Libyan Fiscal Policy

 – Prepared by the researcher 

Abubaker Khalifa Dileab – Faculty of Economics. Omar Al-Mukhtar University

Ali Mansour Ateeyah – Faculty of Economics. Ajdabiya University

Democratic Arab Center

International Journal of Economic Studies : Fifteenth issue – February 2021

A Periodical International Journal published by the “Democratic Arab Center” Germany – Berlin

Abstract 

This paper aims to evaluate Libya’s fiscal policy by scrutinizing the long term relationship between the government’s revenue and expenditure. Numerous versions of cointegration approaches have included the application of the conventional Engle-Granger and Johansen’s cointegration tests. First, the analysis of the data properties is carried out using Augmented Dickey-Fuller unit root test as part of the identification process for the data’s stationarity. The outcomes of both cointegration tests indicate that the long-run equilibrium between government’s revenue and expenditure does not exist, suggesting that the fiscal policy is unmanageable in Libya. The implication of the policy’s results specifies that there is interdependence between government spending and revenue. This shows that there is no formulation in the government as spending and revenue decision is done simultaneously. Thus, the government should look for methods to reorder intertemporal relationship among public revenue and expenditure that proportionate with the economy’s revenue mobilization probability.

Introduction

Currently the utmost significant macroeconomics discourse is the connection between public spending and revenue. Such affiliation created a main concern for both governments and policy makers. Hence, the significance and relevance of empirical researches on interrelation amongst public revenue and expenditure are undeniable as governments in both developed and developing economies have incurred fiscal deficits continuously. This is the result of the connection between public spending and revenue and their effect on fiscal deficit. An accurate fiscal policy is important promoting a stable price and maintaining employment growth and output. In macroeconomic policy discussions, the fiscal policy is greatly seen as an instrument that should be used in mitigating fluctuating output and employment. Furthermore, it could be utilized in bringing economy back to its possible stages. Consequently, it is important to have deep understanding on the connection between public expenditure and revenues when assessing the efficiency of the government’s strategy in resources’ management. Based on the significance of public economics’ matters, particularly

Based on the issue’s significance in public economics, particularly the causality of serious consequences for fiscal deficit, and a large quantity of researches have evaluated the connection between public revenue and expenses throughout the decades. The latest economic crises have shown the effects of fiscal policies being managed poorly and how it has caused great damages to the nation’s welfare. It also shows the importance of an effective handling of fiscal policies benefits economic development.  The Keynesian theory highlights the important role of public spending on the economy and how it functions as the catalyst of economic development in every economic sector and betters the public’s standard of living. In Libya, the fiscal policy is weak among all the economic policies implemented despite the government’s dependence of public expenditure for public projects’ financing. This may have been caused by the dependency on oil as the main source in supplying the public budget needed, thus the dependency has weakened the country’s fiscal policy.

Generally, Libya’s state budget is utilized in reinforcing and achieving the economic targets by guiding public spending in the direction of building infrastructure, education, healthcare and housing to escalate living standards. Government expenses play the role as Libya’s main GDP component that averages about 33% in the past decade. Over the study’s model period, public expenditure increased from LD 37 billion in 1962 when it reaches more than LD 45 billion in fiscal year 2019.

Table 1: Government Revenue and Expenditure for Selected Years (million, L.D)

Source: Central Bank of Libya, Annual Reports.

The consecutive yearly budgets showed a rise in spending on economic activities, social welfare, housing and infrastructure construction as the aim is to better the citizens’ lives. Existing spending rose periodically to meet the salaries’ raised and that shows a high percentage of the state’s budget. Nevertheless, public revenue is procyclical as it is dependent on the oil receipts and price. Alternatively, there were several fiscal years that experience deficits and the deficits were funded from excess attained in previous years or from assets. Consequently, it is imperative to experimentally explore the long-run association between public spending and revenue in Libya from 1962-2019. The study is divided into five sections. Succeeding the introduction is literature review of applicable studies and section three discusses data and methodology that is used in this paper. Section four provides the study’s empirical outcomes. Finally, the paper’s conclusion is presented in section five.

Problem Statement

Fiscal policy plays a pivotal role in the sustenance of macroeconomic stability, as well as  government statistics can be guide to identify the effects of fiscal policy on economic performance (Siegel, 1979; Tanzi and Blejer, 1984). Excessive debt and deficits are among the most increasing concerns in the global . Consequently, they result in the weakening of the economy and compromise its sustainability of finances. In 2020s, the Libyan economy has witness persistent budget deficits coupled with high levels of government debt due to political conflicts that the country went through at the beginning of 2011, which causes a lowering public revenues and an increase public expenditure, espeashly current expenditure. for instance, the public revenue gradually decreased from 61.5 billion  in 2010 to 22.3 billion in 2018 coupled with decline in GDP from 102.5 billion to 36.8 billion in the same time. These changes result in  an increase the  budget deficit to retch about -28% of GDP. According to central bank of Libya,  the government’s spending is growing at a rate greater than its income; in consequence, an ever-increasing gap between public expenditure and revenue may result in an increase in levels of government debt and continuous fiscal deficits, which may eventually create future economic imbalances. This raises a question: Is the fiscal policy in Libya sustainable?

Research Question

In order to achieve the  objective of the study, it must answer the following question:

• Is the Libya’s government fiscal position sustainable?

Research Aim and Objective                                                 

In fact, the increase in government debt and budget deficit in Libya has motivated us to carry out an empirical study to investigate the sustainability of fiscal policy in Libya .Thus, the main objective of the study is:

• To investigate the sustainability of fiscal policy in Malaysia

Hypotheses’ Statement

Based on the empirical findings Based on the previous empirical findings that conducted on developing economies, we can establish our hypothesis as follows:

• The fiscal policy for Libya between 1962 and 2019 is sustainable.

Literature Review

In this section, the paper evaluates the theoretic literature; numerous empirical literatures based on the connection between public spending and revenue. Nonetheless, the agreement on the affiliation between government spending and expenditure revenue did not exist. The empirical studies highlight various types of explanation on the link between these variables.

In this context, many researches have been Conducted on  fiscal policy sustainability in developed and developing economies. For instance, in Arabic countries, namely, Saudi Arabia, the link between government spending and revenue used the unit root test, Engle-Granger cointegration method and an error correction model. (Al-Qudair (2005)) confirmed the presence of bidirectional causality between public revenue and expenditure that supports fiscal synchronization hypothesis. Moreover, in Asia, a study on government’s revenue and expenditure affiliation in Malaysia, the application of  Johansen cointegration and error correction models was to sample data from 1965 – 2002. Wong and Lim (2005) found that the outcomes maintained tax-spend hypothesis, and  public  revenue Granger caused spending. Aisha and Khatoon (2009) evaluated the link between the Pakistani government revenue and outlay from 1972-2007 and applied the methodology of unit root test, Engle-Granger approach of cointegration and error correction model. The results revealed that cointegration exist among government spending and revenue. Ravinthirakumaran (2011) explores the relationship between Sri Lanka’s public revenue and expenditure from 1977-2009. His study used Engle-Granger’s approach of cointegration and error correction model.  The results concluded that bidirectional causality existed between government spending and revenue and there was evidence of a long term balance between Sri Lanka’s two time series variables. Additionally, there is on long term relationship between government expenditure and revenue, as Ali and Shah (2012) studied Pakistan’s case. They used annual data for the period of 1976-2009, and used Johansen cointegration and Granger causality techniques.

In similar, Hoai et al. (2015) found that there was no sustainability of fiscal policy in Vietnam. a same conclusion was found in China by a study of (Cuestas and Regis, 2018). Tung Bui (2020) applied a panel data on 22 developing economies to examine the fiscal policies in these nations. He found that there was no evidence of sustainable fiscal policies for these countries.

Nevertheless, in African counties, and in the brief analysis, Granger causality analysis confirmed that public revenue and spending had no causal link in Pakistan. For similar purposes, utilizing yearly data over the period of 1970 to 2008 and Autoregressive Distributed Lag (ARDL) technique, Aregbeyen and Ibrahim (2012) concluded that there as  a presence of a long term relationship between the government’s expenditures and revenues in Nigeria. These results referred to the similar conclusion of (Fasano & Wang, 2002). Nyamongo et al. (2007) investigated this relationship in South Africa, and they provided different results. Vector Error Correction Model (VECM) was used to analyze monthly data. The results revealed that there are cointegration between government revenue and expenditure. A sample of annual data was transformed to quarterly data from 1983-2007, utilized by Amoah and Loloh (2008) to test the affiliation between government expenditure and revenue in Ghana. The study uses Engle-Granger bivariate methodology; it discovers a cointegration between the two variables. Results attained from Johansen cointegration and Error Correction model methodology showed that there are indication of the presence of a long term relationship between revenue and expenditures.In contrast, a fiscal policy sustainability in Southern African Development Community Countries (SADC ) were found in a study of ( Nzimande and Ngalawa, 2019).

As for developed countries, Zapf and Payne (2009) concluded that  state and local public expenditure revealed that the budget disequilibrium in the long run, and the state and local government revealed to have a noteworthy influence on the state’s and local public’s revenues in the short-run. Equally, Gil-Alana (2009) application of the fractional cointegration and ECM techniques on the US government spending and revenues evaluated the relationship between them. The study discovers no cointegration at any levels. It was also discovered that revenue and expenditures do not impact huge fiscal imbalances (Stoian, 2008). With the employment of Engle Granger cointegration test linked to the threshold autoregressive (TAR) and momentum threshold autoregressive (MTAR) cointegration techniques and error correction model (ECM) on the data, the aggregated state, local public spending and U.S.A revenue.  After the financial crisis of 2008, in European economies, seven states were investigated by (Neaime,2015), he found a weaknesses of fiscal policy sustainability in these countries. Similarity, results for the OECD group showed an inadequacy in their fiscal policy sustainability from the period of 1970‒2010 (Afonso and Jalles, 2015).

The summarization of previous empirical literature may provide an obvious comprehension related to the association between different countries’ government revenues and expenditures specific analysis. Furthermore, the correlated proposition on the connection between the public spending and revenue do not have an identifiable pattern amongst economies, with regards to both advanced and emerging economies. In conclusion, the results attainable are complex to the approximation technique and the data’s nature.

Data Methodology

This paper employs yearly time series data on public revenue and expenditure from 1962 to 2019.  The data is obtained from the Libyan Central Bank yearly reports. Lastly, the variables are in standard terms and shifted into logarithm.

In fact, most variables have a unit root (non-stationary). Unit root tests are beneficial in identifying integration orders of the time series. Characteristically, the empirical test initial stage begins with carrying out unit root tests to recognize a variable’s univariate characteristics to demonstrate if the time series have a unit root. Generally, if variable has no unit root (stationary) it is described as an integrated order zero or I (0). If the time series variable is not stationary at its level but stationary from the initial different form, it is described as an integrated order one or I (1). For this function, the study used the well-known unit root, the Augmented Dickey Fuller (ADF) test. Thus, the model is as follows:

Where X is the inspected series, t signifies the time trend; ρ is the largest lag length engaged to catch any autocorrelation,  signifies the drift term, and Δ is the initial difference operator. The null hypothesis H0: t X is I (1), that is unit root, is vetoed in approval of I(0) if  is discovered  negative and statistically meaningfully altered from zero.

In econometrics, a long term relationship amongst variables exists if they are co-integrated. In our case, it is obvious that the two time series variables (R and EX) are integrated of order one, I (1). If there existing linear combination e.g. the disturbance term, Ủ from regression is of integration’s lower order , I(d-b), and b>0, then Engle and Granger (1987) defines R and EX as co-integrated. Henceforth, the conditions where R and EX are both I(1), and residuals U are I (0) , the two variables will be co-integrated (Harris, 1995). Engle and Granger (1987) performance has two stages to model the relationship amongst co-integrated time series as follows: firstly, unit root tests must be applied to decide the cointegration order. The data consequently be analyzed on Ordinary least square (OLS) technique termed co-integrating regression:

……………………………………………………….. (2)

R and EX are public revenue and expenditure respectively and  is constant and U contributes the error terms.

Next, the researchers will evaluate if the residuals U, from the regression are stationary. If there are no co-integration in R and EX, any of these variables linear combination would be non-stationary, and subsequently the residuals would be non-stationary (Pindyck & Rubinfeld, 1998). For additional strong results on the presence of long-run relationship between the government revenue and expenditure, this research uses Johansen cointegration test method to see if the variables non-stationary set engaged in this research is co-integrated.

Empirical Results

In a stationary test, the study applicable is the Augmented Dickey-Fuller (ADF). Firstly, at level (intercept and Intercept & Trend), the ADF test displays null hypothesis is not accepted for two variables (R, EX) at level (not integrated). but, at the initial difference form (intercept and Intercept & Trend), the test outcomes revealed that both variables utilized in the model have no unit root at 1^st difference, and the null hypothesis refused at the unit root can be established as the computed F statistics values are a smaller amount to the acute value of five per cent significant level.  Consequently, a strong assumption is possible where integration two variables (R, EX) are at order of one I (1). Table (2) recapitulates ADF test outcomes.

Table 2: Augmented Dickey-Fuller (ADF) Unit Root Test Results for Stationarity of Variables

Note: (i) the ADF unit root test is performed by using Eviews 7.1. the lag selection has been chosen automatically by the program with maximum number of lag (9) and the optimal lag lengths of ADF is obtained by using Schwarz Information Criterion (SIC) and the results of the lag order are given parentheses( ). (ii) *,** and *** indicate significance at 0.01, 0.05 and 0.10 level respectively. (iii) Mackinnon (1996) critical values are used.

From the time when the integration of time series variables are stationary and the similar order, the Engle-Granger’s cointegration technique is applicable to reveal the existent of cointegration among government expenditure and revenue. In the next step, the study runs the regression. Table (3) provides a outcome summary of ADF residuals.

Table 3: Engle-Granger Cointegration Test and Long-run Relationship

Note: (i) R and EX are Government Revenue and Expenditure respectively (ii) **Significant at Five percent Mackinnon (1996) p-values.

In Equation (2), the researcher tries to use the ADF test to approximate the long-run link between government revenue and government expenditure as a way to show the residuals’ stationarity. In this equation, the government revenue is seen as a dependent variable while expenditure as an independent. From the results attained from table (3), the assessed residuals are non-stationary, whereas the computed F-statistic (-1.44) value attained from ADF test is superior to MacKinnon critical value at 5% significance level. This shows that the unit-root null hypothesis could not be rejected, and the implication is that there is no indication of a co-integration association between the government revenue and expenditure. Thus, it can be determined that both variables are not co-integrated, and the long-run relationship of both variables does not exist.

Table 4: Johansen’s Cointegration Test and Long-run Relationship

Note: (i) Trace Test and Max-Engen Value Test indicate no cointegration at the 0.05 level, (ii) *denotes rejection of the null hypothesis at 0.05 level (iii) ** Mackinnon-Haug-Michelis (1999) p-values.

Based on the outcomes attained from the Engle-Granger two-steps co-integration test, the co-integration test displays that the long run relationship between both variables could not be recognized. For likelihood to obtain a long-run relationship among the variables, the study analyses the data over a Johansen co-integration method (1988). Table (4) shows results of unlimited cointegration rank test (Trace) and unlimited cointegration rank test (Maximum Eigen value) to prove that long run equilibrium exists among the R and EX. With reference to results of both trace test and the maximum Eigen value tests, no cointegration null hypothesis could be rejected at (5%) level of significance. On the other hand, the null hypothesis that (r ≤1) can be rejected. This result indicates that there is no cointegration and the long-run relationship between government revenue and government expenditure does not exist during the period of approximation the sample study.

To strengthen the study’s outcomes, various diagnostic tests are used and these tests are Serial Correlation (LM) Test, Heteroskedasticity (ARCH) Test and Normality Test. Table (5) summarizes the results of the tests stated below.

Table 5: Variety of Diagnostic Tests for the Estimated Equation

Note: (i) LM test indicates to Lagrange multiplier test of serial correlation; the normality test is performed based on the test of kurtosis and skewness of residuals; ARCH test indicates to Autoregressive Conditinal Heteroscedasticity test of residuals (ii) the order of LM test and the ARCH test are  given in parentheses( ), and the numbers in [ ] represent the probability-value and CHSQ referee to the X².

LM test outcomes showed that serial correlation problems do not exist in the receiving of null hypothesis of no autocorrelation where it is recommended to accept of null hypothesis of no autocorrelation. This indicates that the estimation is dependable. In addition, the null hypothesis Heteroscedasticity do not exist could not be refused based on ARCH test outcomes. To validate outcomes obtained from econometric test, the normality test is frequently used to see if the data of time series is totally modeled by a standard distribution. With reference to the outcomes of Kurtosis, Skewness and Jarque- Bera, the residuals are generally dispersed in assessed equation, which confirms validity of the obtained results.

Conclusion

The significance of fiscal policy has been on the rise since 1930 which was the era known as “The Great Depression”. Most arguments among economists determined that fiscal policy is compulsory to maintain a state’s economic equilibrium. For the past two decades, Libya has risen its public spending in certain socio-economic sectors as a result of the high energy prices. The increase of petrol prices gave the country the opportunity to earn more thus, supporting the country’s fiscal position. Nevertheless, the oil prices instability has weakened the fiscal policy and there is a need to efficiently reassess and implement to evade fiscal imbalances. Deep comprehension of the government revenue and expenditure’s relationship between could highpoint how fiscal authorities inhibit the resources’ distribution in the economy. Consequently, it is important for policymakers to think of the influences of present and past variations in both expenditure and revenue in the formulation of fiscal policy.

This study engages time series econometric methods for example; the ADF unit root test, Engle-Granger co-integration test and Johansen co-integration test  to examine Libya’s long-run relationship between government revenue and government expenditure from 1962-2007. The ADF test found out that the revenue and spending were integrated of order one. However, both, Engle-Granger co-integration test and Johansen co-integration test showed that there was no cointegration among government revenue and expenditure, which means that the long-run relationship between government revenue and expenditure does not exist. Therefore, it can be established that Libya’s fiscal policy is unmanageable, and the economy is suffering from fiscal imbalances in both long-and-short-run. The outcomes mean that the government spending decision is isolated from the revenue decision. Based on this situation, the Libya’s government revenue and spending function as disequilibrium in the cointegration relation. Consequently, this proposes that Libyan fiscal policymakers should plan revenues and spending concurrently.

The fiscal policy commendations for Libya are summarized as following: to obtain a sustainable fiscal policy, the authorities should try to face up the consequence of policies up to this point. To begin with, the public spending should be re-explored to evaluate its influence to an effective distribution of resources within the economy in addition to its ability to finance growth improving expenditure groups. Furthermore, the government should look for ways to rearrange intertemporal association among public revenue and expenditure as a way to proportionate the economy’s revenue mobilization possibility. This eases the intact medium-term budgeting framework and helps the fiscal authority regulate their spending rather than raising the receipts. Therefore, fiscal discipline is reestablished without endangering the buildup of factors and influencing a long-run state economic growth.

Reference

Afonso, A. and Jalles, J.T. (2015), “Fiscal sustainability: a panel assessment for advanced economies”, Applied Economics Letters, Vol. 22 No. 11, pp. 925-929.

Aisha, Z., and Khatoon, S. (2009). Government Expenditure and Tax Revenue, Causality and Cointegration: The Experience of Pakistan (1972-2007). The Pakistan Development Review, 2(3),951-959.

Ali, R., and Shah, M. (2012). The causal relationship between government expenditure and revenue in Pakistan. Interdisciplinary Journal of Contemporary Research in Business, 3(12), 323-328.

Al-Qudair, K. H. (2005). The Relationship between government expenditure and revenues in the kingdom of Saudi Arabia: Testing for cointegration and causality. Journal of King Abdul Aziz University: Islamic Economics, 19(1), 31-43.

Amoah, B., and Loloh, F. W. (2008). Causal Linkages between Government Revenue and Spending: Evidence from Ghana: Working Paper, WP/BOG-2008/08. Bank of Ghana.14(8), 356-386.

Aregbeyen, O., and Ibrahim, T. M. (2012). Testing the Revenue and Expenditure Nexus in Nigeria: An Application of the Bound Test Approach. European Journal of Social Sciences, 27(3), 374-380.

Cuestas, J.C. and Regis, P.J. (2018), “On the dynamics of sovereign debt in China: sustainability and structural change”, Economic Modelling, Vol. 68 No. C, pp. 356-359.

Economic Studies,Vol. 27 No. 1, 2020,pp. 66-80.

Engle, R. F., and Granger, C. W. (1987). Co-integration and error correction: representation, estimation, and testing. Econometrica: journal of the Econometric Society, 7(4), 251-276.

Fasano, U., and Wang, Q. (2002). Testing the relationship between government spending and revenue: Evidence from GCC countries: International Monetary Fund. 6(4), 536-549.

Gil-Alana, L. A. (2009). Government Expenditures and Revenues: Evidence of Fractional Cointegration in an Asymmetric Modeling. International advances in economic research, 15(2), 143-155.

Harris, R. I. (1995). Using Cointegration Analysis in Econometric Modeling. Pearson Education. chapter ( 3 ), Vol.: 2, 2nd ed. London: Oxford University Press, pp.63-88.

Hoai, B.T.M., Thanh, S.D. and Tung, B.D. (2015), “State budget balance, public debt, and international norms”, Journal of Economic Development, Vol. 22 No. 4, pp. 51-75.

Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of economic dynamics and control, 12(2), 231-254.

Neaime, S. (2015), “Sustainability of budget deficits and public debts in selected European Union countries”, The Journal of Economic Asymmetries, Vol. 12 No. 1, pp. 1-21.

Nzimande,N. and Ngalawa,H. (2019). Fiscal Policy Sustainability in SADC Countries. The African Finance Journal, 86 Volume 21, Part 1, 2019.

Pindyck, R. S., and Rubinfeld, D. L. (1998). Econometric models and economic forecasts Vol:4 chapter(4): Irwin/McGraw-Hill Boston,189-223.

Ravinthirakumaran, K. (2011). The Relationship between Government Revenue and Expenditure in Sri Lanka. Second International Research Conference on Business and Information. Faculty of Commerce and Management Studies, University of Kelaniya, Sri Lanka,pp.134-165.

Siegel, J. J. 1979. “Inflation-induced distortions in government and private saving statistics”. The Review of Economics and Statistics. Vol. 61(1). pp. 83-90.

Stoian, A. (2008). Analyzing Causality between Romania’s Public Budget Expenditures and Revenues. Theoretical and Applied Economics, 11(11), 60-64.

Tung Bui, D. (2020).Fiscal sustainability in developing Asia – new evidence from panel correlated common effect model. Journal of Asian Business and

Wong, H. T., and Lim, K. P. (2005). The relationship between government revenue and expenditure in Malaysia. International Journal of Management Studies (IJMS), 12(2), 53-72.

Tanzi, V. & Blejer, M. I. 1984. “Fiscal deficits and balance of payments disequilibrium in imf adjustment programmes”. Adjustment, Conditionality and International Financing. Washington, DC, IMF.

Zapf, M., and Payne, J. E. (2009). Asymmetric modelling of the revenue-expenditure nexus: evidence from aggregate state and local government in the US. Applied Economics Letters, 16(9), 871-876.