ABR-11709.pdf

Page 1 of 9

Archives of Business Research – Vol. 10, No. 2

Publication Date: February 25, 2022

DOI:10.14738/abr.102.11709. Tenny, L. Z., & Ekperiware, M. (2022). The Macroeconomics of Fiscal Policy Behavior in Liberia: An Error Correction Methodology.

Archives of Business Research, 10(02). 17-25.

Services for Science and Education – United Kingdom

The Macroeconomics of Fiscal Policy Behavior in Liberia: An

Error Correction Methodology

Lester Zomatic Tenny, PhD

University of Liberia, Monrovia, Liberia

Moses Ekperiware, PhD

Caleb University, Lagos, Nigeria

ABSTRACT

The debate of the size of government and economic growth has been popular

especially in Africa. This study examined the relationship between fiscal variables,

inflation and economic growth in Liberia. The Vector Error Correction Model

(VECM) is employed. Results from the Impulse Response Function (IRF) analysis

reveal that the response of inflation to growth in the Liberian economy over the

study period, was weak, though significant and negative in the short run. However,

it became positive and normalized in the medium and long runs. This means that

inflation retarded growth only in the short run which is consistent with Barro

(1996) empirical findings that inflation impact growth negatively and significantly.

Also observed is the relationship between government expenditure and economic

growth. For the Liberian economy and despite the interruption of the war,

government expenditure impact on growth is a short run positive event, In medium

to long run, it has a negative effect. This means that government expenditure only

spur growth in the short run slightly but did not bring about growth in the medium

to long run. This makes Keynesian theory relative to the intervention of

government through spending given rise to growth invalid for the Liberian

economy in the long run. However, the impact of growth on expenditure in the

medium and long run is significant and strong. This suggests that indeed Wagner’s

Law of increasing State spending is valid for the Liberian economy. Hence, fiscal

policy is still a mix in stirring economic growth in Liberia. This study recommends

well stirred fiscal policies that would positively impact on long run development in

the Liberian economy.

Keywords: impulse response function, vector error correction, Keynesian model

INTRODUCTION

Fiscal policy has been prominent of the four basic issues arising from economic management

like; self-regulating or laissez faire, allowing for non-government interventions; the use of fiscal

policy or a mixture of both fiscal policy and monetary policy. How fiscal policy behavior reflects

in inflation rate in determining productivity is germane to any government involved economic

management system like the Liberian economy. The intricacies of the interrelationships among

the components of these three macroeconomic variable (inflation, economic growth and

government expenditure) which have generated a healthy debate of how fiscal policy reflects

on inflation and economic growth is the core of this study.

Page 2 of 9

Archives of Business Research (ABR) Vol. 10, Issue 2, February-2022

Services for Science and Education – United Kingdom

There is a wide range of issues on the applicability of Wagner’s Law and the theories of

increasing state spending. There are mixed results from various research done to test the

validity of the Law and varied methodologies have produced different results. It is a research

motivation as mixed results abound when the law is being tested in various regions. Also to

note is that the law seems to be valid for developed countries where the availability of

infrastructure and systems are visible, unlike the developing counterparts.

LITERATURE REVIEW

Studies like Perotti (2005) has examined the subject matter government spending. Blanchard

and Perotti (2012), in their empirical characterization of the dynamic effects of changes in

government spending and taxes on output using the USA mixed structural VAR/EVENT

Approach. The study concluded that positive government shocks have a positive effect on

consumption, while positive tax shock has a negative effect. Private consumption investment

was constantly crowded out by taxation, and crowded in by government spending. Fatas (2003)

looked at the effect of fiscal policy on consumption and employment through theory and

evidence in USA with structural VAR. the duo concluded that positive shocks in government

expenditure are accompanied by strong and relentless increases in consumption and

employment. Doessel and Valadkhani (2003) further examine the effect of government on

economic growth in Fiji, using OLS using Annual time series. The results were mixed: that

government expenditure exerts a strong beneficial impact on economic growth. However,

marginal factor productivity in the government sector is found to be lower than that of the

private sector.

Alexander (1980) studied the relationship between economic growth and government

expenditure in OECD countries (13 OECD countries), using the OLS from panel data. The result

indicates that growth of government expenditure has a negative impact on economic growth

from the study. Gregorious and Ghosh (2008) examine the impact of government expenditure

on economic growth in 9 OECD countries, using the heterogenous panel data employing OLS.

The results show that countries with large expenditure tends to experience higher growth. In

another study, Abizadeh and Yousif (1998) demonstrate the dynamics of public spending and

growth through an empirical analysis in South Korea using annual data employing Granger

causality test. The findings indicate that government expenditure did not contribute to

economic growth in Korea. More so, Singh and Sahni (1984) contributed to the discuss by

looking at the causality between public expenditure and national income with India total

aggregate and disaggregate data using granger causality. The results show a positive direction

between national income and public expenditure between 1950-1981.

Tang, Tuck and Cheong (2001) illustrate the Wagner Law validity in Malaysia in an empirical

analytical approach in Malaysia with annual data from 1960-1998 employing Johnasen

multivariate cointegration approach. No longrun relationship among the non-stantionary

variables existed; however, a unidirectional causality was observed from national income to

government expenditure growth, which supports Wagner’s Law for the short run. Cheng and

Lai (1997)on Wagner’s law did an econometric test for South Korea 1950–1981. Applying Sims

(1980) Johanesen 8-cointegration and Hsiao version of granger causality method. The Wagner

Law was valid for public expenditure and economic growth in the study. Also, Folster and

Henrekson (2001) looked at government spending and economic growth in sample of 22

developed countries between 1970-1995 with various econometric approaches. The findings

Page 3 of 9

Tenny, L. Z., & Ekperiware, M. (2022). The Macroeconomics of Fiscal Policy Behavior in Liberia: An Error Correction Methodology. Archives of Business

Research, 10(02). 17-25.

URL: http://dx.doi.org/10.14738/abr.102.11709

came up with a more meaningful and robust result to validate Wagner Law for public

expenditure and economic growth. Further testing the Validity of Wagner’s Law in Bolivia,

Bojanic (2010) used a cointegration and causality analysis with disaggregated annual time

series data from the periods 1940 to 2010. The results, after testing nine versions of Wagner’s

Law on the Bolivian economy, was consistent with Wagner’s proposition. Also, a bidirectional

granger causality was found between income and government expenditures in six of the nine

versions of the law. The findings also suggest that government expenditures do not exert a

positive influence on growth, hence the need to rethink how public funds are spent on a variety

of public services was recommended.

Overall, few studies exist for both developed and developing countries according to Schclarek

(2007). They have been restricted by relatively short time series and highly aggregated fiscal

data in economy like Liberia. This study contributed to the very scanty literature on fiscal

behavior in the Liberia economy by providing a more detailed analysis of the effects of fiscal

policy actions in Liberia, using Error- Correction Mechanism Model (VECM). The study covered

a forty-four-year period, from 1970 to 2014. The periods were chosen because it was a period

that Liberia experience increased growth from exports of iron ore and other natural resources

in the 70s. It also covers the periods of war and post war economic situation in Liberian

economy. Hence, it will be informing and contributive to the understanding of fiscal policy and

inflation dynamics affecting economic growth. The basic research question is to what extent

fiscal Policy and inflation influenced economic growth in Liberia? Based on the puzzle this

research exercise is concerned, the study is set to examine the trend of fiscal policy and

economic growth and analyze the effect of fiscal policy on the relationship between inflation

and economic growth.

Keynes (1936) and his supporters emphasize the role of fiscal policy as a tools that should be

used during times of recession to boost economic activities, that is expansionary fiscal policy,

expanding public expenditure to raise national output (Aruwa, 2010). Higher government

spending may hinder overall economic performance if the spending comes at a cost of increased

taxes and/or borrowing to finance the government expenditures.

A fiscal framework that asks for the implementation of sound fiscal policies, notably within the

Public Financial Management Act of Liberia is an indispensable tool for the well-functioning of

the country’s economy. However, Liberia track records of complying with the fiscal rules laid

down in the public financial management (PFM) documents are mixed. Therefore, it is relevant

to analyze fiscal behavior or government spending, and revenue as well as inflation and the role

they played in contributing to the improvement of the nation’s economy.

METHODOLOGY

This section discusses the econometric frameworks behind this study. The empirical approach

for this study relied on restricted Vector Auto Regression (VECM) in providing for the effect of

fiscal policy in the Liberian economy. The VECM presents the needed impulse response to

shocks in examining fiscal policy. It is an extension of VAR and it incorporates the possibly

cointegrating error term ( ) into the VAR model.

VAR (j) (3.1)

ECTit-1

Yt Yt j

Yt j ECTt t =a +f + +f +p +e 1 -1 - -1 ...

Page 4 of 9

Archives of Business Research (ABR) Vol. 10, Issue 2, February-2022

Services for Science and Education – United Kingdom

The study employed these models to test the effects of fiscal policy on selected macroeconomic

variables (inflation and output) on the Liberian economy. The model for this study is a four

variable model on log of, real government expenditure on goods and services , total

government revenue (GRt), annual rate of inflation measured by the consumer price

index, and real gross domestic product .

A conceptual depiction of the Wagner’s law and Keynesian hypothesis is analytically depicted

by AD-AS framework, using the national income identity:

� = � + � + � + � − � (3.2)

From the standard Keynesian specification and considering inflation rate

(3.3)

Thus, the vector of endogenous variables included in the reduced-form four-variable

restricted VAR (VECM) representation (sequenced or ordered) (equation 3.1) include

(3.4)

Government revenue comes before government expenditure because it is out of resource

mobilization that the government can initiate spending in an economy. The first two variables

are policy variables and as such macroeconomic variables such as inflation and gross domestic

product depend on the nature of the policy in place in the Liberian economy.

Where:

GRt = Total Government Revenue

= total government spending on goods and services

= inflation (measured by the consumer price index)

= Gross Domestic Product

(3.5)

Where J=0,.......k, are the 4x4 matrices

RESULTS AND DISCUSSIONS

Trend Analysis of Public Finance and Economic Growth in Liberia

Between 1970 and 1974, the years before the war, the policy variable- government expenditure

was rising or maintaining upward trend. The macroeconomic variable GDP also rose in the

same period but continued up to 1988, though with some degree of variations. From 1973 to

1974, there was a sharp fall in government expenditure, but this fall did not impact GDP

adversely. This fall was primarily due to the decline in tax receipt which resulted to low inflow

of revenue and the sharp decline of the country’s major exports, iron ore and rubber. GDP

continued to maintain persistent upward trend between these periods (1970-1980) but slowed

after the military coup of 1980. The country’s GDP continued to maintain persistent upward

trend after 1985 due to the increased in economic activities, such as mining, and budgetary

( ) GEt

( )t INFLA

( ) GDPt

t t t t t LGDP =a 0 +a1LGE +a 2LGE +a 3 INF + μ

( ) t t t t GDPt X = GR ,GE ,INFLA ,

GEt

t INFLA

GDPt

Xt A A Xt Ak Xt k t = + + + + e - - .... 0 1 1

, Aj

Page 5 of 9

Tenny, L. Z., & Ekperiware, M. (2022). The Macroeconomics of Fiscal Policy Behavior in Liberia: An Error Correction Methodology. Archives of Business

Research, 10(02). 17-25.

URL: http://dx.doi.org/10.14738/abr.102.11709

support to the military junta from the US government. Government expenditure gained an

upward trend from 1975 to 1981 when the demand for iron ore and rubber heightened on the

global markets after the global oil crisis was resolved in the mid 70s, but fell from 1982 to 1987

slightly as a result of the uncertainty surrounding the economy, especially after the

controversial 1985 presidential elections where the military junta, now turned civilian, won the

elections. Meanwhile, GDP continued to rise but fell sharply in 1989 to 1994 as a result of the

intensification of the Liberian civil conflict. From 1989 to 1997, government expenditure

maintained a constant flow by being steady. This was due to budgetary support from ECOWAS

to the transitional government in Monrovia. From 1995 to 1997, GDP gained some momentum

by rising slowly but steadily. This was fuelled by the continued exports of the country’s major

export, rubber and gold and timbers as well as improved investments climate in the country

after the cessation of hostilities. From 2000 to 2003 witnessed a sharp decline in government

expenditure. The country opened up to the international system. Investment climate was

created and the post conflict infrastructural rebuilding was heightened by the newly elected

government (UNDP report, 2007). Over 16 billion dollars planned investment came into the

country (Liberia national investment commission, 2010). The country was now qualified for

the huge debt waiver after reaching the heavily indebted poor country (HIPC) completion point

in 2010 (AFDB, 2010). Meanwhile, GDP continued to grow from 2006 up to 2014 (World Bank,

2014).

Figure 4.1: Trend of Government Expenditure on Economic Growth before, during and After the

war Source: Author’s computation, 2022

It should be noted however, that inflation (INF) was stationary at levels. The LGR, LGE and LGDP

were all non-stationary at Levels for different Test (DF, ADF and PP) while INF was stationary

at level, hence, Stationary under the various Tests. Therefore, the variables are combination of

I (0) and I(1) variables. For the purpose of any study, the optimal lag length suggested by Akaike

1970 1975 1980 1985 1990 1995 2000 2005 2010

LGE LGDP

WAR

Page 6 of 9

Archives of Business Research (ABR) Vol. 10, Issue 2, February-2022

Services for Science and Education – United Kingdom

Information Criteria (AIC), Schwarz Information Criteria(SIC), Hannan-Quinn criteria (HQ),

Final prediction error (FPE) as well as Likelihood Ratio test (LR) was chosen. This is because

all these criteria can produce conflicting lag length choices. However, the decision about the lag

structure of a VAR model could be based on the fact that a given criteria produces a white noise

residual and converses of freedom. In this study, the optimal lag length was 2.

The cointegration test of both trace and maximum eigenvalue indicate at most two (2)

cointegrated equations as the null hypothesis of at most two cointegrating equations was not

significant hence cannot be rejected at 5 per cent. Hence, there exists a long-run association

among the variables. Therefore, the restricted VAR or Vector Error Correction Model (VECM) is

more amenable than the unrestricted vector Autoregressive (VAR) model in this study because

at most two cointegrating equation exits.

Impulse Response Function (IRF) Result

The IRFs essentially map out the dynamic response path of a variable due to a one-period

standard deviation shock to another variable. The IRFs are normalized such that zero

represents the steady-state value of the response variable (Masih and Masih, 1996). The

relationship between inflation and growth has been an issue of concern for macroeconomists.

Classical economics recalls supply-side theories, which emphasizes the need for incentives to

save and invest if the nation's economy is to grow; Keynesian theory provided the AD-AS

framework, a more comprehensive model for linking inflation to growth. Monetarism

reemphasized the critical role of monetary growth in determining inflation, while Neoclassical

and Endogenous Growth theories sought to account for the effects of inflation on growth

through its impact on investment and capital accumulation (Gokal and Hani, 2004).

Page 7 of 9

Tenny, L. Z., & Ekperiware, M. (2022). The Macroeconomics of Fiscal Policy Behavior in Liberia: An Error Correction Methodology. Archives of Business

Research, 10(02). 17-25.

URL: http://dx.doi.org/10.14738/abr.102.11709

Figure 4.1: Impulse Response Function

Source: Author’s computation, 2022

The response of inflation to growth in the Liberian economy over the study period, for example,

was weak, significant but negative in the short run but became positive and normalized in the

medium and long runs, which is consistent to studies of Fischer (1993) and De Gregorio (1993).

This means that inflation retarded growth only in the short run which is consistent with Barro

(1996) empirical findings that inflation impact growth negatively and significantly. On the

other hand, the relationship between growth and inflation for the Liberian economy was

significant and positive in all periods, thus validating the unidirectional relationship that the

causality between the two variables ran one-way from GDP growth to inflation (Gokal and Hani,

2004).

Also observed is the relationship between government expenditure to growth. From all

indications, theories have provided two major findings: Wagner’s Law for example posits that

the industrialization of society will spurred series of government intervention so that the need

for more provision of services from government is inevitable. That is to say, as society grows

and expands, the building of additional infrastructure, employing more servicemen, etc.

becomes a necessity. On the other hand, Keynes and his follower believe that it is the

intervention of government in a society that spurs growth. For the Liberian economy and

-.4

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GR) to LOG(GR)

-.4

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GR) to LOG(GE)

-.4

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GR)to INFLATION

-.4

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GR) to LOG(GDP)

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GE) to LOG(GR)

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GE) to LOG(GE)

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GE)to INFLATION

-.2

1 2 3 4 5 6 7 8 9 10

Response of LOG(GE) to LOG(GDP)

-8

-4

1 2 3 4 5 6 7 8 9 10

Response of INFLATION to LOG(GR)

-8

-4

1 2 3 4 5 6 7 8 9 10

Response of INFLATION to LOG(GE)

-8

-4

1 2 3 4 5 6 7 8 9 10

Response of INFLATION to INFLATION

-8

-4

1 2 3 4 5 6 7 8 9 10

Response of INFLATION to LOG(GDP)

-.1

1 2 3 4 5 6 7 8 9 10

Response of LOG(GDP) to LOG(GR)

-.1

1 2 3 4 5 6 7 8 9 10

Response of LOG(GDP) to LOG(GE)

-.1

1 2 3 4 5 6 7 8 9 10

Response of LOG(GDP) to INFLATION

-.1

1 2 3 4 5 6 7 8 9 10

Response of LOG(GDP) to LOG(GDP)

Response to Cholesky One S.D. Innov ations

Page 8 of 9

Archives of Business Research (ABR) Vol. 10, Issue 2, February-2022

Services for Science and Education – United Kingdom

despite the interruption of the war, government expenditure impact on growth is a weak short

run event but in medium to long run, it has a negative effect. This means that government

expenditure only spur growth in the short run slightly but did not bring about growth in the

medium to long run. This makes Keynesian theory relative to the intervention of government

through spending given rise to growth invalid for the Liberian economy. On the other hand, the

impact of growth on expenditure in the medium and long run is significant and strong. This

suggests that indeed Wagner’s Law is valid for the Liberian economy.

CONCLUSION

This study contributed to other literature by examining fiscal policy, inflation and economic

growth. The results from the study showed that indeed Keynesian theories are invalid for the

Liberian economy. Results from the Impulse Response Function (IRF) analysis reveal that the

response of inflation to growth in the Liberian economy over the study period, was weak,

though significant but negative in the short run but became positive and normalized in the

medium and long runs. This means that inflation retarded growth only in the short run which

is consistent with Barro (1996) empirical findings that inflation impact growth negatively and

significantly. The results of the study confirmed the validation of Wagner Law effect that indeed

growth motivates government expenditure in the Liberian economy.

For the Liberian economy and despite the interruption of the war, government expenditure

impact on growth is a short run event but in medium to long run, it has a negative effect. This

means that government expenditure only spur growth in the short run slightly but did not bring

about growth in the medium to long run. However, the impact of growth on expenditure in the

medium and long run is significant and strong. This suggests that indeed Wagner’s Law is valid

for the Liberian economy. Hence, fiscal policy is still a mix in stirring economic growth in

Liberia. This study recommend well stirred fiscal policies that would positively impact on long

run development in the Liberian economy. Thus, the study tested whether empirical evidence

based on annual time series data from Liberia supports the Keynesian view on the effect of

policy variables (government expenditure and government revenue) on macroeconomic

variables (Gross domestic product and inflation).

Reference

Abizadeh, S., Yousefi, M. (1998), “An Empirical Aanalysis of South Korea’s Economic Development and Public

Expenditures Growth.” Journal of Socio-economics, Vol. 27, No. 6, pp. 687-701.

Aruwa, S.A. (2010). The quality of public expenditure in Nigeria. Nigeria Defense Academy, Kaduna. Seminar

series III.

Barro, R. (1996) inflation and growth. Journal of Economic Growth

Blanchard, O., Perotti, R. (1999). An empirical characterization of the dynamic effects of changes of government

spending and taxes on output. NBER Working Paper, 7269. National Bureau of Economic Research, Massachusett.

Bojanic, A.N. (2013) Testing the Validity of Wagner’s Law in Bolivia: A Cointegration and Causality Analysis with

Disaggregated Data.Economic Analysis Review vol. 28 no. 1

Cheng, B. and Lai, T.W. (1997) Government Expenditures and Economic Growth in South Korea: A VAR

Approach. Journal of Economic Development. Volume 22.

Doessel, D.P. and Valadkhani, A. (2003). The effect of Government on economic growth in Fiji. Faculty of

Commerce-Paper, University of Wollogon online research.

Fatas, A. (2003) The case for restricting fiscal policy discretion. The quarterly journal of economics 118(4).

Page 9 of 9

Tenny, L. Z., & Ekperiware, M. (2022). The Macroeconomics of Fiscal Policy Behavior in Liberia: An Error Correction Methodology. Archives of Business

Research, 10(02). 17-25.

URL: http://dx.doi.org/10.14738/abr.102.11709

Gokal, V and Hanif, S. (2004) Relationship between inflation and economic growth. Working Paper. Reserve

Bank of Fiji

Ghosh, S., Gregoriou, A. (2008). The composition of government spending and growth: is current or capital

spending better? Oxford Economic Papers, 60 (3), 484–516. DOI: 10.1093/ oep/gpn005

Henrekson, M. (1993) Wagner’s law: A spurious relationship? Public Finance, 49 (2).

Perotti, R. (2007) In search of the transmission mechanism of fiscal policy. NBER Macroeconomic Annual.

Singh, B. and Sahni, B.S. (1984), “Causality between public expenditure and national income,” The Review of

Economics and Statistics, vol. 66, no. 4.

Tang, T. C., Tuck, L. Cheong, H. (2001) Testing the relationship between government expenditure and national

income in Malaysia. Analisis, 8 (1&2). pp. 37-51. ISSN 0127-8983.