european of economic journal issn 2304-9669. e-issn 2305...

European Journal of Economic Studies, 2015, Vol.(13), Is. 3

119

EUROPEAN of Economic

Journal Studies

Has been issued since 2012. ISSN 2304-9669. E-ISSN 2305-6282

2015. Vol.(13). Is. 3. Issued 4 times a year Impact Factor OAJI 2012 – 0,527

Impact Factor MIAR 2015 – 3,477

EDITORIAL STAFF

PhD Vidishcheva Evgeniya – Sochi State University, Sochi, Russia (Editor-in-Chief) Dr. Simonyan Garnik – Scientific Research Centre of the Russian Academy of Sciences,

Sochi, Russia Dr. Levchenko Tatyana – Sochi State University, Sochi, Russia Dr. Tarakanov Vasilii – Volgograd State University, Volgograd, Russia

EDITORIAL BOARD

Dr. Balatsky Evgeny – Central Economics and Mathematics Institute (RAS), Moscow, Russia Dr. Dinh Tran Ngoc Huy – Banking University HCMC Viet Nam – GSIM, International

University of Japan, Japan Dr. Gerasimenko Viktor – Odessa State Economic University, Odessa, Ukraine Dr. Gvarliani Tatjana - – Sochi State University, Sochi, Russian Federation Dr. Gunare Marina – Baltic International Academy, Riga Dr. Kryshtanovskaya Olga – Institute of Sociology of the Russian Academy of Sciences,

Moscow, Russia Dr. Minakir Pavel – Economic Research Institute of the FarEastern Branch Russian Academy

of Sciences, Khabarovsk, Russia Dr. Papava Vladimir – Ivane Javakhishvili Tbilisi State University, Tbilissi, Georgia Dr. Prokopenko Olga – Sumy State University, Sumy, Ukraine Dr. Vishnevsky Valentine – Institute of Industrial Economics of the National Academy of

Sciences of Ukraine, Donetsk, Ukraine The journal is registered by Federal Service for Supervision of Mass Media,

Communications and Protection of Cultural Heritage (Russia). Registration Certificate ПИ ФС77-50465 4 July 2012.

Journal is indexed by: CrossRef (UK), EBSCOhost Electronic Journals Service (USA), Electronic scientific library (Russia), Global Impact Factor (Australia), Index Copernicus (Poland), Open Academic Journals Index (Russia), ResearchBib (Japan), ULRICH’s WEB (USA).

All manuscripts are peer reviewed by experts in the respective field. Authors of the manuscripts bear responsibility for their content, credibility and reliability.

Editorial board doesn’t expect the manuscripts’ authors to always agree with its opinion.

Postal Address: 26/2 Konstitutcii, Office 6 354000 Sochi, Russia Website: http://ejournal2.com/ E-mail: [email protected]

Founder and Editor: Academic Publishing House Researcher

Passed for printing 15.09.15.

Format 21 29,7/4.

Enamel-paper. Print screen.

Headset Georgia.

Ych. Izd. l. 4,5. Ysl. pech. l. 4,2.

Circulation 250 copies. Order 105.

© European Journal of Economic Studies, 2015

А

Eu

rop

ea

n J

ou

rna

l o

f Ec

on

om

ic S

tud

ies

3

2015

1 2010 Is.

http://globalimpactfactor.com/european-researcher/

mailto:[email protected]


120

ЕВРОПЕЙСКИЙ ЭКОНОМИЧЕСКИХ

Журнал ИССЛЕДОВАНИЙ

Издается с 2012 г. ISSN 2304-9669. E-ISSN 2305-6282 2015. 3 (13). Выходит 4 раза в год.

Impact Factor OAJI 2012 – 0,527 Impact Factor MIAR 2015 – 3,477

РЕДАКЦИОННАЯ КОЛЛЕГИЯ

Видищева Евгения – Сочинский государственный университет, Сочи, Россия

(Гл. редактор) Левченко Татьяна – Сочинский государственный университет, Сочи, Россия Симонян Гарник – Сочинский научно-исследовательский центр Российской академии

наук, Сочи, Россия Тараканов Василий – Волгоградский государственный университет, Волгоград, Россия

РЕДАКЦИОННЫЙ СОВЕТ

Балацкий Евгений – Центральный экономико-математический институт РАН, Москва, Россия

Вишневский Валентин – Институт экономики промышленности Национальной академии наук Украины, Донецк, Украина

Гварлиани Татьяна – Сочинский государственный университет, Сочи, Российская Федерация Герасименко Виктор – Одесский государственный экономический университет, Одесса,

Украина Гунаре Марина – Балтийская международная академия, Рига Динь Чан Нгок Хай – Банковский университет Хошимин Вьетнам - GSIM, Международный

университет Японии, Япония Минакир Павел – Институт экономических исследований ДВО РАН, Хабаровск, Россия Крыштановская Ольга – Институт социологии РАН, Москва, Россия Папава Владимир – Тбилисский государсвенный универстите имени Иване

Джавахишвили, Тбилисси, Грузия Прокопенко Ольга – Сумский государственный университет, Сумы, Украина

Журнал зарегистрирован Федеральной службой по надзору в сфере массовых коммуникаций, связи и охраны культурного наследия (Российская Федерация). Свидетельство о регистрации средства массовой информации ПИ ФС77-50465 от 4 июля 2012 г.

Журнал индексируется в: CrossRef (Великобритания), EBSCOhost Electronic Journals Service (США), Global Impact Factor (Австралия), Index Copernicus (Польша), Научная электронная библиотека (Россия), Open Academic Journals Index (Россия), ResearchBib (Япония), ULRICH’s WEB (США).

Статьи, поступившие в редакцию, рецензируются. За достоверность сведений, изложенных в статьях, ответственность несут авторы публикаций.

Мнение редакции может не совпадать с мнением авторов материалов.

Адрес редакции: 354000, Россия, г. Сочи, ул. Конституции, д. 26/2, оф. 6 Сайт журнала: http://ejournal2.com/ E-mail: [email protected] Учредитель и издатель: ООО «Научный издательский дом "Исследователь"» - Academic Publishing House Researcher

Подписано в печать 15.09.15.

Формат 21 29,7/4.

Бумага офсетная.

Печать трафаретная.

Гарнитура Georgia.

Уч.-изд. л. 4,5. Усл. печ. л. 4,2.

Тираж 250 экз. Заказ 105

© European Journal of Economic Studies, 2015

А

Ев

ро

пе

йс

ки

й ж

урн

ал

эк

он

ом

ич

ес

ки

х и

сс

ле

до

ва

ни

й

3

2015

http://oaji.net/journal-detail.html?number=3

http://globalimpactfactor.com/european-researcher/



121

C O N T E N T S

Factors Affecting Public Investment in Manufacturing Sector of Pakistan Gulzar Ali …………………………………………………………………………………………………………..

122

Impact of Tourism on Spatial Transformation: a Case Study of the Bela Crkva Muncipality (Serbia)

Jovana Boškov, Stefan Kotrla, Dajana Lulić …………………………………………………………..

131

Coal Mining and Indigenous Communities: А Case Study of Jharia Coalfields Sribas Goswami ………………………………………………………………………………………………….

139

Unemployment and Economic Growth of Developing Asian Countries: A Panel Data Analysis

Muhammad Imran, Khurrum S. Mughal, Aneel Salman, Nedim Makarevic ……………..

147

Fiscal Policy and Income Inequality in Pakistan: An ARDL Approach Rana Ejaz Ali Khan, Bushra Jabeen Hashmi ………………………………………………………….

161

Volatility in Sectors and National Income Growth: A Comparative Analysis of Pakistan and South Korea

Rana Ejaz Ali Khan, Tasnim Khan, Nadia Mahtab ………………………………………………….

175


122

Copyright © 2014 by Academic Publishing House Researcher

Published in the Russian Federation European Journal of Economic Studies Has been issued since 2012. ISSN: 2304-9669 E-ISSN: 2305-6282 Vol. 13, Is. 3, pp. 122-130, 2015 DOI: 10.13187/es.2015.13.122

www.ejournal2.com

UDC 33

Factors Affecting Public Investment in Manufacturing Sector of Pakistan

Gulzar Ali

School of Economics, Huazhong University of Science & Technology, P.R. China PhD Scholar E-mail: [email protected]

Abstract Public Investment in manufacturing sector has emerged as one of the most active part all

over the world. In Pakistan public investment had made a significant contribution in different sectors and has played a fundamental role in the financial system of the country. This study has been an attempt to identify the factors, which affect public investment in manufacturing sector significantly in Pakistan. The data used in this study are from 1981 to 2014. Descriptive as well as quantitative techniques are applied to derive the results and advance statistical software E-views are used. In time-series analysis there always remains a suspicion about spurious relationship. This research study is also based on time-series data, that’s why before going to estimate the model, the data are tested by applying Augmented Dickey Fuller (ADF), but the data used in the study did not show any sign of spurious relationship. In order to capture the effect of various factors affecting public investment in manufacturing sector, the investment accelerator model is used and regressed through NLS and ARIMA model. The regression results shows that the variables Value-added in large-scale manufacturing sector (Vm), and Lagged Public Investment in Large-Scale Manufacturing Sector (Igm(-1)), Index of Price of Capital (Ipk), Change in Domestic Credit available to Public Sector (ΔDcp) and dummy (Dps) for political stability and favorable economic conditions of the country plays a significant role in public investment in manufacturing sector. The study recommends that the government should create a sufficient demand by increasing domestic purchasing power, by export expansion, by import substitutions through assets redistribution. The study also found that stable political condition also necessary for the rapid economic and investment growth.

Keywords: public investment, manufacturing sector, NLS and Аrima model, investment accelerator model.

1. Introduction When expenditure on goods and services falls during a recession a lot of the decline is usually

due to a drop in investment spending. Economists study investment to better understand fluctuation in the economy’s output of goods and services. Investment plays two roles in macroeconomics. First because it is a large and the most volatile component of spending, sharp changes in investment can have a major impact on aggregate demand. This in turn affects the output and employment. In addition, investment leads to capital accumulation. Adding to the stock of buildings and equipments increases the nation’s potential output and promotes economic

http://www.ejournal2.com/


123

growth in the long run. Thus investment plays a dual role, affecting short-run output through its impact on aggregate demand and influencing long-run growth the impact of capital formation on potential output and aggregate supply.

Public Investment over the years has been the only consistent variable contributing to the economic growth of nations. The role of capital stock in boosting the development in countries like Korea, China, Malaysia and many others, proves the premises adopted by the investment theorists. Empirical economics has come very close in identifying the broad determinants that play the major role in the entire process of providing the optimal impetus to the economy. Unfortunately, the impact of public investment and foreign direct investment on growth remains more contentious in empirical than theoretical studies, while some of the studies observe a positive impact of public investment on economic growth, other detect a negative relationship between these two variables. The controversy has arisen partially due to data, insufficiency in either cross countries or time series investigation. Developed countries are expected to have a higher level of human capital and hence benefited more from public investment than developing countries.

Public Investment has a composite bundle of capital stock, know-how and technology and can augment the existing stock of knowledge in the recipient economy through labor training, skill acquisition and diffusion and the introduction of alternative management practices and organizational arrangements. However the Neo-classical growth model promotes economic growth by increasing the volume of investment and its efficiency. In the endogenous growth model public investment raises economic growth by generating technological diffusion from developed world to the host country.

The important factor that underlines the need to reformulate public investment theories in the developing country context is the existence of a debt overhang in many countries, which has often been cited as a factor inhibiting private investment. The possibility that confiscatory future taxation will be used to finance future service may need to be reflected in the specification of private investment behavior. The large role of the public capital stock suggests the need to incorporate complementary and substitutability relationship between public and private capital in to private investment decision. The relationship between public and private investment takes on greater importance in the developing world than in industrial countries because of the large role played by the government in the overall process of capital formation. On the one hand, public sector investment can crowd out private investment expenditure if it uses scarce physical and financial resources that would otherwise be available to the private sector. The financing of public sector investment whether through taxes, issuance of debt instruments, or inflation, can reduce the resources available to the private sector and thus depress private investment activity. Moreover, the public sector may produce marketable output that competes with private output. On the other hand, public investment to maintain or expand infrastructure and the provision of public goods is likely to be complementary to private investment. Public investment can enhance the prospects for private investment by raising the productivity of capital. Of course, the empirical relevance of phenomena such as these will differ across countries and at different points in time. However, existing studies of private investment in developing countries have not always taken them into account, even where they seem clearly relevant; and to the context they have done so, it has undertaken in an adhoc fashion, by adding new variable to a regression that specifies investment as a linear function of explanatory variables suggested by the theories described above, rather than reformulating the theory and estimating the revised model of investment.

In under-developed countries, the analysis of this (public investment) highly volatile component of GNP is extremely difficult not only from demand side but also from supply side. In developing countries the sources of investible funds include not only domestic savings but also foreign resources. The sources of domestic saving (voluntary savings, compulsory savings, deficit financing and export) are limited and it is difficult to mobilize them in the desired channel especially in the private sector. As regards foreign capital inflow, it is determined by highly volatile non-economic factors and as such cannot be predicted with any degree of accuracy (Shamshad, 2009).

In most of the studies, undertaken both in the developed and the developing countries, public investment functions have been estimated only for the fixed assets, i.e. expenditure on capital equipment, machinery and structures. Some of the studies also distinguished between capital equipment, machinery and structures, while most of the studies do not. This study will examine investment in manufacturing sector only. The study confines to Gross Fixed Investment in


124

manufacturing economic activities by public sector of Pakistan economy, and its determinants in this activity will be observed and analyzed. Theories have been developed to analyze investment behavior; however, the empirical studies show that none of these theories is universally true. Therefore, it is difficult to specify the variables, which are universally accepted as the determinants of public investment. In fact the nationalization of industries in seventies shook the roots of structure of investment and crowding out effect took place. Political structure of Pakistan has been changing with time, which makes it even difficult to identify the determinant of public investment (Ahmed, 2009).

1.2. Objective of the study The objectives of the present study are to analyze observe the public investment behavior in

Manufacturing sector in Pakistan. Public Investment is highly volatile, it is quite difficult to analyze and forecast this

component of gross national product (GNP). It is affected by a multitude of economic variables, which assume varying degree of importance in different situations, particularly in different phases of the business cycle. Investment, particularly public investment is very sensitive to non-economic variables such as war, political instability and other disturbance inside and outside the economy. Obviously, the effect of these variables on investment behavior is extremely difficult to predict. Public sector investment is mainly determined by the state, keeping in view the social, political and economic priorities as well as financial constraints.

2. Literature review Resek (1966) studied the investment behavior of thirteen industrial groups of United States

of America, applying a fixed lag distribution for all variables, concluded that the ratio of interest and stock price were most significant. However factors such as change in output and the debt capacity were less significant and this significance was not present for all industries. Replacement investment being proportional to the overall capital stock has been captured by the regression intercept. Data used was quarterly ranging from 1953 to 1962. Resek before using a fixed lag distribution for independent variables acquired the weights attributed to lags from an investment expenditure regression made on capital.

Jorgenson and Stephenson (1969) determined the public investment in manufacturing sector of the United States of America and found that only significant factor is the output. The major determinants which were tried in the model included change in output (taken on current prices), capital stock and the prices of capital service provision. The variable capital stock was included to capture the replacement investment. Researcher have tried to use seasonally adjusted quarterly data from 1948 (3rd quarter) to 1960 (4th quarter).

Hickman (1985) studied investment behavior of manufacturing sector in the United States of America. He used conventional variables such as the overall output, capital stock, output prices and the prevalent wage rate. After observing a sample, ranges from 1949 to 1960, he compared alternative specifications and finally reduced his emphasis to a prototype of accelerator model.

Bhaskar and Glyn (1995) found that declining profitability in the 1960s and 1970s accounted for a major part of the investment hold back in manufacturing sectors of Germany and Japan, and also depressed investment in the United States of America. Glyn estimated for twelve OECD countries for two time periods from 1960 to 1973 and from 1973 to 1992 and concluded that a three percent higher profit share is associated with approximately one percent growth of the capital stock.

Constantan (1973) had analyzed the effect of public investment on economic growth. In the study, the main objective was to analyze the relationship between export, import, foreign capital inflow and the rate of domestic growth. The analysis was based on two-gap model and a complete Harrod-Domar model was used. On the basis of empirical analysis and found that there is strong relationship between public investment and the growth of domestic product, in the post world war. But the study did not found any significant relationship between the inflow of foreign capital and the rate of growth of domestic product.

Amjad (1976) using pooled data for the period of 1962-1970 analyzed the investment behavior of manufacturing sector. Dividing the time period in to two; he took data of thirty nine (39) firms for the period 1962-1965 and data of further ninety four (94) firms for the period of 1966-1970. In the later case profit factor seemed to be only significant variables. Then re-specifying


125

model the researcher included two additional variables; price of output and sales. Both variables remained significant for the period of 1961-1965, however, for the period 1965-1970, both stood statistically insignificant.

Zaren (1991), attempts to identify the factors which had played significant role in determining private and public investment in manufacturing sector of Pakistan. Several models have been estimated by applying Ordinary Least Square (OLS) or AR (1) techniques using annual data for the period of 1962-1963 to 1988-1989. The first conclusion that the researcher had drawn was that, the naïve accelerator and flexible accelerator model of Junankar and the Neo-classical model of Jorgenson and Jorgenson-Stephenson are unable to explain the behavior of private and public investment in Pakistan.

However, the Rowley-Trivedi flexible accelerator model and the Keynesian model appear to have reasonable empirical performance. Secondly the results indicate that qualitative factors like government policies (towards nationalization etc) and the political stabilization are the two most prominent determinants of public investment in the manufacturing sector.

3. Description of Data and derivation of the Model 3.1. Data Analysis and Description The data used in this research study are based on annual figures because quarterly data for

most of the variables are not available from any source in case of Pakistan. The time period of the study data is from 1981 to 2014, because data prior to 1981 at constant price are unavailable. There is no direct source to complete data; therefore data are collected from Economic Surveys, Federal Bureau of Statistics, State Bank of Pakistan, Agriculture Development Bank of Pakistan (ZTBL), Cooperatives and Commercial Banks, International Financial Statistics (IFS), Pakistan Institute of Development Economics (PIDE), World Development Report (WDR), National Accounts of Pakistan and from different surveys and reports.

All the variables used in the estimation for all investment function are taken as real and at constant prices. The price index of capital good has been calculated by dividing the value of gross fixed capital formation at current price by corresponding value at constant prices.

3.2. Derivation of the Model and its Justification White (September, 1956) has worked on this approach, which is more or less an extension of

Keynes work. The optimal level of capital stock is determined and then the actual stock adjusted according to that stock. The demand for capital would depend on the present value and the internal rate of return. Lower rate of interest would imply greater levels of investment and vice versa. Hence the demand for capital is negatively associated with the rate of interest.

A natural starting point of discussion of investment is the rationale of the Present Value (PV) criterion and its implication for the determinants of investment. Thus, by reducing current income, the owners can increase future by investing the firms retained earnings. The investment rule, that the firm should maximize its present value by investing in any projects with positive returns. In order to maximize its present value the firm should invest in all projects that have a (PV>0). The present value ranking depends on the market rate of the interest – the rate at which earning can be reinvested.

Keynes also stressed the importance of expectations in determining investment since it is the expectation that determines the rate of return and thereby any change in expectation would shift the Marginal Efficiency of Capital (MEC). Due to frequent changes in expectation the investment behavior shows wide fluctuations. The Keynesian theory explains investment function with respect to the interest rate. It relates the marginal efficiency of capital (m) with the real rate of interest (r). The marginal efficiency of capital is defined as that rate of discount which equates the present value of net returns to the cost of capital. It declines with an increase in the price of capital and increases with the price of output as well as the quantity of output.

m = m (Pk, K, PQ, Q) 3.1 m = marginal efficiency of capital Pk = price of capital K = capital PQ = price of output Q = output


126

The optimal capital stock can be expressed as a function of (r), (Pk), (PQ) and Q K* = K (r, Pk, PQ, Q) 3.2 Hence PQ = f (Q) So the optimal capital stock function equation becomes

K* = K (r, Pk, f (Q), Q) 3.2(a)

Increase in output leads to increase in the level of desired capital stock, hence the partial

derivative of (K*) with respect to (Q) shall be positive. However the partial derivative of (K*) with to the price of output shall be negative. The combined effect of these two variables shall be indeterminate. Hence

dK* / dQ = )(( */ ) ( */ / )k kk Q K P P Q 3.3

dK* / dQ = )( */ ) ( */ kk Q K P / ( )f Q 3.3(a) This equation can be rewritten as K* = h (r, Pk, Q) 3.3(b)

Expressing the function in linear form gives

*

0 1 2 3t kt tK r P Q 3.3(c)

Where,

1 2 30, 0, 0

Net investment can be written as

1 2 3 1 1 2 1 3 1t t kt t t kt tNI r P Q r P Q 3.4

Depreciation is proportional to the capital stock in the previous period

1 0 1 1 2 1 3 1t t t kt tD K r P Q

3.5

Gross investment equals net investment and depreciation therefore

0 1 2 1 1 2 1 3 1(1 ) (1 ) ( )t t kt t kt t tI r P r P Q Q 3.6

0 1 1 1 2 1 3t t t kt tI r r P Q 3.7

1 1 2 30, 0, 0, 0

It is an accelerator model as it shows the relationship between the level of net investment and growth rate of output.

3.3. Model for Public Investment in Large-scale Manufacturing Sector Investment in public sector industries is financed by number of sources: like government

grants, government direct investment through budget, loan and advances (budgetary and non-budgetary) and sponsors own equity. The function of public investment in manufacturing (large-scale) is as follows:

Igm = f (Vm, Igm(-1), Ipk, Dps) 3.8 The corresponding regression/econometric equation of the above given function is given

below:

0 1 2 ( 1) 3 4gmt t t t ps gmtI Vm Igm Ipk D 3. 8(a)

Where, the expected sign of the coefficient are;

1 2 3 40, 0, 0, 0

Here the variables used are; Igm = Public Investment in Large-Scale Manufacturing Sector Vm = Value-Added in Large-Scale Manufacturing Sector Igm = Lag Value Public Investment in Large-Scale Manufacturing Sector Ipk = Index of Price of Capital


127

Dps = Dummy Variable for Political Stability, (D=1 if the observation belongs to political stability period, and (D=0, Otherwise)

4. Results and Discussions 4.1. Applying the Unit Root Test Firstly, to search for the most suitable regression techniques and model to analyze truly the

picture of the data depend upon the stationary of the data that are checked through unit root tests. As this research study consist on the time series data, most suffers from non-stationarity. Therefore, the unit root test ADF is applied to check the stationarity of the data. The variables show their stationarity at level as shown in the table 4.1.

The ADF unit root test is chosen for the stationarity of data the Augmented Dickey Fuller is best in the case of large samples. The best estimator chosen to test significance is the comparison of the ADF test value and the critical value selected at 0.05% or 95% confidence interval. The ADF tests applied on all the variables to check stationarity. The variables show the stationarity at level form (with and with-out trends). The results are incorporated in table

Table 4.1: The unit root test results

Variables in Full form Proxy for

Variable Augmented Dickey Fuller

Critical Value (ADF)

Public Investment in Large-Scale Manufacturing Sector

Log(Igm) -3.84 -2.967

Value-Added in Large-Scale Manufacturing Sector

Log(Vm) -5.65 -2.967

Lag Value Public Investment in Large-Scale Manufacturing Sector

Log(Igm (-1))

-4.98 -2.967

Index of Price of Capital Log(Ipk) -5.21 -2.967

Dummy Variable for Political Stability Log(Dps) -4.72 -2.971

Change in Domestic Credit Log (ΔDcp) -6.71 -2.971

4.2. PUBLIC INVESTMENT IN LARGE-SCALE MANUFACTURING SECTOR Public sector investment in Large-Scale Manufacturing sector is very essential for the country

and plays an important role in the Gross Domestic Product. The investment in large-scale manufacturing sector has been analyzed in terms of Value-added in large-scale manufacturing sector (Vm), and Lagged Public Investment in Large-Scale Manufacturing Sector (Igm(-1)), Index of Price of Capital (Ipk), Change in Domestic Credit available to Public Sector (ΔDcp) and dummy (Dps) for political stability and favorable economic conditions of the country. The regression result of the estimated function for Public Investment in Large-Scale Manufacturing Sector is reported in below table 4.2.

Table: 4.2. Regression Results of Public Investment in Large-Scale Manufacturing Sector as

Dependent Variable are (The variables are taken in their logarithmic form):

Variables Coefficient Std.Error t-statistics Prob.

Constant 2.025491 0.513866 3.999915 0.0005 Log (Vm) 0.050692 0.013229 3.831880 0.0002 Log (Igm(-1)) 0.073417 0.017573 4.177726 0.0000 Log (Ipk) -0.066982 0.025419 -2.627621 0.0002


128

Log (ΔDcp) 0.125472 0.016857 7.443139 0.0000 Log (Dps) 0.193383 0.032415 5.965874 0.0000

R-squared 0.943781 Durbin-Watson stat 1.897688 Adjusted R-squared 0.930995 Prob(F-statistic) 0.000000

The result in table 4.2 shows that the overall model are significant having the R-squared

value is (0.94) and the Durban-Watson value is (1.89), both are very close to the desired values. The coefficient values of Value-Added in Large-Scale Manufacturing Sector and Lagged Public Investment in Large-Scale Manufacturing Sector are statistically significant having positive signs. On the basis of this, it can conclude that these two are the major determinant of Public Investment in Large-Scale Manufacturing Sector. The Dummy variable for stable political and economic condition is highly significant, explaining that a stable government with sound economic policies and implementation will play keen role an increasing the level Investment in Large-Scale Manufacturing Sector. The coefficient value of Value-Added in Large-Scale Manufacturing is (0.05) and that of Lagged Public Investment in Large-Scale Manufacturing is (0.073). One percent increase in these variables will increase the outcome in Public Investment Large-Scale Manufacturing Sector by five (05) and Seven (07) percent. The coefficient of Change in Domestic Credit available to Public Sector is (0.12), which means that one percent increase in credit available will bring twelve (12) percent increase in public investment in Large-Scale Manufacturing Sector. Similarly, the result also shows that Index of Price of Capital is negative effect on private investment in Large-Scale Manufacturing Sector. Index of Price of Capital can also be used as opportunity cost of capital. So, the key factors which seem to have a strong role in determining the private investment in this sector include the price level of capital goods, capital stock and the output level.

The autonomous investment is also positive and statistically significant, means that it has playing an important role in Public Investment in Large-Scale Manufacturing Sector.

Conclusion & Recommendations The research study has been an attempt to identify the factors, that affect domestic

investment significantly and which can be used as policy variables to get the desired results for public investment or capital formation, and in determining the investment behavior in Pakistan. The results reveal that demand is the most important determinant of investment activity, employing that the ‘accelerator model’ explains the investment behavior quite significantly. The result shows that the level of output and costs are quite important in determining the investment behavior in any country.

In addition to these, a number of factors such as credit availability, profitability, government policies, capital stock held by the public manufacturing sector also have significant impact on domestic physical capital formation or investment. The level of political stability and favorable economic condition of a country and their possible effects towards government policies are also captured through dummy variable. The result indicates that this qualitative factor is considered as the important determinant of public investment in manufacturing sector of Pakistan. The private investment declined sharply during early 1970’s and remained very low till early 1980’s due to nationalization policy and the lack of confidence among business community resulting from uncertain political condition and unfavorable economic conditions. After this period, through the process of privatization the private investment regained some strength due to visible tendency towards political stability and inclination towards encouragement of private sector.

Public investment in Large-Scale Manufacturing sector increased significantly during later years in 1990’s and early twentieth century. But there is need of further research to find out the factors which turn out to be relatively more important determinants for investment in Large-Scale Manufacturing sector. The most important factor is the political and economic condition of a country as well as the government policies of investment in public Large-Scale Manufacturing sector. Value-added in large-scale manufacturing sector and previous Investment decision also affect the investment level in this category. By increasing domestic purchasing power, export expansion and import substitutions through assets redistribution the fruitful outcome from public


129

investment in manufacturing sector may be increased more significantly. But all these policies may be difficult to implement in that they have other impacts on the domestic economy. Export expansion and import substitution may be quite desirable for increasing demand. Government should take into consideration not only the conditions in domestic economy but also the international economy and finally the assets redistribution may be difficult politically.

In addition to these pricing and credit policies also seem to be quite important. A democratic political system and its stability are also the necessary conditions to convince the business community that the government’s industrial policy is based on nation’s aspirations rather than being derived from some temporary motives of a dictator. Although the nation at present is experiencing a positive change in our politics, there is a lot more to be done to realize a mature democracy and political stability in the country. A businessman is primarily interested in maximizing his expected profits with the minimum risk. In the absence of political stability the businessman cannot predict future and he would remain quite uncertain of the outcomes of his business adventure. Under such uncertain conditions the businessman would avoid risk that would affect the investment severely and negatively.

References: 1. Ahmad, I. (2009), "Role of Public Expenditures and Macroeconomic Uncertainty in

Determining Private Investment in Large Scale Manufacturing Sector of Pakistan", International Research Journal of Finance & Economics, 14 (5).

2. Almon, S. (1968), “Lags between Investment Decision and their Causes.” Review of Economics and Statistics, 50 (1-4)

3. Amjad, R. (1976), “A Study of Investment Behavior in Pakistan, 1962-1970”. The Pakistan Development Review, 15(2): 134-153.

4. Bourneuf, A. (1964), “Manufacturing Investment, Excess Capacity and the Rate of Growth of Output”, American Economic Review, 54: 607-625.

5. Branson, W. H. (1989), Macroeconomic Theory and Policy, Third Edition, New York: Harper and Row Publisher.

6. Burney, N.A. (1986), “Sources of Pakistan’s Economic Growth”, The Pakistan Development Review, 25.

7. Duesenberry, J. S’. (1965), “Business Cycles and Economic Growth”, New York: McGraw- Hill Company.

8. Amber Fatima. A (2011), "Effects of Macroeconomic Uncertainty on Investment and Economic Growth: Evidence from Pakistan", Transition Studies Review, 25 (3).

9. Gould, J. P. (1968), “Adjustment Cost in Theory of Investment of the Firm”, Review of Economics Studies, 35.

10. Government of Pakistan (Various Issues), Economic Survey, Ministry of Finance, Economic Advisor’s Wing, Islamabad.

11. Government of Pakistan (2002), National Accounts of Pakistan 2001-2002, Federal Bureau of Statistics, Statistics Division, Islamabad.

12. Government of Pakistan (2002), Capital Formation in Pakistan 1999-2000, Federal Bureau of Statistics, Statistics Division, Islamabad.

13. Government of Pakistan (2003), Pakistan Investment Guide, Ministry of Industry and Production 2001-2002, Federal Expert’s Advisory Cell, Islamabad.

14. Griliches, A. and Wallace (1965), “The Determinants of Investment, Revised”, International Economic Review, 6 (3).

15. Grinblatt, M. and Matti, K. (2001), “What Make Investor Trade”, The Journal of Finance, 56 (2): 589-616.

16. Grunfeld, Y. and Grilihes, Z. (1960), “Is Aggregation Necessarily Bad?” Review of Economics and Statistics.

17. Guisinger, S. E. and Shahnaz, K. (1978), “The Rental Cost of Capital for the Manufacturing Sector, 1959-1960 t0 1970-1971”, The Pakistan Development Review, 17 (2):387-407.

18. Kamal, A. R. (1976), “Consistent Time Series Data Relating to Pakistan’s Large scale Manufacturing Industries”, The Pakistan Development Review, 15(1).

19. Kamal, A. R. (1976), “Sectoral Growth Rates and Efficiency of Factors Use in Large scale Manufacturing Sectors in West Pakistan”, The Pakistan Development Review, 15(4).

20. Kenway, P. (1996), “Too Little Investment: Why Investment is Low, Why That Matters and What New Labor Government Could Do about It”, The Review of Economics and Statistics, 36 (2).

21. Koyck, L. M. (1954), “Distributed Lags and Investment Analysis”, North-Holland, Amsterdam.

http://search.ebscohost.com/login.asp?r=29.0062392676006&svr=3&lang=en_us&x?direct=true&db=ECD&AN=39877390&site=ehost-live&EPSource=esi



http://dx.doi.org/10.1007/s11300-011-0188-0

http://dx.doi.org/10.1007/s11300-011-0188-0


130

22. Mamatzakis, E. C. (2001), “Public Spending and Private Investment, Evidence from Greece”, International Economics Journal, 15 (4): 33-46.

23. Resek, R. W. (1966), “Investment by Manufacturing firms: A Quarterly Time Series analysis of Industrial Data”, Review of Economics and Statistics, 48(3):322-333.

24. Siddiqui, R (1993) “ Investment Behavior in Pakistan” The Pakistan Development Review 32:4, 1281–1292.

25. Vass, G. M. (2002), “Public and Private Investment in United States and Canada”, Economic Modeling, 19(4): 641-655.

26. Wai, V. T. and Wong, C. (1982), “Determinants of Public Investment in Developing Countries”, The Journal of Development Studies, 19 (1).

27. Wang, T. (2001), “Equilibrium with New Investment Opportunities”, Journal of Economic Dynamics Control, 25(11): 1751-1773.

28. White, W. H. (1956), "Interest Inelasticity of Investment Demand," American Economic Review, Vol. 46, No. 4, pp. 565—587.

29. World Bank (1997), “Training Manual: Macroeconomic Adjustment and Stabilization”, Vol.2, Washington D. C., Economic Development Institute, World Bank.

30. Zaidi, S.A. (1999), Issues in Pakistan Economy, London: Oxford University Press.

УДК 33

Факторы, влияющие на государственные инвестиции в производственном секторе Пакистана

Гюльзар Али

Школа экономики, Хуажонге университет науки и технологии, Китайская Народная Республика Аспирант E-mail: [email protected]

Аннотация. Государственные инвестиции в производственный сектор набирают обороты во

всем мире. В Пакистане государственные инвестиции внесли существенный вклад в различные секторы экономики и играют основополагающую роль в финансовой системе страны. В этом исследовании была предпринята попытка выявить факторы, наиболее значительно влияющие на государственные инвестиции в производственном секторе в Пакистане. В исследовании использованы данные за период с 1981 по 2014 годы. При анализе временных рядов всегда остается подозрение о фиктивных отношениях. Это исследование также основывается на данных временных рядов, поэтому перед началом процесса оценки модели, данные проверяются путем применения дополненной модели Дики-Фуллера (ADF), но данные, использованные в исследовании, не свидетельствуют ни о каких фиктивных отношениях. Для того чтобы отследить влияние различных факторов на государственные инвестиции в производственном секторе, используется модель акселератора инвестиционного процесса, и регрессируется через модели NLS и Аrima. Исследование рекомендует правительству создать достаточно спроса за счет увеличения покупательной способности внутри страны путем расширения экспорта, а также замещения импорта через перераспределение активов. Исследование показало, что стабильные политические условия также необходимы для быстрого экономического и инвестиционного роста.

Ключевые слова: государственные инвестиции, производственный сектор, и NLS Arima модели, модель акселератора инвестиционного процесса.


131



www.ejournal2.com UDC 33

Impact of Tourism on Spatial Transformation: a Case Study of the Bela Crkva Muncipality (Serbia)

1 Jovana Boškov

2 Stefan Kotrla 3 Dajana Lulić

1-3 University of Novi Sad, Serbia Faculty of Science, Department of Geography, Tourism and Hotel Management Trg Dositeja Obradovića 3, 21 000 Novi Sad E-mail: [email protected]

Abstract Muncipality of Bela Crkva is located in the northeastern part of Serbia, at the southeast end

of the Autonomous Province of Vojvodina and Banat. In recent years, tourism in this Municipality is expanding, creating certain effects. The effects of tourism on spatial transformation are still not noticeably, but there are some changes which are present.

The aim of this paper is to show spatial transformation that is created as a result of the development of tourism in the municipality of Bela Crkva.

Keywords: tourism, spatial transformation, municipality of Bela Crkva, Serbia. Introduction Tourism has become one of the world's leading industries. Growth and development of the

so-called tourism industry, despite numerous positive effects (economic, social, cultural and others), realized a series of negative consequences. These negative effects are primarily related to environmental degradation, transformation of area, as well as problems in the life of local communities in a tourist place. In this regard, there is a plethora and regression of some destinations, destruction of local cultures, traffic problems and dissatisfaction of the local population. People with their activities, including the activities in the sphere of tourism, are exhausting natural and anthropogenic resources much faster than they can recover. Having in mind the fact that most of the tourist activities depends precisely on these resources it is clear that it is necessary to protect and preserve them for the future. Awareness of the negative consequences of tourism on environment comes in late 20th century. Many tourism businesses are turning to a new way of doing business, which is focused on caring for the environment, different programs are applied, bringing documents and plans aimed at the sustainable development of tourism. However, the current development of tourism has greatly influenced the degradation of the environment and transformed the space to the extent that some destinations completely lost their original appearance and function.



132

Serbia can’t be characterized as a country with a developed tourism, yet the problem of spatial transformation due to its development exists. This problem is present to a greater or lesser measure in certain destinations in Serbia and one of them is the municipality of Bela Crkva.

Relationship of tourism and space The space is unique, inseparable, entirety of physical conditions, natural and man-made

resources and values. It represent basic resource of existence, development and survival of human society. Space is a system in which different subsystems coexist (natural, economic, social, technical, infrastructural, political). Not one of them can exist and develop independently by itself and outside particular space (Maksin, 2012).

Tourism is more than any other industry conditioned by space. Space in tourism can be seen as a framework in which tourism is cycling, and that involves tourists traveling from their place of residence to the tourist destination, and returning to their home city. On the other hand, space is the objective of the trip, because it contains the potential attractiveness to tourists and motivating them to travel. It is a place where tourist demand and supply meet and where expectations of all stakeholders in tourism come to expression (Tomka, 2006). Tourism represents a link between the urban centers and areas with native, preserved nature and unchanged space. Increasing pollution of urban and industrial centers reinforces the need for tourist travel aimed at non urban environment (Stojanović, 2011). Tourism is a kind of space consumer, which often uses those objects and surfaces that are not of interest to some other industry. Degradation of these objects and surfaces can lead to irreversible loss of tourist potential, or to disable the tourism activity (Stanković, 1998). If tourism continues to transform the area in which it takes place in the manner and to the extent that it is today, and leads to saturation and decline of tourist destinations survival of this industry is questionable.

Figure 1. Location of Bela Crkva municipality in the Autonomous Province of Vojvodina Source: Boškov at all, 2015

On the other hand, there is a certain dependence of space from tourism. Thought here is

primarily on economic and social dependence. Many tourist destinations, and the population of


133

those destinations achieved great economic benefits thanks to the development of tourism. Tourism in this region encourages the employment of the local population and creates new enterprises. It also leads to cultural exchange, and local population is introduced to different culture that tourists from different parts of the world bring. On the other hand tourists get acquainted with the local culture of the place where they reside. What is also important when it comes to dependence on space and tourism, is the fact that some forms of tourism through its activities contribute to raising awareness of the threats to the natural heritage and allow a better understanding of the need for its protection.

However, despite being largely dependent on space, tourism constantly contributes to its transformation. Under the influence of tourism, receptive place is increasingly changing, gaining specific tourist physiognomy. Tourism has the power to make each potentially attractive space more accessible for tourists. In this way tourism as a factor of transformation of the space plays a major role (Čomić, 2002).

Study area Bela Crkva municipality is located in the northeastern part of Serbia, in the southeast of the

Autonomous Province of Vojvodina and Banat district. It covers an area of 353 km2 and territorially belongs to the South Banat district. On the north it borders with the municipalities of Kovin and Vršac, while on the east it borders with the Republic of Romania (Devrnja et al., 2015).

Despite its peripheral position, the Municipality is well connected with contiguous

municipalities. The main form of transport and the most important communication of this area is road traffic. The road network of this Municipality consists of 26 km of highway, 21 km of regional and 39 km of local roads. The distance from Belgrade, the capital and most important city center is about 95 km. The distance of other centers is: Vršac (37 km), Kovin (47 km), Požarevac (50 km), Smederevo (60 km), Pančevo (80 km). Over the border crossing Kaluđerovo, Bela Crkva is connected with border villages in Republic of Romania. The distance from Timisoara (Romania) is about 110 km (Boškov, 2014b).

On the territory of the Municipality there are numerous natural values, and some of them are protected, such as Special Nature Reserve "Deliblatska sands", Ramsar site Labudovo okno Landscape of exceptional features "Karaš-Nera" (Boškov, 2014a).

Examples of spatial transformation to the impact of tourism When it comes to the transformation that has occurred as a result of the development and

operation of tourism in a given area, it should be noted that it is not only about changes in physical space. The impacts of tourism are complex and should be distinguished ecological, socio-cultural and economic effects of tourism (Stojanović, 2006). It can be said that in the case of the Bela Crkva, socio-cultural effects of tourism are not largely expressed. Culture and lifestyle of the local community has not changed. So far there have been no cases of conflicts between tourists and locals, the deliberate destruction of cultural monuments and architectural value, nor any inappropriate behavior of tourists. Economic effects are express to a certain extent. They are most expressed in period of the most famous events ("Karneval cveća", "Bela Crkva u jabukama", "Lov na Besija"), and in the summer, during the bathing season on Belocrkvanska lakes. On the other hand, tourism in this region realized a number of environmental impacts, and impacts on the physical spatial transformation.

Environmental impacts of tourism Studies on ecological transformation mainly emphasize the changes that tourism makes on

the nature and ecosystems of a landscape (Stojanović, 2011). In the case of the Bela Crkva, there are numerous ecological problems. Some of these problems have been created solely as a consequence of the development of tourism, while most of them are caused by a combination of tourism activities, neglect of the local population, as well as the inactivity of local government in the field of preventing and solving these problems.

Plastic waste in the rivers Nera and Karaš: is caused by great irresponsibility of tourists. Hikers and bathers on these rivers during the summer months, leave behind large amounts of plastic waste, which together with fallen trees retained in certain places, make a kind of dam.


134

It should be noted that tourists are not the only ones to blame for this problem. Local residents showed an equal level of negligence and irresponsibility, as well as local authorities which refuse to take specific measures to address this problem.

The emergence of garbage on the beaches of the Gradsko and Vračevgajsko lake: these two lakes, the most visited tourist sites of Bela Crkva, are also the most polluted. During the bathing season the beaches are full of garbage that tourists leave behind.

Fishing with illegal means: on the river Nera, Karaš and the canal Danube-Tisa-Danube was recorded illegal fishing with electricity, explosives and "hooks", which leads to the question of fish population (Kotrla and Belobabić, 2014). For this problem mostly are guilty locals, however, there are a number of tourists, visitors of these sites, which also contribute to the spread of this problem.

Ramsar site Labudovo okno: this Ramsar site includes the shores of Danube and Nere, islands Žilavu, Čibukliju and Zavojsko, sunken meanders of Karaš, the Nera river confluence of the Danube. Flow of Danube in this sector is slowing down, it has a higher water level, flooding the low coastal terrain and extreme southern edge of the Banat sands. Coastal river swamps have conditions for different types of aquatic biodiversity, and also for species typical for damp areas such as these. For Ramsar site Labudovo okno was declared on 01. 05. 2006. The protected area covers an area of 3,733 hectares (www.ramsar.org). Protection of this area contributed to establishment of tourism infrastructure. At the site there is an information board, viewpoint, and in the immediate vicinity there are several significant anthropogenic values such as remnants of the Smederevo and Ram fortress, Roman Castrum and others. Tourism in this area exists, although the overall potential is not used enough. Tourist visits are realized mostly through school field trips or individuals and small groups who come alone (without the intermediation of tourist organizations and travel agencies). These visitors are largely environmentally conscious so that environmental pollution by tourism is minimal.

Impacts of tourism on the physical transformation of space This group includes the visible impact of the changes that tourism is realized in the

municipality of Bela Crkva, which are due to the construction and adaptation of space so that it subordinated the needs of tourism, such as the construction of certain buildings, concreting works, arranging beaches and resorts. It should be noted that these effects are directly intertwined with environmental impacts that tourism accounts.

Excessive urbanization of Glavno lake: in past few years a large number of prefabricated buildings have been built near the Dečija beach on the Glavno (Gradsko) lake. On the western side of the lake, in the former forest of plane trees, there are numerous apartment buildings. Some of the restaurants on the main lake are: Resort "Rafaelo", bungalows "Jezero" Camping "Belocrkvanska lakes" in which there are basketball, volleyball and football courts, then a restaurant "Jezero", cafes and numerous villas. Most beaches on the lake are made of concrete. In order to prevent further urbanization of the Main Lake, it is necessary to direct the construction of tourist facilities to other lakes (Kotrla and Belobabić, 2014).


135

Figure 2. Tourist villas on the Glavno lake

Photo: J. Boškov

In this way further urbanization of the area around Glavno lake would be stopped and that would also activate tourism on other lakes in Bela Crkva. Balanced tourism development at all (or at least most) Belocrkvanska lakes would contribute to preserving biodiversity at each of these places.

The facilities built here do not fit into the natural environment. The facades of bright, varied colors are not in accordance with the environment or with one another. Colorful tourist villas give the impression of confusion and deviate greatly from the ambient environment.

Urbanization on Vračevgajsko lake grows rapidly: however, in comparison to Glavno lake urbanization is not so expressed here. Auto-camp "Bela Crkva" is located on the lake coast. Capacity is 100 camping units, it has parking space for about 400 vehicles. In the camp there are beaches, football field and volleyball, shop, cafe and restaurant (Boškov, 2014a). As already mentioned, in order to prevent excessive urbanization, it is necessary to divert tourist activity to other lakes in Bela Crkva.

Illegal cottages on the Dunav–Tisa–Dunav: on some parts of the channel, at the dam itself, there are illegally buildings. Currently, there are 15 objects constructed about 50-70 meters from the confluence with the Dunav canal. These objects are damaging the natural environment and represent a danger when floods occur. Waste waters from these facilities are also a problem, alongside waste that remains after staying visitors. Most of this waste ends up in the channel.

Construction of weekend houses in the Siga area: this source of cold water is also known as Pricentalov valley. In Siga today there are remnants of old oak forest, which covered this area. Provincial Institute for Nature Protection has registered seven specimens of old oaks between 200-300 years. Due to the high humidity in the vicinity, large numbers of plants typical of humid areas are present. It is clear that the construction of weekend houses in Siga area, which are 5 at the moment, leads to a distortion of visual beauty of the landscape, but there is also the danger of disrupting the ecosystem in this area. Lack of sewerage network, concreting soil, noise, construction on the former forest glades have negative consequences for the flora and fauna of Siga (Kotrla and Belobabić, 2014).

Impacts of tourism on the cultural heritage In the area of Bela Crkva municipality impacts of tourism on cultural heritage are not largely

expressed. One of possible reasons for this is the fact that tourists in Bela Crkva more visit natural sites (lakes, river banks), and in these localities impacts of tourism are more visible. Numerous cultural values in Bela Crkva, such as archaeological sites, churches (Roman Catholic, Romanian Orthodox, Russian Orthodox, Evangelical), the National Museum, the National Library, the mill on Nera river, monument of technical culture, as well as numerous events have not yet been


136

sufficiently activated for tourism purposes. Careful planning of tourism development and attracting tourists can be achieved by the including these sites in tourist activities. It is of great importance their sustainable use.

Water mill on Nera river: this monument of technical culture today is completely ruined and is a true example of neglect of the authorities for the protection and promotion of cultural heritage. The whole area around the former watermill is very shabby, water mill wheel is missing, there is only building in ruins. Given the current state of this cultural monument, it is clear that it is not used for tourist purposes. During the field survey conducted in July 2014 in Bela Crkva for the purpose of making a study "Geoheritage of Bela Crkva", local residents and tourists answered several questions related to the water mill on the river Nera. Respondents mostly did not even know that this facility exists.

In the 19th century from the basin of Nera river to Stara Palanka about twenty water mills and mills existed. Over time, they were destroyed or burnt down, and this water mill is the only one remaining. The reconstruction and development of this property can be its inclusion in tourism. The inclusion in the tourist offer would allow the protection of monuments of culture and prevent the traditional way of doing business in oblivion. In the case of renewal of the water mill also would be necessary to promote it through various promotional materials and media, because as already mentioned this facility is largely unknown to the public.

Figure 3. Water Mill on Nera

Source: Historical Archives of Bela Crkva

The architecture of the city center: it is a noticeable impact of tourism on the architecture of the old core of Bela Crkva. Tourism development has caused the opening of a large number of restaurants in the city center. In order to allow the construction of cafes, restaurants, fast foods restaurants and similar objects, a number of old buildings were torn down or their original condition was changed. The beautiful facade of the famous kibic windows are very rare now. Instead street in the center of Bela Crkva now adorn the outdoor cafe and shop windows which are completely inept in architectural solutions


137

Figure 4. Details of architecture in Bela Crkva

Photo: J. Boškov

Tourist events as part of the cultural heritage play an important role in tourism of Bela Crkva. The most important and most visited tourist event is "Carnival of Flowers". This event has a tradition of 150 years and it is every year in June. After the carnival in Bela Crkva the most important manifestations are "Hunt on Besi" (first time held in 1988) and "Bela Crkva in apples" (Boškov, 2014a). The impact of tourism development on the events reflected primarily in providing various products for tourists, which originally were not an integral part of the event (souvenirs, food and beverage offer, etc.). Events, especially the "Carnival of Flowers" every year attracts large number of tourists, which is why their importance is growing, and also to achieve greater economic effects. For the "Carnival of Flowers" can be said to represent a kind of brand of Bela Crkva.

Conclusion Bela Crkva has rich natural and anthropogenic resources which have great scientific,

educational, landscape and tourist value. These values are the result of complex geological structure and geomorphologic processes in the course of Earth's history took place in this area, as well as rich and turbulent history of these areas. In the municipality there are numerous natural resources and several of them are protected. There is a plateau Dumače which enters into the composition of the Special Nature Reserve "Deliblatska peščara”, Labudovo okno which is protected as a Ramsar site, and now is during the process of forming a landscape of exceptional characteristics "Karaš-Nera" as a protected area in second category, with a total area of 1.541,27 ha. This area also has a number of anthropogenic values and the most importantes are archaeological sites. The rich heritage provides good conditions for the development of different forms of tourism, and so far in this area to some extent developed swimming, mountaineering, event, sports and recreational tourism. In Bela Crkva tourism has not largely influenced the transformation of the area, but the changes exist. On some sites they are more pronounced, while in some locations are in the beginning. The area around Glavno lake and spring Siga suffered the biggest transformation, as a result of tourism development.

In order to adequately respond to demands of modern tourism, it is necessary for tourist destination to accept the changes that it brings. It is extremely important that there is monitoring of the impact of tourism. Any changes that tourism brings in one area inevitably changes the


138

natural environment, but with careful planning of tourism development that can be prevented, or at least to limit the negative aspects of tourism as an agent of spatial transformation. On the other hand, the careful planning of tourism, a specific area can be revitalized, protected and enhanced.

References: 1. Boškov, J. (2014b) Geoheritage of Bela Crkva municipality. Citizens Association “Aurora”,

Bela Crkva, 91 p. (in Serbian) 2. Boškov, J. (2014b): Tourist presentation of the geoheritage in Bela Crkva municipality.

Department of Geography, Tourism and Hotel Management, University of Novi Sad. (in Serbian) 3. Boškov, J., Kotrla, S., Tomić, N., Jovanović, M., Rvović, I. (2015): Perspectives for

geotourism development in the Bela Crkva municipality (Serbia). Acta Geoturistica, volume 6, 1, 1-10.

4. Devrnja, D., Boškov, J., Kotrla, S., Rvović, I., Belobabić, M. (2015): Bela Crkva municipality as a future destination of adventure tourism in Serbia. Proceedings of the V International Conference of Ecotourism, National ecotourism association, Sremska Mitrovica. (in Serbian)

5. Kotrla, S., Belobabić, M. (2014): Environmental protection of Bela Crkva municipality-perspectives and problems. Citizens Association "Aurora", Bela Crkva.

6. Maksin M. (2012): Tourism and space. Department of Tourism and Hospitality Management, University Singidunum, Belgrade.

7. Stankovic, S. (1998): Tourism and space – C omplementarity and collision. Tourism, Vol. 2. 8. Stojanović, V. (2006). Sustainable Development of Tourism and Environment. University

of Novi Sad, Faculty of Science, Department of geography, tourism and hospitality, Novi Sad. 9. Stojanović, V. (2011): Tourism and Sustainable Development. University of Novi Sad,

Faculty of Science, Department of geography, tourism and hospitality, Novi Sad. 10. Tomka, D. (2006): Tourism basics. Faculty of sport and tourism, Novi Sad. 11. Čomic, Đ. (2002), Geography of closed circle. Higher hotel management school, Belgrade. 12. Historical Archives of Bela Crkva. 13. www.ramsar.org

УДК 33

Влияние туризма на пространственную трансформацию: тематическое исследование муниципалитета Бела-Црква (Сербия)

1 Джована Босков 2 Стефан Котрла 3 Дайана Лулич

1-3 Университет Нови-Сад, Сербия Факультет естественных наук, кафедра географии, туризма и гостиничного менеджмента E-mail: [email protected]

Аннотация. Муниципалитет Бела-Црква находится в северо-восточной части Сербии, на юго-

восточной части автономного края – Воеводине и Банате. В последние годы туризм в этом муниципалитете расширяется, создавая определенные предпосылки. Воздействие туризма на пространственную трансформацию до сих пор не столь заметно, но уже можно отметить какие-то изменения.

Цель данной работы показать пространственную трансформацию, которая создается в результате развития туризма в муниципалитете Бела-Црква.

Ключевые слова: туризм, пространственная трансформация, муниципалитет Бела Црква, Сербия.

http://www.ramsar.org/


139




Coal Mining and Indigenous Communities: А Case Study of Jharia Coalfields

Sribas Goswami Department of Sociology, Serampore College, West Bengal, India PhD, Assistant Professor E-mail: [email protected]

Abstract Mining is indispensable for the individual, for the society, and for the development of the

nations. Unfortunately, mining procedures and operations are often associated with health hazards and environmental deterioration. Present study has been attempted from a socio-economic point of view and the dynamics of the environment of the coal-mining region has been focused upon while keeping in mind what Gerasimov has said, "The purview of ecological approach has been enlarged to digest relevant information and results of studies in biology, sociology and anthropology etc. under such a changed set-up, Geography has equally emphasized aspects of spatial variation and relationship and biological science are no more the sole custodian of ecological approach it has rather displayed a well-marked tendency to become in other fields of science". This study has come up with issues related to harmful effects of mining and how trace elements influence the local environment and may affect human health in the vicinity of the mining area.

Keywords: coal mining, health hazard, environment, respiratory diseases. Introduction The mining industry is one of the three basic industries in the primary sector, the other two

being agriculture, and wildlife and fisheries. Unfortunately, the general opinion of the mining industry is often associated with the accidents: disasters and environmental degradation related to mining and particularly coal mining. Certainly, there are reasons based on incidents for such opinions expressed by the people. Mine disasters receive wide coverage from the media, whether it is an explosion or mine fire or inundations, the lives of people are touched by the personal and societal impacts of these events. In most of the cases enquiries, after these disasters happen, do not deny the fact that the disaster situation was present and could have been detected with thoughtful search. In many cases human error has been found to be the immediate cause, but it could have been avoided, if the management and planning had been more efficient in their approach. In June 2005 it was reported in a national newspaper' that fourteen miners were trapped inside a mine at the central Saunda colliery in Hazaribagh in Jharkhand after water gushed in and roof collapsed. About three million gallons of water rushed into the mine of the Central Coalfields Limited (CCL). Any rescue operation could be carried out only when water was thoroughly pumped out, which took almost a week. The outcome was absolutely no chances of rescuing the trapped miners back, alive.

Apart from the mining disasters, coalminers too are subjected to certain potential health hazards, such as dust, gas, noise, vibration which may not manifest in the form of any immediate



140

danger but sooner or later it may cause grave negative impacts on human health as many studies and researches have shown. Our detection of the potential health hazards for the mineworkers in the mine area is limited only to the extent of present level of medical advancement. As the new studies come up, they reflect certain new perspectives about influences of mine environment on mineworker's health. Of course human civilization has come a. long way from the past when technological advancement was less and so were the medical facilities particularly in the last two centuries. The health effects due to breathing high concentrations of respirable dust in coal mines are slow to develop and can only be controlled by checking high dust concentrations in the mine and making changes in the method and planning for the coal extraction procedure. Professor Ramani, the Head of the Department of Mineral Engineering in Pennsylvania State University holds the view that "The adverse health impacts in coal mines can be slow and long term in developing but once afflicted, debilitation can be progressive and horrific. The specific conditions are most important in determining and controlling the health hazard."(Ramani, 1995)

Mechanization of mining process has not contributed significantly in reducing the risk of health hazards to the mine workers and has resulted in upsetting the environment in our coalmines. "Mechanized mining systems offer high production and high percentage of recovery with improved productivity. But they produce gases and dust at higher rates. Employees of such mining systems are exposed to many health hazards due to high dust concentrations. Even in underground coal mines an airborne dust survey showed that the dust concentration levels at, many places, such as where blasting is done, are beyond statutory limits (Sastry, et al 2000). The coal mining process affects adversely not only its immediate environment but through river channels and air transportation, dust particles can travel quite a distance from the mine area. According to Jones (1993) "The majority of coal-related projects have the potential to affect the environment to a significant degree. Open cast methods of coal extraction can directly affect terrestrial and aquatic ecosystems."

Objectives of the Study: a. Societal impact of coal mining techniques in India b. Environmental impact of coal mining c. Impact of coal mining on human health Methodology The present study is an empirical research conducted in one major coalfield namely Jharia

Coalfields in India. Jharia coalfield has two subsidiaries, BCCL (Bharat Coking Coalfield Ltd) and CCL (Central coalfields Ltd). The methodology of the present study includes collection of research material over the field study and direct observation methods. The present research is based on both primary as well as secondary data. Primary data have been collected from a structured interview schedule with the officers and workers of Coal India Ltd. and secondary data have been collected from CMPDI (Coal Mining Planning And Design Institute) records, monthly journals of IICM (Indian Institute of Coal Management) , books and research paper related to coal mining. The field study was conducted from the Coal India Headquarters in the year of 2014.

Environmental issues related to coal mining: How important are the environmental concerns can be judged by that many scientists

associated the recent Tsunami tragedy to global climatic change. Depletion of our forest cover and burning of fossil fuel has a significant role in increase in CO2 levels and the resultant global warming. Coal mining and its use in allied industries has a major contribution in increase in CO2 levels in the atmosphere. It is being said that recent tsunami tragedy is a window to what earth can do with its devastating capacities and it is well known what can be the repercussions of an unchecked global warming. In the Indian context the fragility of the environment can be judged from the following facts: a) Carbon dioxide emission from India are over 3% of global equivalent emissions of which about 55% are from the energy sector (road transport, burning of biomass fuels, coal mining and fugitive emissions from oil and gas). b) The Industrial sector generates about 100 million tones of non - hazardous solid waste and 2 million tons of hazardous 'waste annually, c) Nearly 23% of India's animal species have become extinct. d) Over 24000 hectare of India's forest


141

cover is lost every year and 25% of the country's area is under threat of desertification." (TERI, 2001, 2002).

Human beings are a very powerful agent of Geographical dynamics and more effective than the collective strength of rest of all the erosional or degradational forces of nature. Coal now faces increasing environmental constraints in all facets of the fuel cycle. Furthermore, the business of coal is compounded by alternatives which are gaining ascendancy in world energy scenarios. It is a formidable task to ensure that coal is mined and used in an economically efficient and ecologically sustainable manner, particularly in those countries and regions which cannot presently meet the costs of environmental protection. On one hand, the future of coal looks bright in India as we have vast resources which can enhance the steel industry and power generation which are two major coal consuming sectors of our economy. On the other side, we cannot afford to ignore another aspect which needs to be addressed equally and urgently and that is coal mining associated with the environment. Not only for India but globally coal is the most important extractive industry. "Worldwide, about 4.5 billion tones coals, all types, are mined annually with a net value of approximately £175 billion. The worlds coal industry is large and expanding, with an average growth rate of about 21.4 % p.a. throughout the

1980s. There is a significant export trade in hard coal, totaling around 0.4 billion tones p.a. which is increasingly influenced by requirements for low ash and sulfur content. The value of UK mineral production in 1988 was £15.3 billion, or a little under4%of gross domestic product (GDP). Coal dominates with some £4.3 billion in value. For the next decade and beyond, ecological sustainability will be one of the principal measures against which economic activities, including the mining of coal and its subsequent utilization, will be assessed. This will pose considerable challenges to those involved in the extraction and consumption of coal and other fossil fuels (Crowson, 1992).

Environmental aspects of coal mining: Coal Mining has multiple adverse impacts on the environment: disturbance of the land

resource, adverse effect on river channels and aesthetical deterioration of the landscape. Mine fire occurring mainly in underground coal seams and the effect on the land, water and air due to refuse created from mining and coal preparation units. The environmental implications of energy use arise from the fact that nearly 90% of the primary energy consumption comes from combustion of fossil fuels. The most direct environmental impact of fossil fuel use is an increase in air pollution levels and production of Green house gases increasing the threat of global warming and this is besides the land degradation due to mining, water pollution, and vibrations due to blasting adverse impact on the health of the mineworkers and of the people living in the adjoining areas. In Jharkhand Coal bearing area is spread over a vast geographical extent. Practically all coalfields are located in major river basins. Damodar river basin shares almost 65 % of the coal reserves located in the river basins. It is estimated that the washery and beneficiation activities amount to dumping of 10-15% of coal into rivers such as Damodar (Kadekodi, 1988)

There are two methods mainly adopted for opencast mining: a) Area strip mining: It is done in relatively flat areas. By this method overburden is removed and piled alongside

the depression until the coal seam is reached. Then progressing further next portion of overburden is removed and filled in the initial depression. This operation is repeated and unless corrective measures are taken land stripped by this method leaves valleys and ridges.

b) Contour mining: It is preferred and practiced in a rather undulating, hilly or mountainous region. In this type

of mining at the coal outcrop in the hillside, the overburden is removed and coal is extracted by following the contour of the hillside and follows till the proportion of the overburden to coal seam thickness makes it uneconomic to mine. In this type overburden is disposed of by casting it down the hillside below the coal seam. Unless the discarded material stabilizes there are chances of erosion and landslides and also it may damage the flora and fauna downwards. "Primarily, the concern of all geographers is with the environment of man. But man cannot exist or be understood in isolation from the other forms of life and from plant life". (Strahler, 1976). "Environment refers to the sum total of conditions which surround man at a given point in space and time". (Park, 1980). Development of coalfields is essential in providing fuel for electricity generation and coke


142

for steel making. Exploitation of shallow coal reserves using open-cast mining techniques in this area involves removing the soil and rock (overburden) from the top of the coal seams by drilling and blasting, followed by removal using large earth – moving equipment (dumpers and dozers). The exposed coal is broken or directly trucked away.

Land degradation: Open cast mining has a large footprint. A mine producing 40 million tons of coal in its

lifetime (approximately 15 years) will leave a scar of about 25 sq km in area."(Herbert and Dutt, 2004). Surface Mining has more potential impact on land than underground mining. More than 80.000 hectare of land in India are affected by various mining activities (Valdiya, 1988)102.It is estimated that a total land of 539 Sq Km is expected to be disturbed through opencast coal mining during the tenth Plan period (L.K. Bose, 2003)103 There are large voids and mountains of overburden and the land is scarred and destroyed unfit for any productive use. The excavations have formed pits typically a few hundred meters long, more than 50 meters wide and up to 80 meters deep, depending on the depth of coal and the thickness of the seam. These pits are usually left as such by the mining company after the coal is exhausted. Most of these pits are filled with water. Near Religada worker's colony there are several such waterlogged pits. The water from these quarries is pumped from these quarries is pumped and supplied directly to the worker's houses. It was complained by many worker's here that there has been no filtering of the supplied water for last several years. There have been instances of dumping dead animals also in these waterlogged pits.

Environment Management plans (EMP) are part and parcel of all projects now and it would be imperative for the project officials to strictly adhere to all the stipulations as per the approved EMP's to safeguard environment in terms of land degradation, water, air, noise pollution and socio-economic issues.

Rehabilitation of mine sites: Rehabilitation of mined areas is a key phase of open cut mining and involves the use of

overburden to refill mined areas, reshaping these areas, replacing top soil, and finally sowing and nurturing vegetation. Inadequate vegetation on rehabilitated areas may result in dust generation, and also water pollution due to soil erosion and the discharge of suspended solids from the premises. Selection of vegetation species will depend on intended land use. Biodiversity of species, if local native flora and fauna are to be a feature of the rehabilitated site, 'will require vegetation to be based on seeds collected from appropriate local species. On the policy level CIL has always maintained its conviction with clarity regarding reclamation in the post mining period but it is seldom done with efficacy. "The commitment to reclamation of mined land in CIL's environmental policy is clear and unambiguous. The policy includes a commitment to progressive reclamation to achieve a post- mine land form and use consistent with the EMP, maximizing backfilling, preservation and re-use of top soil. "(Herbert and Dutt, 2004) The Environmental and Social Review Panel (ESRP 2000) says about Parej East mine (CCL) in Jharkhand, "At present virtually no effort is being made to reclaim mined land ... all the top soil resources of the mined land are being destroyed through burial in overburden dumps ... we have seen little evidence of any fundamental change in attitude to overburden management and reclamation since our first visit"

In Religada Underground mine (Argada Area, South Karanpura) which is under Central Coalfields Limited (CCL) there was underground mine fire and the mine had to be closed down due to this reason on 3151 May 2004.

Air quality: Air quality status at the coalmine site and health of the mineworkers: Sources of emissions

effecting air quality in the area are as follows: 1) Dust from opencast mining operations, for example movement of heavy earth moving machinery, drilling and blasting, etc 2) Exhaust from trucks, dumpers, dozers and shovels. 3) During loading and unloading operations of the coal and dump materials. 4) Dust generation from waste dumps and coal dumps due to wind erosion, and 5) Dust emission due to movement of trucks and vehicles in public roads.

Impact of coal mining on water resource:


143

The drainage pattern map shows 'sub-parallel' trend in the northern, eastern and south -eastern parts, but absence of drainage in the western part. Results of morphometric analysis indicate low topographic relief and high infiltration. For the environmental impact study, coal dumps, overburden dumps, fire zones, subsidence, forest, wasteland, agricultural land, water bodies and built-up areas have been selected as land degradational features. These features have been identified and delineated from individual imagery and compared with the land use/cover map. "The static ground reservoir in the whole south Karanpura Coal field is 97.60 million cubic meters. The impact of mining on hydro geo-logical regime has been assessed to be within 200 meters of the mining activities. The drying of a well beyond this distance is because the wells have been constructed up to the desired depth where aquifers are occurring." (CMPDI, 2013).

Impact of coal mining on forest resource the study reveals large number of coal and overburden dumps causing decrease in forest as well as agricultural lands, increase in waste lands, pollution and shrinkage of water bodies, rapid increase in built-up areas, subsidence, numerous fire zones and wide-spread coal dust deposits on the main drainage, Damodar River. The topsoil of the area has is formed out of outwashes products of the metamorphic gneiss (forming the high altitude hill range around the coalfield area). There are red gravelly and sandy red and yellow soils and at places there is old alluvium occurs in small patches at some places.

Impact of coal mining on land resource: River Damodar is the main source of water for cultivation and sustaining underground water

table. Pumped out water with suspended coal dust from the coalmines may cause phenolic (a toxic organic chemical) contamination to both surface and underground water. Due to coal mining the geomorphological, hydrological and land use pattern has changed. Vast areas have turned into hillocks of overburden all around the abandoned quarries due to unplanned dumping of OB materials. These practices choke the drainage flow and are responsible for partial blockade of nail a and rivulets at many places. If continued for a long time it also may cause change in the original course of the rivers this further upsets the ecological balance and hamper settlements. In many of the streams of Damodar, the surface run has decreased over the years. Surface rainwater at present, transports extensive sand /silt material to the river due to sheet erosion instead of gully erosion, causing geometric changes to the riverbeds.

Workers health and safety concerns: Globally, mining remains a difficult and hazardous job and there have been concerns related

to negative health impacts related to the mineworker." Mining remains one of the most difficult, dirty and hazardous occupations - causing more fatalities than other occupations even in United States or in Europe (2001). In terms of scientific evidence, despite studies showing long term historical improvements, particularly in the middle of the last century, the bulk of the literature focuses on the continued burden of largely preventable health impacts that mine workers sustain not just in their working life but beyond into their old - age.". Mining has been a primarily male dominated profession, needing to employ principally able-bodied individuals to undertake arduous risky work. In the study area for the present study also, there have been all male workers involved in the direct operations related to coal extraction. Though women are employed by CCL but they are either engaged in official work or at the lower level they are engaged in petty cleaning/sweeping work.

There are many potential hazards which can be perceived and their realization will take a longer course, but they definitely have an adverse affect on human health which may take from years to decades to realize. One such example is coal dust affecting human lungs and expressing much later in form of pneumoconiosis or other respiratory ailments. Similarly noise and vibration beyond acceptable limits can be annoying and it may do some harm to health. An accident can be considered as an abrupt realization of a potential hazard. If the hazard is of a larger intensity then it may be said to have' disaster potential’. For example overburden dumps in the mine may let's itself loose or its movement/displacement during rains may cause some accident, or due to air movement can cause a lot of air pollution etc. Therefore the location and the dumping site condition need to be chosen carefully and close monitoring and control is also required. Inundation and Mine fire are some of the realizations of the potential hazards. Various studies show that it is difficult to detect the ailments during the period of working years and mariy ailments are detected


144

after decades. Therefore this study looks into the perception of the mineworkers about their exposure and awareness about various health hazards associated with coal mining. As studies have shown that, there are occupational health hazards in coal mining and there is an association between dust and gas emissions in the coalmines and several respiratory and other forms of ailments. Though situation may differ from place to place, depending on whether sufficient steps have been taken by the mining company to ameliorate any situation of health hazard and adverse impact of coal mining on the local environment. Many a times, there is sheer neglect as far as the implementation of the EMP (Environment Management Plan) in the project area is concerned. Steps taken in accordance with the EMP may transform the coalmining a better work option for various skilled and semi-skilled workforces. Therefore, adopted methodology in this study is to identify the awareness level of the mineworkers to various physical and mental health hazards for that purpose, a perception based analysis based on a field survey of Argada area in South Karanpura Coalfield (SKCF) is incorporated.

Numerous studies have revealed that dust, gas, noise and vibrations in the industries or coalmines do have adverse affects on health. The coal mines are risky and hazardous, this also has been studied by many scholars and scientists and all this has been discussed in detail in the previous chapters, and there is no doubt about it. We have to see what is felt by the mineworkers of the studied coalmines. The idea behind perception based analysis is that, it is believed that risk perception or hazard perception is most essential and a prerequisite for taking actions or safeguards such as identification and elimination of the potential health hazard before its realization. There are several studies which reveal that in the Indian households, a reason of poor health, related to respiratory system especially, is the use of solid fuel and bio-fuel. It is a fact that, in the rural Indian household, bio-fuel is the main fuel for cooking purposes. In the mine areas, most of the workers households use coal as a household fuel. During the survey it was observed that, in the morning and evening the workers colony is totally engulfed in smoke emanating from burning of coal. Most of the officer's households use LPG Stoves or other electric cooking appliances. Use of Coal has been prevalent in the area among the D.R & P.R workers Household. They have been provided with LPG stoves by the company but workers prefer coal as household fuel because it comes free for them as it can be picked up from anywhere in this area. Various studies have correlated bad health conditions with the use of biomass or coal in the developing countries. "National surveys, including the 1991 national census, show that nearly 80% of Indian households use biomass as their primary cooking fuel. As a result, a large portion of the Indian population is potentially exposed to indoor and outdoor levels of pollution produced by cooking stoves."(Smith et al, 2000) Based on risks derived solely from studies of the health effects of individual diseases occurring in biomass-using households in developing countries, many in India itself, it is possible to estimate the total national burden of disease in India from use of these fuels: "Acute respiratory infection: Many studies around the world have found that household use of solid fuels is associated with acute respiratory infection in young children (although, as with all the diseases discussed here, there are other important risk factors, including malnutrition and crowding). Lung cancer, which is also dominated by smoking in industrialized countries, has been found to result from long-term experiment sure to cooking with coal in more than 20 studies in China. No such effect has been shown or biomass fuels, however. In India 400-800 women under 45 suffer from lung cancer linked to solid fuel use; the number is small because households rarely use coal."(Smith, 2000) Disorders of respiratory system among Indian children are also associated with solid fuels like coal. "Acute respiratory infection is the leading cause of death of the world's children and the largest category of ill health in the world in terms of disease burden. Almost 9 percent of the global burden of ill health and 12 percent of India's is due to acute respiratory infection. Acute respiratory infection linked to solid fuel use is estimated to cause 290,000-440,000 premature deaths a year in Indian children under 5." (Murray, et al, 1996) Tuberculosis has been associated with household solid fuel use in a national survey in India involving nearly 90,000 households as well as in smaller studies. Although this relationship is not yet established with complete certainty, it would be highly significant because tuberculosis is on the rise in many developing countries due to HIV infection and the increase in drug resistant strains. In India 50,000-130,000 cases of tuberculosis in women under 15 years, are associated with solid fuel use. Chronic respiratory disease, such as chronic bronchitis, is almost entirely due to smoking in the industrialized world. But studies in Asia and Latin America have found the chronic respiratory


145

disease develops in omen after long years of cooking with solid fuels. In India 19,000-34,000 women under 45 years suffer from chronic respiratory disease linked to solid fuel use e.g. cardiovascular disease. Although there are apparently no studies biomass-using households, studies of urban air pollution suggest that in India 50,000-190,000 women less than 30 years suffer from pollution-related heart disease, adverse pregnancy outcomes. Stillbirth and low birth weight have been associated with solid fuel use by pregnant women in Latin America and India. Low birth weight is a big problem in developing countries because it is a risk factor for a range of health problems.

Conclusion Different regions/nations are reaping the adverse consequences of growth - generated

activities in various forms as a result of modem technology based development. In many cases natural resources such as minerals are mined to the last limit: Mining is one of the chief economic activities in the Chotanagpur region. Though successive governments have been largely benefited from the abundant mineral resources of this region but little attention are paid to environmental considerations whose negligence often leads to degradation of the environment and sometimes directly and drastically affecting the surroundings. In Dhanbad "The reckless mining by the BCCL (Bharat Coking Coal Ltd.) which owns the right of mining in Jharia caused Chasnala or Gajlitand disaster. In this incident, land subsidence took place and Cracks developed in the houses of the entire area of Husainabad locality of Jharia". Though the finer points of the cause of such land subsidence could not ascertain, it is almost certain that it was caused due to the heavy underground blasting of coal in the Jharia coal field. There is certainly not unanimity as far as the ongoing discourse on economic and environmental sustainability of mining activities is concerned. The mining industry currently finds itself increasingly squeezed by environmental concerns, covering the whole range of operations from grass-roots exploration to final end-use of mineral products. This development may be inextricably linked to rising standards of living, and the associated trend towards paying more attention to the quality of life, as basic material needs gradually become satisfied. Rich societies can afford to minimize the disruptive impact of development on the environment, even to the extent of foregoing the financial benefits of that development. The same standards cannot, however, be applied in countries still blighted by poverty and disease; such impositions by well-meaning elements within rich countries are resulting in negative side effects, not only for the mining industry. The industry has to adopt the highest possible health, safety, and environmental standards that are consistent with its long-run viability, wherever it operates. It constantly has to distinguish between extreme views and reasoned legitimate concerns. It can, and should do no more.

References: 1. Biswas AK. Environmental Impact Assessment for Developing Countries. 1st ed. Tycolly

International, London, 2007; 232. 2. Boliga BP. Challenges of environmental management. 4th ed. Indian Institute of Mines,

Dhanbad, India, 2010;25. 3. Banerjee SP. Land Reclamation in Mined Areas. Gondwana. 2010; 63: 73. 4. Bose AK. Environmental Problem in coal Mining areas, Impact assessment and

Management strategies- case study in India. Elsevier, Amsterdam,1989;4:243. 5. Chadwik MJ. Environmental impacts of coal mining and utilization. 1st ed. Pergamum

Press Oxford, New Delhi, 2007-211. 6. Dhar BB. Environmental impact and abatement of noise pollution. 1st ed. ISM press,

Varanasi, India, 2000; 168-204. 7. Ghose KM. Effect of opencast mining on soil fertility. Centre of mining environment,

I.S.M, Dhanbad, India, 2004;12:255. 8. Goswami S. Need for Clean Coal Mining in India. Environmental Research, Engineering

and Management, 2013; 4:66: 79-84. 9. Goswami S. Coal Mining, Environment and Contemporary Indian Society. Global

Journal of Human Social Science. 2013; 13: 6:17-26. 10. Goswami S. Coal Mining, Communities and the Environment. 1st ed. New Delhi

Publishers, New Delhi, 2014; 103.


146

11. Goswami, S. Impact of Coal Mining on Environment” published in European Researcher, Vol. 92, Is. 3, pp. 185-196, 2015.

12. Hill P. Improving Efficiencies of Coal Fired Plants in Developing Countries. 10th ed. Gemini House, London, 2003; 10-18.

13. Penz P. Displacement by Development. Ethics, Rights and Responsibilities. 1st ed. Cambridge University Press, Cambridge 2011;32.

14. Singh TN. Clean coal initiatives. 2nd ed. Scientific Publishers, Jodhpur, 2012;12. 15. Wathern P. An introductory guide to EIA in: Environmental impacts assessments. 2nd

ed. London, U.K, 1988; 3-28 16. www.cea.nic.in. New Delhi, Central Electricity Authority, 11th Plan Shelf of Thermal

Power Projects, Inc,; 2007-12 [updated 2014 April 02; cited 2014 May 10 ]. Available from –www.cea.nic.in УДК 33

Добыча угля и коренные общины: социологическое исследование

угольных месторождений Джахарии

Шрибас Госвами Факультет социологии, Серампур колледж, Западная Бенгалия, Индия PhD, доцент E-mail: [email protected]

Аннотация. Горнодобывающая промышленность является неотъемлемой частью

жизнедеятельности общества и развития наций. К сожалению, процесс добычи и последующих операций часто ассоциируется с опасностями для здоровья человека и ухудшением состояния окружающей среды.

Это исследование презентует вопросы, связанные с вредным воздействием добычи, влиянием микроэлементов на местную окружающую среду, а также на здоровье человека в непосредственной близости от горной области.

Ключевые слова: добыча угля, опасность для здоровья, окружающая среда, шахта пожар, болезни органов дыхания.


147




Unemployment and Economic Growth of Developing Asian Countries: A Panel Data Analysis

1 Muhammad Imran

2 Khurrum S. Mughal 3 Aneel Salman

4 Nedim Makarevic 1 IQRA University, Islamabad Campus, Pakistan E-mail: [email protected] 2 COMSATS Institute of Information Technology, Islamabad E-mail: [email protected] 3 COMSATS Institute of Information Technology, Islamabad E-mail: [email protected] 4 Embassy of Bosnia and Herzegovina in Pakistan, Pakistan E-mail: [email protected]

Abstract This study presents the new regression estimates of the relationship between unemployment

and economic growth for 12 selected Asian countries over the period 1982-2011. Fixed effect and Pooled OLS techniques are used to analyze the panel data for measuring individual country effects, group effects and time effects while exploring the relationship between Unemployment rate and the Economic Growth. The results showed that higher unemployment rate has significant negative impact on GDP per capita growth (a proxy for economic growth). The results also investigated that economic growth seems to be significantly affected by traditional determinants such as Inflation (consumer price index), Population growth, Gross Capital Formation, Trade openness etc. Based on our results the author has concluded that reduction in unemployment rate would be a better option for more and sustained economic growth and also improving the welfare of the people.

Keywords: unemployment, economic growth, developing countries, panel data, fixed effect model.

1. Introduction Labor markets in Asia are characterized by pervasive unemployment and under-employment.

Asian countries vary in size and complexity. The nature, size and structure of population of Asia region have been changing qualitatively and quantitatively. From 7 most populous countries of the world, 6 of them (China, India, Brazil, Indonesia, Pakistan and Bangladesh) are located in Asia region. Economic growth, development and low level of unemployment are a dream that has become authenticity for some countries in the west, and also a few Asian countries like China, Japan, India and many other countries also. Man has constantly investigated to develop his material state through effectual use of resources, such as improving economic growth and low level of un-employment, price







148

stability, stable currency value etc. Unemployment and Economic growth have been found to influence each other, but so far this aspect is normally neglected in studying the comparative analysis of developing Asian countries. Unemployment is a continuing concern of every economy and economic growth is driven by country’s structural changes. The structural changes can not contribute in economic growth if social costs of structural changes are high and one of them is persistent unemployment. Unemployment rate has negative consequences for the economic well-being of human being (Levine, 2012). According to ILO population report in 2012, the number of unemployed individuals in the world has increased by 4 million in 2012 with the total reaching to_197 million. This year it is expected that it will reach up to 5.1 million and further more 3 million people will be jobless in 2014. For over three decades there has been massive amount of exploration on both theoretical and empirical effects of unemployment on economic growth of developing countries but little more has been done to investigate the relative relationship between economic growth and unemployment in Asian countries like Bangladesh, Cyprus, India, Indonesia, Korea (south), Kuwait, Pakistan, Philippines, Sri-Lanka, Syria Arab, Thailand and Turkey.

After five years of world financial crisis, economic growth has decelerated with a rise in unemployment. Rises in unemployment rate of Asia is mainly due to increase in labor force. According the World Bank report in 2011, unemployment rates in 2011 was 5, 6 and 10 percent for Bangladesh, Pakistan and India respectively. According to Economic Survey of Pakistan in 2011-12, from now, in the past few years, industrial load shedding accounts for loss of 400,000 jobs in Pakistan. It is an economic reality that country’s qualitative and quantitative nature of workforce directly impacts its GDP per capita growth rate. Workforce of any country is not only a productive agent of goods and services but these also play a role in country’s purchasing power which in-turn is a fuel for economic growth. According to World Bank statistics in 2012, at the end of 1980 Asian countries unemployment was very low; however, in 2000s it started to increased and was high in 2011 and is still high today. The unemployment situation in Asia has become critical. There are misleading arguments that there is no negative relationship between unemployment and other economic indicators with economic growth because each indicator including rate of unemployment and Gross Domestic Product (GDP) are rising in the long run. Asia always presents highly contrasting economic images. Economic growth is a problem in Asia due to unemployment strain and other weak economic indicators lead by defective government policies and corruption. The degree to which persistent Unemployment influence the economic growth of Asia region needs to be investigated, especially in the period where there is decline in overall economic growth (real GDP growth per capita) of Asia region.

Economic growth is the main objective of every economy. It is a standard fact that countries with good economic conditions are operationally efficient. A survey of global financial and economic practices suggests that current economic conditions of Asia countries are not optimal. The author has critically reviewed some of important empirical researches to develop main objectives in the environment of Asian countries and further, to utilize it and to draw important conclusions and recommendations for policy making. Osinubi (2005) explore the possibility of relationship among unemployment, poverty and economic growth. The results have been found by using multi-equation model by collecting the time series data for 31 years from 1970 to 2000. He concluded that increase in employment will lead to increase the output and hence cause economic growth. On the other hand, a decrease in employment rate will decrease the output and then economic growth. Blanchard (2006) conducts the study about European unemployment on evolution of facts and ideas. From survey reports, he found that European Unemployment started to increase in 1970s; further increased in 1980s and it reached a plateau in 1990s and is still high. He considered the 30 years data from 15 European countries and found that total factor productivity growth started to decline.

Wang & Abrams (2007) constructed a simple model of government outlays, growth and unemployment, by taking data of 20 OECD countries during recent three decades started from 1970 to 1999. They examined that the negative relationship between unemployment and growth is due to another cause called government outlays. Adjemian et al. (2010) examine the relationship that how labor market institutions affect unemployment and then economic growth. The data set covers 183 European regions and period from 1980 to 2003. They show that high labor costs and trade union power lead to higher unemployment rate and lower economic growth rate. Ahmed et al. (2011) explore the relationship among unemployment and growth (GDP) of Nigerian Economy,


149

by taking the secondary data for just 9 years from 2000 to 2008. They used regression techniques and showed that unemployment effect is 65.5 percent on the Nigerian GDP growth and there exist a negative relationship between unemployment and economic growth. Stephen (2012) explored the relationship between urban unemployment crisis on economic growth of Nigerian economy, also combining with inflation rate and investment level. Estimates showed that there exists a negative relationship between urban unemployment and economic growth. Stephen suggested that integrated vocational training programs and economic activities toward self-reliance and self-employment should be encouraged so that the unemployment rate can be minimized.

2. Data Description and Methodology The data set consists of the period 1982 to 2011, which is thirty (30) years. The observed data

was time series as well as cross sectional data, which is converted to Panel data/Pooled data. For this purpose we have already normalized the data for each country by using them as percentage of respective GDP in case the variable was in monetary terms. In our data set all the values of variables are presented, some of the observations were missing that have been attained by interpolation technique because missing values lower the quality of panel data.

Table 2.1: Descriptive statistic

Variables Mean Std. Dev. Minimum Maximum Observations

GDP_PC 3.0419 4.2349 -16.3 22.5 N = 360

UNEM 5.3052 3.4465 0.5 15.2 N = 360

INF 10.2711 14.5906 -3.0 88.1 N = 360

FDI 1.2536 1.7426 -2.8 10.5 N = 360

GCF 23.9313 6.1490 10.7 42.8 N = 360

TRD 63.3936 29.7340 12 150.3 N = 360

DCB 66.0208 47.2448 13.5 330.1 N = 360

PG 1.8522 0.9787 -2.8 5.4 N = 360

GS 25.4955 8.8311 6.7 64.7 N = 360

TNRR 8.0116 13.0090 0 63.7 N = 360

GFCE 12.5291 6.5161 4.1 76.2 N = 360

RIR 4.4180 6.2690 -24.6 46.2 N = 360 Total number of observations were 360 because there are twelve countries (n=12) and thirty

years’ time period (T=30). The mean value for unemployment is 5.3052 and the minimum value of the series is 0.5 and belongs to Kuwait for the year 1984-87 and 1984-1992. The maximum value of Unemployment15.2 belongs to Syria for the year 1997.

Graphical presentation for unemployment rate and economic growth are presented in Appendix Figure A2.1 and Appendix Figure A2.2. And Appendix Table A2.3 describes the matrix of correlation coefficients which shows that our studied data is free from the threat of high multicollinearity. Here GDP per capita growth is a dependent variable. GDP is a good measure of average real income in a country (Akbar et al, 2011).

The methodology adopted for this study is empirical and experimental. This research study has aim to examine whether unemployment has an impact on the economic growth of the selected twelve Asian countries. Now suppose variable factors of production only determine the output level in an economy, and the model presented by Tiwari & Mutascu in 2011 as follows:-

--- (i) Where, Y is output level (i.e. Per Capita GDP), L denotes the labor amount (measured by

Labor force of the country) and K denotes the capital (measured by Gross Capital Formation), it can be said that an increase in the amount of employed labor and capital will increase the output level of an economy. Then following above for our research study extended model after including the other explanatory variables, the model would be as follows:


150

GDP_PCit = f (UNEMit, GCFit, PGit ,TRDit, RIRit, DCBit, INFit, GFCEit, TNRRit, GSit, FDIit)

…(ii) where

GDP_PCit = GDP per capita growth (annual %)

UNEMit = Unemployment, total (% of total labor force)

GCFit = Gross capital formation (% of GDP)

PGit = Population growth (annual %)

TRDit = Trade Openness (% of GDP)

RIRit = Real interest rate (%)

DCBit = Domestic credit provided by banking sector (% of GDP)

INFit = Inflation, consumer prices (annual %)

GFCEit = General government final consumption expenditure (% of GDP)

TNRRit = Total natural resources rents (% of GDP)

GSit = Gross savings (% of GDP)

FDIit = Foreign direct investment, net inflows (% of GDP)

Here, i show country effects in explanatory variables, and t shows time effects in explanatory

variables and the assumptions of is that , i.e. errors are independently identically distributed with zero mean and stable variances. Where i denote a particular country and t denotes a particular time.

The adopted methodology is distributed in four sections. First: - Group effects where all coefficients are constant across time and countries. Second: - Slope coefficient constant but intercept varies across countries. Third: - Slope coefficients constant but the intercept varies over countries as well as time. Fourth: - All coefficients (intercept and slope) vary across countries.

3. Results After conducting a panel data analysis represented by econometric models presented in the

methodology section, we see some interesting results. For choosing the best model between FEM and REM, Hausman test is used, which has favored FEM (Fixed Effect Model), detailed test results are presented in Appendix table A3.1. The results are distributed further in four sections.

3.1. Group effects where all coefficients are constant across time and countries The results for all coefficients constant across individual and/or time are presented in table

3.1. It is concluded that we cannot reject the null hypothesis that unemployment does not explain the GDP per capita growth (GDP_PC) and selected determinants considered enough in order to explain the economic growth. In Model-1a, in case of zero Unemployment Rate (UNEM), zero Gross Capital Formation (GCF), zero Population Growth (PG) and zero Trade Openness (TRD) for each country (from twelve selected countries) is expected to have 2.6970 GDP per capita growth (p<.0000).

Table 3.1: Results with OLS & Fixed Effect Model for period 1982-2011. DV is GDP per capita

growth (GDP_PC)

Model-1a Model-1b Model-1c Model-2a Model-2b Model-2c

(OLS) (OLS) (OLS) (Fixed Effect) (Fixed Effect) (Fixed Effect)

UNEM -0.1219** -0.1279** -0.0764* -0.1157** -0.1059* -0.0478*

GCF 0.1774*** 0.1578*** 0.1802*** 0.1613*** 0.1339*** 0.1528***

PG -1.3341*** -1.4711*** -1.6065*** -1.3246*** -1.4578*** -1.6287***

TRD -0.0124* -0.0067 -0.0198** -0.0109 -0.0061 -0.0198***

RIR 0.0130 0.0172 0.0571* 0.0611

DCB -0.1208** -0.0179** -0.0091* -0.0115***


151

INF -0.0379** -0.0373** -0.0306** -0.0296**

GFCE 0.0887** 0.0847**

TNRR 0.0290 0.0270

GS 0.0021 0.0121

FDI 0.1470 0.1649

Intercept 2.6970* 4.2214** 3.0675* 2.9414** 4.1540** 2.9115*

F Test 27.24*** 17.69*** 12.51*** 25.2*** 16.69*** 11.88***

Adj. R2 .5132 .6218 .7113 .5415 .6372 .7215

Obs. 360 360 360 360 360 360

***, **, and * denote significance at 1%, 5% and 10% respectively.

And for 1 percent increase in unemployment rate (UNEM), the total GDP per capita growth

(GDP_PC) for selected countries is expected to decrease by 0.1219 percent, holding all other variables constant. In Model-1a unemployment rate (UNEM), population growth (PG) and trade openness (TRD) are negatively correlated to GDP per capita growth (GDP_PC) only Gross capital formation (GCF) is positively correlated. The signs of unemployment (UNEM) co-efficient are consistently negative across specifications and in all models it is statistically significant. Further the coefficient values of unemployment (UNEM) across specifications are nearly similar, ranging between 0,0478 and -.1279. A good-nees of fit measure Adjusted R2 is increasing with the addition of more regressors which means that the included variables are going to response more for better explanation of the model. Adjusted R2 of .7113 in Pooled OLS Model-1c means that this model accounts for 71 percent of the total variance in the GDP per capita growth (GDP_PC) rate of twelve selected countries and Adjusted R2 of .7215 in Fixed Effect Model with “with-in” effects mean that model accounts for 72 percent of total variances in the GDP per capita growth (GDP_PC) rate of selected Asian countries.

3.2. Slope coefficient constant but intercept varies across countries Appendix Table A3.2 presents the results by using Least Squre Dummy Variabel (LSDV) a

technique of Fixed Effect Model. Here we examine the fixed group effects by introducing group (country) dummy variables. The dummy variable c1 is set for Bangladesh and zero for other countries, similarly for other countries. There is no dummy for turkey as Turkey is a comparison country, in other words intercept for baseline in models are representing the intercept of Turkey. Akbar et al, in 2011 used Pakistan as a comparison country. LSDV fits the data better as Adjusted R2 increases from .5817 to .6671 and from .6671 to .7329. Each of c1-c11 dummy intercepts has deviation from its group specific intercept that is the baseline intercept (intercept for Turkey). These differences in country intercepts are due to the unique features of managerial talent or managerial style etc. after considering the Model-3a, we can write it in the equation form as follows:-

Bangladesh:

Cyprus:

India:

Indonesia:

Korea:

Kuwait: Pakistan:

Philippines:

Sri-Lanka:


152

Syria:

Thailand:

Turkey: 3.3. Slope coefficients constant but the intercept varies over countries as well as

time Results for panel data models have been presented in Appendix table A3.3. The results

presented in Model-4c appear to be more robust and have higher value of adjusted R2 and making the prediction that 73% variances in economic growth are explained by the studied explanatory variables, country dummies and time dummies regressors. In all three models, individual time dummies were individually statistically significant as they include year’s 1983, 1985, 1986, 1987, 1989, 1990, 1992, 1995, 2000, 2003, 2004, 2006, 2007 and 2010 which suggest that GDP per capita growth have changed much over a time. Here, also some of the individual country effects were also statistically significant like as Indonesia, Korea (south), Kuwait, Philippines and Sri-Lanka. If all of these were statistically significant, then no reason for polling (Gujrati, 2003). The overall conclusion from the Appendix table A3.3 was that there was propound individual country effects and also individual time effects. In other words, the GDP per capita growth functions for twelve selected countries have changed due to explanatory variables effects, individual country effects and as well as time effects.

3.4. All coefficients (intercept and slope) vary across countries Appendix Table A3.4 presents the estimated GDP per capita growth where all the studied co-

efficients vary across countries. In our models the differential slope coefficients were different for different countries. For unemployment rate (UNEM), the relationship for GDP per capita growth (GDP_PC) and unemployment (UNEM) is negative for all countries which is showing that with increase in unemployment (UNEM) the GDP per capita growth (GDP_PC) will be lowers. Some of the differential slope coefficients are also statistically significant (Gross capital formation in Kuwait, Gross capital formation in Turkey, Population growth in Kuwait, Real interest rate in Syria, Domestic credit provided by baking sector in Cyprus, Inflation in Korea (south), Inflation in Syria, Gross savings in Turkey, Foreign Direct investment in Kuwait and Foreign direct investment in Turkey, we can say that the variable introduced in the model influences the GDP per capita growth rate.

The relationship between Inflation (INF) and GDP per capita growth (GDP_PC) also presents the mix nature. Some countries have positive slope differential and some countries have negative slope differential. In last, the relationship for foreign direct investment (FDI) and GDP per capita growth (GDP_PC) have also mix nature for slope differential intercepts.

Limitations In terms of policy implications, the issues that are central in the exploration of the

unemployment should also be investigated, which will also be closely linked with the question of reduced unemployment. Although analysis presented and empirical models constructed for research are as complete and comprehensive as possible but still there are some limitations causing further suggestions for future research. First:

– Analysis covers only twelve (12) Asian countries thus the results only presents the realities of twelve selected countries only. Second;

– Main explanatory variable is unemployment rate that have different causes for different countries which needs to be explored in depth.

4. Conclusion We have used a panel data of twelve selected developing countries from Asia to capture in

time and country effects of unemployment rate on economic growth. Considering our data set of twelve countries between 1982 and 2011 periods, we have consistently found that high unemployment causes the decrease in economic growth in all models. Research study first presents


153

the importance of unemployment phenomenon toward economic growth. As we saw the unemployment is very heatedly discussed in national as well as international level. Growth of Asian developing countries is influenced by unemployment rate, especially among some Asian countries namely India, Indonesia, Sri-Lanka, and Thailand which have the highest unemployment rates when compared with other studied countries from Asia region. The above discussion clearly makes Kuwait, India and Turkey at the top but Pakistan, Sri-Lanka and Thailand at last in order while comparing for economic growth. All in all, the research study supports the view that there is some scope for developing countries in order to correcting and maintaining the economic development indicators, so the economic growth would be sustainable. Research conclusion underlines the importance of unemployment rate to the economic growth, both on global and as well as on local level. Hence, the conclusion indicates that increased unemployment rate decrease the economic growth rate in the long-run.

References: 1. Adjemian, S., Langot, F., & Rojas, C. Q. (2010). How do Labor Market Institutions affect the link

between Growth and Unemployment: The case of the European Countries. The European Journal of Comparative Economics, 7(2), 347-371.

2. Akbar, A., Imdadullah, M., Aman. U. M., & Aslam. M. (2011). Determinants of Economic growth in Asian countries: a panel data perspective. Pakistan journal of social sciences, 32(1), 145-157.

3. Blanchard, O. (2006). European Unemployment: the Evolution of facts and ideas. Economic Policy in Great Britain. 5-59.

4. Gujarati, D. N. (2003). Basic Econometrics. 3ra Edition, Mc Grawhill International Editions, Economic Series.

5. Hausman, J. A. (1978). Specification Tests in Econometrics. Econometrica, Vol.46, 69-85. 6. International Labor Office (ILO) (2012). Global Employment Trends. The Challenge of a Jobs

Recovery. 7. Labor Force Survey of Pakistan 2010-11. Government of Pakistan, Islamabad. 8. Levine, l. (2012). Economic Growth and Unemployment rate. Congressional Research Service. 7-

5700. 9. Osinubi, T. S. (2005). Macroeconometric Analysis of Growth, Unemployment and Poverty in

Nigeria. Pakistan Economic and Social Review, XLIII(2), 249-269. 10. Stephen, B. A. (2012). Stabilization Policy, Unemployment Crises and Economic Growth in

Nigeria. Universal Journal of Management and Social Sciences, 2(4), 55-63. 11. Tiwari, A. K., & Matascu, M. (2011). Economic growth and FDI in Asia: A panel-data approach.

Economic analysis and policy, Vol. 41. 12. The World Bank (2012). World Development Report. World Bank, Washington, DC 13. Wang, S., & Abrams, B. A. (2007). Government Outlays, Economic Growth and Unemployment: A

VAR Model. Working paper, University of Delaware, New York.

УДК 33

Безработица и экономический рост в развивающихся азиатских странах: панель анализа данных

1 Махаммад Имран

2 Харум Магал 3 Анель Салман

4 Недим Макаревик 1 IQRA университет, Исламабад Кампус, Пакистан E-mail: [email protected] 2 КОМСАТС Институт информационных технологий в Исламабаде E-mail: [email protected] 3 КОМСАТС Институт информационных технологий в Исламабаде E-mail: [email protected] 4 Посольство Боснии и Герцеговины в Пакистане, Пакистан E-mail: [email protected]






154

Аннотация. В этом исследовании представлены новые оценки регрессии взаимосвязи безработицы и экономического роста для 12 выбранных стран Азии за период 1982-2011 годы. Фиксированный эффект и объединенные МНК методы используются для анализа панельных данных, оценивающих отдельные страновые эффекты, групповые эффекты и временные эффекты, исследуя взаимосвязь между уровнем безработицы и экономическим ростом. Результаты показали, что более высокий уровень безработицы оказывает значительное негативное влияние на ВВП на душу населения (аппроксимация процессов экономического роста). Результаты также указывают, что на экономический рост, похоже, оказывают существенное воздействие такие традиционные детерминанты как инфляция (индекс потребительских цен), рост численности населения, валовое накопление капитала, степень открытости торговли и т.д. На основе полученных результатов автор пришел к выводу, что снижение уровня безработицы будет благоприятным фактором для поддержания устойчивого экономического роста и повышения благосостояния людей.

Ключевые слова: безработица, экономический рост, развивающиеся страны, панельные данные, модель с фиксированными эффектами.

Appendix

Appendix Figure A2.1 Graphical presentation of Unemployment rate of 12 selected countries

05

10

15

05

10

15

05

10

15

1980 1990 2000 2010 1980 1990 2000 2010 1980 1990 2000 2010 1980 1990 2000 2010

Bangladesh Cyprus India Indonesia

Korea Kuwait Pakistan Philippines

Sri Lanka Syrian Arab Thailand Turkey

Une

mplo

yme

nt ra

te, (%

of to

tal l

abo

r fo

rce

)

Years

Figure A2.2 Graphical presentation of GDP per capita growth of 12 selected countries


155

-20

-10

01

02

0-2

0-1

0

01

02

0-2

0-1

0

01

02

0

1980 1990 2000 2010 1980 1990 2000 2010 1980 1990 2000 2010 1980 1990 2000 2010

Bangladesh Cyprus India Indonesia

Korea, Rep. Kuwait Pakistan Philippines

Sri Lanka Syrian Arab Thailand Turkey

GD

P p

er

capita

gro

wth

(an

nu

al %

)

Years

Table A2.3

Matrix of Correlation Coefficients

GDP_PC UNEM GCF PG TRD RIR DCB INF GFCE TNRR GS FDI

GDP_PC 1

UNEM -0.0927* 1

GCF 0.4039* -0.1384* 1

PG -0.4010* -0.1043* -

0.4689* 1

TRD -0.0238 -0.1151* 0.1396* -0.2051* 1

RIR 0.1055* -0.1513* 0.1521* 0.0139 -0.0457 1

DCB -0.0232 -0.2916* 0.0574 -0.1838* 0.5810* -0.0171 1

INF -0.1354* 0.2617* -0.0943 -0.0070 -0.2761* -0.2756* -0.2526* 1

GFCE 0.0776 -0.2748* -0.0511 -0.0254 0.4265* -0.0112 0.4206* -0.1151* 1

TNRR 0.0932 -0.2425 -0.2758 0.4662 0.2134 -0.0689 -0.0694 -0.1596* 0.4222* 1

GS 0.1114* -0.4381* 0.2660* 0.1384* 0.2487* 0.0908 -0.0449 -0.2459* 0.1388* 0.5466* 1

FDI 0.0060 0.0230 0.0475 -0.2044* 0.4319* -0.0230 0.6283* -0.1728* 0.0629 -

0.2126* -

0.2139* 1

1. Source: World Bank Development Indicators, Economic surveys of selected respective countries.

2. * denote significance at 5%.

Table A3.1 Hausman specification test answers for best model by comparing the Fixed Effect Model and

Random Effect Model. The Hausman test results for all models are presented in Appendix Table


156

A3.1: Hausman test results for all models

Sr. No Model Hausman test value Significant or not Sig.

1 Model-2a 0.021 Significant

2 Model-2b 0.029 Significant

3 Model-2c 0.000 Significant










If Hausman test value <0.05 then statistically significant.

Table A3.2 Results with Fixed Effect Model for period 1982-2011. DV is GDP per capita growth

(GDP_PC)

Model-3a Model-3b Model-3c

(Fixed Effect) (Fixed Effect) (Fixed Effect)

UNEM -0.0139** -0.0919** -0.0755* GCF 0.2634*** 0.2448*** 0.2637*** PG -1.6982*** -1.8264*** -1.7092*** TRD -0.0188 -0.0157 -0.0160 RIR 0.0131 0.0198 DCB -0.0179** -0.0171** INF -0.0733*** -0.0695*** GFCE 0.0583 TNRR 0.0939 GS -0.0242 FDI -0.0263 c1( Bangladesh) -2.5255 -2.8014* -2.3472 c2(Cyprus) -0.4755 -0.8217 -0.8895 c3(India) -0.0052 -2.7381* -2.9286* c4(Indonesia) -2.7986* -3.5749** -4.2766** c5(Korea) -2.0614* -3.9480** -3.5129** c6(Kuwait) 3.4611** 2.4294 -3.8314 c7(Pakistan) 1.3720 -1.3164 -1.3736 c8(Philippines) -0.2636** -2.9141** -2.8032** c9(Sri-Lanka) -0.4139** -2.8794** -2.7710** c10(Syria) 0.2709 -1.9280 -3.9371** c11(Thailand) -0.8235 -3.0770** -2.8632 Intercept(baseline) 0.9216 2.0549** 2.8666*

for Turkey F Test 8.62*** 8.68*** 7.15*** Adj. R2 .5817 .6671 .7329 Obs. 360 360 360



157

Table A3.3

Results with Fixed Effect Model for period 1982-2011. DV is GDP per capita growth (GDP_PC)


(Fixed Effect) (Fixed Effect) (Fixed Effect) UNEM -0.0121** -0.0574** -0.0606* GCF 0.2449*** 0.2179*** 0.2452*** PG -1.7085*** -1.8327*** -1.8128*** TRD -0.0104 -0.0088 -0.0091 RIR 0.0621* 0.0677* DCB -0.0123* -0.0150 INF -0.0607** -0.0582** GFCE 0.0288 TNRR 0.0922 GS 0.0230 FDI 0.0208 c1 ( Bangladesh) 0.7721 -2.3213 -2.1547 c 2 (Cyprus) 0.1480 -1.2652 -1.6489 c 3 (India) 0.3765 -2.1168 -2.4829 c4 (Indonesia) -0.6980* -3.1904** -4.1328** c5 (Korea) -0.9773* -3.4135** -3.2971* c6 (Kuwait) 3.1942** 0.4200** -3.4118** c7 (Pakistan) 1.4910 -0.7393 -0.8302 c8 (Philippines) -0.5230* -2.8496** -2.8049** c9 (Sri-Lanka) -0.6208* -2.6955** -2.6957** c10 (Syria) 0.1767 -1.5119 -3.3900 c11 (Thailand) -0.9464 -3.1337* -3.3883 t2 (1983) 3.0664** 3.0262** 2.9891** t3 (1984) 1.5313 1.9425 1.9555 t4 (1985) 0.9723 0.9397 1.0027 t5 (1986) 2.5660* 2.2876 2.5937 t6 (1987) 3.7237** 4.0869** 4.3635** t7 (1988) 3.7575** 4.2176** 4.5024** t8 (1989) 3.5211** 3.6466** 3.7726** t9 (1990) 4.8532*** 5.2528*** 5.2276*** t10 (1991) 1.4859 1.9191 1.9554 t11 (1992) 3.0041** 3.1187** 3.1027** t12 (1993) 2.0082 2.1318 2.1521 t13 (1994) 1.7217 2.0490 2.0970 t14 (1995) 2.9422** 3.3744** 3.4093** t15 (1996) 1.7791 2.1739 2.1770 t16 (1997) 0.8383 1.1100 1.1675 t17 (1998) -1.5393 -0.8454 -0.5613 t18 (1999) 1.2500 1.3962 1.6320 t19 (2000) 3.1279** 3.2325** 3.2710** t20 (2001) 0.1276 0.2056 0.2888 t21 (2002) 2.1627 2.1802 2.3257 t22 (2003) 3.1952** 3.1735** 3.2556** t23 (2004) 3.8936** 4.0151** 4.0469** t24 (2005) 3.3942** 3.6274** 3.4600** t25 (2006) 2.9080* 3.0924** 2.9498* t26 (2007) 2.6135* 2.7415* 2.6087* t27 (2008) 0.3916 1.0841 0.7452


158

t28 (2009) -1.2535 -1.3039 -1.2148 t29 (2010) 2.9952** 3.3571* 3.4178** t30 (2011) 1.5965 2.0812 2.0644 Intercept(combined) (Turkey + 1982)

1.4220 2.8408 2.0526

F Test 4.70*** 5.04*** 4.65*** Adj. R2 .5915 .6711 .7388 Obs. 360 360 360 ***, **, and * denote significance at 1%, 5% and 10% respectively.

Table A3.4 Results with Fixed Effect Model for period 1982-2011. DV is GDP per capita growth

(GDP_PC)


(Fixed Effect) (Fixed Effect) (Fixed Effect) UNEM -0.0731** -0.1736** -0.2109***

GCF 0.2947*** 0.3170*** 0.3508** PG -0.3653* -0.0032* -0.2158*** TRD 0.0147* 0.0391** 0.0605** RIR -0.0305 -0.0269 DCB -0.0722*** -0.0783** INF -0.0723** -0.0840 GFCE 0.0938 TNRR 0.1157 GS -0.0687 FDI -0.3975 c1 ( Bangladesh) -54.9464 -54.9464 -54.9464 c2 (Cyprus) 28.3735 28.3735 28.3735 c3 (India) -8.4057 -8.4057 -8.4057 c4 (Indonesia) -24.1146 -24.1146 -24.1146 c5 (Korea) -12.7651 -12.7651 -12.7651 c6 (Kuwait) -21.0859 -17.7851 -14.5294 c7 (Pakistan) -24.1161 -21.1258 -19.1920 c8 (Philippines) -22.9338 -19.2020 -17.5053 c9 (Sri-Lanka) -23.7749 -20.2757 -15.4650 c10(Syria) -23.7749 -17.6983 -15.3115 c11 (Thailand) -21.9153 -15.6808 -13.2670 c1UNEM ( Bangladesh) -0.8446 -1.5191 -1.1287 c2 UNEM (Cyprus) -0.6047 -0.8515 -0.8888 c3 UNEM (India) -0.2002 -0.4470 -0.4843 c4 UNEM (Indonesia) -0.0094 -0.2374 -0.2747 c5 UNEM (Korea) -1.4889 -1.2421 -1.2048 c6 UNEM (Kuwait) -0.8561 -1.1030 -1.1403 c7 UNEM (Pakistan) -0.0072 -0.0525 -0.1116 c8 UNEM (Philippines) -0.2078 -0.2676 -0.3267 c9 UNEM (Sri-Lanka) -0.7810 -0.8407 -0.8999 c10 UNEM (Syria) -0.0352 -0.0244 -0.0836 c11 UNEM (Thailand) -0.4508 -0.3916 -0.3319 c12 UNEM (Turkey) -0.3013 -0.2416 -0.1824 c1GCF ( Bangladesh) 0.5269 0.5492 0.5830 c2 GCF (Cyprus) 0.0646 0.0423 0.0085 c3 GCF (India) 1.3662 1.3439 1.3101 c4 GCF (Indonesia) 0.0658 0.0434 0.0096 c5 GCF (Korea) 0.0735 0.0959 0.1297 c6 GCF (Kuwait) 0.7392** 0.7615** 0.7953** c7 GCF (Pakistan) 0.6310 0.5555 0.5167 c8 GCF (Philippines) 0.0450 0.0304 0.0693 c9 GCF (Sri-Lanka) 0.2139 0.2894 0.3282


159

c10 GCF (Syria) 0.1227 0.0472 0.0083 c11 GCF (Thailand) 0.0736 0.1491 0.1879 c12 GCF (Turkey) 1.2357*** 1.1311*** 1.3501*** c1PG ( Bangladesh) -3.2355 -3.5976 -3.3850 c2 PG (Cyprus) 0.6811 0.3190 0.5316 c3 PG (India) -4.6530 -5.0151 -4.8025 c4 PG (Indonesia) -2.8331 -3.1952 -2.9826 c5 PG (Korea) 3.2162 2.8541 3.0667 c6 PG (Kuwait) -3.2426** -3.6047*** -3.3921*** c7 PG (Pakistan) -1.1203 -1.1932 -1.2964 c8 PG (Philippines) -2.4704 -2.5433 -2.6466 c9 PG (Sri-Lanka) -2.0235 -2.0964 -2.1997 c10 PG (Syria) 3.5426 3.6155 3.7187 c11 PG (Thailand) 8.6917 8.7646 8.8679 c12 PG (Turkey) 3.4735 3.5464 3.6497 c1TRD ( Bangladesh) -0.0035 -0.0280 -0.0493 c2 TRD (Cyprus) 0.2562 0.2317 0.2104 c3 TRD (India) 0.0412 0.0168 -0.0045 c4 TRD (Indonesia) -0.1193 -0.1437 -0.1651 c5 TRD (Korea) -0.0820 -0.1065 -0.1278 c6 TRD (Kuwait) 0.1677 0.1432 0.1218 c7 TRD (Pakistan) 0.0097 0.0279 -0.0197 c8 TRD (Philippines) 0.1728 0.1350 0.1432 c9 TRD (Sri-Lanka) 0.0980 0.0603 0.0684 c10 TRD (Syria) 0.1987 0.1609 0.1691 c11 TRD (Thailand) 0.1835 0.1457 0.1539 c12 TRD (Turkey) 0.0776 0.0398 0.0480 c1RIR ( Bangladesh) 0.1289 0.1595 0.1559 c2 RIR (Cyprus) -0.3636 -0.3330 -0.3366 c3 RIR (India) 0.0764 0.1070 0.1034 c4 RIR (Indonesia) 0.3510 0.3816 0.3780 c5 RIR (Korea) -0.5785 -0.5480 -0.5515 c6 RIR (Kuwait) 0.0490 0.0795 0.0760 c7 RIR (Pakistan) 0.0415 0.0581 0.0444 c8 RIR (Philippines) 0.0210 0.0376 0.0239 c9 RIR (Sri-Lanka) -0.1271 -0.1104 -0.1241 c10 RIR (Syria) 0.3616** 0.3782** 0.3645** c11 RIR (Thailand) -0.0909 -0.0742 -0.0879

c12 RIR (Turkey) -0.1588 -0.1421 -0.1558

c1DCB ( Bangladesh) 0.0041 0.0763 0.0493 c2 DCB (Cyprus) 0.0084* 0.0806** 0.0867** c3 DCB (India) -0.0426 0.0295 0.0356 c4 DCB (Indonesia) -0.0065 0.0656 0.0717 c5 DCB (Korea) 0.1825 0.2547 0.2608 c6 DCB (Kuwait) 0.0132 0.0854 0.0915 c7 DCB (Pakistan) 0.2041 0.2149 0.2129 c8 DCB (Philippines) -0.0472 -0.0365 -0.0384 c9 DCB (Sri-Lanka) -0.1266 -0.1158 -0.1178 c10 DCB (Syria) -0.0150 -0.0042 -0.0062 c11 DCB (Thailand) -0.0990 -0.0882 -0.0901 c12 DCB (Turkey) -0.0085 0.0022 0.0002 c1INF ( Bangladesh) 0.1370 0.2093 0.2174 c2 INF (Cyprus) -0.9583 -0.8866 -0.8779 c3 INF (India) -0.0179 0.0543 0.0624 c4 INF (Indonesia) -0.0758 -0.0035 0.0045 c5 INF (Korea) -1.0790** -1.0066** -0.9986** c6 INF (Kuwait) 0.0101 0.0824 0.0905 c7 INF (Pakistan) -0.1435 0.0815 0.0666 c8 INF (Philippines) -0.2306 -0.0054 -0.0203 c9 INF (Sri-Lanka) -0.0043 0.2208 0.2059 c10 INF (Syria) 0.2134** 0.4385*** 0.4236***


160

c11 INF (Thailand) -0.0828 -0.6577 -0.6726 c12 INF (Turkey) -0.0788 0.1462 0.1313 c1GFCE ( Bangladesh) -0.6576 -0.6576 -0.5637 c2 GFCE (Cyprus) 0.0460 0.0460 0.1399 c3 GFCE (India) -2.1654 -2.1654 -2.0716 c4 GFCE (Indonesia) -1.9960 -1.9960 -1.9022 c5 GFCE (Korea) -1.4322 -1.4322 -1.3384 c6 GFCE (Kuwait) -0.1443 -0.1443 -0.0505 c7 GFCE (Pakistan) 0.0719 0.0719 0.0244 c8 GFCE (Philippines) 0.2346 0.2346 0.1871 c9 GFCE (Sri-Lanka) 0.5419 0.5419 0.4944 c10 GFCE (Syria) -0.4025 -0.4025 -0.4500 c11 GFCE (Thailand) -0.5555 -0.5555 -0.6030 c12 GFCE (Turkey) -0.6995 -0.6995 -0.7470 c1TNRR ( Bangladesh) -0.1338 -0.1338 -0.1813 c2 TNRR (Cyprus) -18.5468 -18.5468 -18.4311 c3 TNRR (India) -1.7412 -1.7412 -1.6254 c4 TNRR (Indonesia) 0.6234 0.6234 0.7391 c5 TNRR (Korea) 15.2901 15.2901 15.4058 c6 TNRR (Kuwait) -0.0901 -0.0901 0.0255 c7 TNRR (Pakistan) 0.3168 0.3168 0.2711 c8 TNRR (Philippines) 2.1654 2.1654 2.1197 c9 TNRR (Sri-Lanka) -3.1801 -3.1801 -3.2250 c10 TNRR (Syria) 0.0636 0.0636 0.0179 c11 TNRR (Thailand) -0.6454 -0.6454 -0.6911 c12 TNRR (Turkey) -6.2859 -6.2859 -6.3316 c1GS ( Bangladesh) 0.2087 0.2087 0.1399 c2 GS (Cyprus) 0.2162 0.2162 0.2849 c3 GS (India) 1.6726 1.6726 1.6039 c4 GS (Indonesia) 0.0918 0.0918 0.1606 c5 GS (Korea) 0.6190 0.6190 0.6878 c6 GS (Kuwait) 0.1335 0.1335 0.2023 c7 GS (Pakistan) 0.3779 0.3779 0.3195 c8 GS (Philippines) 0.0162 0.0162 0.0421 c9 GS (Sri-Lanka) 0.0479 0.0479 0.0104 c10 GS (Syria) 0.1864 0.1864 0.2448 c11 GS (Thailand) 0.4317 0.4317 0.3733 c12 GS (Turkey) -1.2835** -1.2835** -1.3419** c1FDI ( Bangladesh) -1.5925 -1.5925 1.9901 c2 FDI (Cyprus) -0.0512 -0.0512 0.3462 c3 FDI (India) -0.0434 -0.0434 0.3541 c4 FDI (Indonesia) 0.1206 0.1206 0.5182 c5 FDI (Korea) 2.3491 2.3491 2.7467 c6 FDI (Kuwait) 10.1035*** 10.1035*** 9.7059*** c7 FDI (Pakistan) -0.4627 -0.4627 -0.4343 c8 FDI (Philippines) 0.3960 0.3960 0.3676 c9 FDI (Sri-Lanka) 0.3151 0.3151 0.2867 c10 FDI (Syria) -0.1073 -0.1073 -0.1357 c11 FDI (Thailand) -0.6333 -0.6333 -0.6618 c12 FDI (Turkey) -1.9384 -1.9384* -1.9668* Intercept (baseline) 4.1160** 6.8959*** 6.1994** F Test 2.77*** 3.20*** 3.09*** Adj. R2 .6478 .6937 .7549 Obs. 360 360 360



161




Fiscal Policy and Income Inequality in Pakistan: An ARDL Approach

1 Rana Ejaz Ali Khan 2 Bushra Jabeen Hashmi

1 The Islamia University of Bahawalpur, Pakistan Associate Professor and Chairman, Department of Economics E-mail: [email protected] 2 The Islamia University of Bahawalpur, Pakistan M. Phil. Candidate, Department of Economics E-mail: [email protected]

Abstract This study is an attempt to capture the impact of fiscal policy on income inequality in

Pakistan. It employed Autoregressive Distributed Lag (ARDL) model on annual time series data from 1980 to 2012. The stationarity of data is checked by Augmented Dickey Fuller unit root test. Short-run dynamics are tested by error correction model. Model reliability is tested with the help of the diagnostic tests. Chow test is applied to detect structural breaks and Gregory-Hansen technique is employed as a remedial measure of the structural breaks. Results indicate that development expenditures and financial development has diminishing effect on income inequality. On the other hand fiscal deficit and urbanization are affecting the income inequality positively. The current expenditures and indirect tax has no influence on Gini-coefficient. The study recommends that fiscal deficit should be diminished by reducing current expenditures. The development expenditures require an increase to decrease income inequality. For the financing revenue from indirect taxes may be increased, it will not hurt the income inequality. The financial development as a tool for decreasing inequality is also proposed.

JEL Classification: E62, D3, H3 Keywords: income disparity, fiscal policy, gini-coefficient, ardl, development expenditures,

fiscal deficit, financial development, urbanization. 1. Introduction

Debate on income inequality is not new in the economic literature. Earlier Marxists focused on the theory of social classes and blamed capitalistic system for having unequal societies. They argued that propertied class suppress labor class and generate inequality. Classical economists were mainly concerned with the income distribution between factors of production. Now the economists are conscious about income distribution at household and personal level.

With the passage of time there emerged a number of determinants of income inequality in an economy. For instance, role of urbanization in income inequality was introduced by Kuznets (1955). He took the urbanization and industrialization as two complimentary processes. The urbanization lead to industrialization and it increases inequality at the initial stage of





162

industrialization. The industrialization increases the income gap between urban and rural population, until the benefits of industrialization are reached to rural households. Austrian school of thought believes that inflation generating through monetary phenomenon increase inequality. Patrick (1966) claimed that financial sector development can reduce the income inequality because financial development increases provision of credit and productive capacity of the households having no or comparatively less assets.

The co-existence of inequality and economic growth is linked with Kuznets’s (1955) views that at initial level of development of an economy it faces high income inequality and then it starts decreasing, that is income inequality has a U-shaped curve against economic growth. Kaldor (1956) also linked income inequality with development process and accepted U-curve hypothesis. Barro (2000) proved the same phenomenon empirically* but with certain limitations and reservations. Sylwester (2002) proved an inverse relationship between education expenditures and income inequality in OECD, East Asian, Latin American and African countries. The relationship was stronger in OECD countries but it was existed to some extent in developing countries. Angello and Sousa (2012) revealed that degree of openness of trade reduces inequality in industrialized economies. Brenneman and Kerf (2002) and Caldern and Serven (2004) attempted to estimate the impact of infrastructure development on poverty and inequality. Both of the studies proved an inverse association between infrastructure development and income inequality.

1.1. Fiscal Policy and Income Inequality Neoclassical growth models limits the role of fiscal policy in income inequality and believe

that fiscal policy has temporary effects on growth so it becomes difficult to decide the effect of fiscal policy on income inequality in an economy. But analyses in the framework of endogenous growth models made it possible to see the role of fiscal policy in income inequality. In this way endogenous growth theory opened new horizon for the role of fiscal policy in income inequality. Under these models fiscal variables impact the level and growth rate of output and change the temporary impact (under neoclassical models) into permanent impact. So fiscal policy can be used as an important tool for redistribution, although the exogenous or endogenous growth models do not include distributional issues directly. There emerged an indirect relationship between economic growth and income inequality.

The government expenditures as a component of fiscal policy in the forms of subsidies, social welfare expenditures, infrastructure expenditures, expenditures on poverty reduction programs and expenditures on food and health instigate the income inequality to decrease (Ramos and Roca-Sagales 2008). The governments need resource to finance these expenditures and generally tax revenue is the major source for meeting these expenditures. If the taxes are regressive these expenditures do not work for inequality reduction and if taxes are progressive the inequality would be reduced.

Regarding the current expenditures and development expenditures it is evidenced that development expenditures reduce the income inequality and current expenditures enhance the income inequality (Ali and Ahmed, 2010). Gallo and Sagales (2013) evidenced that current expenditures increase income inequality, although the current expenditures have certain spending such as pensions and other benefits which are considered as decreasing the income inequality. Ramos and Roca-Sagales (2008) suggested that government should increase the public spending to improve the situation of income inequality but it may happen at the cost of growth.

The empirical research evidenced that fiscal policy tools are more effective in developed and advanced economies. The evidences from OECD countries showed that Gini coefficient was reduced by 15 percent by effective working of fiscal tools (Brandolini and Smeeding, 2009; Barnard and Atta, 2010). The public spendings on housing, food, health and education decrease inequality (Decoster et al. 2009; Donoaghue et al. 2004). The governments in these economies usually believe on transfer payments and public expenditures for reducing inequality. But role of the taxes becomes important in the perspective of efficiency. The indirect taxes cause to increase income

* However, Samanta and Cerf (2009) have shown that higher income inequality has positive impact on GDP growth

rate.


163

inequality because they are highly regressive*. However, Galo and Sagales (2013) found that direct taxes increase the income inequality.

The income redistribution effect of fiscal policy in UK has been examined by Ramos and Roca-Sagales (2008), and they explained that public expenditures improve the distribution of income more than taxes even though the tax system is progressive in the country. Taxes deteriorate the income distribution and particularly indirect taxes negatively affect the distribution of income.

1.2. Income Inequality in Pakistan In Pakistan historical trend of income distribution shows a persistent existence of income

inequality. It is not only harming the growth process of the country rather making a burden in the social, cultural and political development of the economy along with enhancing the numerical strength of the deprived section of the society†. The income inequality has an increasing trend since 1980 as shown in figure 1.

Figure: 1 Income Inequality in Pakistan (1980-2012)

In 1980s the income inequality increases but comparatively at a lower rate. The Middle East phenomena (overseas employment in Middle East) not only reduced the poverty but also help to contain the sharp rise in income inequality. Power of the Middle East phenomena may be expressed by the figure that in 1982/83 remittances were about 10 percent of the GDP of Pakistan. In 1990s inequality rise sharply possibly due to Structural Adjustment Programs (SAP). Implementation of SAP leads to withdrawal of subsidies, reduction in public sector program and increasing the tax which enhanced the burden on common man. During the period of 2000-2010, initially the income inequality increased but after 2004 it decreased. Overall in 2000-10 the income inequality remained almost constant but at a higher level. After 2010, it again rises. The figure shows that in 1993-94, during 2000 and 2004, and in 2012 the income inequality was comparatively higher and greater than 0.4.

The literature has identified a number of factors affecting income inequality in Pakistan. They include the urbanization, financial development, government spendings, loan from IMF, adaptation of SAP, tax structure and political inefficiency in allocation of expenditures (Sherazi et al. 2001; Li and Zou, 2002; Ali and Ahmed, 2010; Shahbaz and Islam, 2011). Pakistan’s tax structure is regressive as the major emphasis is on indirect taxes. Tax net is loose where upper class is escaping while middle class is paying the tax. Government expenditures are politically induced and inefficiently allocated (Sherazi et al. 2001; Ali and Ahmed, 2010; Shahbaz and Islam, 2011). It means fiscal policy may be one of the factors of income inequality in Pakistan. On the other hand

* However, Galo and Sagales (2013) found that direct taxes increase the income inequality. † In some economies, the income inequality negatively impacts the fiscal multiplier and its effectiveness (see Samanta

and Cerf, 2009)


164

theoretically one of the objectives of fiscal policy is to mitigate the income inequality in an economy. There are diverging empirical evidences on achieving this objective of fiscal policy. For Pakistan the fiscal policy is successful for achieving this target needs attention of the researchers.

In the light of above, the current study attempt to find the role of fiscal policy in income inequality in Pakistan. The core objective of the study is to empirically estimate the impact of the components of the fiscal policy on income inequality. The components of fiscal policy are reducing income inequality, or not, is the research question to be answered. Based on the results some policy proposals would be framed.

2. Literature Review A variety of literature exists on fiscal policy and its impact on poverty and income inequality.

Shirazi et al. (2001) used the micro data from Household Integrated Economic Survey to see the redistributive effect of fiscal policy in Pakistan. Fiscal policy was incorporated in the analytical model in the form of public expenditures and taxes. Public expenditures were categorized into education, defense, health, agriculture and general administration. They were also divided into urban and rural areas expenditures. Taxes were decomposed into indirect tax, import and export duties as well as categorized into the tax burden faced by rural and urban population. Results explained that upper class is getting least benefits from government expenditure and tax burden is also higher on upper class. It causes the income inequality to decrease. The study also concluded that urban households are getting more benefits as compared to rural households. It was recommended that fiscal policy in Pakistan should be more pro-poor and focused on low income group of the economy.

Samanta and Cerf (2009) focused on fiscal policy and income inequality in the perspective of welfare impact of government expenditures in 10 transitional economies. The study used time series data and employed OLS and 2SLS models. The estimation witnessed that more unequal income distribution needed more government expenditure to increase income and ultimately to overcome income inequality. It was recommended for transitional economies to follow privatization, openness of trade and more progressive tax system to improve the economic situation.

The role of public expenditures in reducing poverty in Pakistan has been examined by Ali and Ahmed (2010) by splitting expenditures into current and development expenditures and employing ARDL technique on annual time series data. Results revealed that development expenditures reduce poverty but current expenditures increase the poverty. However both types of expenditures affect the poverty through inequality. Development expenditures slide down the income inequality and current expenditure increases the income inequality. The study suggested that government should diminish the proportion of current expenditures and increase the ratio of development expenditures to have the reduction in poverty and inequality.

Muinelo-Gallo and Roca-Sagalés (2011) attempted to estimate the impact of fiscal instruments on economic growth and income inequality in 43 countries by using annual time series data and employing OLS model. The results showed that current expenditures and direst taxes reduce economic growth and income inequality. It was proposed that public investment size should be increased to reduce inequality as trade off between growth and equity can be eliminated.

Probing the role of fiscal consolidation in income inequality in 18 industrialized countries, Angello and Sousa (2012) found that fiscal consolidation has a positive impact on income inequality. The study further explained that during the period of consolidation there was high income inequality and the size of the fiscal consolidation program has an increasing impact on income inequality. The spending cuts were found extremely damaging for income distribution while tax hikes were found helpful in improving the income distribution. Inflation and low economic growth rate enhance the effect of fiscal consolidation on inequality.

Claus et al. (2014) investigated the impact of fiscal policy on income inequality in 15 Asian countries including Pakistan. In the part of fiscal revenues personal income tax, corporate tax, payroll tax, social security contribution, custom and excise duties were included. The expenditures included were health expenditure, education expenditures and social protection expenditures. The results explained that even progressive taxation system is playing a minor role in redistribution of income. Education and health expenditures are reducing inequalities but


165

surprisingly social protection expenditures and housing expenditure are increasing income inequality. It explained that social protection expenditures target only a selected segment of the society while others remain deprived. The study proposed that these countries should increase the tax base and reduce the tax rates. To increase the tax base reductions in tax concessions and tax holidays are proposed.

The impact of different fiscal instruments on income distribution and economic growth has also been estimated by Gallo and Sagales (2013) for Uruguay. Vector autoregressive technique was employed on annual time series data. The study explained that current expenditures and direct taxes increase income inequality. However public investment decreases the income inequality. The results revealed that fiscal policy is responsible for increasing income inequality in Uruguay.

In the literature different aspects of the fiscal policy in relevance to income inequality have been probed by the researchers using various econometric techniques. We are going to analyze the impact of fiscal policy on income inequality by including four aspects of fiscal policy like current expenditures, development expenditures, indirect taxes and fiscal deficit along with supportive variables of financial development and urbanization*. We will use the fresh data and see the impact of fiscal policy on income inequality in the presence of financial development and urbanization. So the current study will be an addition to the literature in the focused area.

3. Methodology 3.1. Sources of Data and Model Specification The study used annual time series data for the period 1980 to 2012. The data has been taken

from various issues of Economic Survey of Pakistan by State Bank of Pakistan (SBP various years) and 50 Years of Pakistan in Statistics by Federal Bureau of Statistics (FBS 1999).

To incorporate the fiscal policy in the model government expenditures categories into current expenditures and development expenditures are included. Current expenditures are comprised of interest payments, subsidies, general administration, defense, pension’s grants, etc. while development expenditures are comprised of expenditures on public sector development programs and other developmental projects regarding human capital and infrastructure.

Revenue is collected through direct taxes and indirect taxes. In Pakistan indirect taxes are the major source of revenue and it has different real impact on purchasing power of the people in different income groups. So indirect tax revenue is included in the analysis to see its contribution in income inequality. Fiscal deficit is a prominent feature of Pakistan’s economy. The gap between expenditures and revenue is bridged by fiscal deficit that may affect the people of different income group differently and influence income inequality. The fourth variable related with fiscal policy included in the model is fiscal deficit. The supporting variables are financial development and urbanization.

The financial development has emerged as a new area affecting a large number of macroeconomic indicators in Pakistan (Khan and Hye, 2013). It is based on the fact that in the last three decades financial sector has tremendous growth rate. The credit distributed to private sector has been taken as an indicator of financial development. It includes all the sources of credit including loans, trade credits, purchase of securities and all credit advances by banking and non-banking†. Urbanization is the phenomena involved with income inequality and there exists income inequality between urban and rural population of the economy. Urbanization as a factor of income inequality has been analyzed by a wide range of researchers (Dutt, 2001; Davis and Henderson, 2003).

The functional form of the model is expressed in equation 1. GINI = f (CEXP, DEXP, INDT, FISCD, FINAND, URBAN) ……. (1)

GINI = β0 + β1CEXP + β2DEXP +β3INDT + β4FISCD + β5FINAND + β6URBAN + e …… (2) GINI = Gini coefficient

* The control variables in the analysis impact of fiscal policy on income inequality has a wide range including

population growth, dependency burden of youth and old age, globalization, corruption and education (Claus, et. al.

2012). † Fishman and Love (2003) and Petersen and Rajan (1997) have used broad money as the percentage of GDP and the

trade credit for provision of financial development. Khan and Hye (3013) have created an index for financial

development.


166

CEXP = Current expenditures (Government current expenditures in million rupees) DEXP = Development expenditures (Government development expenditures in million

rupees) INDT = Indirect tax revenue (Government revenues from indirect taxes in million rupees) FISCD = Fiscal deficit (Government fiscal deficit as percentage of GDP) FINAND = Financial development (Credit disbursed to private sector as percentage of GDP) URBAN = Urbanization (Urban population as the percentage of total population) The current expenditures, development expenditures, indirect tax revenues and urbanization

are taken in log form. 3.2. Econometric Estimation 3.2.1. Unit Root Test In time series analysis data stationarity is a necessary condition. If time series data is not

stationary, shocks in the data will exist disappear and results will be aggravated. On the other hand in stationary time series data shocks are eliminated and data turn back to its mean value. Indication of stationarity is that by increasing the lags correlogram is declined. In non-stationary series autocorrelation plot would expand. If we have the model

Y = ΩY-1 + µ ……. (3) Where µ = white noise error Condition of stationarity is |Ω| < 1. If it is not, time series will be non-stationary. So the hypothesis is Ω = 1 (series has a unit root) While alternative hypothesis is Ω < 1 We will apply the Augmented Dickey Fuller (ADF) test to check the stationarity of the data. 3.2.2. Autoregressive Distributed Lag Model (ARDL) General test to estimate the cointegration among the variables is Johansen cointegration test

or Engle Granger approach. These methods have two main problems. First one is that it requires the data to be integrated of same order. Secondly small data estimation becomes difficult under such techniques. To overcome these problems Autoregressive Distributed Lag (ARDL) test was introduced by Pesaran and Shin (1999). In this test dependent variable is regressed upon its own lag value including current and lag values of other explanatory variables. As the model is going to be tested through ARDL bounds technique that depends on F statistics, the Bounds test will show that either the variables are co-integrated or not.

H0 = There is no cointegration among the variables. That is β0 = β1 = β2 = β3 = β4 = β5 = β6 = 0 H1 =There is cointegration among the variables. That is β0 = β1 = β2 = β3 = β4 = β5 = β6 ≠ 0 We will use the same technique for current analysis. The equation 2 will be treated under ARDL approach as:

ΔGINI = α + Σα1iΔGINIt-i + Σα2iΔLCEXPt-I + Σα3iLDEXPt-i + Σα4iLINDTt-i + Σα5iFISCDt-i+ Σα6iFINANDt-I + Σα7iURBANt-i + β1GINIt-1 + β2LCEXPt-1 + β3LDEXPt-1 + β4LINDTt-1 + β5FISCDt-1 +

β6FINANDt-1 + β7URBANt-1 + et …….. (4)

3.2.3. Error Correction Model (ECM) If we have two variables Y and X and if both are co-integrated, we may write their relation

with ECM specification as: ΔY = α0 + β1ΔX – π µt-1 +et ------------ (5)

This is ECM equation which has the information of both short-run and the long-run relation. Where β1 is impact multiplier and π is adjustment factor. Impact multiplier expresses the short-run effect of change in Y due to change in X while adjustment factor shows that how much of the past period disequilibrium is adjusted in the current period. We will apply ECM estimation to see the impact of fiscal policy components along with the supporting variables on income inequality in Pakistan.


167

3.2.4. Diagnostic Tests Reliability of regression results is very important even the results show the good and

significant relationship. It is possible that the regression does not qualify the certain diagnostic tests. There is a large number of diagnostic tests which can suggest that either empirical findings are correct or not. These tests are related to the assumptions of classical linear regression model and they are for no autocorrelation, functional form, normality and homoscedasticity. In the current study we will employ LM Lagrange multiplier for no autocorrelation, Ramsey’s Reset test for appropriate functional form and model specification, Jarque-Bera test for the normality of residuals and heteroscedasticity test based on the regression of squared residuals and squared fitted values.

3.2.5. Stability Test for ARDL Model To test the structural change in the ARDL model the graphs of CUSUM and CUSUM Squares

are examined to know that they lies in the critical bounds or not. The graph proves the model stability and the long-run estimates. We will employ the CUSUM and CUSUMSQ to check the stability of the model in current analysis.

If the stability test of the given ARDL model will not be satisfied and CUSUM or CUSUMSQ

values lie outside the critical bounds then Chow test may confirm the existence of structural break. We will employ Chow test and adopt the Gregory-Hansen (Gregory and Hansen, 1996a, 1996b) approach to diagnose the structural break in the model and to have the remedial measure for the structural breaks and shocks in the given time series data.

4. Empirical Estimates In this section the empirical estimation results for the procedure to see the impact of fiscal

policy on income inequality in Pakistan are presented. 4.1. Results of Augmented Dickey Fuller Test To check the stationarity of the data ADF test wad applied. The results of ADF test are shown

in table 1.

Table 1: Results of Augmented Dickey Fuller Test

Variables Level First differenc

e

Critical values of unit root Decision Order of integration

1% 5% 10%

GINI 2.02508 -5.6008* -4.3393 -3.5875 -3.22923 Nonstationary at level but stationary at 1st difference

I (1)

CEXP 6.048036

-4.65813* -3.6210 -2.9434 -2.6102 Nonstationary at level but stationary at 1st difference

I (1)

DEXP 1.97046 -3.767919* -3.6329 -2.94840 -2.61287 Nonstationary at level but stationary at 1st difference

I (1)

INDT 5.43997 -1.7866*** -2.6416 -1.9520 -1.61040 Nonstationary at level but stationary at 1st difference

I (1)

FISCD -0.90741 -7.58927* -2.63921

-1.95168 -1.61057 Nonstationary at level but stationary at 1st difference

I (1)

FINAND -0.38522 -4.336305*

-2.64167

-1.95206 -1.61040 Nonstationary at level but stationary at 1st difference

I (1)

URBAN -5.4062* - -4.28458

-3.56288 -3.21536 Stationary at level I (0)

*Significant at 1%, ** Significant at 5% and *** Significant at 10%

Results of ADF show that the variable of urbanization (URBAN) is stationary at level while

other variables are integrated at first difference. The estimation of this model through ARDL is fully justified as ARDL is the cointegration technique that can handle the issue of data stationarity at different orders. However if the data is stationary at second difference then the results would be unreliable. In the results of ADF test none of the variable is integrated at 2nd difference or above so results of ARDL would be reliable.

4.2. Results of ARDL Technique


168

The results of ARDL technique for cointegration based on Schwarz Bayesian Criterion (1,0,0,1,0,0,1) are shown in table 2.

Table 2: Results of ARDL Technique for Gini Coefficient

Variable Coefficients T-ratio, p-Value GINI(-1) 0.74682 10.636, (0.000)* CEXP -1.0417 -1.2824, (0.214) CEXP(-1) 1.7833 2.1027, (0.048)** DEXP -0.55029 -2.2661, (0.035)** INDT -0.15872 -0.2362, (0.816) FISCD 0.10713 1.8980, (0.072)** FINAND 0.010150 0.24561, (0.808) FINAND(-1) -0.10488 -2.7784, (0.012)** URBAN 3.0631 4.1604, (0.000)* R square = 0.98433 Adjusted R square = 0.97806 D.W statistic = 1.97 F-statistics = 157.0000 [.000] *Significant at 1%, **Significant at 5%

We have used the ARDL bounds test approach so the evidence of cointegration among the

variables is given by F-statistic. If it lies above the upper critical bounds then there will be cointegration. Results of F statistic shows that there is cointegration between the variables, as the value of F-statistic lies above the upper bound. So the null hypothesis of no cointegration is rejected.

The zero value of Gini coefficient shows perfect equality and 1 shows perfect inequality, the positive signs of the coefficients of explanatory variables represent the increasing income inequality, so their negative signs are desirable. ARDL cointegration estimation proves that lag value of inequality is causing inequality in current period. The lag value of current expenditures, fiscal deficit and urbanization are responsible for increase in income inequality. However, development expenditures, lag value of financial development are reducing inequality. Impact of the indirect tax revenue is negative but it is insignificant. The results express that the model has a good explanatory power, i.e. 98 percent of variations in the Gini coefficient is caused by the explanatory variables.

Table 3: Results of ARDL Estimates for Long-run

Variables Coefficients T-ratio, p-Value CEXP 2.9292 1.5699, (0.132) DEXP -2.1735 -2.745, (0.012)** INDT -.62690 -.22945, (0.821) FISCD .42313 1.7079, (0.103)***

FINAND -.37416 -2.0900, (0.050)**

URBAN 12.0985 3.1870, (0.005)* F-statistic = 5.9514 Significant at 95% Lower bound = 2.4829 Upper bound = 3.9494 *Significant at 1%, **Significant at 5% and ***Significant at 10%

The ARDL estimates for the long-run explain that fiscal deficit and urbanization have positive

impact on income inequality. However the development expenditures and financial development are emerged as the decreasing factor of income inequality. Current expenditures and indirect taxes have shown insignificant effect.

4.3. Results of Error Correction Model (ECM) After knowing that the variables in the model have long-run relationship, the next step was to

investigate the short-run estimates. The short-run dynamics of the model are tested by ECM which


169

also explains that how much of the previous disequilibrium can be adjusted in current time period. The results of the ECM are shown in table 4.

Table 4: Results of Error Correction Model for Gini Coefficient

Variables Coefficients T-ratio, p-Value dCEXP -1.0417 -1.2824, (0.214) dDEXP -0.55029 -2.2661, (0.035)** dINDT -0.15872 0.2362, (0.816) dFISCD 0.10713 1.8980, (0.072)** dFINAND 0.010150 0.24561, (0.808) dURBAN 3.0631 4.1604, (0.000)* ECM(-1) -0.25318 -3.4973, (0.002)* R square = 0.7780 Adjusted R square = 0.6892 Durbin Watson Stat = 1.97 F statistics = 11.68(0.000) *Significant at 1% and **Significant at 5%

The short-run estimates by ECM have shown that fiscal deficit and urbanization are the

increasing factors of income inequality. The ARDL estimation has already shown same type of effect in the long-run. So fiscal deficit and urbanization are responsible for increasing income inequality in Pakistan. The development expenditures have negative effect on income inequality in the short-run. Same type of effect is shown in the long-run ARDL results. It explains that fiscal policy with component of development expenditures is playing its role in reducing income inequality. The value of ECM is .25318 which shows that 25 percent disequilibrium will be adjusted in current period.

4.4. Results of Diagnostic Test Diagnostic tests are essential for finding the accuracy and reliability of the empirical findings.

The results of diagnostic tests are shown in table 5.

Table 5: Results of Diagnostic Test for ARDL Model for Gini-Coefficient

Test Co-efficient, p-Value LM Lagrange Multiplier for no autocorrelation

1.6418, (0.215)

Ramsey reset test for functional form 1.1715, (0.293) Jarque-Bera test for normality 0.40919, (0.815) Regression of squared residual and square fitted residual for heteroscedasticity

1.4409, (0.240)

The results of the diagnostic tests suggested that there is no autocorrelation in the model.

The functional form of the model is correctly specified given by the Ramsey’s reset test. Jarque-Bera results show the acceptation of null hypothesis that all the estimates of model are normally distributed. It is also confirmed that there is no problem of heteroscedasticity in the given ARDL model.

4.5. Results for Stability of the Model Stability of the model is tested by CUSUM and CUSUMSQ test. The graph of CUSUM

indicates that it remains within the critical bounds but the graph of CUSUMSQ is out of the critical bounds so the model has the structural break (see Appendix A).

For the detection of structural break we employed the Chow Test. Results of Chow test are

shown in table 6.

Table 6: Results of Chow Test


170

F-statistics = 149973 Prob. F(9,11) = 0.0000

Log Likelihood ratio = 339.8090 Prob. Chi-Square(9) = 0.0003 Wald Statistics = 134976 Prob. Chi-Square(9) = 0.0000

The results have expressed the existence of structural break at 2001. F statistic, log likelihood

and Wald statistic indicate same results. So the alternative hypothesis is accepted, i.e. the existence of structural break.

4.6. Results of Gregory-Hansen Test For confronting the problem of structural instability we incorporated Gregory-Hansen test. Dtk = 0 if t < k Dtk = 1 if t > k Model 1: Cointegration with level shift

GINI = µ1 + µ2Dtk + β1CEXP + β2DEXP + β3INDT + β4FISCD+ β5FINAND + β6URBAN + et ….. (6) Model 2: Cointegration with regime shift

GINI = µ1 + µ2Dtk + β1CEXP + β11CEXPDtk + β2DEXP + β22DEXPDtk + β3INDT + β33INDTDtk + β4FISCD + β44FISCDDtk + β5FINAND + β55FINANDDtk + β6URBAN + β66URBAND + et …….. (7)

In the model of level shift a dummy is incorporated that shows a change in intercept after the

structural break. The regime shift model shows the impact on each coefficient when dummy is multiplied by each variable included in the model.

The Gregory-Hansen Technique by incorporating dummy variable in the OLS test is applied for the remedy of structural break. The results are shown in table 7 and 8.

Table 7: Results of Gregory-Hansen Test (with level shift) for Structural Break at 2001

Variable Coefficient t-Statistic Prob. CEXP -6.041169 -3.704122 0.0012* DEXP -2.402314 -4.366101 0.0002* INDT 3.208536 2.057053 0.0517** FISCD 0.321851 2.148204 0.0430**

FINAND 0.071962 0.947536 0.3537 URBAN 125.8862 4.240853 0.0003* DUMMY -2.067890 -2.604604 0.0162**

C -335.1303 -4.180753 0.0004* R2 = 0.899294 Adj R2 = 0.867251

F – stat = 28.06543 F- stat Prob = 0.000000 *significant at 1% and ** significant at 5%

Table 8: Results of Gregory-Hansen Test (with regime shift) for Structural Break at 2001

Variables Coefficients t-stat Prob. CEXP 0.527437 4.400917 0.0005* DEXP 0.057728 2.047926 0.0585**

INDT 0.211642 3.059660 0.0079* FISCD 0.002348 0.385924 0.7050

FINAND -0.003372 -0.677849 0.5082 URBAN 30.31396 16.88762 0.0000* DUMMY 75.63643 4.824409 0.0002* DUM_GI 1.000000 13.27957 0.0000*

DUM_CEXP -0.527437 -1.175275 0.2582 DUM_DEXP -0.057728 -0.232086 0.8196 DUM_INDT -0.211642 -0.468687 0.6460 DUM_FISCD -0.002348 -0.090047 0.9294

DUM_FINAND 0.003372 0.311728 0.7595


171

DUM_URBAN -30.31396 -4.961765 0.0002* C -75.63643 -15.12549 0.0000* R2 0.999923 Adj.R2 0.999850

F-stat 13847.04 F-stat Prob 0.000000 *Significant at 1%, **Significant at 5%

The results of Gregory-Hansen test show that dummy coefficient is having negative sign.

It shows that after structural break intercepts change downwards. Results of regime shift where intercept and slope coefficient change show that variables of Gini-coefficient and urbanization are significant while other variables are insignificant.

5. Discussion The discussion of the findings is based on the results of table 3 where the explanatory

variables have shown the effect in the long-run through ARDL technique. The results have shown that development expenditures have a negative impact on income

inequality in Pakistan. It is supported by the findings of Ali and Ahmed (2010). The explanation is based on the notion that the development affects are reaching the deprived class of the economy. The development expenditures create jobs and increase the incomes of the general people. It results into decreasing income inequality.

The fiscal deficit has increasing effect on income inequality in Pakistan. The fiscal deficit makes the government to take borrowings for deficit financing. The borrowing is not properly utilized which cause pushing down the growth rate and increasing the inflation. The phenomena enhance the income inequality in the economy. The other way of deficit financing used in Pakistan is printing of currency through State Bank of Pakistan. It also creates inflation and income inequality through lowering the purchasing power of the fixed income labor class.

The financial development has shown negative impact on income inequality. It explains that due to the financial development more credit is distributed to the private sector which increases the employment opportunities and choices. It reduces the income inequality. Furthermore, the financial development makes the loaning available to lower class which results into decreasing inequality in the economy.

Urbanization has a positive impact on income inequality. It explains the rural urban disparity. Urbanization rate in Pakistan is highest in South Asian countries. It is expected that urban and rural population would be equal in 2030. Wage differentials and good quality of life in cities compel rural people to migrate to urban areas. But the job opportunities in cities are already limited and migrants build pressure on the constrained job opportunities of the cities. The situation compels the migrants to live in slums and creating new slums in the cities. Deficit of houses and other social problems like lack of clean water and sanitation facilities make the lives of people miserable. So the urbanization process creates unequal classes even in the cities (Li and Zou, 2002).

6. Conclusion and Policy Implications We have attempted to see the impact of fiscal policy on income inequality in Pakistan by

ARDL approach. The empirical estimation shows that development expenditures as a component of fiscal policy is reducing income inequality in Pakistan. Financial development has also shown negative impact on Gini coefficient. On the other hand fiscal deficit and urbanization have shown increasing impact on income inequality.

To control the existing trend of income inequality in Pakistan it is recommended to enhance the ratio of development expenditures in the budget. Currently the ratio of such type of expenditures is not encouraging as in the last decade the development expenditures were 20 percent of the budget.

These development expenditures are distributed into public sector development program, rural support program, infrastructure, education and health projects. They may have the spillover effects, as in the current analysis urbanization has shown positive impact on income inequality. By providing the share of rural support program and allocating a significant part of the development expenditures in rural areas the income inequality may be declined.

The financial development has also emerged as an important tool to slide down the income inequality in Pakistan. The financial policy makers should further focus on financial development.


172

Fiscal deficit has also shown an increasing effect on income inequality. To narrow down the fiscal deficit, the current expenditures should be decreased as major part of the current expenditures goes to interest payments, defense expenditures and administration. The interaction of fiscal deficit, current expenditures and development expenditures with income inequality consequent on the point that current expenditures is the area needs government’s attention. It will show a spreading effect on decreasing the income inequality in the economy. However, for decreasing fiscal deficit the tax revenue may also be increased.

Urbanization process is increasing income inequality in the economy. To reduce such type of effect development projects should be introduced in the rural areas.

References: 1. Agnello, L. and Sousa, R. M. (2012) How Does Fiscal Consolidation Impact on Income

Inequality?. Banque De France No.382. 2. Ali, S. and Ahmad, N. (2010) The Effects of Fiscal Policy on Economic Growth: Empirical

Evidence based on the Time Series data from Pakistan. The Pakistan Development Review, 49 (4):497-512. 3. Barro, R. (2000) Inequality and Growth in a Panel of Countries. Journal of Economic Growth,

5:5-32. 4. Bernard, A. and Atta, D. (2010) Distributional Effects of Direct Taxes and Social Transfers (Cash

Benefits). In Income and Living Conditions in Europe (ed.) A. Atkinson and B. Malier. Brussels: European Union.

5. Brandolini, A. and Smeeding, T. (2009) Income Inequality in Richer and OECD Countries. In The Oxford Handbook of Economic Inequality (ed.) Oxford University Press.

6. Brenneman, A. and Kerf, M. (2002) Infrastructure and Poverty Linkages, A Literature Review. Washington, D.C.: The World Bank.

7. Calderon, C. and Serven, L. (2004) The Effect of Infrastructure Development on Growth and Income Distribution. Central Bank of Chile Working Papers No. 19, 20.

8. Claus, I., Martinez-Vazquez and Vulovic, V. (2014) Government Fiscal Policies and Redistribution in Asian Countries. In Kanbur, R., Rhee, C. and Z. Zhuang (eds.) Inequality in Asia and Pacific: Trends, Drivers and Policy Implications. New York: Asian Development Bank and Routledge, pp-173-201.

9. Davis, J. C. and Henderson, J. V. (2003) Evidence on the Political Economy of the Urbanization Process. Journal of Urban Economics, 53(1):98-125.

10. Decoster, A., Loughrey, J., O’Donoghue, C. and Verwerft, D. (2009) Micro Simulation of Indirect Taxes, AIM-AP-Project Paper, International Journal of Microsimulation, 4 (2): 41-56.

11. Donoaghue, C., Baldini, M. and Mantovani, D. (2004) Modelling the Redistributive Impact of Indirect Taxes in Europe :An Application of EUROMOD. Euromod Working Paper EM 7/01.

12. Dutt, A. K. (2001) Global Urbanization: Trends, form and Density Gradients. Allahabad: Professor Foundation Publications. Working Paper No. 21.

13. Fishman, R. and Love, I. (2004) Financial Development and Growth in Short and Long Run. National Bureau of Economic Research, Working Paper No. 10236.

14. Gallo, L, and Sagales, O. (2013) Is the Fiscal Policy Increasing Income Inequalities in Uruguay?. Journal of Faculty of Economics and Administration, 27.

15. Gregory, A. W. and B. E. Hansen (1996a) Residual-based Tests for Cointegration in Models with Regime Shifts. Journal of Econometrics, 70(1):99-126.

16. Gregory, A. W. and B. E. Hansen (1996b) Test for Cointegration in Models with Regime and Trend Shits. Oxford Bulletin of Economics and Statistics, 58(3):555-559.

17. Kaldor, N. (1956) Alternative Theories of Distribution, Review of Economic Studies, 23(2):94-100. 18. Khan, R. E. A. and Hye, Q. M. A. (2913) Financial Liberalization and Demand for Money: A Case

of Pakistan. Journal of Developing Areas, 47(2):175-198. 19. Kuznets, S. (1955) Economic Growth and Income Inequality. American Economic Review, 45(1):1-28. 20. Li, H. and Zou, H. (2002) Inflation, Growth and Income Distribution: A Cross Country Study.

Analysis of Economics and Finance, 3:85-101. 21. Patrick, H. T. (1966) Financial Development and Economic Growth in Underdeveloped Countries.

Economic Development and Cultural Change, 14:174-89. 22. Pesaran, M. H. and Shin, Y. (1999) An Autoregressive Distributed Lag Modeling Approach to Co-

integration Analysis, Cambridge University Press, Cambridge. 23. Petersen, M. A. and Rajan, R. G. (1997) Trade Credit: Theories and Evidence. Review of Financial

Studies, 10(3):661-691. 24. Muinelo-Gallo, L. and Roca-Sagalés, O. (2011) Economic Growth and Income Inequality: The

Role of Fiscal Policies. Australian Economic Papers, 50,74-97.


173

25. Ramos, X. and Roca-Sagales, O. (2008) Long Term Effects of Fiscal Policy on the Size and Distribution of Pie in the UK. Fiscal Studies, 29(3):387-411.

26. Samanta, K. and Cerf, G. (2009) Income Distribution and the Effectiveness of Fiscal Policy: Evidence from Some Transitional Economies. East-west Journal of Economics and Business, 12 (1):21-38.

27. Shahbaz, M. and Islam, F. (2011) Financial Development and Income Inequality in Pakistan: ARDL Approach, Journal of Economics Development, 3(1):1-22.

28. Shirazi, N. S., Ilyas, M. and Ahmed, M. (2001) Redistributive Effect of Fiscal Policy Across the Income Groups in the Urban Rural Areas of Pakistan. The Pakistan Development Review, 40(4):519-533.

29. Sylwester, K. (2002) Can Education Expenditure Reduce Income Inequality?. Economics of Education Review, 21:43-52. УДК 33

Налогово-бюджетная политика и неравенство доходов в Пакистане:

ARDL подход

1 Рана Эджаз Али Хан 2 Бухра Джабин Хашми

1-2 Исламский университет Бахавалпур, Пакистан 1 Доцент, заведующий кафедрой экономики E-mail: [email protected] 2 PhD E-mail: [email protected]

Аннотация. Данное исследование – это попытка выявить влияние фискальной политики на

неравенство доходов в Пакистане. Краткосрочная динамика контролируется моделью коррекции ошибок. Надежность модели протестирована с помощью диагностических тестов. Тест Чоу и техника Грегори-Хансена применяется для выявления структурных сдвигов. Результаты указывают на то, что расходы на финансовое развитие оказывают влияние на уменьшение неравенства в доходах. С другой стороны бюджетный дефицит и урбанизация влияют на неравенство доходов положительно. Текущие расходы от косвенных налогов не оказывают никакого влияния на коэффициент Джини. Исходя из полученных результатов исследования, рекомендуется следующее: бюджетный дефицит должен быть уменьшен за счет сокращения текущих расходов. Увеличение расходов по развитию потребует увеличение расходов по сокращению неравенства доходов. Возможность обеспечения финансового развития как инструмента снижения неравенства также рассматривается.

Ключевые слова: неравенство доходов, налогово-бюджетная политика, коэффициент Джини, ARDL, расходы по развитию, финансовый дефицит, финансовое развитие, урбанизация.




174

Appendix A

CUSUM of the model (Fiscal Policy and Income Inequality)

CUSUMSQ of the model (Fiscal Policy and Income Inequality)


175




Volatility in Sectors and National Income Growth: A Comparative Analysis of Pakistan and South Korea

1 Rana Ejaz Ali Khan

2 Tasnim Khan 3 Nadia Mahtab

1 The Islamia University of Bahawalpur, Pakistan Associate Professor and Chairman, Department of Economics E-mail: [email protected] 2 The Islamia University of Bahawalpur, Pakistan Associate Professor and Chairman, Department of Economics 3 The Islamia University of Bahawalpur, Pakistan M. Phil. Candidate, Department of Economics

Abstract The study empirically investigated the impact of sectoral volatility on fluctuations in

economic growth of Pakistan and South Korea (Korea). ADF unit root test is used to check the stationery of the data. Autoregressive Conditional Heteroscadasticity (ARCH) and Generalized Autoregressive Conditional Heteroscadasticity (GARCH) have been used for estimating the volatility in variables under analysis. The results revealed that there exists almost equal level of volatility in GDP growth rate of both countries. However, volatility in agriculture, industry, services, export and import sector varies for two nations. Greater volatility shocks exist in agriculture sector of Pakistan as compared to Korea. Volatility in industrial sector persists in Korean economy only but not to the greater extent. Almost equal level of volatility shocks have been observed in services and import sector of both economies. However, export sector has shown greater volatility shocks in Korea as compared to Pakistan. The results of regression analysis have shown that volatility in agriculture sector contributes more towards volatility in GDP growth of Pakistan as compared to Korea. The volatility in industrial sector almost equally contributes to volatility in GDP of both countries. On the other hand the volatility in services sector contributes more volatility in GDP of Korea as compared to Pakistan. The export sector volatility contributes to volatility in GDP of only Pakistan. Finally the imports sector volatility negatively impacts the GDP volatility in Pakistan but positively in Korea.

Keywords: international trade, agriculture sector, services sector, industrial sector, economic fluctuations, fluctuations in GDP.

JEL Classification: E32, O14, O40. Introduction The growth of the economies depends on characteristics of different economic sectors such as

export, import, agriculture, industry and services etc. The structural changes in these sectors



176

contribute volatility in growth rate of the economies (Koren and Tenreyro 2005). The share of agriculture sector in GDP has traditionally been dominated in pre-industrialized economies while the industrial and service sectors’ shares remain comparatively modest. In the industrialized countries like Japan, U.S.A and certain number of the European countries, the share of agriculture sector in GDP remain as little as 2 percent. The remaining share of production originates from the industrial and service sectors (Dutta 2009). For the industrializing economies like Taiwan, China and South Korea (Korea in the coming pages), etc. the share of industrial sector is increasing. On the other hand there is variety of the developing economies where agriculture sector is having a greater share in national output. However, Singapore is the economy having largest share of exports. Fluctuations in growth of these sectors may oscillate the GDP growth. For instance a major chunk in agriculture sector of Pakistan due to virus attack on cotton decreased the agricultural production in Pakistan. Consequently GDP growth rate of the economy remained lower for the early years of 1990s. Similarly the floods devastated the agricultural land in Southern Punjab and Sind in 2010 which affected the GDP growth rate of Pakistan economy.

An important ingredient of development is growth stability. Instability in the national income becomes costly for developing countries as it deters growth rate in the long-run (Mobarak 2005). It is attributed to volatility in major sectors of the economy*. The analysis of volatility in growth of economies and the factors behind this phenomenon can be useful for formulation of the policies for mitigating the volatility in national income growth.

However, in the empirical literature there are contradicting evidences of relationship between fluctuations in sectoral growth and fluctuations in GDP growth rate. Some of the researchers have found positive association between sectoral growth and economic growth volatility (Imbs 2007) while others (Aizenman and Marion 1993; Ramey and Ramey 1995) have found that sectoral volatility decrease the economic growth. These contradictions may be due to varying estimation techniques, nature and structure of the economies and sectors as well as proxies and variables used in the models. It makes the notion of contribution of sectoral volatility in economic growth volatility ambiguous.

Pakistan is a developing economy with average growth rate of 4.62, 3.66 and 6.28 percent in the last three decades respectively. On the other hand Korea is one of the Asian economic tigers having growth rate of 4.17, 6.18 and 8.74 percent in the last three decades. The largest sector of the Pakistan and Korean economies is services (See Ahmed and Ahsan 2011 for contribution of services sector in Pakistan). Pakistan is taking Korea as role model and trying to follow the initiatives that were taken by Korean economy during its past track record in the last sixty years. Actually Pakistan and Korea have started their economic progress since 1961 from per-capita GDP of $86.87 and $91.62 respectively. Currently the GDP per-capita of these economies is $1256.65 and $2259.15 respectively. That is why the current study concerned with these two economies†. For the relationship between fluctuations in sectoral growth and GDP growth fluctuations, the agriculture, industry, services, export and import sectors have been included in the analysis‡.

We will attempt to see the effect of volatility in growth of sectors on volatility in economic growth of two countries, i.e. Pakistan and Korea. The core objective of the study is to see the role of volatility in sectoral growth in volatility in GDP growth and making a comparison of implications of sectors as well as economies.

Review of Literature In the literature, there are evidences that the sectors in which economies specialize have a

significantly large effect on production and trade of these economies (Koren and Tenreyro 2005; Krishna and Levchenko 2009). Plethora of the studies have estimated the sectoral volatility and its effect on economic performance of the countries by using various econometric techniques and models (Hnatkovska and Loayza 2003; Mobarak 2005; Imbs 2007).

* Political instability, socio-cultural fluctuations in the form of ethnic disruptions, terrorism and strategic disturbances

along with natural and environmental changes also make fluctuations in growth rate of the economies (Iyigun and Owen

2004; Aisen and Veiga 2011). † See also, Khan, et. al. (2013a) for adaptation of Korean growth model for Pakistan. ‡ Although other sectors like financial sector also contribute towards volatility in GDP (Azid, et. al. 2006).

http://www.google.com.pk/search?tbo=p&tbm=bks&q=inauthor:%22M.+Dutta%22


177

The cross country association among volatility and economic growth rate was empirically investigated by Ramey and Ramey (1995).Two samples of countries were chosen for analysis. The first sample was consisted of 92 countries while the second was consisted of 24 OECD countries. The study used the data for the period 1960-1985 for the first sample and 1950-1988 for the second sample. They calculated the mean and standard deviation of per-capita annual growth rate over time period for each country and see its effect on GDP growth. They found that economies with higher volatility in per-capita growth rate had lower economic growth rate. The results also revealed that the investment as a share of GDP played a little part in association between volatility of per-capita annual growth rate to economic growth. However, the government spending volatility has shown negative impact on economic growth.

Azid, et al. (2006) investigated the impact of sectoral volatility on economic growth of Pakistan. The quarterly dataset for the time period 1971-72 to 2002-2003 was used. Volatility was estimated by using rolling variance of the series and GARCH. The output (GDP) as dependent variable and value added of agriculture, finance and insurance, services, industry and whole sale and retail as independent variables have been taken in the analysis. They found that every sector has a significant impact on the volatility of output growth except financial sector. However, association exists only for the short-run. There exists no long-run relationship between volatility of growth rate of different sectors of the economy and fluctuations in growth rate of output.

Imbs (2007) has analyzed the relationship between sectoral fluctuations and economic growth across countries. The study used the data for the time period 1963-19996 for 47 countries to demonstrate that velocity at the sectoral level and economic growth correlate in the same direction. Data includes yearly value added of different sectors, employment and factor of production in manufacturing activities published by United Nations Industrial Development Organization (UNIDO). It was found that across countries the relationship between economic growth and volatility was positive.

Koren and Tenreyro (2010) investigated volatility, diversification and development in the Gulf Cooperation Council (GCC) countries. The volatility was estimated as the deviation of output growth rate of a specified sector in a specified country from average growth rate of that sector over the time period. A positive covariance among sectoral shocks and volatility in GDP of specific country was found. The study concluded that GCC countries were more fluctuated as compared to other countries at the same level of economic development and it was due to their strong dependence on oil. The high levels of country specific fluctuations suggested that macroeconomic policies should be enhanced to alleviate volatility.

Babatunde (2013) investigated the relative contributions of stock market volatility on economic growth in Nigeria. The study used quarterly time series data for the years 1980:1 to 2010:4 for stock price index, real GDP, consumer price index as measure of economic activities, inflation and short-term interest rate. Volatility was estimated by Exponential Generalized Autoregressive Conditional Heteroskedasticity (EGARCH). The study revealed that the volatility shock was quite persistent in Nigeria and it might distort growth of the economy. The study recommended to make the stock market less volatile.

The reviewed studies have applied various analytical techniques to find the association between sectoral volatility and economic growth (as well as fluctuations in economic growth). We are focusing on the comparative analysis of volatility effect of some sectors of Pakistan and Korea on fluctuation of their GDP growth rates using the same time period data for both countries and employing ARCH and GARCH technique for volatility. To check structural breakpoint Chow test is applied.

Methodology Annual time series data for the years 1971-2010 taken from World Development Indicators

(World Bank 2012) is used for estimation of volatility and OLS regression. Time-series data often contains a unit root or non-stationarity. Ordinary least square estimates are impractical if in a model all the variables are non-stationary on level or if integration orders of all the variables are not zero. To check stationary properties of time-series data Augmented Dickey Fuller (ADF) unit root test is applied. Volatility of the variables under discussion, i.e. growth rates of national income, value added of agriculture, value added of industry, value added of services, exports, and imports is examined by applying Autoregressive Conditional Heteroscedasticity (ARCH) and Generalized


178

Autoregressive Conditional Heteroscedasticity (GARCH). After estimating volatility of all the variables Ordinary Least Square method is used to check the relationship between volatility in growth of different sectors and volatility in growth of output in both countries. To diagnose multicollinearity Variance Inflation Factor (VIF) is used. To check structural breakpoint or parameter stability of regression models Chow test is employed, furthermore Cumulative Sum (CUSUM) and Cumulative Sum of Squares (CUSUMSQ) tests are also employed to check stability of model. After observing structural breakpoint we apply OLS by using Gregory and Hansen methodology (Gregory and Hansen 1996a and 1996b). Finally, to check the effect of the shock in the volatility of one sector on the volatility in growth rate of output as a whole Impulse Response Function (IRF) is employed.

Theoretical Framework for Chow Test and Gregory and Hansen Methodology To check structural breakpoint we employ the Chow test. When regression model is used that

is involving time series data, it may happen that there is a structural change in the relationship between the regressand and the regressors. By structural change, it means that the values of the parameters of the model do not remain the same through the entire time period. After analyzing the structural break point we apply OLS using Greogry and Henson methodology.

The four models of Gregory and Hansen (1996a and 1996b) with assumptions about structural breaks and their specifications with two variables, for simplicity, are as follows:

Model 1: Level Shift

1 2 1t tk t tY d X e ...................................... (1)

Model 2: Level Shift with Trend

1 2 1 1t tk t tY d t X e …………………………. (2)

Model 3: Regime Shift where Intercept and Slope coefficients change

1 2 1 1 2t tk t t tk tY d t X X d e ………..…… (3)

Model 4: Regime Shift where Intercept, Slope coefficients and Trend change

1 2 1 2 1 2t tk tk t t tk tY d t d X X d e ……. (4)

Where t = time subscript and k = break date. Model Specifications Stationarity of the variables are checked under the step of unit root. If mean, variance and

auto covariance of a variable remains same no matter at what point we compute them, then variable is called stationary. In literature many tests are offered to detect that whether a series has a unit root or not. If stochastic terms are not correlated then Dickey Fuller test is applicable. But Dickey Fuller test is ineffective if the stochastic term is correlated. To solve this problem Augmented Dickey Fuller (ADF) test is being presented. ADF test solve this issue by augmenting the equations of DF test by adding the lagged values of the endogenous variable. We apply Augmented Dickey Fuller (ADF) test on growth rate of the variables to check the stationary property of the variables. The primary objective of this analysis is to test out the impact of volatility of various sectors on the growth rate of output. In this regard we also apply unit root test on the volatility variables obtained from ARCH and GARCH process as well. We can write ADF test in equation form as:

None (i.e. without intercept and Trend)

1 1t t t tY Y Y ……………..………………... (5)

With Intercept and no Trend

1 1 1t t t tY Y Y ………………………….. (6)

With Intercept and Trend

1 2 1 1t t t t tY Y Y ………………….. (7)

H0: = 1 for non-stationary process (null hypothesis).


179

H1: < 1 for stationary process (alternative hypothesis). To check volatility in the variables GARCH (1, 1) model (Bollerslev 1986) is used. General

form of GARCH (p, q) model is:

t t tY X u

t| ~ N(0,h )t tu

2

t 0

1 1

hp q

i t i j t j

i j

h u

which explains that the value of ht (i.e. variance parameter) depends on past values of the shocks (expressed by the lagged squared residuals terms) and on past values of variance (expressed by lagged ht terms). GARCH (1, 1) is the simplest form of GARCH (p, q) model. Variance equation for GARCH (1,1) model is:

2

0 1 1 1 1t t th h u

Where p shows the order of GARCH term and q shows the order of ARCH term. Model for growth rate of national output of Pakistan and Korea are:

GR_YPak = β0 + β1 GR_Yt-1+ ut ………………………………… (8) GR_YKor = π0 + π1 GR_Yt-1+ ut ………………………………… (9)

In the same way to check volatility of different sectors of both countries GARCH (1, 1) model

is applied on all independent variables. The operational definitions of the variables have been given in table-1.

Table 1: Operational Definitions of the Variables

Variables Operational Definitions GR_Y (Growth rate of GDP) Annual percentage growth rate of Gross Domestic

Product (GDP) at market prices based on constant local currency

GR_AGR (Growth rate of agriculture)

Annual growth rate of agricultural value added based on constant local currency

GR_IND (Growth rate of industry)

Annual growth rate of industrial value added based on constant local currency

GR_SER (Growth rate of services)

Annual growth rate of services value added based on constant local currency

GR_EXP (Growth rate of exports)

Annual growth rate of exports of goods and services based on constant local currency

GR_IMP (Growth rate of imports)

Annual growth rate of imports of goods and services based on constant local currency

Ordinary Least Square (OLS) methods for Pakistan and Korea are as:

VOL_YPak = β0 + β1VOL_AGR + β2VOL_IND + β3VOL_SER + β4VOL_EXP + β5VOL_IMP ………… (10)

VOL_YKor = π0 + π1VOL_AGR + π2VOL_IND + π3VOL_SER + π4VOL_EXP + π5VOL_IMP ……….. (11)

Where VOL_Y = Volatility in growth rate of GDP (for Pakistan and Korea) VOL_AGR = Volatility in growth rate of agriculture VOL_IND = Volatility in growth rate of industry VOL_SER = Volatility in growth rate of services VOL_EXP = Volatility in growth rate of exports VOL_IMP = Volatility in growth rate of imports


180

In our analysis for remedy of structural break we incorporate a dummy variable such that: Dtk = 0 if t ≤ k, Dtk = 1 if t > k

The model specification after incorporating dummy variable for Pakistan is as: Model 1: Level Shift VOL_Yt = µ1 + µ2Dtk + α1VOL_AGR + α2VOL_IND + α3VOL_SER + α4VOL_EXP +

α5VOL_IMP +et …..(12) Model 2: Regime Shift where Intercept and Slope Coefficient Change VOL_Yt = µ1+µ2Dtk+ α1VOL_AGR+ α11VOL_AGR Dtk + α2VOL_IND+ α22VOL_IND

Dtk+α3VOL_SER + α33VOL_SER Dtk+ α4VOL_EXP + α44VOL_EXP Dtk + α5VOL_IMP + α55VOL_IMP Dtk + et ……………. (13)

In the same way model specification after incorporating dummy variable for Korea is as: Model 1: Level Shift

VOL_Yt = σ1+ σ2 Dtk +β1VOL_AGR + β2VOL_IND + β3VOL_SER + β4VOL_EXP + β5VOL_IMP +et … (14)

Model 2: Regime Shift where Intercept and Slope Coefficient Change VOL_Yt = σ1+ σ2Dtk+β1VOL_AGR+ β11VOL_AGR Dtk + β2VOL_IND + β22VOL_IND Dtk +

β3VOL_SER + β33VOL_SER Dtk + β4VOL_EXP + β44VOL_EXP Dtk + β5VOL_IMP + β55VOL_IMP Dtk + et …............... (15)

Empirical Results ARCH and GARCH Estimates for Volatility Before doing the ARCH and GARCH process to check the stationarity of the time series ADF

unit root test has been applied on all the variables. The results of ADF unit root test of all the variables about Pakistan and Korea are reported in table 2 and 3 respectively. The results express that all the variables for both countries are stationary at level 1 percent at Mackinon Critical values, with intercept, and with trend and intercept. It means that all of the variables are integrated of order zero or I (0) for both countries.

Table 2: Unit Root Test of Variables for Pakistan

ADF Statistics

Variables Intercept Trend and Intercept

None

Level Level Level First Difference

GR_Y -4.846 -4.964 -1.436 -10.172 GR_AGR -8.098 -7.966 -0.926 -7.429 GR_IND -5.127 -3.679 -0.635 -3.119 GR_SER -4.230 -4.769 -1.343 -7.230 GR_EXP -6.355 -6.300 -5.202 -8.398 GR_IMP -5.799 -5.636 -5.368 -8.265

Critical values (1%)

-3.610 -3.211 -2.625 -2.628


-2.938 -3.529 -1.949 -1.950


-2.607 -3.196 -1.611 -1.611

When unit root is applied on without trend and intercept data is stationary at first difference.

This means that without trend and intercept data of the variables is integrated of order one or I (1).


181

Table 3: Unit Root Test of Variables for Korea

ADF Statistics Variables Intercept Trend and

Intercept None

Level Level Level First Difference

GR_Y -4.979 -5.672 -0.859 -6.128 GR_AGR -9.618 -5.537 -8.376 -6.432 GR_IND -4.471 -5.580 -1.741 -5.708 GR_SER -4.029 -4.506 -1.062 -9.590 GR_EXP -4.994 -5.472 -2.737 -8.732 GR_IMP -6.273 -6.467 -1.826 -6.597


-3.610 -4.211 -2.632 -2.632


-2.938 -3.529 -1.950 -1.950


-2.607 -3.196 -1.611 -1.611

The variables under analysis are stationary at level with intercept, and with trend and

intercept. The ADF statistics are found significant at 1 percent. Without trend and intercept data is stationary at first difference and all variables are found significant at 1 percent.

The results of ARCH and GARCH process for Pakistan and Korea are reported in Table 4 and 5 respectively (see Annexure I and II for conditional variance graph).

Table 4: Results (Variance) of GARCH (1, 1) Model for Pakistan

Variable Coefficient z- statistic Prob.

Dependent variable: GR_YPak

Variance Equation Constant 0.41832 4.197325 0.0000*

RESID(-1)^2 -0.20277 -318554 0.0014* GARCH(-1) 1.1576 16.6187 0.0000*

Dependent variable: GR_AGRPak Variance Equation

Constant 1.51948 1.7756 0.0758** RESID(-1)^2 -0.23768 -2.5206 0.0117* GARCH(-1) 1.10551 29.4014 0.0000*

Dependent variable: GR_INDPak Variance Equation

Constant 3.75196 0.32315 0.7466 RESID(-1)^2 0.02583 0.08488 0.9324 GARCH(-1) 0.64601 0.56433 0.5725

Dependent variable: GR_SERPak Variance Equation

Constant 0.11619 1.10872 0.2675* RESID(-1)^2 -0.16663 -1.93261 0.0533* GARCH(-1) 1.13293 9.53239 0.0000

Dependent variable: GR_EXPPak Variance Equation

Constant 91.4484 1.79961 0.0719** RESID(-1)^2 -0.25416 -2.64880 0.0081* GARCH(-1) 0.62807 1.74888 0.0803**


182

Dependent variable: GR_IMPPak Variance Equation

Constant 23.1593 0.89456 0.3710* RESID(-1)^2 -0.25387 -2.75603 0.0059* GARCH(-1) 1.08164 4.81253 0.0000*

* represents 5 percent level and ** represents 10 percent level of significance No. of observations = 39

Table 5: Results (Variance) of GARCH (1, 1) Model for Korea

Variable Coefficient z- statistic Prob. Dependent variable: GR_YKor

Variance Equation Constant 0.26624 0.08485 0.9324

RESID(-1)^2 -0.19411 -2.05200 0.0402* GARCH(-1) 1.16574 4.13717 0.0000*

Dependent variable: GR_AGRKor Variance Equation

Constant 9.22264 0.39134 0.6955 RESID(-1)^2 0.21266 0.49421 0.6212 GARCH(-1) 0.53168 0.57478 0.5654

Dependent variable: GR_INDKor Variance Equation


Dependent variable: GR_SERKor Variance Equation

Constant -0.32524 -0.29942 0.7646 RESID(-1)^2 -0.22158 -1.34765 0.1778 GARCH(-1) 1.20465 4.11883 0.0000*

Dependent variable: GR_EXPKor Variance Equation


Dependent variable: GR_IMPKor Variance Equation


* represents 5 percent and ** represents 10 percent level of significance. No. of observations = 39 Empirical results reported above are obtained from GARCH (1, 1) model. For convenience we

denote ARCH parameter by α and GARCH parameter by β. To check volatility we add the ARCH and GARCH coefficients (α + β). If the sum is very close to 1, it indicates that volatility shocks are persistent to the greatest extent and if the sum is very close to 0 it indicates that there is no persistent of volatility shocks. From tables 4 and 5 variance equation for GR_YPak and GR_YKor shows that the sum of the lag squared error term and lagged value of variance i.e. (α + β) is equal to 0.95483 and 0.97163 respectively for Pakistan and Korea, which indicated that volatility shocks in GDP growth rate of Pakistan and Korea are persistent to a greater extent. The estimates of ARCH and GARCH coefficients are also highly significant.


183

Similarly the variance equation for GR_AGR, GR_IND, GR_SER, GR_EXP and GR_IMP show that the sum of the lagged square error term and lagged value of variance is equal to 0.86783 and 0.97163, 0.67184 and 0.5049, 0.9663 and 0.98307, 0.37391 and 0.87266, and 0.8277 and 0.9816 respectively for Pakistan and Korea which indicates that the volatility in growth rate of agriculture, services and imports for Pakistan and Korea are persistent to a greater extent while volatility in growth rate of exports are not much more persistent for Pakistan, and the estimates of ARCH and GARCH coefficients are also highly significant except coefficient of industry for Pakistan.

4.2 OLS Estimates

We have applied the unit root test on volatility variables. The results of the ADF unit root tests of all the variables about Pakistan and Korea are presented in table 6 and 7 respectively.

Table 6: Unit Root Test of the Volatility in Variables for Pakistan

ADF Statistics

Variables Intercept Trend and Intercept

None

VOL_Y -8.426 -8.895 -8.522 VOL_AGR -6.045 -5.953 -6.069 VOL_IND -5.105 -5.114 -6.589 VOL_SER -6.322 -6.743 -6.335 VOL_EXP -6.150 -6.055 -6.160 VOL_IMP -6.684 -6.556 -6.741


-3.615 -4.219 -2.627


-2.941 -3.533 -1.949


-2.609 -3.198 -1.611

Table 7: Unit Root Test of the Volatility in Variables for Korea

ADF Statistics Variables Intercept Trend and

Intercept None

VOL_Y -5.762 -6.619 -5.795 VOL_AGR -4.757 -4.803 -4.793 VOL_IND -5.489 -6.196 -5.567 VOL_SER -5.970 -6.656 -6.053 VOL_EXP -5.667 -5.881 -5.737 VOL_IMP -5.463 -5.850 -5.539


-3.615 -4.219 -2.627


-2.941 -3.533 -1.949


-2.609 -3.198 -1.611

The results of table 6 and 7 show that with intercept or with trend and intercept or without

trend and intercept all the variables are stationary at level one or I (0) i.e. integrated of order zero.


184

Before employing the OLS we listed results of multicollinearity for both countries. To diagnose multicollinearity we have employed Variance Inflation Factor (VIF) test, and the results are presented in table 8 and 9 for Pakistan and Korea respectively.

Table 8: Test of Multicollinearity for Pakistan

Variables Coefficient Variance

Uncentered VIF

Centered VIF

Constant 0.04123 1.08190 NA VOL_AGR 0.00359 1.11153 1.10368 VOL_IND 0.00363 1.24991 1.24908 VOL_SER 0.00929 1.20871 1.19634 VOL_EXP 0.00026 1.11606 1.08341 VOL_IMP 0.00023 1.13029 1.10552

The larger is the value of VIF the higher collinear in the variable Xj. In the limit VIF can be

finite in case of perfect colinearity. A rule of thumb, if the VIF of a variable is greater than 10, which will happen if R2j is higher than 0.90, that variable is said be highly collinear. Our results show that the value of VIF of variables VOL_AGR, VOL_IND, VOL_SER, VOL_EXP and VOL_IMP are 1.106, 4.491, 2.438, 1.755 and 3.077 respectively. It shows that as explanatory variables are not collinear with each other. Multicollinearity doesn’t exist in the model.

Table 9: Test of Multicollinearity for South Korea

Variables Coefficient

Variance Uncentered

VIF Centered

VIF Constant 0.02504 1.02946 NA

VOL_AGR 0.00075 1.10613 1.09659 VOL_IND 0.00277 4.49174 4.49085 VOL_SER 0.00957 2.43806 2.43417 VOL_EXP 0.00032 1.75586 1.73359 VOL_IMP 0.00062 3.07731 3.07714

The results of multicollinearity for Korea are presented in Table 9. The results show that the

value of VIF of variables VOL_AGR, VOL_IND, VOL_SER, VOL_EXP and VOL_IMP are 1.106, 4.491, 2.438, 1.755 and 3.077 respectively. It shows that explanatory variables are not collinear with each other and multicollinearity doesn’t exist in our results.

To check the relationship between volatility in growth of different sectors and volatility in GDP growth for Pakistan and Korea, OLS models have been applied. The results of multivariate regression analysis are presented in Table 10. Dependent variable is volatility in GDP growth rate and independent variables are volatility in growth rate of different sectors.

Table 10: Regression Analyses of Volatility for Pakistan and Korea

Pakistan South Korea

Model Coefficient (Prob.)

t-values Coefficient (Prob.)

t-values

Constant -0.047 (0.848) -0.192 -0.336 (0.006)* -2.885 VOL_AG

R 0.201 (0.009)* 2.778 0.095 (0.000)* 4.719

VOL_IND 0.249 (0.001)* 2.456 0.207 (0.000)* 10.439 VOL_SER 0.285 (0.019)* 3.434 0.751 (0.000)* 5.343 VOL_EXP 0.035 (0.080)* 1.801 -0.017 (0.199) -1.308 VOL_IMP -0.033

(0.080)** -1.800 0.059 (0.002)* -2.886


185

R2 = 0.582 Adj. R2 = 0.519 F-stat = 9.194

R2 = 0.966 Adj. R2 = 0.961 F-stat = 189.71

Dependent Variable: VOL_Y * represents 5 percent and ** represents 10 percent level of significance. No. of observations = 39 Results obtained from multivariate regression analyses of both countries show that volatility

in each sector has a significant impact on volatility in GDP except volatility in export sector in Korea.

Structural or Parameter Stability: The Chow Test In order to investigate the structural breakpoint we have employed the Chow test. The results

of Chow test are presented in table 11 and 12 for Pakistan and Korea respectively.

Table 11: Estimates of Chow Test for Pakistan

F-statistic 2.085676 Prob. F(12,21) 0.0883 Log likelihood ratio 14.85196 Prob. Chi-Square(12) 0.0214

Wald Statistic 12.51406 Prob. Chi-Square(12) 0.0514 Chow Breakpoint Test: 1986

Table 12: Estimates of Chow Test for South Korea

F-statistic 3.762077 Prob. F(12,21) 0.0075

Log likelihood ratio 23.69635 Prob. Chi-Square(12) 0.0006 Wald Statistic 22.57246 Prob. Chi-Square(12) 0.0010

Chow Breakpoint Test: 1998 Results in tables 11 and 12 show that there exists structural breaks in both economies of

Pakistan and Korea. The F-statistics is significant at 5 percent level so we reject the null hypothesis (i.e. there is no structural breakpoint). Breakpoint for Pakistan is 1986 while for Korea it is 1998. For remedy we run OLS by using Gregory and Henson methodology. The results of both models (i.e. level shift as well as intercept and slope coefficient change) for Pakistan are presented in table 13 and 14.

Table 13: Multi Regression Estimates of Pakistan after Incorporating Dummy

Model 1 (DUM 1986)

Variables Coefficient t-Statistic Prob. Constant 5.701549 * 17.71394 0.0000

VOL_AGR 0.235593 * 4.232902 0.0002 VOL_IND 0.293027 * 5.279901 0.0000 VOL_SER 0.193430 * 2.111782 0.0426 VOL_EXP 0.047494 * 3.160354 0.0034 VOL_IMP -0.003513 -0.249951 0.8042

DUM -1.056151 * -2.651447 0.0124 R2 = 0.775 Adj. R2 = 0.732 F-stat. = 18.38

* indicates 5 percent level of significance. The year relevant to dummy variable is indicated in the first row in the parentheses. DUM 1986 means that the dummy is unity after that year and so on.


186

Table 14: Multi Regression Estimates of Pakistan after Multiplying Dummy with all Regressors

Model 2 (DUM 1986)


VOL_AGR 0.195641 ** 1.942740 0.0625 VOL_IND 0.335071 * 3.266962 0.0030 VOL_SER 0.124453 1.087456 0.2864 VOL_EXP 0.076193 * 3.303988 0.0027 VOL_IMP -0.021035 -0.964484 0.3434

DUM -1.174015 * -2.872100 0.0078 DUM ×

VOL_AGR 0.047814 0.394128 0.6966 DUM ×

VOL_IND -0.078683 -0.628946 0.5347 DUM ×

VOL_SER 0.207101 0.914496 0.3686 DUM ×

VOL_EXP -0.055612 ** -1.820512 0.0798 DUM ×

VOL_IMP 0.024377 0.801886 0.4296 R2 = 0.812 Adj. R2 = 0.736 F-stat. = 10.64

* and ** indicate 5 percent and 10 percent level of significance respectively. The year relevant to dummy variable is indicated in the first row in the parentheses. DUM 1986 means that the dummy is unity after that year and so on.

The results of both models (i.e. level shift as well as intercept and slope coefficient change)

for Korea are presented in table 15 and 16.

Table 15: Multi Regression Estimates of Korea after Incorporating Dummy

Model 1 (DUM 1998) Variables Coefficient t-Statistic Prob. Constant 6.982710 * 39.34048 0.0000

VOL_AGR 0.090700 * 3.725465 0.0008 VOL_IND 0.181400 * 3.887832 0.0005 VOL_SER 0.647208 * 7.051567 0.0000

VOL_EXP -0.030888

** -1.915704 0.0644 VOL_IMP 0.089770 * 4.058623 0.0003

DUM -1.037664 * -3.175781 0.0033 R2 = 0.954 Adj. R2 = 0.946 F-stat. = 112.30

* and ** indicate 5 percent and 10 percent level of significance respectively. The year relevant to dummy variable is indicated in the first row in the parentheses. DUM 1998 means that the dummy is unity after that year and so on.

Table 16: Multi Regression Estimates of Korea after Multiplying Dummy with all Regressors

Model 2 (DUM 1998)



187

VOL_AGR 0.109409 * 4.442454 0.0001 VOL_IND 0.174628 * 3.734498 0.0009 VOL_SER 0.851075 * 6.639564 0.0000 VOL_EXP -0.035645 * -2.232308 0.0341 VOL_IMP 0.076586 * 3.271359 0.0029

DUM -1.178206 * -3.713899 0.0009 DUM ×

VOL_AGR -0.132187 * -1.976347 0.0584 DUM ×

VOL_IND -0.090618 -0.635177 0.5307 DUM ×

VOL_SER -0.366302 ** -1.708792 0.0990 DUM ×

VOL_EXP -0.050455 -0.934637 0.3583 DUM ×

VOL_IMP 0.116796 ** 1.826955 0.0788 R2 = 0.967 Adj. R2 = 0.954 F-stat. = 73.12

* and ** indicates 5 percent and 10 percent level of significance respectively. The year relevant to dummy variable is indicated in the first row in the parentheses. DUM 1998 means that the dummy is unity after that year and so on.

The empirical results of table 13 and 15 show that after incorporating dummy variable

coefficients of (DUM × intercept) are significant and having negative sign for both countries, which show that after structural break, intercept change downwards. Results of model 2 (i.e. regime shift where intercept and slope coefficient change) for Pakistan and Korea presented in table 14 and 16 show that for Pakistan, after structural break the estimates of all slope coefficients are insignificant except exports, and for Korea, the estimates of agricultural, services and exports are significant while coefficient of industrial and imports sector insignificant.

Stability of the Models In order to investigate the stability of models we have employed cumulative sum (CUSUM)

and cumulative sum of squares (CUSUMSQ) tests. Pesaran, et al. (1999, 2001) proposed CUSUM and CUSUMSQ tests for estimating the stability of long and short-run estimates. Figures 1 indicates that plots for CUSUM and CUSUMSQ are between the critical boundaries at 5 percent level of significance for Pakistan.

Similarly, figures 2 indicates that plots for CUSUM and CUSUMSQ are between the critical boundaries at 5 percent level of significance for Korea. Both of the figures demonstrate the stability of models for both countries.

Figure 1: CUSUM Charts for Pakistan


188

Figure 2: CUSUM Charts for Korea

Discussion The estimates of ARCH and GARCH, and OLS are quite significant and need economic

explanation. The results regarding the volatility in the sectors and GDP growth rates and the impact of volatility in sectoral growth rates on volatility in growth rate of GDP of Pakistan and Korea are discussed here.

Volatility in Agriculture Sector The results demonstrate that agriculture sector in Pakistan and Korea contains volatilities in their

growth rates. Volatility in growth rate of GDP is positively influenced by volatility in growth of agriculture in Pakistan and Korea. In Pakistan 20 percent volatility of growth rate of output is caused by one percent volatility in growth rate of agricultural sector while in Korea 9 percent volatility of growth rate of output is caused by one percent volatility in growth rate of agricultural sector. Higher volatility effect is shown in Pakistan. It may be due to the fact that the share of agriculture sector in GDP is higher in Pakistan as compared to Korea. On the other hand the share of value added of agriculture in GDP has been significantly declining (Kniivila 2007) in Korea. The agriculture is a slow-moving sector of the economy as compared to industrial and services sector.


189

Volatility in Industrial Sector Industrial growth has a vital role in the economic growth of countries like Indonesia, China,

Taiwan and Korea. The OLS estimates for Pakistan and Korea have shown that volatility in growth rate of industry is positively affecting the volatility in growth of GDP in Korea only. The 24 percent increase in volatility of GDP growth in Pakistan and 20 percent in Korea is caused by one percent volatility in growth of industrial sector. The results are supported by Medyawati and Yunanto (2011) for Indonesia and Azid, et. al. (2006) for Pakistan. It partially explains that industrial sector is still the “engine” of economic growth (see also Linden and Mahmood 2007 for Schengen countries).

Volatility in Services Sector The role of services in economic growth has been empirically evidenced by a number of

studies (Linden and Mahmood 2007) along with theoretical support from Kuznet (1957). Our results have shown that in Pakistan approximately 29 percent volatility in GDP growth rate is caused by one percent volatility in services sector while in Korea 75 percent volatility in growth of output is caused by one percent volatility in growth of services sector. It explains that volatility shocks in services sector have higher effect on volatility of growth of output in Korea as compared to Pakistan. The explanation may be that services sector contribute highest ratio to GDP in Korea. The sector has also strong correlation to other economic sectors. It provides necessary skilled labor force to agriculture as well as manufacturing sector (Ahmed and Ahsan 2011 for Pakistan). Our results have further revealed that volatility in growth of services sector contributes highest to volatility in growth of output as compared to other sectors of the economy like agriculture, industry and exports in both Korea and Pakistan (see also Azid, et. al. 2006 for Pakistan).

Volatility in Exports The literature explains that exports play an important role in economic growth (Fosu 1990;

Zang and Baimbridge 2012; Gilbert, et. al. 2013; Khan, et. al. 2013b). Our results have shown that in Pakistan 3 percent volatility in growth of output is caused by one percent volatility in growth of exports. However, volatility in exports contributes lowest to volatility of growth of output in Pakistan as compared to other sectors. This may be due to the fact that the share of exports in GDP is comparatively lower than other sectors like services and agriculture.

Volatility in Imports Our results have shown that volatility shocks in growth of imports have significantly affected the

volatility of growth of output in Korea. The estimates have shown that 5 percent volatility of growth of output is caused by one percent volatility in growth of imports. It may be explained on the fact that imports play an important role in growth of output through different channels. For Pakistan volatility in growth of output is negatively influenced by volatility of growth of imports. The explanation may be that Pakistan’s imports are highly concentrated with raw material like machinery, petroleum, chemicals, edible oil, transport equipment, iron and steel and fertilizer. The most important is the import of cotton for textile sector in Pakistan that is the largest manufacturing sector of the country. All the raw material for textile industry is produced domestically but whenever it is needed due to lower production of cotton is imported. It makes the imports volume fluctuated but smoothing the GDP growth rate. The imported raw material is used in the production of final goods, the volatility in imports may decrease in volatility of growth rate of output.

Conclusion and Policy Recommendation The empirical analysis through ARCH and GARCH technique made us to conclude that

There exists approximately equal level of volatility in GDP growth rate of both countries, i.e. Pakistan and Korea.

The agriculture growth has shown greater volatility shocks in Pakistan as compared to Korea.

Volatility in growth of industrial sector also existed only in Korea but not to a greater extent.

Volatility in services sector indicated the existence of approximately equal level of shocks in both of the economies.

Volatility shocks in export sector have shown higher level in Korea as compared to Pakistan.


190

The volatility in growth of imports indicated approximately equal level of volatility shocks in Pakistan and Korea.

The OLS results for the influence of volatility in different sectors on volatility in GDP growth rate are concluded as:

The volatility in agriculture sector is causing relatively higher volatility in GDP growth rate of Pakistan than that of Korea.

The results of volatility in industry show that it contributes to volatility in growth of output in Korea.

The volatility in growth rate of services sector contributes more to volatility in growth rate of output in Korea than Pakistan.

The volatility of services sector is causing high volatility in GDP growth in Pakistan as well as Korea as compared to other sectors under analysis. The empirical evidences have shown that the share of services sector increases as the economy passes through the stages of development. There is a need to stabilize the services sector, which contribute to stabilize economic growth of the country. It may be recommend that there is need to equip labor force with education, skill and advance technical knowhow which reduces the unemployment and increases the productivity. Information technology may be a good tool to equip the labor force and stabilize the services sector and consequently the GDP growth. Human resource development should be stressed in areas like health, nutrition, training and education.

The multivariate analysis has shown positive effect of volatility in exports on volatility in GDP growth in Pakistan only.

The results of multivariate regression show that volatility in growth of imports effect volatility in growth of output negatively in Pakistan while positively in Korea.

The major imports in Pakistan are raw material for industrial sector and inputs for agricultural sector. The agricultural sector again produces the raw material to industrial sector. There is a need to stabilize the agricultural sector in Pakistan. It is also needed to establish its linkage with industries and to provide advance technology to this sector. There is also needed to stabilize the industrial sector. Industrial sector in Pakistan is largely depending upon the imported raw material in the form of fertilizers, chemicals, pesticides, etc. along with occasionally imported raw cotton (for the years when domestic cotton crop is damaged). So the industrial sector production is linked with imports. The smooth functioning of the industrial sector requires the imports at the needed level. So flexibility in imports to meet the needs of the industrial sector may result into smooth production of industrial sector. The empirical estimates have shown that imports are negatively affecting volatility in GDP growth in Pakistan. Raw material requirements comprising chemicals, fertilizers and cotton are fulfilled by imports if they are not flexible for the requirements, the GDP will fluctuate. If the imports are fixed by quota or tariff the GDP will fluctuate. The notion ultimately supports the proponents of WTO and globalization where liberal imports would result into smoothing the GDP growth rate.

References: 1. Ahmed, A. and Ahsan, H. (2011) Contribution of Services Sector in the Economy of Pakistan.

Working Paper No.79. Pakistan Institute of Development Economics, Islamabad. 2. Aisen, A. and Veiga, F. J. (2011) How Does Political Instability Affect Economic Growth? IMF

Working Paper No.12. Middle East and Central Asia Department. IMF Washington, D.C. 3. Aizenman, J. and Marion, N. (1993) Policy Uncertainty, Persistence and Growth. Review of

International Economics, 1(2):145-63. 4. Azid, T., Khaliq, N., and Jamil, M. (2006) Sectoral Volatility, Development and Governance: A

Case Study of Pakistan. The Pakistan Development Review, 45(4):797-817. 5. Babatunde, O. A. (2013) Stock Market Volatility and Economic Growth in Nigeria (1980-2010).

International Review of Management and Business Research, 2(1):201-209. 6. Bollerslev, T. (1986) Generalized Autoregressive Conditional Heteroskedasticity. Journal of

Econometrics, 32:307-327. 7. Dutta, M. (2009) The Asian Economy and Asian Money. Emerald Group Publishers. 8. Fosu, A. K. (1990) Export Composition and the Impact of Export on Economic Growth of

Developing Economies. Economic Letters, 34:67-71.


191

9. Gilbert, N.A., Linyong, S. G., and Divine, M. G. (2013) Impact of Agricultural Export on Economic Growth in Cameron: Case of Banana, Coffee and Cocoa. International Journal of Business and Management Review, 1(1):44-71.

10. Gregory, A. W. and Hansen, B. E. (1996a) Residual-based Tests for C cointegration in Models with Regime Shifts, Journal of Econometrics, 70(1):99-126.

11. Gregory, A. W. and Hansen, B. E. (1996b) Tests for Cointegration in Models with Regime and Trend Shifts. Oxford Bulletin of Economics and Statistics, 58(3):555-60.

12. Hnatkovska, V. and Loayza, N. (2003) Volatility and Growth. Policy Research Working Paper Series No. 3184. The World Bank, Washington, D.C.

13. Imbs, J. (2007) Growth and Volatility. Journal of Monetary Economics, 54(7):1848-1862. 14. Iyigun, M. F. and Owen, A. L. (2004) Income, Inequality, Financial Development and

Macroeconomic Fluctuations. The Economic Journal, 114(495):352-375. 15. Khan, R. E. A., Batool, N. and Sarwar, J. (2013a) Adoptability of Korean Growth Model to

Developing Economies: The Case of Pakistan. Middle East Journal of Scientific Research, 13:43-49. 16. Khan, R. E. A., Gill, A. R and Bashir, H. N. (2013b) Primary Export Led Growth Hypothesis: A

Case Study of Pakistan. Actual Problems of Economics, 141(3):465-472. 17. Kniivila, M. (2007) Industrial Development and Economic Growth: Implications for Poverty

Reduction and Income Inequality. Working Paper. Department of Economic and Social Affairs. United Nations.

18. Koren, M. and Tenreyro, S. (2005) Volatility and Development. Centre for Economic Policy Research. 5307. London.

19. Koren, M. and Tenreyro, S. (2010) Volatility, Diversification and Development in the Gulf Cooperation Council Countries. Research Paper. Kuwait Programme on Development, Governance and Globalisation in the Gulf States.

20. Krishna, P. and Levchenko, A. A. (2009) Comparative Advantage, Complexity, and Volatility. Working Paper No.14965. National Bureau of Economic Research.

21. Kuznets, S. (1957) Quantitative Aspects of the Economic Growth of Nations: II, Industrial Distribution of National Product and Labour Forces. Economic Development and Cultural Change, 5(4).

22. Linden, M. and Mahmood, T. (2007) Long Run Relationships Between Sector Shares and Economic Growth–A Panel Data Analysis of the Schengen Region. Discussion Paper No.50. University of Joensuu.

23. Medyawati, H. and Yunanto, M. (2011) The Role of Banking, Agriculture and Manufacturing Sector in Economic Growth in Indonesia: Do They Influence?. International Journal of Trade, Economics and Finance, 2(4):312-317.

24. Mobarak, A. M. (2005) Democracy, Volatility, and Economic Development. The Review of Economics and Statistics, 87(2):348-361.

25. Pesaran, M. H., Shin, Y. (1999) An Autoregressive Distributed lag Modeling Approach to Cointegration Analysis. Chapter 11 in Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch Centennial Symposium, Strom S. (ed.). Cambridge University Press, Cambridge.

26. Pesaran, M. H., Shin, Y. and Smith, R. J. (2001) Bounds Testing Approaches to the Analysis of Level Relationships. Journal of Applied Econometrics, 16(3): 289-326.

27. Ramey, G. and Ramey, V. A. (1995) Cross-Country Evidence on the Link between Volatility and Growth. The American Economic Review, 85(5):1138-1151.

28. World Bank (2012) World Bank Development Indicator. World Bank Washington, D.C. 29. Zang, W. and Baimbridge, M. (2012) Exports, Imports and Economic Growth in South Korea and

Japan: A Tale of Two Economies. Applied Economics, 44(3):361-372.

УДК 33

Волатильность в секторах и рост национального дохода: сравнительный анализ

Пакистана и Южной Кореи

1 Рана Эджаз Али Хан 2 Тасним Хан

3 Надиа Махтаб 1-3 Исламский университета Бахавалпур, Пакистан 1 Доцент, заведующий кафедрой экономики E-mail: [email protected]


192

Аннотация. Авторы эмпирически исследовали влияние отраслевой волатильности на колебания экономического роста Пакистана и Южной Кореи. АDF-тест используется для проверки стационарных данных. Методы ARCH и GARCH были использованы для оценки волатильности анализируемых переменных. Результаты показали, что существует почти одинаковый уровень волатильности темпов роста ВВП в обеих странах. При этом волатильность в сельском хозяйстве, промышленности, услугах, торговле варьируется у обеих стран. Существуют большие сдвиги в волатильности сельскохозяйственного сектора Пакистана по сравнению с Кореей. Волатильность в промышленном секторе сохраняется в корейской экономики, но нельзя, что в большей степени в сравнении с Пакистаном. Почти одинаковый уровень волатильности был обнаружен в сфере услуг и секторе импорта обеих стран. Тем не менее, экспортный сектор показал больший уровень волатильности в Корее по сравнению с Пакистаном. Результаты регрессионного анализа показали, что волатильность в сельскохозяйственном секторе способствует большей волатильности роста ВВП Пакистана по сравнению с Кореей. Волатильность в промышленном секторе практически одинаково способствует нестабильности в ВВП обеих стран. С другой стороны, волатильность в секторе услуг способствует большей волатильности в ВВП Кореи по сравнению с Пакистаном. Волатильность экспортного сектора способствует нестабильности в ВВП только Пакистана. Наконец, волатильность сектора импорта негативно влияет на волатильность ВВП в Пакистане, но при этом положительно в Корее.

Ключевые слова: международная торговля, сектор сельского хозяйства, сектор услуг, промышленный сектор, экономические колебания, колебания ВВП.

Annexure I

(Conditional Variance Graphs for Pakistan)

Сonditional Variance Graph for GARCH (1, 1) Model of the GR_Y

Conditional Variance Graph for GARCH (1, 1) Model of the GR_AGR

0

5

10

15

20

25

30

1975 1980 1985 1990 1995 2000 2005 2010

Conditional variance

Conditional Variance Graph for GARCH (1, 1) Model of the GR_IND

0

1

2

3

4

5

6

7

8

1975 1980 1985 1990 1995 2000 2005 2010


10

12

14

16

18

20

22

24

1975 1980 1985 1990 1995 2000 2005 2010



193

Conditional Variance Graph for GARCH (1, 1) Model of the GR_SER

0

2

4

6

8

10

12

14

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_EXP

0

40

80

120

160

200

240

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_IMP

0

100

200

300

400

500

1975 1980 1985 1990 1995 2000 2005 2010


Annexure II

(Conditional Variance Graphs for Korea)

Conditional Variance Graph for GARCH (1, 1) Model of the GR_Y

0

5

10

15

20

25

30

35

40

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_AGR


194

20

30

40

50

60

70

80

90

100

110

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_IND

0

10

20

30

40

50

60

70

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_SER

0

4

8

12

16

20

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_EXP

0

100

200

300

400

500

600

1975 1980 1985 1990 1995 2000 2005 2010


Conditional Variance Graph for GARCH (1, 1) Model of the GR_IMP

0

50

100

150

200

250

1975 1980 1985 1990 1995 2000 2005 2010


european of economic journal issn 2304-9669. e-issn 2305...

Documents