-Transformed Correlation Matrix")
The regression model explains 98.26% of the variance in GDP (adjusted \(R^2 = 0.9808\)), indicating a very strong fit. The F-statistic (\(536.7\), \(p < 2.2e^{-16}\)) confirms the overall model significance. Among the predictors, Administration (\(\beta = 0.95097\), \(p < 0.001\)) and Transfers (\(\beta = 0.14408\), \(p < 0.01\)) significantly contribute to GDP, while Economic Services and Social Community Services are not statistically significant (\(p > 0.05\)). Residuals appear well-behaved with a standard error of 0.3451. However, the high \(R^2\) and significant predictors suggest possible multicollinearity, as indicated by the large variance inflation factors (VIFs): Administration (47.63), Economic Services (19.50), and Social Community Services (31.65). This multicollinearity may inflate standard errors, impacting individual predictor reliability.
# OLS Regression
ols_model <- lm(GDP ~ Administration + Economic_Services + Social_Community_Services + Transfers, data = data_log)
summary(ols_model) ##
## Call:
## lm(formula = GDP ~ Administration + Economic_Services + Social_Community_Services +
## Transfers, data = data_log)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.57168 -0.23894 -0.07839 0.20849 1.08693
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.52020 0.12892 35.063 < 2e-16 ***
## Administration 0.95097 0.17115 5.556 2.31e-06 ***
## Economic_Services -0.09942 0.10517 -0.945 0.35051
## Social_Community_Services 0.23650 0.15885 1.489 0.14477
## Transfers 0.14408 0.04340 3.320 0.00199 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3451 on 38 degrees of freedom
## Multiple R-squared: 0.9826, Adjusted R-squared: 0.9808
## F-statistic: 536.7 on 4 and 38 DF, p-value: < 2.2e-16# Check multicollinearity with VIF
vif(ols_model)## Administration Economic_Services Social_Community_Services
## 47.633071 19.501780 31.650129
## Transfers
## 2.257349The extended regression model explains 99.87% of the variance in GDP (\(R^2 = 0.9987\), adjusted \(R^2 = 0.9984\)), indicating an excellent fit. The lagged GDP terms are highly significant: \(GDP_{lag1}\) (\(\beta = 1.416\), \(p < 0.001\)) positively impacts GDP, while \(GDP_{lag2}\) (\(\beta = -0.489\), \(p < 0.01\)) has a negative effect. Among the other predictors, Administration shows marginal significance (\(p = 0.067\)), while Economic Services, Social Community Services, and Transfers are not significant (\(p > 0.05\)), likely due to multicollinearity or the dominant influence of lagged GDP. The residual standard error (0.09455) indicates precise predictions, and the overall model is highly significant (\(F = 4227\), \(p < 2.2e^{-16}\)). This suggests GDP is strongly influenced by its past values, overshadowing other predictors.
# Time Series Regression
# Add lagged variables
data_log <- data_log %>%
mutate(
GDP_lag1 = lag(GDP, 1),
GDP_lag2 = lag(GDP, 2)
) %>%
na.omit()
ts_model <- lm(GDP ~ Administration + Economic_Services + Social_Community_Services + Transfers + GDP_lag1 + GDP_lag2, data = data_log)
summary(ts_model)##
## Call:
## lm(formula = GDP ~ Administration + Economic_Services + Social_Community_Services +
## Transfers + GDP_lag1 + GDP_lag2, data = data_log)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.192641 -0.045109 -0.008072 0.029246 0.301767
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.457814 0.194616 2.352 0.02458 *
## Administration 0.118207 0.062558 1.890 0.06737 .
## Economic_Services -0.033422 0.031026 -1.077 0.28897
## Social_Community_Services -0.012658 0.048133 -0.263 0.79416
## Transfers 0.006868 0.013564 0.506 0.61590
## GDP_lag1 1.415573 0.156114 9.068 1.34e-10 ***
## GDP_lag2 -0.489396 0.144614 -3.384 0.00181 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.09455 on 34 degrees of freedom
## Multiple R-squared: 0.9987, Adjusted R-squared: 0.9984
## F-statistic: 4227 on 6 and 34 DF, p-value: < 2.2e-16The ARIMA forecast for GDP, based on the ARIMA(0,2,0) model, projects a continued upward trend in GDP with a widening confidence interval as the forecast horizon increases, reflecting increased uncertainty. The model assumes a second-order difference (\(d=2\)), indicating that GDP growth is modeled as an acceleration process. The fitted model’s training error metrics include a mean absolute error (MAE) of 1844.98 and a root mean square error (RMSE) of 3294.01, suggesting the model captures general trends but may have difficulty with smaller fluctuations. The MAPE of 6.55% indicates a relatively low average percentage error, suggesting reasonable forecast accuracy. However, the residual autocorrelation (\(ACF_1 = -0.25\)) suggests some degree of residual structure, which may indicate room for model improvement.
# ARIMA Modeling
# Convert GDP to a time series object
gdp_ts <- ts(data$GDP, start = 1981, frequency = 1)
# Fit ARIMA model
arima_model <- auto.arima(gdp_ts)
summary(arima_model)## Series: gdp_ts
## ARIMA(0,2,0)
##
## sigma^2 = 11379759: log likelihood = -391.25
## AIC=784.49 AICc=784.6 BIC=786.21
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE ACF1
## Training set 745.3731 3294.005 1844.979 2.960245 6.550829 0.330745 -0.2534202forecast_arima <- forecast(arima_model, h = 5)
plot(forecast_arima, main = "ARIMA Forecast of GDP")
The Granger causality test evaluates whether lags of Administration improve the predictive power of GDP beyond the lags of GDP itself. The test compares two models: - Model 1 includes GDP lags and Administration lags. - Model 2 includes only GDP lags.
The \(F\)-statistic (0.5102, \(p = 0.6049\)) is not statistically significant, indicating that lags of Administration do not Granger-cause GDP at the given significance level. This suggests that Administration does not provide additional predictive information about GDP beyond what is already captured by GDP’s own past values.
# Granger Causality Test
# Test if Administration Granger-causes GDP
granger_test <- grangertest(GDP ~ Administration, order = 2, data = data_log)
granger_test## Granger causality test
##
## Model 1: GDP ~ Lags(GDP, 1:2) + Lags(Administration, 1:2)
## Model 2: GDP ~ Lags(GDP, 1:2)
## Res.Df Df F Pr(>F)
## 1 34
## 2 36 -2 0.5102 0.6049The residual plot of the time series model shows random fluctuations around zero, suggesting the absence of obvious patterns like trends or cycles in the residuals. However, there is some variance clustering visible, particularly in the earlier and middle time points, which may indicate heteroskedasticity. While the residuals appear approximately symmetric, further diagnostic checks (e.g., autocorrelation function) are necessary to confirm whether the model adequately captures the underlying structure of the data.
The ACF plot of residuals shows that most autocorrelation values fall within the 95% confidence intervals (blue dashed lines), indicating no significant autocorrelation in the residuals. This suggests that the time series model sufficiently captures the structure of the data, and the residuals behave like white noise. Any significant spikes outside the intervals would indicate model inadequacy, but none are present here.
# Residual Diagnostics
# Plot residuals of the time series model
plot(ts_model$residuals, main = "Residuals of Time Series Model", ylab = "Residuals", xlab = "Time")
acf(ts_model$residuals, main = "ACF of Residuals")
Ridge and Lasso regression are needed due to the multicollinearity observed in the model, where predictors such as GDP lag values and other economic variables are highly correlated, affecting the stability of coefficient estimates. These regularization techniques help mitigate multicollinearity and overfitting by shrinking coefficients and allowing for better generalization in predicting GDP.
For Ridge Regression (\(\lambda = 16268.97\)), all coefficients are shrunk but remain non-zero, indicating that the model includes all predictors, though their impacts are regularized. For Lasso Regression (\(\lambda = 6269.29\)), some coefficients (e.g., for Economic Services and Social/Community Services) are reduced to zero, effectively excluding these variables from the model, promoting sparsity for feature selection. Both approaches retain the intercept, and Ridge distributes importance among all variables, while Lasso emphasizes Administration and Transfers.
The Ridge Regression and Lasso Regression plots illustrate the relationship between the logarithm of the penalty term (\(\lambda\)) and the mean squared error (MSE). Both methods identify an optimal \(\lambda\), balancing model complexity and error minimization.
# Prepare the data
X <- as.matrix(data[, c("Administration", "Economic_Services", "Social_Community_Services", "Transfers")]) # Predictor variables
Y <- as.numeric(data$GDP) # Dependent variable
# Standardize the predictor variables (important for Ridge and Lasso)
X_standardized <- scale(X)
# Ridge Regression
set.seed(123) # For reproducibility
ridge_model <- glmnet(X_standardized, Y, alpha = 0) # alpha = 0 for Ridge
cv_ridge <- cv.glmnet(X_standardized, Y, alpha = 0) # Perform cross-validation
ridge_best_lambda <- cv_ridge$lambda.min # Optimal lambda value
# Print Ridge regression results
cat("Ridge Regression: Optimal Lambda =", ridge_best_lambda, "\n")## Ridge Regression: Optimal Lambda = 16268.97ridge_coefficients <- coef(cv_ridge, s = "lambda.min") # Coefficients at optimal lambda
print(ridge_coefficients)## 5 x 1 sparse Matrix of class "dgCMatrix"
## s1
## (Intercept) 45962.28
## Administration 15545.60
## Economic_Services 12237.79
## Social_Community_Services 13353.01
## Transfers 15072.99# Lasso Regression
lasso_model <- glmnet(X_standardized, Y, alpha = 1) # alpha = 1 for Lasso
cv_lasso <- cv.glmnet(X_standardized, Y, alpha = 1) # Perform cross-validation
lasso_best_lambda <- cv_lasso$lambda.min # Optimal lambda value
# Print Lasso regression results
cat("Lasso Regression: Optimal Lambda =", lasso_best_lambda, "\n")## Lasso Regression: Optimal Lambda = 6269.294lasso_coefficients <- coef(cv_lasso, s = "lambda.min") # Coefficients at optimal lambda
print(lasso_coefficients)## 5 x 1 sparse Matrix of class "dgCMatrix"
## s1
## (Intercept) 45962.28
## Administration 39794.17
## Economic_Services .
## Social_Community_Services .
## Transfers 13886.65# Visualize Cross-Validation for Ridge and Lasso
par(mfrow = c(1, 2))
plot(cv_ridge, main = "Ridge Regression")
plot(cv_lasso, main = "Lasso Regression")
Testing for Heteroskedasticity using the Breusch-Pagan Test and White Test.
The studentized Breusch-Pagan test result (BP = 0.80302, p-value = 0.938) indicates that there is no significant evidence of heteroskedasticity in the model, as the p-value is well above the typical significance threshold of 0.05.
library(lmtest)
# Perform the Breusch-Pagan Test
bptest(data_log_model)##
## studentized Breusch-Pagan test
##
## data: data_log_model
## BP = 0.80302, df = 4, p-value = 0.938Since the p-value is 0.8908, we fail to reject the null hypothesis of the Breusch-Pagan test. The null hypothesis assumes that the variance of residuals is constant (homoskedasticity). Thus, there is no evidence of heteroskedasticity in the model based on this test. The model appears to satisfy the assumption of homoskedasticity, so no further adjustments for heteroskedasticity (such as weighted regression) are necessary.
Discussion
Based on the analysis of the models, the model that includes lagged values of GDP (Model 2 with lags of GDP) appears to be the best in modeling expenditures and economic growth, as it shows a higher significance of the lagged GDP variables in predicting GDP growth compared to the model with additional lags of Administration, Economic Services, and Social Community Services (Model 1). The Granger causality test results (p = 0.6049 for Model 1) suggest that adding Administration lags does not significantly improve the model’s predictive power, while the residual analysis indicates that the model with only GDP lags captures the underlying trend without significant heteroskedasticity or autocorrelation. Therefore, the simpler model with GDP lags is more effective, as it offers better explanatory power and stability.
In practice, these findings suggest that for modeling and predicting Nigeria’s economic growth (GDP), the most important factor is the past performance of GDP itself, rather than other variables like government expenditures (Administration, Economic Services, Social Community Services). This means that past economic growth trends are strong indicators of future growth, and policy makers may focus more on understanding and responding to these trends rather than heavily relying on adjusting government spending. For Nigeria, this could imply that historical economic data and growth patterns play a larger role in shaping future projections and economic strategies, while changes in expenditure alone may not significantly drive immediate GDP growth.
The present study expands on past research by addressing gaps in the sustainability and long-term impacts of capital expenditure on Nigeria’s economic growth, which have often been overlooked in previous works. While studies by Chandana et al. (2020) and Iriabije et al. (2023) emphasize the positive effects of capital expenditure on economic growth in both the short and long run, the present study integrates sustainability-focused metrics that consider not only economic outcomes but also socio-economic equity and environmental factors. This broader approach acknowledges that the role of capital expenditure in development extends beyond immediate GDP growth, emphasizing the need for balanced, long-term policy planning. In comparison to studies like Okonkwo et al. (2023) and Ukoh and Nwaoha (2024), which focus primarily on administrative and economic services expenditures, this research brings a more holistic view by considering how these expenditures interact with sustainability factors, which have often been underexplored.
Furthermore, while past studies typically relied on models like ARDL and focused on specific sectors or economic indicators, the present study uses advanced econometric techniques to assess the broader impact of capital expenditure across sectors. This includes a more nuanced understanding of the sectoral interaction between administration, economic services, and social/community services expenditures, which previous studies did not fully address. The findings of the current research fill the critical gap identified by examining the enduring and comprehensive role of capital expenditure in driving sustainable economic development. This approach provides more relevant insights for policy-making, particularly for Nigeria, where the long-term impacts of government expenditure on economic development remain a pressing concern amidst challenges such as fiscal constraints and environmental sustainability.
Recommendation
Prioritize Capital Expenditure on Productive Sectors
The Nigerian government should allocate a significant portion of capital expenditure towards sectors with high potential for long-term economic growth, such as agriculture, manufacturing, and infrastructure. Investments in these sectors not only create jobs but also stimulate productivity and innovation, driving sustainable economic growth.Ensure Transparent and Efficient Allocation of Resources
Capital expenditure should be carefully monitored and evaluated to avoid inefficiencies and corruption. The government should enhance transparency by providing clear and accessible reports on expenditure allocation and outcomes, ensuring that funds are effectively directed towards priority projects that directly benefit the public and the economy.Integrate Sustainability Considerations in Expenditure Planning
The government should incorporate sustainability metrics in the planning and evaluation of capital projects, ensuring that investments are not only economically beneficial but also socially and environmentally responsible. For instance, green energy, environmental conservation, and social equity should be considered in the design and execution of infrastructure projects.Support Public-Private Partnerships (PPP) for Infrastructure Development
To reduce the burden on public finances, the government should encourage public-private partnerships (PPPs) to fund large-scale infrastructure projects. This would allow for more efficient use of resources, leveraging private sector expertise and capital, while still ensuring that projects align with national development goals.Focus on Human Capital Development in Line with Expenditures
Capital expenditures should be directed towards projects that build human capital, such as education, healthcare, and vocational training. Investments in education and skill development will ensure that the workforce can take full advantage of economic opportunities created by infrastructure development.Encourage Long-Term Planning and Sustainability of Projects
The government should adopt long-term planning frameworks that focus on the sustainability and maintenance of capital projects. Short-term gains should not overshadow the need for projects to be maintained and scaled over time to achieve enduring impacts, particularly in critical infrastructure sectors.Leverage Data-Driven Decision Making in Budget Allocation
Using advanced data analytics and econometric models, the Nigerian government should analyze the economic impact of past capital expenditures to make more informed decisions. This data-driven approach will help optimize spending, identify successful projects, and adjust policies to focus on the most impactful areas.Strengthen Monitoring and Evaluation Frameworks
Establish a robust framework for monitoring and evaluating capital expenditure projects. Regular assessments of project outcomes will ensure that funds are used effectively and allow for adjustments in real-time to prevent waste or misallocation, ensuring that every naira spent contributes to economic development.
By implementing these recommendations, the Nigerian government can drive more impactful capital expenditures that will not only stimulate economic growth but also ensure that this growth is sustainable, equitable, and aligned with long-term development goals.
Conclusion
In conclusion, this study highlights the significant role of capital expenditure in driving Nigeria’s economic growth, emphasizing the need for careful allocation of funds to key sectors like agriculture, manufacturing, and infrastructure. It underscores the importance of transparency, sustainability, and long-term planning in capital projects, ensuring that investments contribute not only to short-term economic gains but also to long-term development. By focusing on productive sectors and incorporating data-driven decision-making, the government can foster a more robust and sustainable economy that benefits all Nigerians.
REFERENCES
Chandana, Aluthge; Adamu, Jibir; and Musa, Abdu (2020) “Impact of Government Expenditure on Economic Growth in Nigeria, 1970-2019,” CBN Journal of Applied Statistics (JAS): Vol. 11: No. 2, Article 6. Available at: https://dc.cbn.gov.ng/jas/vol11/iss2/6
IRIABIJE Alex Oisaozoje & ETTAH Bassey Essien & NWOSU Nkemjika, 2023. “Capital Expenditure and Economic Development: Implications for Economic Planning in Nigeria,” International Journal of Research and Innovation in Social Science, International Journal of Research and Innovation in Social Science (IJRISS), vol. 7(9), pages 823-837, September.
Mohammed Saleh Auwalu, Aigbedion I. Marvelous,& Aiyedogbon, O. John. (2024). Government Capital Expenditures and Balance of Payment in Nigeria: 1990-2022. International Journal of Development Strategies in Humanities, Management and Social Sciences Vol. 14, No. 2.
Nwaba, Evans & Chika, Odukwu & Eke, Promise. (2023). Capital Expenditure Cost and Economic Development in Nigeria. 6. 13-24.
Okonkwo, Osmond & Ojima, Davis & Echeta, Desmond & Duru, Erasmus & Ejike, & Okechukwu, Akamike & Manasseh, Charles. (2023). IMPACT OF GOVERNMENT CAPITAL EXPENDITURE ON THE ECONOMIC GROWTH RATE OF NIGERIA. 8. 335-348.
Ukoh, J. E. ., & Nwaoha, C. D. (2024). Effect of Federal Government Capital Expenditure on the Performance of Nigerian Economy (2007-2022). American Research Journal of Contemporary Issues, 2(3), 82-93. https://www.openjournals.ijaar.org/index.php/arjci/article/view/689
Quiver Analytics Nigeria, timoyebamiji@gmail.com↩︎
Middle Tennessee State University Murfreesboro Tennessee USA, bolajimoses1@gmail.com↩︎