1

CHAPTER ONE

INTRODUCTION

1.1 Background of the Study

All over the world, inflation has been an area of interest amongst policy makers,

economist and monetary authorities of both developed and developing economies. Its impact of

it to economic growth has been one of the most debated concepts in macroeconomics. Inflation,

interest and exchange rates are fundamental macroeconomic variables, capable of changing the

direction and growth pattern of a country’s economic growth and development.

Inflation is seen as persistent and an appreciable increase in the general prices of

goods and services due to the volume of money in circulation (Udude 2014). Aminu and Anono

(2012), opined inflation as a persistence rise in the general price level of broad spectrum of

goods and services in a country over a long period of time, they attributed inflation to a popular

say that inflation is too much money chasing too few goods. This is an increase in the money

and credit relative to available goods and services resulting in a general price level. Inflation is

an important variable that is used in assessing the performance of the economy. For this reason,

every country aspires to have price stability as one of its core policy objectives.

Both economists and other policy maker favour this policy objective because

fluctuations in prices bring uncertainty and instability to the economy. Rising and falling prices

are both bad because they bring unnecessary loss to some and undue advantage to others. Again

they are associated with business cycles. So a policy of prices stability keeps the value of money

stable, eliminates cyclical fluctuations, brings economic stability, helps in reducing inequalities

of income and wealth, secures social justice and promotes economic welfare.
 2

In a dynamic economy, price changes take place constantly. Sometimes, the price of a

particular good or service may exhibit an upward or downward trend that can last for months,

years or even decades. Initial price changes as a result of shifts in the supply and demand for

particular goods and services do not imply any change in the general price level. A change in the

average price level takes place if there is a strong tendency for all prices to move up or down in

proportion to one another. Such rise in the average price level is Inflation.

Modern countries have all, almost without exception, suffered periods of inflation.

Generally, attempts have been made to control this inflation, both in severity and in their

duration. In some other instance, the acceleration in the rate of increase in prices had been

seeming unmanageable. The German inflation of 1923 (Chu and Feltenstein, 1979) saw price

increases rise to a peak of more than 30,000 percent a month, and the economy was reduced to

function on the basis of barter. However, in 2018, the inflation rate in Zimbabwe rose to 10.6

percent, and is projected to jump dramatically to 319.04 percent in 2020. Although no other

cases of such violent inflation has been recorded in the most recent time, few cases have been

where for a considerable length of time, the rate of price increases tended to be out of control.

The debate on inflation-growth nexus has remained perennial and has attracted

substantial theoretical and empirical efforts. For instance, the structuralists argue that inflation is

crucial for economic growth while the monetarists posit that it is harmful to economic growth

(Doguwa, 2012). The two basic aspects of the debate relate to the presence as well as nature of

relationship between inflation and growth and the direction of causality. Commenting on the

inconclusive nature of the relationship between inflation and economic growth, Friedman (1973)

noted that some countries have experienced inflation with and without development and vice

versa. And Wai (1959) also argues that there is no relationship between inflation and economic
 3

growth noting that growth has been possible without inflation in some countries while in others;

there have been inflation without growth. Similarly, Johanson (1967) posits that there is no

convincing evidence of any clear association, positive or negative, between the rate of inflation

and the rate of economic growth. He argues that it is not inflation that determines economic

growth but the application of knowledge, through technical and managerial change and the

improvement of human capacities. Nell (2000) also opined that single digit inflation may be

beneficial; on the other hand, double digit inflation imposes slower growth. Anochiwa and

Maduka (2015) said the ability to manage the growth of inflation to single digit may be an

important factor to accelerate economic growth.

Inflation affects different people or economic agents differently. Broadly, there are two

economic groups in every society, the fixed income group and the flexible income group. During

inflation, those in the first group lose while those in the second group gain. The reason is that the

price movement of different goods and services are not uniform. During inflation, most prices

rise, but the rate of increase of individual prices differs. Prices of some goods and services rise

faster than others while some may even remain unchanged. The poor and the middle classes

suffer because their wages and salaries are more or less fixed but the prices of commodities

continue to rise. On the other hand, the businessmen, industrialists, traders, real estate holders,

speculators and others with variable incomes gain during rising prices. The latter category of

persons becomes rich at the cost of the former group. There is transfer of income and wealth

from the poor to the rich. It also causes the opportunity cost of holding money to increase

resulting to inefficient use of real resources in transactions (Onyeiwu, 2012).

More generally, which income group of the society gains or loss from inflation depends

on who anticipates inflation and who does not. Those who correctly anticipate inflation can
 4

adjust their present earnings, buying, borrowing and lending activities against the loss of income

and wealth as a result of inflation.

In the economic literature, several factors explain how inflation lowers the output growth

and welfare. Fischer (1984), for instance, identifies several channels through which inflation can

exert a substantial cost on output growth. First, since money remains the most important source

of transactions in modern world, changes in its supply influence the welfare of an economy.

Agents economize the use of money for transaction purposes which consequently entails welfare

losses for the country. The proponents of high inflation, however, establish that these welfare

losses are not so important because money is used for many illegal transactions and therefore

taxing its use has some important re-distributional effects. Further, these welfare effects are also

small because the use of money has become limited after the innovation of new transaction

technologies. Second, inflation is considered costly because it increases the frequency of

transactions among agents. As real balances lose their value quickly in inflation, to avoid this

loss, agents make rapid transactions and spend more time on these activities. This particular

aspect of effects of inflation is named as “shoe-leather” cost in the literature. Nonetheless, this

frequency of transaction is not influenced at low or moderate levels of inflation rate. Only

hyperinflation influences this frequency.

The other important cost of inflation appears through its effects on interest rate and tax

burden. Inflation reduces interest rate earnings of the depositors and these losses are particularly

large when the nominal interest rate is not adjusted accordingly. Fischer posits that the nominal

interest rate paid by the financial institutions exhibits certain controls or has some ceilings in

most of the banking systems, which discourages the deposits and causes resource misallocation.
 5

In fact, these effects are not only confined to interest rates but also transferred to the principal

amount of creditors. A redistribution of wealth takes place between creditors and debtors in favor

of the latter group. Further, inflation also increases the tax burden of agents since it is hard to

implement complete tax indexation, due to administrative problems. This leaves space for

inflation to bring more people in the tax nexus, based on their nominal income, whitch

discourages economic activity.

1.2 Statement of the Problem

Since the attainment of independence of 1960, economic policies have been concerned

basically with anti-inflationary measures aimed at achieving price stability. Indeed, the monetary

policy framework adopted by Nigeria since 1993 has an overriding objective and that is the

achievement of single-digit inflation.

Monetary and fiscal policies as well as wage freeze, price control, exchange rate and other

measures have been employed from time to time to stem the tide of sustained increase in the

general price level. In retrospect, it appears that in spite of these efforts; the achievement of price

stability objective has been limited. Inflation undermines the role of money as a store of value. It

also frustrates investments and growth.

Experience has shown, however, that a continuing high level of inflation has so reduced

the purchasing power o f the Naira from year to year, that the monetary values in successive final

accounts and capital budgets do not give realistic up-to-date values for proper assessment of net

wealth and profitability in Nigeria economy. In Nigeria, rate of inflation is fairly increasing

unlike that of Ghana. The Nigeria inflation rate as of 2013 was 13.7%, which is 0.017% greater

than that of 2012. In 2014, the inflation rate increased to 15.4% and this is 0.017% greater than

2013 rate of inflation. It also reduces by 0.0034% in 2015. Nigeria inflation rate in 2016 and
 6

2017 was 11.8% and 10.3% respectively. In 2018 it also increased to 11.44%. (CBN, 2018 ).

But in other neighboring countries like Ghana the inflation rate is pretty fair. For instance, Ghana

inflation rate as of 2017 was 12.37%, a 5.08% decline from 2016. In 2018, it also decline by

4.56%. These results have shown that the rate at which inflation reduces in Ghana is faster than

that of Nigeria.

Furthermore, it can be noted that economic growth in Nigeria has fluctuated significantly

showing different patterns that contradict with both assertions about the linkage that exist

between inflation and economic growth. This entails that economic growth has been exhibiting

different responses to changes in inflation. This therefore sought to establish the impact of

inflation on economic growth with regards to Nigeria from 1981-2018.

1.3 Objective of the Study

In this study, critical attention is devoted towards analyzing the impact of inflation on economic

growth in Nigeria from 1981-2018. However, the specific objectives are

(i) To estimate impact of inflation on economic growth in Nigeria.

(ii) To examine the causality relationship between inflation and economic growth in

Nigeria.

1.4 Research Question

In order to accomplish the research objectives stated above, this study attempts to answer the

following research questions:

(i) What is the impact of inflation on economic growth in Nigeria?

(ii) What is the causality relationship between inflation and economic growth in

Nigeria?
 7

1.5 Hypothesis of the Study

The following hypothesis will be tested;

(i) Ho: inflation has no significant impact on economic growth in Nigeria.

(ii) Ho: There is no significant causality relationship between inflation and economic

growth in Nigeria.

1.6 Scope of the Study

The undertaking of this study is mainly centered on analyzing the impacts that are posed by

inflation on economic growth. This will be aided by the utilization of secondary data which runs

from 1981 to 2018. The choice of this period was informed by data availability and the need for

a precise time series analysis. The data used here will be gotten basically from Central Bank of

Nigeria (CBN) 2019 statistical bulletin.

1.7 Significance of the Study

The research study is on the evaluation of the impact of inflation on economic growth in Nigeria.

The study will be significant to the following groups:

 Government: To ensure the efficient and effective control of the money in the

economics, this is possible if government or government agencies accept and implement

its policies base on the recommendations stated in this work. Base on the findings and

recommendations made in this study, it will enable the government to come up with more

effective policies that will help to tackle the issues of inflation through effective

monetary policy.
 8

  Lecturers/students: This research work is a very good teaching/learning material for

both lecturers and students in higher institutions who wish to learn about inflation and

its effect on the economic growth of Nigeria.

 Researchers/business tycoons: The study serves as a reference material for further

research in this field. It as well serves as a guide to producers and sellers in different

markets.

 General Public: Interested publics can use the result of this research work as a means of

gaining a better insight into monetary policy of Nigeria and economic growth in Nigeria.

 Policy makers: This study will help in policy making and investment. It will also serve

as a guide to foreign and local investors to have a clear and deeper understanding of the

different Nigerian monetary policies measures aimed at curbing inflation and prepare for

change of policy in the future.

 9

CHAPTER TWO

LITERATURE REVIEW

2.0 Introduction

This chapter deals with the literature review and the chapter is divided into the following

subheading: conceptual framework, theoretical framework and empirical review

2.1 Conceptual Framework

According to Madueme (2011), conceptual framework of literature review is an

articulation and definition of key concepts of the study as they suit your work. Therefore, we

proceed as follows.

2.1.1 The Concept of Inflation

Inflation is one of the most frequently used terms in economic discussions, yet the concept is

variously misconstrued. There are various schools of thought on inflation, but there is a

consensus among economists that inflation is a continuous rise in the prices.

Simply put, inflation depicts an economic situation where there is a general rise in the prices of

goods and services, continuously. It could be defined as ‘a continuing rise in prices as measured

by an index such as the consumer price index (CPI) or by the implicit price deflator for Gross

National Product (GNP)’. Inflation is frequently described as a state where “too much money is

chasing too few goods”. When there is inflation, the currency loses purchasing power. The

purchasing power of a given amount of naira will be smaller over time when there is inflation in

the economy. In the words of Friedman (1968), "inflation is always and everywhere a monetary

phenomenon; and can be produced only by a more rapid increase in quaintly of money than

output". They regarded inflation "as a destroying disease born out of lack of monetary control

whose result undermined the rules of business, creating havoc in the markets and financial ruin

of even the products. According to Jhingan (2009) inflation is a persistent and appreciable rise
 10

in the general level of prices in an economy. Not every rise in the price level is termed inflation.

Therefore, for a rise in the general price level to be considered inflation, such a rise must be

constant, enduring and sustained. Thus, a practical definition of inflation would be persistent

increase in the general price level at a rate considered too high and therefore unacceptable

(Ogboru, 2010). The rise in the price should affect almost every commodity and should not be

temporal. But Demberg and McDougall are more explicit referring to inflation as a continuing

rise in prices as measured by an index such as the Consumer Price Index (CPI) or by the implicit

price deflator for Gross National Product.

In the definition of inflation, two key words must be borne in mind. First, is aggregate or general,

which implies that the rise in prices that constitutes inflation must cover the entire basket of

goods in the economy as distinct from an isolated rise in the prices of a single commodity or

group of commodities. The implication here is that changes in the individual prices or any

combination of the prices cannot be considered as the occurrence of inflation. However, a

situation may arise such that a change in an individual price could cause the other prices to rise.

An example is petroleum product prices in Nigeria. This again does not signal inflation unless

the price adjustment in the basket is such that the aggregate price level is induced to rise. Second,

the rise in the aggregate level of prices must be continuous for inflation to be said to have

occurred. The aggregate price level must show a tendency of a sustained and continuous rise

over different time periods. This must be separated from a situation of a one-off rise in the price

level.

One major focus of monetary authorities the world over is the effective and efficient

management of inflation and money supply, to achieve steady growth of the economy. Price

stability is a cardinal objective of the government macroeconomic goals, given that it bears direct
 11

impact on the standard of living as well as cost of living of the citizenry. Hence, the CBN as a

government institution is charged with the primary responsibility of managing or controlling

money supply to achieve stable prices of goods and services in the economy. According to Sarah

(2014), the CBN takes whatever growth and ination levels that the Federal Government desire to

achieve, to determine how much money would be adequate to grow the economy. In other

words, in observing the growth rates of GDP and Inflation, the CBN determines the extent of

money supply that matches the Government budgetary objectives. The CBN adopts fiscal and

monetary policy coordination to ensure: financial stability, moderate interest rate, and stable

exchange and inflation rates with no adverse effect on the economy.

In an inflationary economy, it is difficult for the national currency to act as medium of exchange

and a store of value without having an adverse effect on income distribution, output and

employment (CBN, 1984). Inflation is characterized by a fall in the value of the country’s

currency and a rise in her exchange rate with other nation’s currencies. This is quite obvious in

the case of the value of the Naira (N), which was N1 to $1 (one US Dollar) in 1981, but has now

fallen to N160 to $1 in 2013, N380 to $1 and N450 to $1 in 2017 and 2019 respectively.

(http://www.oanda.com/convert/classic). This decline in the value of the Naira coincides with the

period of inflationary growth in Nigeria, and is an unwholesome development that has led to a

drastic decline in the living standard of the average Nigerian.

There are three approaches to measure inflation. These are the Gross National Product (GNP)

implicit deflator, the Consumer Price Index (CPI) and the wholesome or producer price index

(WPI or PPI). The period to period changes in these two latter approaches (CPI and WPI) are

regarded as direct measures of inflation. There is no single-one of the three that rather uniquely

best measures inflation. The Consumer Price Index (CPI) approach, though it is the least efficient
 12

of the three is used to measure inflation rates in Nigeria as it is easily and currently available on

monthly, quarterly and annual basis . This study views inflation as a function of monetary policy.

This means that keeping inflation at tolerable level depends on the effectiveness of monetary

policy. Broadly, inflation can be grouped into six types, according to its magnitude.

1. Creeping Inflation: This occurs when the rise in price is very slow. A sustained annual rise in

prices of less than 3 per cent per annum falls under this category. Such an increase in prices is

regarded safe and essential for economic growth.

2. Walking Inflation: Walking inflation occurs when prices rise moderately and annual inflation

rate is a single digit. This occurs when the rate of rise in prices is in the intermediate range of 3

to less than 10 per cent. Inflation of this rate is a warning signal for the government to control it

before it turns into running inflation.

3. Running Inflation: When prices rise rapidly at the rate of 10 to 20 per cent per annum, it is

called running inflation. This type of inflation has tremendous adverse effects on the poor and

middle class. Its control requires strong monetary and fiscal measures.

4. Hyperinflation: Hyperinflation occurs when prices rise very fast at double or triple digit rates.

This could get to a situation where the inflation rate can no longer be measurable and absolutely

uncontrollable. Prices could rise many times every day. Such a situation brings a total collapse of

the monetary system because of the continuous fall in the purchasing power of money.

5. Demand-pull inflation: demand pull inflation is an inflation that occur when there is an

increase in the conditions of demand. These could either be an increase in the ability to buy

goods or an increase in the willingness to do so. According to Chankreusna (2017) demand pull

inflation is concerned with aggregate demand as the determinant of inflation. Aggregate demand
 13

includes consumption, investment and government expenditure. If aggregate demand is higher

than aggregate supply, it generates output gap which fuels inflationary pressure

6. Cost-push inflation: this inflation arises from anything that causes the conditions of supply to

decrease. Some of these factors include a rise in the cost of production, an increase in

government taxation and a decrease in quantity of goods produced. Jhingan (2010) also noted

that cost – push inflation is caused by wage increases enforced by employers.

2.1.1.1 Cause of Inflation in Nigeria

The following are identified as the real contemporary factors responsible for rising rate of

inflation in Nigeria

1. Incessant increase in the price of fuel (Petrol and Gas): Frequent increase in the price

of fuel is the principal factor responsible for the ever-rising prices of product and services

in Nigeria. Fuel being a macroeconomic product, it prices seems to be the springboard for

prices of other product. This link stems from the influence of fuel price on transportation

cost. Experience has shown that there is a direct relationship between the price of fuel,

transport cost and prices of products. For instance, from 2001 to date the pump price of

petrol (Prime motor spirit) has consistently increased over six (6) times. From N22 per

litre in 2001 to N29 per litre in 2002, N45 in 2003, N65 in 2005 (all official rate) CBN

(2007). Between 2006 and 2009, the economy has experienced various treat of fuel price

increase. However, pump price in 2009 is N72. This has directly or indirectly increased

the prices of product and services over these years.

2. Growth in Money Politics: Excess supply of money is one of the traditional causes of

inflation. Politics in Nigeria over the years have become so monetized that only the rich

can dream of political positions. The flagrant display of money during the electioneering
 14

campaigns leading to increasing financial profligacy. The implication is that the marginal

propensity to consume on free gotten money seems to be higher than for money worked

for. Thus prices of products often get higher and remain at the peak during elections due

to excess circulation of money and hardly recede thereafter.

3. Poor Commitment of Government to Agricultural Policies: The price of food and raw

materials has been the major indices for inflation. It has surfaced over the years due to

government neglect of Agriculture. This was as a result of the assurance of oil revenue.

The economy experienced inflation due to shortage in food supply as compared to higher

demand.

4. High Rate of Urbanization: In search for greener pasture, urbanization leads to influx of

people into the town giving rise to high cost of living in the cities while at the same time

creating shortage of productive labour in the rural areas.

5. Rising Population: Nigeria is currently passing through various phases of demographic

transition model characterized by increasing birth rate and declining death rate. This can

be viewed from the various development programmes such as National Economic

Empowerment and Development Strategy (NEEDS), The Millennium Development

Goals (MDG's) e.t.c. In view of the above programmes, population growth rate in Nigeria

is higher than the growth of means of sustenance. Thus excess demand over supply of

basic necessities such as food, housing among others has resulted in rise in the prices of

these basic necessities.

Politicizing of Wage Increases: In a bid to score political points, several Nigerian governments

at both the Federal and State levels, often make public pronouncements about wage increases

(some of which were never implemented). This has a double impact on price rise. As soon as
 15

pronouncement is made, businessmen quickly adjust prices to take advantage of the increase then

after implementation; increased demand by affected workers gives another boost to prices.

2.1.1.2 Effects of Inflation in Nigeria

Inflation affects different people or economic agents differently. Broadly, there are two

economic groups in every society, the fixed income group and the flexible income group.

During inflation, those in the first group lose while those in the second group gain. The reason is

that the price movement of different goods and services are not uniform.

During inflation, most prices rise, but the rate of increase of individuals prices differ. Prices of

some goods and services rise faster than others while some may even remain unchanged. The

poor and the middle classes suffer because their wages and salaries are more or less fixed but the

prices of commodities continue to rise. On the other hand, the businessmen, industrialists,

traders, real estate holders, speculators and others with variable incomes gain during rising

prices. The latter category of persons becomes rich at the cost of the former group. There is

transfer of income and wealth from the poor to the rich.

More generally, which income group of the society gains or loses from inflation depends on who

anticipates inflation and who does not. Those who correctly anticipate inflation can adjust their

present earnings, buying, borrowing and lending activities against the loss of income and wealth

as a result of inflation.

To further determine the effect of inflation on individuals, it will be necessary to discuss the

effect of inflation on different groups.

a) Creditors and Debtors: When there is inflation, creditors are generally worse off because,

the real value of their future claims is reduced to the extent of the rate of inflation. On the other
 16

hand, when inflation occurs, debtors tend to pay less in real terms than they had borrowed.

Therefore, it could be said that inflation favours debtors at the detriment of creditors.

b) Salaried Persons: Those with white-collar jobs lose during inflation because their salaries are

slow to adjust when prices are rising.

c) Wage Earners: Wage earners may gain or lose depending on the speed with which their

wages adjust to rising prices. If their union is strong, they may get their wages linked to the cost

of living index. In this way, they may be able to protect themselves from the negative effects of

inflation. Most often in real life there is a time lag between the rise in the wages of employees

and the rise in price.

d) Fixed Income Group: These are recipients of transfer payments such as pensions,

unemployment insurance, social security, etc. Recipients of interest and rent also live on fixed

incomes. These people lose because they receive fixed payments while the value of money

continues to fall with rising prices.

e) Equity Holders and Investors: These group of people gain during inflation as the rising

prices expand the business activities of the companies and, consequently, increase profit. Thus,

dividends on equities also increase.

However, those who invest in debentures, bonds, etc, which carry fixed interest rates, lose during

inflation because, they receive fixed sum while purchasing power is falling.

f) Businessmen: Producers, traders, and real estate holders gain during periods of rising prices.

On the contrary, their costs do not rise to the extent of the rise in prices of their goods. When

prices rise, the value of the producer’s inventories rise in the same proportion. The same goes for

traders in the short run. The holders of real estates also make profit during inflation because the

prices of landed property increase much faster than the general price level.
 17

However, business decisions are difficult in an environment of unstable price. In the long-run,

there could be an increase in wages which will reduce profit thereby, having an adverse effect on

future investment.

g) Agriculturalists: Agriculturalists are of three types, namely, landlords, peasant proprietors

and landless agricultural workers. Landlords lose during rising prices because they get fixed

rents. Peasant proprietors who own and cultivate their farms gain. Prices of farm products

increase more than the cost of production.

Prices of inputs and land revenue do not rise to the same extent as the rise in the prices of farm

products. On the other hand, the wages of the landless agricultural workers are not raised by the

farm owners, because trade unionism is absent among them. But the prices of consumer goods

rise rapidly. So landless agricultural workers are losers.

h) Government: Inflation will have both positive and negative effects on the government. The

government as a debtor gains at the expense of households who are its principal creditors. This is

because interest rates on government bonds are fixed and are not raised to offset expected rise in

prices. The government in turn levies less tax to service and retire its debt. With inflation, even

the real value of taxes is reduced. Inflation helps the government in financing its activities

through inflationary finance. As the money income of people increases, government collects that

in the form of taxes on incomes and commodities. So the revenue of the government increases

during rising prices.

 18

2.1.1.3 Control of Inflation

How inflation is controlled in an economy depends on the causes and the type of inflation the

economy is experiencing

1. Use of Fiscal Policy

Fiscal policy is one of the two main macroeconomic policies used to control aggregate demand

and thereby achieve economic stability. Fiscal measures relate to taxation. Government

expenditure and public debt management, which seek to influence the level of aggregate demand

in an economy.

There are three main tools of fiscal policy viz. government spending (G), the income tax rate (t)

and government transfer payments (Tr). In times of demand pull inflation these tools are used to

reduce aggregate demand. An increase in tax rate, decrease in government expenditure and

decline in government transfer payment will reduce aggregate expenditure in the economy.

2. Use of Monetary Policy

Monetary policy is that part of macroeconomic policy which regulates the changes in money

supply in order to maintain price stability.

Tools of monetary policy are changing discount rate (d); changing required ratio (rr) and open

market operations (OMO). Increased required reserve ratio (rr) reduces the extent to which

commercial banks create credit hence reduces money supply. When the discount rate is increased

short term interest rates increased and this discourages borrowing to finance investment

spending. This invariably reduces aggregate demand. Central bank selling of its own government

securities to the general public reduces money supply which reduces aggregate demand.

3. Control measures

These measures may take the form of wage freeze, linking wage increases to increase in

productivity. Price controls may also be used. Maximum prices are used in this case. These
 19

prices are the highest possible legal prices for scarce goods. However, these prices may lead to

queues rationing and black marketing in scarce products.

4. Supply Side Policies.

In addition to the demand management policies, supply side policies could also be used in

controlling inflation. This however is a long – term measure. The following may increase

aggregate supply: increasing productivity in all sectors of the economy. Increases in productivity

may increase output, which will subsequently increase supply. This may be achieved by the

retraining labour, improving technology, removing all structural rigidities e.g. land tenure

system, poor road infrastructure etc.

2.1.2 Concept of Economic Growth

To an economist, economic growth is the sustained increase in the National Income (NI) or the

total output of all goods and services produced in an economy. It is an increase in the capacity of

an economy to produce goods and services, compared from one period of time to another.

Economic growth is defined as the expansion in a nations real output or it can be define as the

expansion in a nations capability to produce goods and services its people want. Economic

growth also refers to an increase in real aggregate output (real GDP) reflected in increased real

per capital income. The rate of economic growth is measured as the percentage increase in real

GDP overtime. Economic growth can equally be defined as increase in a nation‘s output which is

identifiable by sustainable increase in real per capita income.

Kuznets (1973), a Nobel laureate in economics, defined a country’s economic growth as “a long-

term rise in capacity to supply increasingly diverse economic goods to its population, this

growing capacity based on advancing technology and the institutional and ideological
 20

adjustments that it demands”. This means that for an economy to achieve growth there should be

advancement in technology accompanied by institutional and attitudinal adjustments.

Economic growth according to Todaro and Smith (2006), is the steady process by which the

productive capacity of the economy is increased over time to bring about rising levels of national

output and income.

Economic growth therefore occurs whenever people take resources and efficiently rearrange

them in ways that make them more productive overtime Paul Romer, (2007). It is the continuous

improvement in the capacity to satisfy the demand for goods and services, resulting from

increased production scale, and improved productivity i.e. innovations in products and processes.

According to Ominde and Ejiogu, (1972), economic growth includes the rate of which new

investment and new resources could be brought into productive use by the population.

Stanlake and Grant, (1995) defines economic growth as any increase in the Gross National

Product (GNP) or Gross Domestic Product (GDP), but for several reasons this is a rather

misleading use of the term. In “Economics” “economic growth” or economic growth theory”

typically refers to growth of potential output i.e. production at full employment” rather than

growth of aggregate demand (David Begg, 1994). “Balanced growth” is an important

consideration for a community or region. If the economy of a community or region depends

heavily on one industry, it will feel the effects of the peaks and valley of the business cycle of the

industry. By encouraging the industry to expand into a number of different geographic markets

or attracting different industries into the community or regions, the “boom” and “bust” cycle can

be managed. Through this diversification, the impact of a single event in one market or industry

on the local economy can be made less dramatic.

 21

2.2 Theoretical Literature.

For the purpose of understanding the impact of inflation on economic growth in Nigeria, it will

be very important to first understand the theoretical underpinning.

2.2.1 Theories of Inflation and Economic Growth

Gokal and Hanif (2004) identified the following theories that explain the relationship between

inflation and economic growth.

2.2.1.1 Neo- Classical Theory

The Neo-classical theory is based on the idea by Mundell (1963) which outlines that there is a

linkage between economic growth and inflation. Mundell asserts that changes in inflation or

inflationary expectations have an effect on wealth. An increase in inflation is thus said to reduce

wealth through a decrease in the rate of return. Mundell posit that the need to acquire more assets

causes people to save and in the process the prices of assets rise as their demand increases

causing interest rates to fall. However, the higher the savings available the higher the level of

capital accumulation and thus a high level of growth.

Tobin (1965) made improvements to the Neo-classical theory to come up with what is known as

the “Tobin Effect‟. This model outlines that consumers postpone current consumption by either

investing in capital or holding money. Thus individuals are assumed to hold money for either

speculative or precautionary motives.

The model suggests that as people switch from money they switch to capital which causes an

increase in capital stock which causes the steady state to increase as well. The increase in output

is temporary as the economy is assumed to be in a period of adjustment or going through a

transition. The changes caused by inflation on capital accumulation and economic growth are

termed the „lazy dog effect‟ were it causes were both capital accumulation and economic growth
 22

will rise but will fall when the rate of return on capital starts to decline. Inflation is thus said to

be having an upward effect on economic growth through capital accumulation.

However, recent models have exhibited that a bilateral association between inflation and

economic growth can also exist. For instance, Stockman (1991) argued that inflation causes the

steady state to decline. This is based on the notion that inflation erodes the purchasing power of

both capital and consumer goods so individuals will cut down their purchases and as a result the

level of the steady state falls. It can also be noted that inflation negatively impacts the labor

returns and cause individuals to substitute leisure for consumption. The marginal rate of return of

labor falls in line with the increase in inflation and both the level of the steady state and capital

returns will decline.

The neoclassical models can produce models with different theoretical results about the

association between economic growth and inflation. For instance, the „Stockman Effect‟

contends that an upswing in inflation causes output to fall. Other assertions argue that output will

not change while the „Tobin Effect‟ contends that output will increase. These differences can

cause researchers to adopt different approaches which may make it difficult to compare or apply

study results.

2.2.1.2 Keynesian Theory

This theory is posits that the association between economic growth and inflation can be analyzed

using aggregate supply and aggregate demand curves. It is based on the idea that the short run

aggregate supply curve is upward sloping and hence changes in demand will only cause a change

in prices. Thus shifts in the AS curve will effect changes in both output and prices (Dornbusch et

al., 1997). This applies in the short run period because output and inflation are determined by a

lot of factors such as monetary and or fiscal policy, changes in labor force and expectations.
 23

The model assumes that as the economy enters the long run „steady state‟, factors such as

monetary and or fiscal policy, changes in labor force and expectations will have a balancing

effect. The steady state thus implies that there are no changes but adjustments in the AS and AD

curves result in what is known as the „adjustment path‟. The model further asserts that there will

be a positive relationship between inflation and economic growth in the „adjustment path‟ or

during the „adjustment period‟. A negative relationship can thus be only witnessed after the

„adjustment period‟ or „adjustment path‟.

The positive relationship between inflation and economic growth is as a result of time

inconsistency. This means that producers will be perceiving that their prices are higher than

those of other producers in the economy and yet all prices have gone up causing them to

continue to produce more output. The positive relationship between inflation and economic

growth can be attributed to market agreements between suppliers and consumers to supply goods

at a later date. Thus changes in prices of goods will not cause a change in output since the

supplier has to supply the agreed quantity of goods at the agreed price (Blanchard and Kiyotaki,

1987).

It can also be ascertained that during the „adjustment period‟ the bilateral relationship between

inflation and economic growth is termed stagflation. This is a situation which occurs when prices

rise but causing output to either fall or remain the same. The model also suggests that during the

„adjustment period‟ inflation does not necessarily increase but follows an „adjustment path‟ of

temporal increase and then it falls.

 24

2.2.1.3 Monetarism Theory

Monetarism is an idea developed by Milton Friedman and centers on long run supply. Friedman

contends that there are long run supply elements that can be used to relate money supply to

growth (Gomme, 1993). For instance, the Quantity Theory of Money establishes a linkage

between economic growth and inflation by equating the total amount of money in circulation to

the economy‟s total spending. According to Friedman, the equation can be specified as follows

MV = PY

Where M = money stock in circulation

V = velocity of circulation

P = price

Y = output

Using the above equation the inflation rate can be determined as follows; p = v + m –y

Where p = inflation rate

V=velocity of circulation m= money stock

y = output growth rate

From this equation, Friedman postulated that was aggravated by increases in velocity and supply

of money that are greater than the prevailing level of economic growth.

He further argues that the effects of inflation on economic growth dependent on whether

inflation is anticipated or not. Anticipated inflation causes consumers to adjust their patterns of

consumption and lobby for wages increases such that the increase in inflation might match the

increase in wages (Gomme, 1993). When this is the case an increase in inflation will have no

effect on either employment or growth and this condition is known as neutrality of money. In
 25

this case inflation can be said to be harmless. It can therefore be deduced from the monetarism

approach that money growth affects long run prices and not growth and that inflation occurs as a

result of money supply being higher than the level of economic growth.

2.3 Empirical Literature

Over the years, different scholars have researched on topics that relate to this study, some

domestic and some foreign. We shall look into some of them below.

Hossain et al. (2012) examined the inflation and economic growth in Bangladesh. The study

used time series data from 1978to 2010. The objective of the study was to find out the long run

relationship between inflation and economic growth. The variables used in the study include

GDP deflator (GDPD) as a proxy for inflation and GDP as a proxy for economic growth. The

study employed co-integration and granger causality test. The Johansen –Juselius co-integration

result shows that there was no co-integration between inflation and economic growth in

Bangladesh. The result of causality at lag two (2) shows unidirectional causality was seen

running from inflation to economic growth. Further test at lag four (4) supported the first by

showing unidirectional causality running from inflation to economic growth.

The major limitation of this study is that it failed to capture necessary post estimation to

determine the robustness of the model.

Ziaur (2013) examined relationship between inflation and economic growth in Bangladesh. The

study used time series data between1976 to 2011. The objective was to investigate the empirical

relationship between inflation and economic growth in Bangladesh. The variables include GDP

growth (GDPgr), inflation, trade openness and remittance growth. The study used several

econometric techniques which includes unit root test, stationary test, co-integrated test, VAR

model, VAR Granger Causality test, impulse response function and variance decomposition of

error term. The result shows statistical significant negative relationship between inflation and
 26

economic growth in Bangladesh. The negative relationship between economic growth and

inflation is in line with the finding of Ferdinand and Isadora (2014), and Inyiama (2013).

Max and Mark (2008) examined the effect of inflation on Growth using panel of transition

countries. The study used panel data evidence for 13 transition countries from 1990 to 2013. The

objective was to examine the effect of inflation on growth in transition countries. The variables

used in the study include. Real GDP Growth in local currency units (LCU) Natural log of GDP

deflator percentage growth rate ln (π), mz/GDP: income normalized money demand (money

demand), product of normalized money demand and in (inflation rate) i.e. Ln(π)1 (money

demand), Czech/other (real GDP, Czech republic/real GDP, other country) l/GDP i.e.

investment/GDP at market prices each in LCU, PopGr (population Growth Rate: each in LCU).

The result shows a negative influence of inflation on economic growth. This finding is in line

with Rahman (2013). The major limitation of this study is that the time frame is not sufficient to

give better analysis.

Manoel (2010) investigated relationship between inflation and economic growth in Latin

America. The study used panel data from 1970 to 2007 for four Latin American countries,

namely Argentina, Bolivia, Brazil and Peru. The objective of this study was to investigate the

role of macroeconomic performance, in terms of inflation rates, in determining economic growth

in panel of Latin American countries that experience hyperinflation episode in the1980s and

early 1990s. The variables used in the study include GROW which is the growth rate of real

GDPs and it serves as a dependent variable. While the independent variables include: inflation

(INFLAT), government’s share in the real GDP (Gov), which proxies for the size of government,

the ratio of exports and imports to real GDP (OPEN), as proxy for economic openness, the ratio
 27

of investment to real GDP (INN), measure of financial development i.e. the ratio of liquid

liabilities to GDP (Mz), index of structural development (DEV) which is measured by the level

of education and urbanization, political regime (POL)which consist of common factors of

DEMOC, XCONST and POLCOMP. The study employed pooled ordinary least square, fixed

effect (FE) and random coefficient estimators (RC). The result shows a significant negative

relationship between inflation and economic growth. This is in line with the finding of Rahman

(2013). The major limitation of this study is inability to carry out necessary post estimation test

to determine the robustness of the model.

Jaganath (2014) examined the impact of inflation on economic growth in six South Asian

countries. The study used time series data for the period 1980 to 2012. The main objective of this

study was to investigate the impact of inflation on economic growth in six South Asian countries.

The variables include GDP as an indicator of economic growth and CPI as a proxy for inflation.

The study used co-integrated test and error correction mechanism, causality test and unrestricted

VAR model. It also employed correlation analysis. The result shows that there is high positive

correction between inflation and economic growth for all the countries in this study. The co-

integration result suggests that there is long run relationship exist for Malaysia. However, the rest

of the countries have no long run relationship between inflation and economic growth.

The Granger causality result shows that there is unidirectional causality run from GDP to CPI for

Bangladesh, Bhutan, and India. It also shows unidirectional causality run from CPI to GDP in the

context of Nepal. However, there is no causality between GDP and CPI for Maldives and Sri

Lanka. The use of correlation does not really explain the effect of inflation on economic growth,

rather a regression analysis would have been used, the time frame for this study is not sufficient

to give better analysis.

 28

Ferdinand and Isidore (2014) examined short – run and long run inflation and Economic Growth

nexus in Ghana using quarterly data from 1986Q1 to 2012Q4. The major objective was to

examine the link between inflation and economic growth in Ghana. The variables include

economic growth (y) as a dependent variable, while the independent variables include stock of

labour (L), stock of capital, (K), government expenditure (GEXP), interest rate (INT), money

supply (M2) and consumer price index (CPI). The study employed co -integration and error

correction mechanism. The result shows a negative relationship between economic growth and

inflation. Interest was also found to have negative impact on economic growth. The granger

causality test shows there was no causation between economic growth and inflation. The result

of negative relationship between inflation and economic growth and also no causality between

inflation and economic growth are in line with Kasidi and Mwakanmela (2013). The major

limitation of this study is that it fails to carry out post estimation test in order to determine the

robustness of the model.

Kasidi and Mwakanmela (2013) examine the impact of inflation on economic growth in

Tanzania using annual time series data for the period of 1990 to 2011. The objectives of the

study were to examine the impact of inflation on economic growth in Tanzania, to measure the

degree of responsive of economic growth in Tanzania to change in general price level and to

establish relationship between inflation and economic growth in Tanzania. The variables used in

the study include GDP which served as a dependent while inflation served as independent

variable. The study used reduced form regression equation to investigate the impact of inflation

on economic growth. Co-integration was applied to measure whether the two variables moved

together in the long run. The result from regression analysis revealed that inflation has negative

impact on economic growth in Tanzania. Correlation coefficient and co-integration test using
 29

Johansen Co-integrating relationship between inflation and economic growth shows that there is

no significant long-run relationship between inflation and economic growth in Tanzania. Only

short term negative statistical significant. The negative relationship between inflation and

economic growth is in line with the finding of Inyiama (2013). The limitations of this study are:

the period covered by this study is not enough to give better analysis. The study also failed to

establish the causation between inflation and economic growth.

Aminu and Anono (2012) carried out empirical analysis of the effect of inflation on the Growth

and Development of the Nigerian economy. The study used time series data from 1970 to 2010.

The objective was to investigate the impact of inflation on economic growth and development in

Nigeria. The variables used in the study include GDP which is the Gross Domestic product

(output) and also serves as a dependent variable while inflation serves as independent variable.

The study used Augmented Dickey – fuller technique in testing the unit root property of the

series and Granger causality test of causation between GDP and inflation. The result shows that

inflation is statistically insignificant and positive. The positive impact of inflation on economic

growth in Nigeria is in line with the finding of Olu and Idih (2015). The result of causality

suggests that GDP causes inflation and not inflation causing GDP. The unidirectional causation

of GDP causing inflation is in contrast with other finding, such as Inyiama (2013). The model

was not robust as autocorrelation was visible due to very low Durbin-Watson statistic of 0.031.

The unit root test shows that the variables are I (1), this means loss of long run information. The

right model for this study is Johansen co integration test.

Inyiama (2013) investigated whether inflation weakens Economic Growth? Using evidence from

Nigeria from 1970 to 2010. The objective was to evaluate the link between inflation and

Economic Growth in Nigeria. It also examined the nature and form of association between
 30

inflation rate and exchange rate as well as interest rate. The variables used are GDP, inflation,

interest rate and exchange rate. Ordinary least square approach in the form of multiple

regressions was adopted in examining the relationship among the variables while causality was

evaluated using Granger causality test. Johansen Co-integrated test was also adopted to check

whether short term relationship would be maintained in the long run. It was found that inflation

is negatively related with the real GDP. This is sustained even in the long run. On causality, at

both lag 2 and lag 4, the study revealed that there is no causality between inflation rate and real

GDP.

Oladipo et al. (2015), examined the inflation, interest rate and economic growth in Nigeria using

annual time series data from 1981 to 2014. The variables used for this study includes Real Gross

Domestic Product (RGDP), Inflation at consumer prices, Interest Rate (INTR), Net Domestic

Credit (NDC), Transfer Payment (TRF). This used Augmented Dickey Fuller test to test the unit

root properties of the series. The result of the unit root shows that all the variables are stationary

at first difference but inflation is stationary at level. The study adopts the Ordinary Least Square

(OLS) method. The long run relationship among the variables was tested using Johansen co

integration test and causality test was also carried out. The OLS result shows that both inflation

and interest rate have negative impact on the economic growth. Johansen co integration shows

that there is long run relationship among the variables under consideration. The Granger

causality test shows that both inflation and interest rate do not Granger cause the economic

growth in Nigeria. The limitations of this study: It did not carry out post estimation test to

ascertain the robustness of the model, Johansen co integration test used to test long run

relationship is not the appropriate model for I (0) and I (1). The right model for this is

Autoregressive Distributive Lag (ARDL).

 31

Bakare, Kareem and Oyelekan (2015), examined the effects of inflation rate on economic growth

in Nigeria (1986-2014). The variables used for this study are: Gross Domestic Product (GDP) as

a dependent Variable and inflation rate as an independent variable. The Augmented Dickey

Fuller unit root test was used to test the stationarity of the variables. The study used regression

analysis to determine the effect of inflation on economic growth, while Granger causality test

was used to test the causation between inflation and economic growth. The result shows that

inflation has negative impact on the economic growth. The Granger causality shows that GDP

cause inflation but inflation does not cause GDP. The major limitation of this study is that the

variables were differenced which leads to loss of long run inflation but this study did not

consider long run relationship.

Olu and Idih (2015), investigated the nature of the relationship between inflation and economic

growth in Nigeria using annual time series data from 1980 to 2013. The variables used for the

study are Gross Domestic Product (GDP) as a dependent variable, while the independent

variables are: Inflation rate, Exchange Rate (EXCHR), input of labour and Capital. The study

used the Ordinary Least Square to capture the impact of the dependent variable on the

independent variables. The result shows that inflation has positive impact on the economic

growth in Nigeria. The positive impact of inflation on economic growth is in line with the

finding of Aminu and Anono (2012). The major limitation of this study is that it fails to test unit

root properties of the series.

Emerenini and Eke (2014) investigated the determinants of inflation in Nigeria using a monthly

data from January 2007 to August 2014. The ordinary least square (OLS) method was adopted

because of its best linear unbiased estimator (BLUE) property. The result showed that expected

inflation, exchange rate and money supply influenced inflation, while annual treasury bill rate
 32

and monetary policy rate though rightly signed did not influence inflation in Nigeria within the

period under investigation. The estimated model displayed that all the explanatory variables used

for the analysis accounted for 90% variation in explaining the direction of inflation as regards to

increase or decrease. The co-integration test showed that a long term relationship existed among

the variables and they were stationary at order one.

Enu and Havi (2014) studied the macroeconomic determinants of inflation in Ghana using a

cointegration approach. The found out that in the long run, population growth and service output

affect inflation positively. However, foreign direct investment, foreign aid and agricultural

output increase inflation impact negatively. Also, in the short run, the past two years inflation

had a significant impact on the current inflation. The population growth and foreign direct

investments past records had both positive and negative impact on current inflation.

Nevertheless, they were not significant.

Maku and Adelowokan (2013) in their work using annual data from 1970 to 2011 examined the

determinants of inflation in Nigeria by employing the partial adjustment model. The result

indicated that fiscal deficit and interest rate exert decelerating pressure on dynamics of inflation

rate in Nigeria. While, other macroeconomic indicators such as real output growth rate, broad

money supply growth rate, and previous level of inflation rate further exert increasing pressure

on inflation rate in Nigeria. The real output growth and fiscal deficit were found to be significant

determinants of inflation rate in Nigeria during the period.

Iya and Aminu (2014) investigated the determinants of inflation in Nigeria between 1980 and

2012 using the ordinary least square method. The result revealed that money supply and interest

rate influenced inflation positively, while government expenditure and exchange rate influenced

inflation negatively. They suggested that for a good performance of the economy in terms of
 33

price stability may be achieved by reducing money supply and interest rate and also increase

government expenditure and exchange rate in the country.

Hossain and Islam (2013) examined the determinants of inflation using data from 1990 to 2010

in Bangladesh with the ordinary least square method. The empirical result showed that money

supply, one year lagged value of interest rate positively and significantly affect inflation. The

result also indicated that one year lagged value of money supply and one year lagged value of

fiscal deficit significantly and negatively influence over inflation rate. There was an insignificant

relationship between interest, fiscal deficit and nominal exchange rate. The explanatory variables

accounted for 87 percent of the variation of inflation in during the period.

Moses, et al., (2015) examines monetary growth and inflation dynamics in Nigeria. The

motivation for the study is derived from the perceived weakening relationship between money

and inflation in recent times. The methodology was a Vector Auto regressive (VAR) model.

Three variants of OLS - ordinary least square, fully modify OLS, and dynamic OLS – techniques

were used in estimating the data. Results from the study revealed that that the coefficients of

money supply were positive and significant at 1, 5, and 10 per cent, respectively in the inflation

equation for the full sample period, suggesting that money supply bears a long run positive

relationship with inflation. Based on the coefficient stability results obtained from the Chow test,

the entire sample was divided into two sub samples with the first one covering the period 1982q1

to 1996q4 while the second sub sample covered the period 1996q1 to 2012q4. The equation was

re-estimated for the two sub-samples.

 34

CHAPTER THREE

RESEARCH METHODOLOGY

3.1 Introduction

This section takes care of the method used to source for data, the model specification, and the

statistical instrument used to analyze the data obtained.

3.1 Research Design

The research design adopted for this work is ex-post facto research design. Ex post facto design

is a quasi-experimental study examining how an independent variable presents prior to the study

affects a dependent variable. This design ideally fits this work as it is not possible or permissible

to manipulate the characteristics of the variables under study. Simon and Goes (2013) see ex post

facto research as one which is based on a fact or event that has already happened and at the same

time employs the investigation and basic logic of enquiring like the experimental method.

The choice of the ex post facto method is justified by the fact that the cost of collecting data is

much lower than in the experiment type.

3.2 Data Analysis Techniques

The ordinary least square (OLS) technique is employed in obtaining the numerical estimates of

the coefficient parameters, the OLS is chosen because of its BLUE (best linear unbiased

estimator) properties, according to (Gujarati 2009).

This method has some very attractive statistical properties that have made it one of the best and

most powerful methods of regression. It is intuitively appealing and mathematically much

simpler than any other econometric techniques. E-views 9 software package will be employed in
 35

this analysis to test non violation of the basic assumption of the OLS model. And the numerical

data to be used will be gotten from Central Bank of Nigeria Statistical Bulletin, 2018 edition.

3.3 Model Specification

The model is specified as directed by Gujarati and Porter (2009). The model is specified as:

Functional form of the Model

RGDP = F( INFR, INTR, EXR, TSV) (3.1)

Mathematical form of the model:
RGDPt =β0 −β 1 INFR t −β2 INTRt + β 3 TSV t + β 4 EXRt (3.2)

Econometric specification of the model:

RGDPt =β0 −β 1 INFR t −β2 INTRt + β 3 TSV t + β 4 EXRt + μt (3.3)

Where

RGDP = Real gross domestic product

INFR = Inflation Rate

EXR= Exchange rate

TSV= Total Savings

INTR= Interest rate.

μ1 t = Stochastic Error term

β 0 … β 3.= Parameters

t = Times series analysis

 36

3.4 A Priori Expectation

This has to do with the sign expectation set by economic theory. The economic a priori

expectation test evaluates the parameters whether they meet standard economic theory, this

include the sign and sizes.

Table 1 A priori expectation

Parameters Variables Expected Signs

β0 CONSTANT +
β1 INFR -
β2 INTR -
β3 TSV +
β4 EXR +
3.5 Econometric Test

3.5.1 Pre-Estimation Test

In this study, some pre-estimation tests will be carried out to ensure that a valid result will be

obtained. This will be done to test the time series properties of the data intended to be used for

the estimations since time series data often lead to spurious result if the problem of stationarity is

left unaddressed. Besides it is necessary to check the long run relationship among the time series

variables used before performing the actual estimations.

3.5.1.1 Stationarity Test

 37

Most macroeconomic time series data are often not stationary at level in view of this, the

variables of interest are tested for stationarity using Augmented Dickey- Fuller (ADF) unit root

test. According to Gujarati et. al (2012), the test involves the estimation of the following model:
m

β β
∑
ΔYt = 1 + 2 + δYt-1+ i=1 αiΔYt-i + εt

where

Y is a time series, t is a linear time trend, Δ is the first difference operator, βs are parameters, m is

the optimum number of lags in the dependent variable and εt is error term.

If ADF statistic > ADF critical, we reject the Null hypothesis that the variable is non-stationary

at the chosen level of significance, but if otherwise, we do not reject the null and conclude that

the variable is stationary.

3.5.1.2 Co-integration Test

The co-integration test is conducted to test whether there is a long-run equilibrium relationship

among the variables in the model. The Johansen co-integration test will be used in this study. If

the model is co-integrated, the vector Error correction mechanism will be carried out to ascertain

the speed of adjustment of the dependent variable when there is a change in the independent

variables. If not co-integrated Vector Autoregression model will be used. The co-integration

model is specified as follows;

Δȗt =α0 + α1ut-1 + ut

Decision Rule: If the ADF test statistic is greater than the critical value at 5%, then the variables

are co-integrated (values are checked in absolute term).

3.5.1.3 Granger Causality Test

 38

Granger Causality Test is conducted to establish the direction of causality among the variables of

in the model and to investigate whether there is a degree of causation of one variable on the

other. Engle and Granger (1987) noted that if two variables are cointegrated, the possibility of

causality between the two exists, at least in one direction. Granger causality test for the series is

expressed in general form as:

k k
Yt = ∑ δ 11iYt-1 + ∑ δ 12iYt-1 + U1t
i=1 i=1

k k
Xt = ∑ δ 21iYt-1 + ∑ δ 22iYt-1 + U2t
i=1 i=1

where Y = dependent variable, X = independent variables in the model, t = the current period of

the variables and t-i = the lagged period of the variables, δ 11 to δ 22 = the coefficients of the lagged

variables and U1 and U2 = mutually uncorrelated white noise error terms. The Granger causality

analysis decision rule follows F-distribution. Reject null hypothesis if the p(F-statistic) < 0.05;

otherwise accept.

3.5.1.4 Vector Error Correction Mechanism

If there exist a long run relationship (co-integration) among the time series variables, the Vector

Error correction mechanism will be estimated to know the rate at which the dependent variable

returns to equilibrium to the independent variable after some levels of variations i.e to derive the

numerical value of the magnitude of the short run dynamics or disequilibrium. The model is as

below,

ΔYt= b0 + b1ΔXt + b2ΔXt-1 + μt

Decision Rule
 39

In conducting VECM, the expected sign of the result should be negative. A positive VECM

implies a model misspecification or an indication of structural changes and will not give us the

rate of these change in the dependent and independent variables

3.5.2 Post Estimation Test

3.5.2.1 Multicollinearity Test

In this study, the test for linear relationship among the variables used in the model would be

performed. This is in line with Assumption of the Classical Linear Regression Model (CLRM) of

“no high or perfect multi-collinearity”. The essence of this is to see if there is high collinearity

among variables or not. The correlation matrix will be used for this test. If the correlation

coefficient between two variables exceeds 0.8, then such variables have high multi- collinearity.

3.5.2.2 Normality Test

This is used to test whether the error term is normally distributed. The normality test adopted in

this work is that of Jarque-Bera (JB) statistic which follows the Chi-square distribution. If JB

statistic < JB critical value we do not reject the null hypothesis that the error term is normally

distributed at the chosen level of significance, but if otherwise, we reject. We can as well use its

probability value to judge the result.

3.5.2.3 Heteroscedasticity Test

One of the assumptions of the random variable U t is that its probability distribution should be

constant over all observations of Xi, that is, the variance of each disturbance term is the same for

all values of the explanatory variables. The aim of this test is to see whether the error variance of

each observation is constant or not. Non-constant variance can cause estimated model to yield a

biased result. White’s general heteroscedasticity test would be adopted for this purpose (Gujarati

et.al. 2012).
 40

3.5.2.4 Autocorrelation Test

This is used to test whether the error terms corresponding to different observations are

uncorrelated. The Durbin Watson d-Static will be used. It is based on the assumption that the

model includes the intercept term and that the explanatory variables are non – stochastic. The

Durbin Watson values usually range from 0 to 4. The Durbin Watson d distribution table is used

to verify the exact value of the Durbin Watson tabulated.

3.6 Sources of Data

Annual time series data on the variables under study covering the period 1981-2018 are used in

this study for the estimation of the function. The data used for this study are obtained from

Central Bank of Nigeria 2018 Statistical Bulletin.

 41

CHAPTER FOUR

DATA PRESENTATION AND ANALYSIS

4.0 Introduction

This chapter presents the regression results and interprets the various economic, statistical and

econometric tests in study. The hypotheses posed in the work are also examined based on the

empirical results.

4.1 Presentation and Analyses of Data

4.1.1 Unit Root Test Result

The result of the Augmented Dickey-Fuller (ADF) Unit Root Test to check the stationarity of the

variables in the model is presented in table 4.1 below. The result showed that variables such as

(INFR, INTR, and TSV) are stationary at level form, while, RGDP, and EXR are stationary at

first difference.

Table 4.1: Result of Unit Root Test

t-statistic ADF Test Order of Test Result

Variables 5% Level Statistic Integration

RGDP -2.948404 -5.298095 I(1) Stationary at 1st difference

INFR -2.943427 -3.225648 I(0) Stationary at level form

INTR -2.943427 -3.573088 I(0) Stationary at level form

EXR 2.948404 -3.303324 I(1) Stationary at 1st difference
TSV 25.31753 -2.971853 I(0) Stationary at level form

The test was conducted on the basis of the following hypothesis:

HO: variable contains unit root and therefore is non-stationary.

H1: variable does not contain unit root and hence is stationary.
 42

Since all the variables are not stationary at level, a co integration test was conducted to find out if

the variables have a long run relationship, that is, whether or not the variables are co integrated.

4.1.2 Co-integration Test Result

The co-integration test was carried out using the Johansen co-integration test and under the
Johansen co-integration test, co-integration is said to exist if the values of the computed Eigen
values are significantly different from zero or if the trace statistics is greater than the critical
values at 5% level of significance.
Table 4.2 Johanson Co-integration Test
Date: 12/05/20 Time: 15:05
Sample (adjusted): 1983 2017
Included observations: 35 after adjustments
Trend assumption: Linear deterministic trend
Series: RGDP INFR INTR EXR TSV 
Lags interval (in first differences): 1 to 1

Unrestricted Cointegration Rank Test (Trace)

Hypothesized Trace 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None *  0.646569  80.44511  69.81889  0.0056

At most 1  0.514408  44.04276  47.85613  0.1091
At most 2  0.290865  18.75926  29.79707  0.5103
At most 3  0.132746  6.729414  15.49471  0.6093
At most 4  0.048624  1.744601  3.841466  0.1866

 Trace test indicates 1 cointegrating eqn(s) at the 0.05 level

 * denotes rejection of the hypothesis at the 0.05 level
 **MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None *  0.646569  36.40236  33.87687  0.0244

At most 1  0.514408  25.28350  27.58434  0.0958
At most 2  0.290865  12.02984  21.13162  0.5447
At most 3  0.132746  4.984813  14.26460  0.7438
At most 4  0.048624  1.744601  3.841466  0.1866

 Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level

 * denotes rejection of the hypothesis at the 0.05 level
 **MacKinnon-Haug-Michelis (1999) p-values

The result of the co-integration test as shown in table 4.2 above indicates one co-integration

equation. This is because the trace statistics is greater than the critical values at 5% level of
 43

significance in one of the hypothesized equations. Similarly, the computed Eigen value is

significantly different from zero in one of the hypothesized equations. Hence, one of the

hypothesized equations satisfies this condition and therefore, the null hypothesis of no co-

integration among the variables is rejected in at least one equation.

4.2 Presentation and Interpretation of Result

4.2.1 Vector Error Correction Model Result

Having satisfied the condition for long run equilibrium relationship as was revealed by the co-

integration which indicated one co-integrating equations, the next step is to construct an vector

error correction model (ECM) so as to estimate the short run relationship that exists among the

specified variables and equally the speed of adjustment having lost information about long run

relationship through differencing.

The VECM result is presented in table 4.3

 Vector Error Correction Estimates

 Date: 12/05/20 Time: 15:45
 Sample (adjusted): 1984 2017
 Included observations: 34 after adjustments
 Standard errors in ( ) & t-statistics in [ ]

Cointegrating Eq:  CointEq1

RGDP(-1)  1.000000

INFR(-1)  61.37320
 (42.1490)
[ 1.45610]

EXR(-1) -97.43913
 (16.4923)
[-5.90815]

INTR(-1) -441.8700
 (131.357)
[-3.36389]

TSV(-1) -9.709172
 (1.05099)
 44

[-9.23814]

C -11699.63

Error Correction: D(RGDP) D(INFR) D(EXR) D(INTR) D(TSV)

CointEq1 -0.793481 -0.001340  0.002544 -9.65E-05 -0.001116

 (0.15283)  (0.00137)  (0.00123)  (0.00028)  (0.00720)
[-5.19196] [-0.97507] [ 2.07486] [-0.34809] [-0.15498]

D(RGDP(-1)) -0.028397  0.000504 -0.002442 -0.001445 -0.007860

 (0.17831)  (0.00160)  (0.00143)  (0.00032)  (0.00840)
[-0.15926] [ 0.31415] [-1.70692] [-4.46874] [-0.93514]

D(RGDP(-2)) -0.471929 -0.000269  0.003659  0.000955  0.014163

 (0.42171)  (0.00379)  (0.00338)  (0.00076)  (0.01988)
[-1.11907] [-0.07079] [ 1.08159] [ 1.24835] [ 0.71250]

D(INFR(-1))  17.23151 -0.020648 -0.029106  0.078055 -0.252903

 (22.0244)  (0.19810)  (0.17668)  (0.03994)  (1.03811)
[ 0.78238] [-0.10423] [-0.16474] [ 1.95430] [-0.24362]

D(INFR(-2))  9.025752 -0.346429 -0.048026 -0.016693 -0.240892

 (23.6125)  (0.21238)  (0.18942)  (0.04282)  (1.11297)
[ 0.38225] [-1.63117] [-0.25354] [-0.38983] [-0.21644]

D(EXR(-1)) -83.74532  0.039472  0.332598 -0.035903  0.584931

 (32.6154)  (0.29336)  (0.26164)  (0.05915)  (1.53732)
[-2.56766] [ 0.13455] [ 1.27118] [-0.60702] [ 0.38049]

D(EXR(-2)) -59.97103 -0.163469  0.339867 -0.030510  0.659303

 (34.1852)  (0.30748)  (0.27424)  (0.06199)  (1.61131)
[-1.75430] [-0.53165] [ 1.23931] [-0.49215] [ 0.40917]

D(INTR(-1)) -203.5210 -0.028750  1.304636 -0.420096  2.329316

 (132.104)  (1.18820)  (1.05976)  (0.23957)  (6.22671)
[-1.54061] [-0.02420] [ 1.23107] [-1.75357] [ 0.37408]

D(INTR(-2)) -87.96133  0.878913 -0.112188 -0.047146 -0.561675

 (113.472)  (1.02061)  (0.91028)  (0.20578)  (5.34845)
[-0.77518] [ 0.86116] [-0.12324] [-0.22911] [-0.10502]

D(TSV(-1))  20.48322  0.015152 -0.039959  0.002522  0.319715

 (5.05623)  (0.04548)  (0.04056)  (0.00917)  (0.23832)
[ 4.05109] [ 0.33317] [-0.98513] [ 0.27507] [ 1.34151]

D(TSV(-2))  20.67963  0.036972 -0.042723  0.002401  0.405235

 (5.75368)  (0.05175)  (0.04616)  (0.01043)  (0.27120)
[ 3.59416] [ 0.71441] [-0.92561] [ 0.23015] [ 1.49424]

C  113.3546 -3.967503  7.381136  1.195474  5.225253

 (647.826)  (5.82683)  (5.19694)  (1.17481)  (30.5351)
[ 0.17498] [-0.68090] [ 1.42028] [ 1.01759] [ 0.17112]

 R-squared  0.665248  0.205957  0.395497  0.676295  0.474361

 Adj. R-squared  0.497872 -0.191065  0.093245  0.514443  0.211542
 Sum sq. resids  1.06E+08  8598.808  6840.226  349.5487  236141.5
 45

 S.E. equation  2198.031  19.77005  17.63291  3.986048  103.6036

 F-statistic  3.974573  0.518754  1.308501  4.178465  1.804893
 Log likelihood -302.4843 -142.3052 -138.4156 -87.85871 -198.6230
 Akaike AIC  18.49908  9.076778  8.847975  5.874042  12.38959
 Schwarz SC  19.03779  9.615493  9.386690  6.412757  12.92830
 Mean dependent  1284.156 -0.379412  8.972525  0.854853  79.27746
 S.D. dependent  3101.892  18.11507  18.51735  5.720345  116.6771

 Determinant resid covariance (dof adj.)  4.11E+16

 Determinant resid covariance  4.66E+15
 Log likelihood -854.5284
 Akaike information criterion  54.08991
 Schwarz criterion  57.00795

The VECM co-integration equation is negative and statistically significant as theoretically

expected. The coefficient of VECM co-integration equation is -0.793481 units.

This shows that the speed of vector adjustment between the short-run and long-run equilibrium is

approximately 79% annually. In other words, the system corrects its previous period

disequilibrium at a speed of 75% annually. With a negative sign and a statistically significant

VECM (-1) as shown by the probability value of -5.19196, it is obvious that the model has a

significant speed of adjustment.

4.2.2 Evaluation of regression results

4.2.2.1 Evaluation based on economic criteria

This subsection is concerned with evaluating the regression results based on a priori

expectations. The signs and magnitude of each variable coefficient was evaluated against

theoretical expectations.

Table 4.3 Result of A priori Test

 46

Variable Expected Sign Post-Test Sign Test Result

INFR -Ve -Ve CWES
EXR +Ve +Ve CWES
INTR -Ve -Ve CWES
TSV +Ve -Ve NCWES
CWES= conforms with the expected sign

NCWES= not conforms with the expected sign

The signs of the variables (INFR, INTR and EX) coefficient from the estimated model are in line

with a priori expectations or theoretical underpinnings total savings. Inflation rates, interests

rates, and total savings exert negative impact on economic growth in Nigeria over the period

covered, and not statistically significant. Exchange rate impacted positively on economic growth

for the same period under review.

Furthermore, the estimated constant term was found to be 113.3546 units, implying that the

model passes through the point 113.3546 mechanically and however, it also implies that the real

gross domestic product (economic growth) would increase by a constant amount of 113.3546 if

inflation rate, interest rate, exchange rate and total savings were equal to zero.

The estimated elasticity coefficient of inflation rate is -0.001340 was found to be negatively

related to gross domestic product and not statistically significant. This implies that if other

variables affecting economic growth are held constant, a unit increase in inflation rate would

bring about -0.001340 decreases in gross domestic product on the average and this is in

congruent with a priori expectation. That is, inflation rates have negative effect on economic

growth.

Similarly, the estimated elasticity coefficient of interest rate is -0.000096 was found to be

negative and indicating that interest rate has a negative relationship with gross domestic product.
 47

This also implies that if the interest rate goes up by 1%, on the average, the gross domestic

product would decrease by 0.000096 with all other variables affecting gross domestic product

held constant.

The estimated coefficient for exchange rate (EXR) 0.002544 units was found to be positively

related to gross domestic product which implies that if we hold all other variables affecting gross

domestic product (economic growth) constant, a unit increase in exchange rate will lead to a

0.002544 increase in economic growth on the average. And this is statistically significant for the

same period under review.

Furthermore, the estimated coefficient for total savings -0.01116 was also found to be negatively

related to gross domestic product which implies that if we hold all other variables affecting

economic growth constant, a unit increase in total savings will lead to a 0.01116 decrease in

economic growth on the average. Thus, gross domestic product is very responsive to changes in

total savings in Nigeria.

4.2.3 Evaluation Based on Statistical Criteria.

This subsection applies the R2, the t–test and the f–test to determine the statistical reliability of

the estimated parameters. These tests are performed as follows:

4.2.3.1 The coefficient of determination (R2):

From our regression result, the R2 is given as 0.665248. This implies that (100 x 0.665248) =

66.5 % of the variations in economic growth is being explained by the changes in inflation rates,
 48

interest rate, exchange rate and total savings. The remaining percent is attributed to other factors

affecting economic growth. That is the error term or the white noise.

4.2.3.2 Interpretation of the t-Test Result

The result of the t-test of significance is shown in table 4.5 below:

The t-test is used to measure the individual statistical significance of the explanatory variables,

for a two tailed test, we use (t a/2). We also employ the 95% confidence interval or 5% level of

significance (i.e.α=0.05) and 37 as our degree of freedom.

Decision Rule: Reject H0 if /tcal/ > tα/2 otherwise, accept.

α = 0.05/2 =t0.025

df = n-k = 37- 5= 32

Therefore, t(0.025, 33) = 2.037 (from the t - distribution table)

Table 4.6 of t-test

Variables t-calculated (tcal) t-tabulated (tα/2) Result

INFR -0.975071 2.037 Statistically insignificant

INTR -0.34809 2.037 Statistically insignificant

EXR 2.07486 2.037 Statistically significant

TSV -0.15498 2.037 Statistically insignificant

From the result above, the computed t-value for;

Inflation rate (INFR): (tcal > tα/2) i.e, (-0.975071< 2.037).Therefore we do not reject the null

hypothesis and reject the alternative hypothesis. Hence, inflation rate has insignificant impact on

economic growth in Nigeria.

 49

Interest rate (INT): the computed t-test is less than t-tabulated (-0.34809< 2.037).Therefore, we

accept the null hypothesis and reject the alternative hypothesis and conclude that interest rate has

no significant impact on economic growth in Nigeria.

Exchange rate (EXR): (tα/2 > tcal) that is, (2.07486< 2.037). Therefore, we accept the alternative

hypothesis and conclude that exchange rate has significant impact on economic growth in

Nigeria.

Also, total savings computed t-test is less than t-tabulated (-0.15498< 2.037). Hence, we accept

the null hypothesis and conclude that total savings has no significant impact on economic growth

in Nigeria.

4.2.3.3 Result and Interpretation of f–Test of Significance

The F-test of significance is used to measure the statistical significance of the entire regression

plane or the joint impact of the independent variables on the dependent variable. The degrees of

freedom for the numerator (v1) and for the denominator (v2) are given as k –1 and n–k where n is

the sample size and k is the number of parameters including the constant term.

If f* > f0.05 we will reject the Null hypothesis and accept the alternative. Otherwise, the

alternative hypothesis H1 will be rejected and the null hypothesis Ho be accepted.

V1 = 5 – 1= 4, V2 = 37 – 5= 32, df= (4, 32)

At 5% level of significance and df= (4, 32),

f0.05 = 2.69 and f* = 3.672810

From the result, the f* > f0.05 (3.672810 >2.69), we reject the null hypothesis and conclude that

the variables (INFR, INT, EXR and TSV), are significant on the entire regression plane.

4.2.4.1 Autocorrelation Test

 50

VEC Residual Serial Correlation LM Tests

Null Hypothesis: no serial correlation at lag
order h
Date: 12/05/20 Time: 16:10
Sample: 1981 2018
Included observations: 34

Lags LM-Stat Prob

1  19.91067  0.7515
2  12.37540  0.9833
3  28.64760  0.2789
4  19.70616  0.7621
5  25.14712  0.4542
6  40.12037  0.0283
7  26.42418  0.3852

Probs from chi-square with 25 df.

the result of VEC residual serial correlation LM tests above indicated that there is no presence of

auto-correlation in the model above. That is, the variables captures by the error term are not

correlated over the period under review.

4.2.5.2 Normality test result:

The normality test that is used for this study is the Jarque-Bera (JB) test of normality. This

follows a chi-square distribution with 4 degree of freedom (4df).

H0: δ1 = 0 (the random variables are normally distributed)

H1: δ1 ≠ 0 (the random variable are not normally distributed)

Decision rule:

Reject H0 if |JBcal| < |JBα (2df)| otherwise, accept H0 and reject H1 at 5% level of significance

Table 4.3
VEC Residual Normality Tests
Orthogonalization: Cholesky (Lutkepohl)
Null Hypothesis: residuals are multivariate normal
Date: 12/05/20 Time: 15:10
 51

Sample: 1981 2018

Included observations: 35

Component Skewness Chi-sq Df Prob.

1 -1.992532  23.15941 1  0.0000

2  0.736363  3.163007 1  0.0753
3 -0.058694  0.020096 1  0.8873
4  1.343800  10.53382 1  0.0012

Joint  36.87634 4  0.0000

Component Kurtosis Chi-sq Df Prob.

1  10.35601  78.91166 1  0.0000

2  4.260410  2.316758 1  0.1280
3  3.878257  1.124863 1  0.2889
4  5.966744  12.83562 1  0.0003

Joint  95.18890 4  0.0000

Component Jarque-Bera Df Prob.

1  102.0711 2  0.0000
2  5.479765 2  0.0646
3  1.144959 2  0.5641
4  23.36944 2  0.0000

Joint  132.0652 8  0.2100

The result of the normality test as indicated in the table 4.2 above shows that the error term

under consideration is normally distributed since the probability value of Jarque-Bera is 0.2100

which is greater than 0.05 at 5% critical value. Normality can be a problem when the sample size

is small (<50). Highly skewed data create problems.

 52

4.2.3.3 Heteroskedasticity Test

The purpose of this test is to see whether the error variance of each observation is constant or
not. Non-constant variance can cause estimated model to yield a biased result. White’s general
heteroscedasticity test would be adopted for this purpose at 5% level of significance (Gujarati
et.al. 2012).
H0: presence of homoscedasticity
H1: presence of heteroscedasticity
Decision rule: we reject H0 if the probability value Of Chi-Square is less than 0.05, we do not
reject if otherwise OR reject H0 if n.R2 >χ2 tab do not reject if otherwise.

VEC Residual Heteroskedasticity Tests: No Cross Terms (only levels and squares)
Date: 12/05/20 Time: 16:07
Sample: 1981 2018
Included observations: 34

   Joint test:

Chi-sq df Prob.

 330.9037 330  0.4756

   Individual components:

Dependent R-squared F(22,11) Prob. Chi-sq(22) Prob.

res1*res1  0.877532  3.582691  0.0163  29.83608  0.1225

res2*res2  0.623038  0.826392  0.6633  21.18328  0.5094
res3*res3  0.363282  0.285277  0.9941  12.35160  0.9497
res4*res4  0.758987  1.574578  0.2197  25.80556  0.2601
res5*res5  0.897134  4.360715  0.0075  30.50257  0.1068
res2*res1  0.797949  1.974627  0.1214  27.13028  0.2063
res3*res1  0.627550  0.842461  0.6497  21.33669  0.5000
res3*res2  0.690282  1.114372  0.4426  23.46959  0.3756
res4*res1  0.809206  2.120628  0.0986  27.51300  0.1924
res4*res2  0.636187  0.874333  0.6230  21.63037  0.4821
res4*res3  0.539201  0.585073  0.8629  18.33285  0.6861
res5*res1  0.797226  1.965799  0.1229  27.10568  0.2072
res5*res2  0.637925  0.880929  0.6176  21.68944  0.4786
res5*res3  0.785246  1.828243  0.1502  26.69836  0.2229
res5*res4  0.880567  3.686452  0.0146  29.93928  0.1199
 53

Since the probability value Of Chi-Square is greater than 0.05 i.e 0.4.756 > 0.05 we do not reject

H0 and conclude that there is no presence of heteroscedasticity in the regression result.

4.2.3.4 Multi-collinearity Test.

This test is to check if there is perfect correlation (collinearity) between independent variables.

This will be conducted using the correlation matrix. According to Gujarati (2009), if the

correlation coefficient between any pair of regressors exceeds 0.8, then there is multi-collinearity

between the two variables. The result of the correlation matrix is given below:

RGDP INFR EXR INTR TSV

RGDP  1.000000 -0.239337 -0.337678  0.138452  0.570518

INFR -0.239337  1.000000  0.000316 -0.130184 -0.003790
EXR -0.337678  0.000316  1.000000  0.279327 -0.464363
INTR  0.138452 -0.130184  0.279327  1.000000 -0.054272
TSV  0.570518 -0.003790 -0.464363 -0.054272  1.000000

The correlation matrix indicted that there is no presence of multi-collinearity among the

independent variables in the model. That is, there is a weak correlation among: INFR and EXR,

INTR and EXR, TSV and INFR and others.

4.2.3.5 Causal Analysis (GRANGER CAUSALITY TEST).

This test is undertaken to investigate whether there is a degree of causation of one variable on

the other.

Decision rule: If the computed F value exceeds the critical F value at the chosen level of

significance, we reject the null hypothesis; otherwise, we do not reject it. The results of the

granger causality test are summarized in the table below.

 54

VEC Granger Causality/Block Exogeneity Wald Tests

Date: 12/05/20 Time: 18:14
Sample: 1981 2018
Included observations: 34

Dependent variable: D(RGDP)

Excluded Chi-sq df Prob.

D(INFR)  0.710644 2  0.7009

D(EXR)  9.681067 2  0.0079
D(INTR)  2.383543 2  0.3037
D(TSV)  31.55381 2  0.0000

All  39.05870 8  0.0000

Dependent variable: D(INFR)

Excluded Chi-sq df Prob.

D(RGDP)  0.098994 2  0.9517

D(EXR)  0.300583 2  0.8605
D(INTR)  1.110154 2  0.5740
D(TSV)  0.658470 2  0.7195

All  2.451912 8  0.9639

Dependent variable: D(EXR)

Excluded Chi-sq df Prob.

D(RGDP)  3.557246 2  0.1689

D(INFR)  0.084658 2  0.9586
D(INTR)  2.465421 2  0.2915
D(TSV)  1.966511 2  0.3741

All  7.154293 8  0.5201

Dependent variable: D(INTR)

Excluded Chi-sq df Prob.

D(RGDP)  20.21330 2  0.0000

D(INFR)  4.139591 2  0.1262
D(EXR)  0.611392 2  0.7366
D(TSV)  0.138312 2  0.9332

All  30.08930 8  0.0002

Dependent variable: D(TSV)

 55

Excluded Chi-sq df Prob.

D(RGDP)  1.189508 2  0.5517

D(INFR)  0.097586 2  0.9524
D(EXR)  0.312560 2  0.8553
D(INTR)  0.282469 2  0.8683

All  2.989944 8  0.9350

From the VEC granger causality tests result in the table above, there is no directional causality

relationship between inflation and economic growth (proxied by RGDP) in Nigeria over the

period covered.

4.3 Test of Hypotheses

The results from the statistical tests conducted (especially the F-stat tests), indicates that the

overall result are significant in explaining variations in the dependent variables. Hence,

inferences and conclusions drawn from the model are both sound empirically and reliable for

policy making. This research work is based on the following hypothesis:

Ho1: Inflation has no significant impact on economic growth in Nigeria

Ho2: There is no significant causality relationship between inflation and economic

growth in Nigeria.

From the t-test results in table 4.3, and based on hypotheses stated earlier, we do not reject the

null hypotheses (H0) on inflation (INFR) and conclude that inflation has no significant impact on

economic growth in Nigeria. Also, we do not reject the null hypotheses (Ho2) that there is no

significant directional causality relationship between inflation and economic growth in Nigeria

and reject the alternative hypotheses (H1).

 56

4.4 Discussion of Major Findings

From this study, the result indicates that inflation has negative insignificant impact on economic

growth in Nigeria; this implies that inflation influences economic growth negatively, that is,

increase in the rate of inflation, retard economic growth in Nigeria. Therefore, there is need for

monetary authority to look for every possible means to control the rate of inflation before it too

late.

Also, interest rate, and total savings have a negative insignificant on economic growth, while

exchange rate has positive relationship with economic growth, and it is statistically significant

for the same period under review. This implies that an increase in interest rate will discourage

potential investors from borrowing, hence, decreases economic growth. Interest rate and total

savings have no significant impact on economic growth. These are attributable to instability in

interest rates and poor monetary policies by monetary authorities. Exchange rate in the other

hand has positive significant impact on economic growth in Nigeria. This implies that, when

naira (Nigerian currency) appreciate in the economy, it will help to increase the purchasing

power of the currency, which will in turn leads to more production of goods and services thereby

causing the economy to grow.

More so, there is no significant directional causality relationship between economic growth and

inflation in Nigeria over the period studied.

 57

CHAPTER FIVE

SUMMARY, CONCLUSION AND RECOMMENDATION

5.1 Summary

The results from this study indicate that inflation rate has negative effect on economic growth in

Nigeria economy. And the effect is statistically insignificant for the period of 1981-2018.

Interests rate and total savings have negative relationship with economic growth while exchange

rate has positive relationship with economic growth in Nigeria. Exchange rate has positive and

significant impact on economic growth in Nigeria. Also, interest rate and cash reserve has no

significant impact on economic growth in Nigeria. The results further indicate that there is no

significant directional causality relationship between inflation and economic growth in Nigeria

over the period covered.

5.2Conclusion

This study critically examined the impact of inflation on economic growth in Nigeria for the

period 1981-2018. The specific objectives are to: examine impact of inflation on economic

growth in Nigeria and to ascertain the directional causality relationship between inflation and

economic growth in Nigeria over the period of 1981-2018. To carry out this research work,

annual time series data on inflation, interest rate, exchange rate, total savings and economic

growth (RGDP) for the period 1981-2018 were collected from Central Bank of Nigeria Annual

Statistical Bulletin, 2018. And error correction model (ECM) and Granger causality test were

used for the analysis.

The result showed that inflation rate, exchange rate, interest rate and total savings have long-run

equilibrium relationship with economic growth within the period under study. The study also
 58

indicates that inflation, interest rate and total savings have the potential of impacting negatively

on the economic growth in Nigeria. That is when the rate of inflation rises in the economy, it

will on the average brings decrease in economic growth of Nigeria.

There is no significant directional causality relationship between inflation and economic growth

in Nigeria over the period covered.

5.3 Recommendations

Based on the findings, the researcher recommends the following;

Government should takes both fiscal and monetary policies aimed at controlling inflation in

Nigeria since inflation has a negative insignificant impact on economic growth and they are

negatively correlated. Hence, if left uncontrolled, inflation in Nigeria will bring about a

devastating effect on the economic growth of the country.

More so, monetary authorities should make policies which would help to improve the saving

culture of the people such as increase in the deposit rate which would lure the people to deposit

their money in banks thereby increasing the total savings and supply of loan-able funds. This

would lead to a fall in interest rate and eventually rise in investment in the economy.

Furthermore, government should maintain a properly managed fixed exchange rate system at a

rate which gives our local currency a better value than what it is today, so its appreciation will

exerts more positive significant impact on economic growth in Nigeria.

And Central Bank of Nigeria should maintain a lower interest rate on loans because

manufacturers and other investors who make up the deficit unit of the economy are encouraged

to borrow more when the interest rate is low thus leading to increased investment and growth.

The Federal government under the current managed flexible exchange rate regime should

formulate policies that will strengthen the domestic currency (Naira) against the global market
 59

currency (US Dollars). Policies such as diversifying the productive base of the economy,

encouraging the patronage of locally produced goods etc. These will go a long way towards

strengthening the local currency.

 60

REFERENCES

Aminu, U. & Anono, A.Z. (2012) An empirical analysis of the relationship between
unemployment and inflation in Nigeria: 1977 – 2009. Economic and Financial Review Vol.
1 (12) 42 – 61.
Anochiwa, L.I. & Maduka, A. (2015). Inflation and Economic growth in Nigeria Empirical
Evidence? Journal of Economics and Sustainable Development Vol. 6 No.20.
Bakare, H, R., Kareem & B. Oyelekan (2015), Effect of Inflation Rate on Economic Growth in
Nigeria (1986-2014). Developing Country Study. Vol.5, No8
Central Bank of Nigeria (2016) quarterly statistical bulletin vol5, no3, September.
Central Bank of Nigeria (2014) statistical bulletin vol25, December
Central Bank of Nigeria (2005) statistical bulletin vol16, December
Chankreusna, I. (2017). Determinants of inflation in Cambodia. Research paper, Graduate
School of Public Policy, University of Tokyo, Japan. 1-5
Chu, K. & Feltenstein, A. (1979). Extra-ordinary Inflation: The Argentine Experience Finance
and Development -16(2) : -- 32 35
Doguwa S. I (2012). Ination and Economic Growth in Nigeria: Detecting the Threshold Level.
CBN Journal of Applied Statistics, vol 3, N0 2, pp 99-124.
Dornbush, R., Stanley, F. and Paul, S. (1977). Comparative Advantage, Trade and Payment in a
Ricardian Model with Continuum of Goods. American Economic Review. Vol. 67; No
5:823-39.
Emerenini, I.,& Eke, T.(2014) ). Effect of Inflation and Unemployment on Economic Growth in
Pakistan. Journal of Economic and Sustainable Development.
Engle, R. & Granger, C. (1987) Co-Integration and Error Correction: Representation,
Estimation, and Testing, Econometrica, Vol. 55, pp. 251–276.
Enu, S.T.,& Havi (2014) An Empirical Investigation into the Effect of Unemployment and Inflation on
Economic Growth in Nigeria. Interdisciplinary Journal of Research in Business. Vol.2. Issue.12,
pp.01-14
Ferdinand, A. & Isidore, A. A. (2014). Short-run and Long-run Inflation and Economic Growth
Nexus in Ghana. Ghana Journal of Development Studies. Vol.11, No2
Friedman, M., (1977). Discussion of the Monetarists Controversy. Federal Reserve Bank of
San Francisco Economic Review Spring, 3(1, Suppl.), 12-26.
Gokal, V. &. Hanif. S. (2014), Relationship between Inflation and Economic Growth.
Economics Department, Reserve Bank of Fiji, working paper 04.
Gujarati, D. N, Porter, D.C & Gunasakar, S. (2012). Basic Econometrics 5th Ed. New Delhi Mc
Graw Hill Education Private Limited.
Hossain. E, B.C. Ghosh, K. & Islam (2012), Inflation and Economic Growth in Bangladesh.
Journal of Arts. Science and Commerce. Vo. -III, issue 4(2).
Inyiama, O. I. (2013). Does Inflation Weaken Economic growth? Evidence from Nigeria.
European Journal of Accounting Auditing and Finance Research Vol. 1, Issue 4, pp. 139-
150.
Iya, I. B. & Aminu, U. (2014). An empirical analysis if determinants of inflation in Nigeria.
Journal of Economics and Sustainable Development, 5(1): 140-150
Jaganath, B. (2014). Inflation and its Impact on Economic Growth. Evidence from Six South
Asian countries Journal of Economic and Sustainable Development
 61

Jhingan, M.L (2009), Advanced Macroeconomic Theory 13th Edition Delhi: Virinda
Publication (P) LTO.
Kasidi, F. &. Mwakanemela, K. (2013) Impact of Inflation on Economic Growth: A case study
of Tanzania, Asian Journal of Empirical Research Vol. 3, No. 4, PP 363-380.
Manoel, B. (2010). Inflation and Economic Growth in Latin America; Some Panel Time Series
Evidence. University of Pretoria Department of Economics working paper series.

Maku, A. O. & Adelowokan, O. A. (2013). Dynamics of ination in Nigeria: An autoregressive

approach. European Journal of Humanities and Social Sciences, Vol. 22, pp. 1-7.
Max, C. & Mark, N.H (2008). The Effect of Inflation on Growth: Evidence from a Panel of
Transition Countries, Cardiff Business School Working Paper Series.
Moese, C. N. O., E. A. Essien, A. O. Adenuga, P. N. Omanukwue, M. C. Ononugbo, A. A.
Oguntade, M. O. Abeng, O. M. Ajao (2015): “The Dynamics of Inflation in Nigeria”, CBN
Occasional Paper 32, August.
Moses, K.T, Obioma, E.C, Okpanachi, U.M, Odeniran, S.O & Olaoye, O.J. (2015). Monetary
growth and inflation dynamics in Nigeria. Central Bank of Nigeria/WPS/1(2)
Mundell, R. (1963). Inflation and real interest. The Journal of Political Economy, 280-283.
Nell, K. (2002) Is Low Inflation a Precondition for Faster Growth: The Case Study of South
Africa, Department of Economics University of Kent, and United Kingdom. No.11
Onyeiwu, C., (2012). Monetary policy and economic growth of Nigeria. Journal of Economics
and Sustainable Development, 3(7): 62 -70.
Oladapo, E., Ekunimi, R. & Joshua (2015), Inflation, Interest Rate and Economic Growth in
Nigeria. European Journal of Business Management. Vol.7, No.30.
Olu, J.F., & Idih, E.O. (2015), Inflation and Economic Growth in Nigeria. Journal of Economics
and International Business Management. Vol.3 (1), pp. 20-30.
Ominde, T.A.& Ejiogu (1972). Deficit Financing, Inflation and Capital Formation: The Analysis of
the Nigerian Economy, 1957 – 1970, NJESS, Volume 14.
Sarah, O. & Alade (2014). Keynote Address to the Committee Retreat on Fiscal and Monetary
Policy Coordination for Optimal Macroeconomic Management, Abuja. April 28-May 1
2013. Vol 1:1; 1-5
Udude C. C. (2014). Monetary policy and economic growth of Nigeria (1981-2012), Journal of
Policy and Development Studies. 9(1)
Wai, U. T. (1995). “The Relationship between Inflation and Economic Development: A atistical
Inductive Study”, IMF Staff Papers, 7: 302-317.
Ziaur, R. (2013). Inflation and Economic Growth in Bangladesh: An Empirical Analysis by
Using VAR model. Asian Journal of Empirical Research. 4(8); 404-420.
 62

APPENDIX 1
PRESENTATION OF DATA
YEAR RGDP INFR TSV INTR EXR
0.61002
1981 15258 20.9 1.9792 7.75 5
14985.0 0.67286
1982 8 7.7 2.3212 10.25 7
13849.7 0.72414
1983 3 23.2 2.8793 10.02 2
13779.2 0.76494
1984 6 39.6 3.3613 12.5 2
14953.9
1985 1 1.03 3.6999 9.25 0.89375
15237.9 2.02057
1986 9 13.67 4.2702 10.5 5
15263.9 4.01794
1987 3 9.69 5.2067 17.5 2
16215.3 4.53673
1988 7 61.21 7.1227 16.5 3
17294.6 7.39155
1989 8 44.67 9.2378 26.8 8
19305.6 8.03780
1990 3 3.61 13.0135 25.5 8
19199.0 9.90949
1991 6 22.96 19.3953 20.01 2
19620.1 17.2984
1992 9 48.8 26.0711 29.8 3
19927.9 22.0510
1993 9 61.26 37.0548 18.32 6
19979.1
1994 2 76.76 49.6011 21.02 21.8861
1995 20353.2 51.59 62.135 20.18 21.8861
21177.9
1996 2 14.31 68.7769 19.735 21.8861
1997 21789.1 10.21 84.0995 13.5425 21.8861
22332.8 101.373
1998 7 11.91 5 18.2925 21.8861
22449.4 128.365 92.6933
1999 1 0.22 8 21.32 5
23688.2 164.624 102.105
2000 8 14.53 2 17.98 2
 63

25267.5 216.509 111.943

2001 4 16.49 4 18.2925 3
28957.7 244.064 120.970
2002 1 12.17 1 24.85 2
31709.4 312.368 129.356
2003 5 23.81 9 20.71 5
35020.5 359.311 133.500
2004 5 10.01 2 19.18 4
37474.9 401.986
2005 5 11.57 8 17.95 132.147
592.514 128.651
2006 39995.5 8.55 8 17.26 6
42922.4 753.868 125.833
2007 1 6.56 8 16.9375 1
46012.5 1091.81 15.1354 118.566
2008 2 15.06 2 3 9
1171.91 18.9908 148.880
2009 49856.1 13.93 8 3 2
54612.2 1589.17 17.5856
2010 6 11.8 5 2 150.298
57511.0 1861.41 16.0213 153.861
2011 4 10.3 1 1 6
59929.8 2017.84 16.7903 157.499
2012 9 12 5 1 4
63218.7 2365.03 16.7228 157.311
2013 2 13.7 3 3 2
67152.7 2471.74 16.5483 158.552
2014 9 15.4 3 9 6
69023.9 2465.87 193.279
2015 3 15.06 4 16.85 2
54612.2 2453.62 253.492
2016 6 11.8 9 16.87 3
57511.0 2698.31
2017 4 10.3 3 39.085 305.79
58412,0 38493.2
2018 5 11.44 1 27.053 380.852
Source: Central Bank of Nigeria Statistical Bulletin, 2018 edition

APPENDIX 2

STATIONARITY TEST

RGDP

Null Hypothesis: D(RGDP) has a unit root

Exogenous: Constant
 64

Lag Length: 0 (Automatic - based on SIC, maxlag=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -5.298095  0.0001

Test critical values: 1% level -3.632900
5% level -2.948404
10% level -2.612874

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(RGDP,2)
Method: Least Squares
Date: 12/05/20 Time: 10:01
Sample (adjusted): 1983 2017
Included observations: 35 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

D(RGDP(-1)) -0.920677 0.173775 -5.298095 0.0000

C 1125.836 562.3641 2.001970 0.0536

R-squared 0.459635    Mean dependent var 90.62000

Adjusted R-squared 0.443260    S.D. dependent var 4181.080
S.E. of regression 3119.713    Akaike info criterion 18.98431
Sum squared resid 3.21E+08    Schwarz criterion 19.07319
Log likelihood -330.2255    Hannan-Quinn criter. 19.01500
F-statistic 28.06981    Durbin-Watson stat 1.981097
Prob(F-statistic) 0.000008

source: e-view 9

INFR

Null Hypothesis: INFR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -3.225648  0.0263

Test critical values: 1% level -3.621023
5% level -2.943427
10% level -2.610263
 65

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INFR)
Method: Least Squares
Date: 12/05/20 Time: 10:03
Sample (adjusted): 1982 2018
Included observations: 37 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

INFR(-1) -0.461530 0.143081 -3.225648 0.0027

C 9.178747 3.905063 2.350474 0.0245

R-squared 0.229156    Mean dependent var -0.255676

Adjusted R-squared 0.207132    S.D. dependent var 17.67563
S.E. of regression 15.73893    Akaike info criterion 8.402690
Sum squared resid 8669.990    Schwarz criterion 8.489767
Log likelihood -153.4498    Hannan-Quinn criter. 8.433389
F-statistic 10.40480    Durbin-Watson stat 1.777472
Prob(F-statistic) 0.002725

source: e-view 9

INTR

Null Hypothesis: INTR has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=9)

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic -3.573088  0.0113

Test critical values: 1% level -3.621023
 66

5% level -2.943427
10% level -2.610263

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(INTR)
Method: Least Squares
Date: 12/05/20 Time: 10:09
Sample (adjusted): 1982 2018
Included observations: 37 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

INTR(-1) -0.517917 0.144949 -3.573088 0.0011

C 9.935887 2.765270 3.593099 0.0010

R-squared 0.267276    Mean dependent var 0.521703

Adjusted R-squared 0.246341    S.D. dependent var 5.882378
S.E. of regression 5.106700    Akaike info criterion 6.151522
Sum squared resid 912.7436    Schwarz criterion 6.238599
Log likelihood -111.8032    Hannan-Quinn criter. 6.182221
F-statistic 12.76696    Durbin-Watson stat 2.226554
Prob(F-statistic) 0.001052

source: e-view 9

EXR

Null Hypothesis: D(EXR) has a unit root

Exogenous: Constant
Lag Length: 0 (Automatic - based on SIC, maxlag=9)

t-Statistic   Prob.*
 67

Augmented Dickey-Fuller test statistic -3.303324  0.0223

Test critical values: 1% level -3.632900
5% level -2.948404
10% level -2.612874

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(EXR,2)
Method: Least Squares
Date: 12/05/20 Time: 10:11
Sample (adjusted): 1983 2017
Included observations: 35 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

D(EXR(-1)) -0.582501 0.176338 -3.303324 0.0023

C 5.701114 3.170944 1.797923 0.0813

R-squared 0.248496    Mean dependent var 1.492426

Adjusted R-squared 0.225723    S.D. dependent var 19.52277
S.E. of regression 17.17867    Akaike info criterion 8.580659
Sum squared resid 9738.517    Schwarz criterion 8.669536
Log likelihood -148.1615    Hannan-Quinn criter. 8.611339
F-statistic 10.91195    Durbin-Watson stat 1.996871
Prob(F-statistic) 0.002305

source: e-view 9

TSV

Null Hypothesis: TSV has a unit root

Exogenous: Constant
Lag Length: 9 (Automatic - based on SIC, maxlag=9)
 68

t-Statistic   Prob.*

Augmented Dickey-Fuller test statistic  25.31753  0.9999

Test critical values: 1% level -3.689194
5% level -2.971853
10% level -2.625121

*MacKinnon (1996) one-sided p-values.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(TSV)
Method: Least Squares
Date: 12/05/20 Time: 10:40
Sample (adjusted): 1991 2018
Included observations: 28 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.  

TSV(-1) 39.71884 1.568828 25.31753 0.0000

D(TSV(-1)) -39.39934 2.569909 -15.33102 0.0000
D(TSV(-2)) -36.17414 2.020487 -17.90368 0.0000
D(TSV(-3)) -53.56849 1.906386 -28.09950 0.0000
D(TSV(-4)) -45.06204 2.505325 -17.98650 0.0000
D(TSV(-5)) -43.22546 3.071929 -14.07111 0.0000
D(TSV(-6)) -60.31131 3.637927 -16.57848 0.0000
D(TSV(-7)) -37.89318 4.035111 -9.390865 0.0000
D(TSV(-8)) -35.58983 4.129402 -8.618640 0.0000
D(TSV(-9)) -102.4736 4.190908 -24.45140 0.0000
C 3.904279 134.9320 0.028935 0.9773

R-squared 0.996683    Mean dependent var 1374.293

Adjusted R-squared 0.994731    S.D. dependent var 6746.892
S.E. of regression 489.7305    Akaike info criterion 15.51231
Sum squared resid 4077212.    Schwarz criterion 16.03568
Log likelihood -206.1723    Hannan-Quinn criter. 15.67231
F-statistic 510.7564    Durbin-Watson stat 2.285901
Prob(F-statistic) 0.000000

source: e-view 9

APPENDIX 3

CO-INTEGRATION TEST
Date: 12/05/20 Time: 15:05
Sample (adjusted): 1983 2017
Included observations: 35 after adjustments
Trend assumption: Linear deterministic trend 69
Series: RGDP INFR INTR EXR TSV 
Lags interval (in first differences): 1 to 1

Unrestricted Cointegration Rank Test (Trace)

Source:
Hypothesized Trace 0.05 e-view 9
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None *  0.646569  80.44511  69.81889  0.0056

At most 1  0.514408  44.04276  47.85613  0.1091
At most 2  0.290865  18.75926  29.79707  0.5103
At most 3  0.132746  6.729414  15.49471  0.6093
At most 4  0.048624  1.744601  3.841466  0.1866

 Trace test indicates 1 cointegrating eqn(s) at the 0.05 level

 * denotes rejection of the hypothesis at the 0.05 level
 **MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None *  0.646569  36.40236  33.87687  0.0244

At most 1  0.514408  25.28350  27.58434  0.0958
At most 2  0.290865  12.02984  21.13162  0.5447
At most 3  0.132746  4.984813  14.26460  0.7438
At most 4  0.048624  1.744601  3.841466  0.1866

 Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level

 * denotes rejection of the hypothesis at the 0.05 level
 **MacKinnon-Haug-Michelis (1999) p-values

 Unrestricted Cointegrating Coefficients (normalized by b'*S11*b=I): 

RGDP INFR INTR EXR TSV

-0.000317  0.026225  0.032757  0.020497  0.003764
 0.000135  0.078910 -0.162208  0.005077 -0.002429
-9.19E-05  0.006817 -0.067755 -0.016166  0.003159
-0.000152  0.008285 -0.233388  0.028215  0.001127
 0.000374 -0.000860 -0.131721 -0.034341 -0.006260

 Unrestricted Adjustment Coefficients (alpha): 

D(RGDP)  871.5775 -256.4396 -1249.138  314.8498 -20.66403

D(INFR) -2.355539 -11.18875  0.325363 -0.679319  1.551096
D(INTR) -0.693377 -0.326476  0.277595  1.063945 -0.290895
D(EXR) -7.468946 -0.950174  2.410327 -1.256132 -2.638836
D(TSV) -40.34052  15.71531 -31.39854  8.886937  8.699942

1 Cointegrating Equation(s):  Log likelihood -894.8423

Normalized cointegrating coefficients (standard error in parentheses)

RGDP INFR INTR EXR TSV
 1.000000 -82.69915 -103.2956 -64.63686 -11.86975
 (36.9438)  (138.159)  (14.5816)  (1.08463)

Adjustment coefficients (standard error in parentheses)

D(RGDP) -0.276390
 (0.15469)
D(INFR)  0.000747
 (0.00104)
D(INTR)  0.000220
 (0.00020)
 70

APPENDIX 4

VECTOR ERROR CORRECTION MODEL

 Vector Error Correction Estimates

 Date: 12/05/20 Time: 15:45
 Sample (adjusted): 1984 2017
 Included observations: 34 after adjustments
 Standard errors in ( ) & t-statistics in [ ]

Cointegrating Eq:  CointEq1

RGDP(-1)  1.000000

INFR(-1)  61.37320
 (42.1490)
[ 1.45610]

EXR(-1) -97.43913
 (16.4923)
[-5.90815]

INTR(-1) -441.8700
 (131.357)
[-3.36389]

TSV(-1) -9.709172
 (1.05099)
[-9.23814]

C -11699.63

Error Correction: D(RGDP) D(INFR) D(EXR) D(INTR) D(TSV)

CointEq1 -0.793481 -0.001340  0.002544 -9.65E-05 -0.001116

 (0.15283)  (0.00137)  (0.00123)  (0.00028)  (0.00720)
[-5.19196] [-0.97507] [ 2.07486] [-0.34809] [-0.15498]

D(RGDP(-1)) -0.028397  0.000504 -0.002442 -0.001445 -0.007860

 (0.17831)  (0.00160)  (0.00143)  (0.00032)  (0.00840)
[-0.15926] [ 0.31415] [-1.70692] [-4.46874] [-0.93514]

D(RGDP(-2)) -0.471929 -0.000269  0.003659  0.000955  0.014163

 (0.42171)  (0.00379)  (0.00338)  (0.00076)  (0.01988)
[-1.11907] [-0.07079] [ 1.08159] [ 1.24835] [ 0.71250]

D(INFR(-1))  17.23151 -0.020648 -0.029106  0.078055 -0.252903

 (22.0244)  (0.19810)  (0.17668)  (0.03994)  (1.03811)
[ 0.78238] [-0.10423] [-0.16474] [ 1.95430] [-0.24362]

D(INFR(-2))  9.025752 -0.346429 -0.048026 -0.016693 -0.240892

 (23.6125)  (0.21238)  (0.18942)  (0.04282)  (1.11297)
[ 0.38225] [-1.63117] [-0.25354] [-0.38983] [-0.21644]

D(EXR(-1)) -83.74532  0.039472  0.332598 -0.035903  0.584931

 (32.6154)  (0.29336)  (0.26164)  (0.05915)  (1.53732)
 71

[-2.56766] [ 0.13455] [ 1.27118] [-0.60702] [ 0.38049]

D(EXR(-2)) -59.97103 -0.163469  0.339867 -0.030510  0.659303

 (34.1852)  (0.30748)  (0.27424)  (0.06199)  (1.61131)
[-1.75430] [-0.53165] [ 1.23931] [-0.49215] [ 0.40917]

D(INTR(-1)) -203.5210 -0.028750  1.304636 -0.420096  2.329316

 (132.104)  (1.18820)  (1.05976)  (0.23957)  (6.22671)
[-1.54061] [-0.02420] [ 1.23107] [-1.75357] [ 0.37408]

D(INTR(-2)) -87.96133  0.878913 -0.112188 -0.047146 -0.561675

 (113.472)  (1.02061)  (0.91028)  (0.20578)  (5.34845)
[-0.77518] [ 0.86116] [-0.12324] [-0.22911] [-0.10502]

D(TSV(-1))  20.48322  0.015152 -0.039959  0.002522  0.319715

 (5.05623)  (0.04548)  (0.04056)  (0.00917)  (0.23832)
[ 4.05109] [ 0.33317] [-0.98513] [ 0.27507] [ 1.34151]

D(TSV(-2))  20.67963  0.036972 -0.042723  0.002401  0.405235

 (5.75368)  (0.05175)  (0.04616)  (0.01043)  (0.27120)
[ 3.59416] [ 0.71441] [-0.92561] [ 0.23015] [ 1.49424]

C  113.3546 -3.967503  7.381136  1.195474  5.225253

 (647.826)  (5.82683)  (5.19694)  (1.17481)  (30.5351)
[ 0.17498] [-0.68090] [ 1.42028] [ 1.01759] [ 0.17112]

 R-squared  0.665248  0.205957  0.395497  0.676295  0.474361

 Adj. R-squared  0.497872 -0.191065  0.093245  0.514443  0.211542
 Sum sq. resids  1.06E+08  8598.808  6840.226  349.5487  236141.5
 S.E. equation  2198.031  19.77005  17.63291  3.986048  103.6036
 F-statistic  3.974573  0.518754  1.308501  4.178465  1.804893
 Log likelihood -302.4843 -142.3052 -138.4156 -87.85871 -198.6230
 Akaike AIC  18.49908  9.076778  8.847975  5.874042  12.38959
 Schwarz SC  19.03779  9.615493  9.386690  6.412757  12.92830
 Mean dependent  1284.156 -0.379412  8.972525  0.854853  79.27746
 S.D. dependent  3101.892  18.11507  18.51735  5.720345  116.6771

 Determinant resid covariance (dof adj.)  4.11E+16

 Determinant resid covariance  4.66E+15
 Log likelihood -854.5284
 Akaike information criterion  54.08991
 Schwarz criterion  57.00795

source: e-view 9
 72

APPENDIX 5

HETROSKEDASTICITY TEST

VEC Residual Heteroskedasticity Tests: No Cross Terms (only levels and squares)
Date: 12/05/20 Time: 16:07
Sample: 1981 2018
Included observations: 34

   Joint test:

Chi-sq df Prob.

 330.9037 330  0.4756

   Individual components:

Dependent R-squared F(22,11) Prob. Chi-sq(22) Prob.

res1*res1  0.877532  3.582691  0.0163  29.83608  0.1225

res2*res2  0.623038  0.826392  0.6633  21.18328  0.5094
res3*res3  0.363282  0.285277  0.9941  12.35160  0.9497
res4*res4  0.758987  1.574578  0.2197  25.80556  0.2601
res5*res5  0.897134  4.360715  0.0075  30.50257  0.1068
res2*res1  0.797949  1.974627  0.1214  27.13028  0.2063
res3*res1  0.627550  0.842461  0.6497  21.33669  0.5000
res3*res2  0.690282  1.114372  0.4426  23.46959  0.3756
res4*res1  0.809206  2.120628  0.0986  27.51300  0.1924
res4*res2  0.636187  0.874333  0.6230  21.63037  0.4821
res4*res3  0.539201  0.585073  0.8629  18.33285  0.6861
res5*res1  0.797226  1.965799  0.1229  27.10568  0.2072
res5*res2  0.637925  0.880929  0.6176  21.68944  0.4786
res5*res3  0.785246  1.828243  0.1502  26.69836  0.2229
res5*res4  0.880567  3.686452  0.0146  29.93928  0.1199

source: e-view 9
 73

APPENDIX 6

AUTO-CORRELATION TEST

VEC Residual Serial Correlation LM Tests

Null Hypothesis: no serial correlation at lag
order h
Date: 12/05/20 Time: 16:10
Sample: 1981 2018
Included observations: 34

Lags LM-Stat Prob

1  19.91067  0.7515
2  12.37540  0.9833
3  28.64760  0.2789
4  19.70616  0.7621
5  25.14712  0.4542
6  40.12037  0.0283
7  26.42418  0.3852

Probs from chi-square with 25 df.

source: e-view 9
 74

APPENDIX 7

NORMALITY TEST

VEC Residual Normality Tests

Orthogonalization: Cholesky (Lutkepohl)
Null Hypothesis: residuals are multivariate normal
Date: 12/05/20 Time: 15:10
Sample: 1981 2018
Included observations: 35

Component Skewness Chi-sq Df Prob.

1 -1.992532  23.15941 1  0.0000

2  0.736363  3.163007 1  0.0753
3 -0.058694  0.020096 1  0.8873
4  1.343800  10.53382 1  0.0012

Joint  36.87634 4  0.0000

Component Kurtosis Chi-sq Df Prob.

1  10.35601  78.91166 1  0.0000

2  4.260410  2.316758 1  0.1280
3  3.878257  1.124863 1  0.2889
4  5.966744  12.83562 1  0.0003

Joint  95.18890 4  0.0000

Component Jarque-Bera Df Prob.

1  102.0711 2  0.0000
2  5.479765 2  0.0646
3  1.144959 2  0.5641
4  23.36944 2  0.0000

Joint  132.0652 8  0.2100

 75

APPENDIX8

MULTI-COLLERINEARITY TEST

RGDP INFR EXR INTR TSV

RGDP  1.000000 -0.239337 -0.337678  0.138452  0.570518

INFR -0.239337  1.000000  0.000316 -0.130184 -0.003790
EXR -0.337678  0.000316  1.000000  0.279327 -0.464363
INTR  0.138452 -0.130184  0.279327  1.000000 -0.054272
TSV  0.570518 -0.003790 -0.464363 -0.054272  1.000000

Source: e-view
9
 76

APPENDIX 9

GRANGER CAUSALITY TEST

VEC Granger Causality/Block Exogeneity Wald Tests

Date: 12/05/20 Time: 18:14
Sample: 1981 2018
Included observations: 34

Dependent variable: D(RGDP)

Excluded Chi-sq df Prob.

D(INFR)  0.710644 2  0.7009

D(EXR)  9.681067 2  0.0079
D(INTR)  2.383543 2  0.3037
D(TSV)  31.55381 2  0.0000

All  39.05870 8  0.0000

Dependent variable: D(INFR)

Excluded Chi-sq df Prob.

D(RGDP)  0.098994 2  0.9517

D(EXR)  0.300583 2  0.8605
D(INTR)  1.110154 2  0.5740
D(TSV)  0.658470 2  0.7195

All  2.451912 8  0.9639

Dependent variable: D(EXR)

Excluded Chi-sq df Prob.

D(RGDP)  3.557246 2  0.1689

D(INFR)  0.084658 2  0.9586
D(INTR)  2.465421 2  0.2915
D(TSV)  1.966511 2  0.3741

All  7.154293 8  0.5201

Dependent variable: D(INTR)

Excluded Chi-sq df Prob.

D(RGDP)  20.21330 2  0.0000

D(INFR)  4.139591 2  0.1262
D(EXR)  0.611392 2  0.7366
D(TSV)  0.138312 2  0.9332

All  30.08930 8  0.0002

 77

Dependent variable: D(TSV)

Excluded Chi-sq df Prob.

D(RGDP)  1.189508 2  0.5517

D(INFR)  0.097586 2  0.9524
D(EXR)  0.312560 2  0.8553
D(INTR)  0.282469 2  0.8683

All  2.989944 8  0.9350

Source: e-view 9