THE IMPACT OF DIFFERENT MACROECONOMIC VARIABLES ON POVERTY IN PAKISTAN *Dr. Hazoor Muhammad Sabir and **Safdar Hussain Tahir

Transcription

1 THE IMPACT OF DIFFERENT MACROECONOMIC VARIABLES ON POVERTY IN PAKISTAN *Dr. Hazoor Muhammad Sabir and **Safdar Hussain Tahir Abstract The objective of the study was to investigate the impact of different macroeconomic variables on the welfare of the poor in Pakistan, through annual time series data from In this study, through annual time series data from , the multiple regression technique was applied to detect the relation between macroeconomic variables and poverty. Inflation, GDP growth, population growth, major crops, minor crops, livestock and per capita income were taken as independent variables while poverty (HCI) as dependent variable. The study results revealed that GDP growth rate per capita income, major crops, minor crops and livestock had negative impact while inflation and population growth rate had positive impact upon poverty. The conclusions drawn from the study are that in the long run, the reduction in poverty in Pakistan is to be driven by the changes in the macroeconomic variables. Key words: Macroeconomic variables, Poverty, Pakistan 1. Introduction: Poverty is a multidimensional phenomenon. Poverty is pronounced deprivation in well being, and comprises many dimensions. It includes low income and the inability to acquire the basic goods and services necessary for the survival. Poverty also encompasses low level of health and education, poor a access to clean water and sanitation, inadequate physical security, lack of voice and insufficient capacity and opportunity to better one s life. The definition of poverty may vary from country to country. There are two kinds of poverty absolute or relative. In Pakistan, both absolute and relative poverty exist. Absolute poverty refers to the lack of basic needs, education health, clothing shelter etc. Relative poverty refers to the lack of socially acceptable level of income or other resources as compared to other countries or societies. Absolute poverty can be alleviated but relative poverty is a vital concept and exists in all parts of the world. It involves comparison between groups. The poverty largely measures in monetary terms. The causes of poverty are also multidimensional. It may be physical, psychological, economic and socio cultural. Physical factors for poverty are lake of basic physical and economic infrastructure and unfavorable natural environment and may also relate to poor health and malnutrition. Psychological factors refer to stress, depression, loss of self esteem, loss of ambition and aspiration. Usually poverty is measured by the poverty line. Poverty line is defined as the minimum level of income deemed necessary to achieve an adequate standard of living in a given country. To determine the poverty line there are two methods per capita income and consumption expenditures and per capita calories requirements. *Associate Professor, Department of Economics, GC University Faisalabad. (Corresponding Author) **Assistant Professor, Department of Banking & Finance, GC University Faisalabad. 788

2 Per capita income and consumption expenditure refers to the total income that an adult human consume in one year at all the essential resources. Per capita calories requirements refers to the 2350 calories per adult per day. International calories requirement is 2550 calories per adult per day. The reason of adopting low calories requirement is to reduce the percentage of the population below the poverty line. (Economic Survey of Pakistan ) Pakistan is a low income developing country. From the last three decades Pakistan has experienced periods of high economic growth accompanied with increase in poverty, period of low economic growth and reduction in poverty and also the period in which economic growth has a positive impact on poverty. The differential of economic growth on poverty is based on basic structural characteristics of the economy and economic policy administration which were assertive during these different time periods. Economic policies play an important role in changing the economic structure and also the economic growth process. The main hindrance in achievement of high sustained rate of growth of per capita income and reduction in poverty is slow progress in education especially of women high infant and child mortality rate that resulted in very high rate of population. Inflation refers to the increase in general price level of goods and services in an economy over time. Inflation is the main indicator of the economy. It provides important insight on the economic state of a country and sound macroeconomic policies to control it. Stable prices not only provide a developed environment for the economic growth but also elevate the poor who are the most accessible in society. In Pakistan inflation has become a serious problem and a main hurdle in the way of progress. Inflation is a key variable that has a significant impact on poverty. In economics growth means the increasing capacity of the economy to satisfy the needs and wants of the society. Economic growth can be achieved by increasing the productivity of the economy. GDP growth means that economy is growing and developing, technically it means the increase or decrease in the GDP compared with previous years. It is the most important indicator of the economy. If it is growing then the growth rate is positive and if it is decreasing then the growth rate is negative and the economy is in recession. The component of the GDP growth is retail expenditures, government spending and imports and exports. Population growth means the increase in the proportion of the individuals in a country using per unit time for measurement. In demographics population growth refers to the increase in the number of individuals within a country in a given time period. It is determined by the four factors birth (B), death (D), immigrant (I) emigrant (E). If the growth rate is increasing then it is called positive growth rate and if it is decreasing then it is called negative growth rate. Population growth and the poverty are correlated. Agriculture growth is also an important component that affecting the poverty trends in a country. Agriculture sector include the production of major crops, miner crops and livestock. The major crops of Pakistan are divided in to food crops and non food crops. The food crops include wheat, maize, grains, grams and other pulses and the cash crops are cotton, sugarcane, tobacco and mustard. Historically it is proved that agriculture 789

3 growth and production played a very important role in the process of development and growth and helps to reduce poverty. Per capita income is the mean income of per person within an economic aggregate such as a city or a country. It does not mean the distribution of income or wealth of a country. It is also the measure of wealth of the population as compared to other nations. It is calculated by taking the aggregate income and then dividing it by the total population. Per capita income also has impact on poverty. 1.2 Objectives of the study: Following are the main objectives of the study: a) To see the poverty situation in Pakistan over time b) To see the impact of different macroeconomic variables growth on poverty c) To work out on the growth rate of different macroeconomic variables d) Suggestion to improve the situation 2. Literature Review: Richad (2002) in his study Macroeconomic adjustment the poor: analytical issues and cross country evidence showed with the cross country regression analysis that there is a link between poverty and macroeconomic adjustment by using some macroeconomic and structural variables. The results shows that negative growth rates, illiteracy, income inequality tends to increase the poverty while the reduction in output growth and real exchange rate tends to reduce the poverty levels. Jamal (2006) analyzed at the macro level the relation between growth, poverty and inequality for 1979 to 2002 for Pakistan. He concluded that there was a positive relation between GDP per capita and the income inequality and found that inflation, sectoral wage gap and the terms of trade worsen the inequality. He also explained that low level of income inequality helped in poverty alleviation and suggested adapted policies to control and reduce the poverty and inequality. Adeyemi, Ijaiya and Raheem (2009) in their study Determinants of poverty in Sub Sahara Africa analyzed the determinants of poverty in sub Sahara Africa by using cross country data of 48 countries and used multiple regression technique. The results showed that the factor like increase in population, inflation, external debt, lack of safe water, gender discrimination and ethic and religious clash causes increase in the level of poverty in the sub-region. The results suggested that the measures like debt forgiveness, use of family planning and stable macroeconomic variables like inflation and exchange rate are the possible solution for poverty alleviation Chaudhary, Malik and Hassan (2009) in there study The impact of socio economic and demographic variables on poverty: A village study analyzed the impact of socio economic and demographic variables on poverty. They used two approaches poverty profile and an econometric approach for empirical analysis by using primary data. The results show that house hold size, landholdings, dependency on household and number of livestock has a significant impact on the incidence of poverty. They concluded that there should be the need to improve the socio economic and demographic factors and land should be allotted to landless households. Chani (2011) in his study Poverty, inflation and economic growth: empirical evidence from Pakistan investigated the relationship between the economic growth, inflation and poverty. With the help of ARDL testing approach there exist long run 790

4 relationship between the variables poverty, economic growth, inflation, investment and trade openness over the period of The result showed that inflation had positive impact on poverty while the economic growth and investment negatively affect the poverty. He analyzed that trade openness insignificantly effect the poverty. The short run analyses showed that economic growth had negative and inflation had positive impact on poverty while the effect of investment and trade openness on poverty is not significant in short run. Egbe and Clement (2011) in their study The Impact of Macroeconomic Policies and Programs on Poverty Problems discussed the impact of some macroeconomic policies on poverty in Nigeria during They analyzed the determinants of poverty in the country instead of the measures taken by the government to conquer the incidence of poverty. For the analysis of data they used two regression equation model based on poverty and GDP and concluded that in Nigeria the policies and programs that based on macroeconomic variables have not transmit the upward trends of poverty in the country. Zaman et al. (2011) used poverty growth model for five selected SAARC countries (Bangladesh, Nepal, Pakistan, India and Sri-Lanka) for the year He used head count ratio as proxy for poverty as dependent variable and GDP as economic growth, GINI co-efficient as income inequality, health and educational expenditures and foreign direct investment and (export + imports) as proxy for trade openness as independent variables. All variables were measured in log form. For empirical analyses pooled OLS and least squares dummy variable (LSDV) methods were used. The results showed that 1 percent increase in income inequality reduce poverty by.67 percent and 1 percent increase in economic growth reduce poverty by 0.1 percent while trade openness and health expenditures were insignificant in poverty reduction in SAARC countries. 3.0 Methodology: This paper explains the various tools and techniques for the econometric analyses related to the research problem to be investigated. It concentrates on to analyze the impact of inflation rate, GDP growth rate, population growth rate, major crops, minor crops, livestock and per capita income on poverty in Pakistan during the period 1981 to To find out the impact of these macroeconomic variables on poverty, multiple regression technique was applied. To check the problem of autocorrelation-durbin Watson d and Breusch-Godfary tests, for Hetroscadasticity-White s test, for structural breaks Chow test and Jarque-Bera tests for the normality in residuals were used. In order to check intra variables correlation matrix was also applied. In this study time series data from the year 1981 to 2010 were used on annual basis. The main source of these data was statistical hand book and economic surveys of Pakistan. 3.1 Variables of the model In this study in total eight variables were included in the model.hci was taken as measures of poverty as dependent variable where as inflation rate, GDP growth rate, population growth rate, major crops, minor crops, livestock and per capita income as explanatory variables in the model. All the variables were taken in percentage form (growth rate). 791

5 The functional form of the model: Y = β 0 + β 1 INF+β 2 GDP+β 3 PGR+β 4 MAJC+β 5 MINC+β 6 LS+β 7 PCI+ µi Where Y = Poverty INF = Inflation rate GDP = GDP growth rate POP = Population growth rate MAJC = Major crops growth rate MINC = Minor crops growth rate LS = Livestock growth rate PCI = Per capita income growth rate µi = Error term 3.2 Hypothesis making: Following hypothesis are tested to check the impact of different variables on poverty a. Inflation has a significant impact on poverty b. There is a significant impact of GDP growth on the poverty c. Population growth has significant impact to reduce poverty d. Major crops causes significant effect on poverty reduction e. Miner crops helps in poverty reduction f. Livestock significantly affect poverty g. Per capita income causes change in the incidence of poverty 3.3 Test of Autocorrelation: Autocorrelation means correlation among the member of the series of observation ordered in time or in space. To check the problem of autocorrelation in the error term Durbin-Watson d and Breusch-Godfary tests are used. 3.4 Durbin-Watson d test: This test is used to detect the autocorrelation between the residuals or error term in regression analysis which is defined as: d = = 2 t n ( u t = t u 2 t 1 ) t = n 2 u t = 1 t This is the ratio of the sum of squared differences in successive residuals to the RSS. 3.5 Breusch-Godfrey test: This test is used to check the autocorrelation in error term in a regression model that is being considered for regression analysis. H 0 =there is no autocorrelation in the error term H 1 =there exists autocorrelation in the error term 792

6 3.7 White Test: the White test is a statistical test that establishes whether the residual variance of a variable in a regression model is constant: that is for Homoscadasticity. Where LM is Lagrange Multiplier and n is sample size 3.8 Chow Breakpoint test In econometrics, the Chow test is most commonly used in time series analysis to test for the presence of a structural break. In program evaluation, the Chow test is often used to determine whether the independent variables have different impacts on different subgroups of the population or not. The hypothesis about the test is as H 0 =there is no structural break H 1 = there exists structural breaks 3.9 Jarque- Bera(JB) Test JB test is used to check that whether the data is drawn from a normal distribution or not or whether the data have skewness and kurtosis equal to normal distribution. Skewness and kurtosis of the OLS residuals are computed through following expression. 2 2 S ( K 3) JB = n Where n=sample size, S = Skewness coefficient, and K=kurtosis coefficient 3.10 Correlation Matrix: Correlation is a statistical technique that can show whether and how strongly pairs of variables are related. The main result of a correlation is called the correlation coefficient (or "r"). It ranges from -1.0 to The closer r is to +1 or -1, the more closely the two variables are related. 4.0 Result and discussion: 4.1 Breusch-Godfery serial correlation LM test: Table 4.1 Breusch-Godfrey Serial Correlation LM Test: F-statistic Probability Obs*R-squared Probability The results shown in the table 4.1 illustrate that by applying the Berusch Godfery test on the model the value of Obs* R-squared which is the value of Chi-square is greater then α at 1%, 5% and 10% level of significance so the null hypothesis is accepted that there is no auto correlation in the error term. 4.2 White Hetroscadasticity test: Table 4.2 White Heteroscedasticity Test: F-statistic Probability Obs*R-squared Probability

7 By applying the white Hetroscadasticity test on the regression model the results have been shown in the table 4.2 that the value of the probability is greater then α, at 1%, 5% and 10% level of significance, so that the null hypothesis is accepted as the error term is homoscedastic and there is no problem of Hetroscadasticity in the model. 4.3 Chow test: Table 4.3 Chow break point test: Year F-statistic Probability In the Table 4.3, data results show that by applying the chow test on the variables from the years the value of the probability of F-statistics is greater than α, at 1%, 5% and 10%, so the null hypothesis is accepted and concluded that there is no structural break occurs in these years. In the year the probability of F-statistic is less than at α 1% and 5% level of significance so the null hypothesis is rejected at 1% and 5% and concluded that there exist structural break during these years. The value of probability of F-statistics of the year 1996 is less than α, at 1% level of significance so the null hypothesis is rejected at 1% level of significance. When we apply the test on the variables from the year the derived values of probability of F-statistics is less than α, at 1%, 5% and 10% level of significance so we accept the alternative hypothesis that there is structural break occurs in these years. In 1995 WTO regime was introduced. It had a great impact on Pakistan economy. During this time period the political situation of our country also remained uncertain. 1n 1996 Benazir government was dismissed and Nawaz Sharif comes into power in Again in 1999 Pervez Musharaf coming into power and our government changed from democratic government to Military government. In the year /11 incidents occurred and a war on terror begins. So due to the above mentioned factors a structural break occurs in the variables from the year

8 4.4 Jarque-Bera test: Table 4.4 Jarque-Bera test of normality: Series: Residuals Sample Observations 29 Mean 5.82E-16 Median Maximum Minimum Std. Dev Skewness Kurtosis Jarque-Bera Probability The results of J.B statistics are shown in the table 4.4. The value of the Jarque- Bera probability is lesser than the α at 1%, 5% and 10% level of significance, so the null hypothesis is accepted that the residuals are normally distributed. 4.5 Correlation Table 4.5 Correlation matrix Correlation INFLATION GDP PGR MAJOR MINOR LS PCI AGR INFLATION 1 GDP PGR MAJOR MINER LS PCI AGR Table 4.5 shows the results of the degree of the relationship among the variables and to check the problem of Multicollinearity. Population growth rate, major crops and per capita income are negatively correlated with inflation while GDP growth rate, minor crops and livestock are positively correlated. Per capita income negatively correlated with all the variables. The results show that there is a high correlation between major crops and agriculture growth rate and there exist problem of Multicollinearity in the model. So to remove the problem of Multicollinearity the study exclude the variable of agriculture growth from the model and take only its sub sectors such as major crops, miner crops and livestock. 795

9 4.6 Application of OLS: Y = β 0 + β 1 INF+β 2 GDP+β 3 PGR+β 4 MAJC+β 5 MINC+β 6 LS+β 7 PCI+ µi Table 4.6 Parameter Esimates Through OLS Variable Coefficient Std. Error t-statistic Prob. INFLATION GDP PGR MAJC(-1) MINC(-1) LIFESTOCK PERCAPITA C R-squared Mean dependent var Adjusted R-squared S.D. dependent var Durbin-Watson stat Prob(F-statistic) The table 4.6 shows the regression results. The variable Y is dependent variable which is poverty and the macroeconomic variables such as inflation, GDP growth, population growth, major crops, miner crops, livestock and per capita income were all explanatory variables. The coefficient of determination R 2 shows that 68% variation caused by the explanatory variables on the explained variable. By applying the Durbin Watson test, (d) value was 2.01 which implies that there was no problem of auto correlation in the error term. The co efficient of inflation was positive implying that there was a direct relationship between poverty and inflation. Statistically it was highly significant.so, we can reject the null hypothesis and accept the alternative that inflation had significant impact on poverty. The reason behind was that as the general level of prices of agricommodities rise, the purchasing power of the people decline and hence low consumption level leading to higher levels of poverty. The estimation results stay against the findings of Agenor (1998) who concluded that inflation has negative impact on poverty. The co efficient of GDP growth had highly significant negative impact on poverty. So, we the null hypothesis was rejected and concluded that there was a significant impact of GDP growth on poverty. However the GDP growth rate is negatively related with poverty and verifies our theoretical findings that when GDP growth rate is increasing poverty is reduced. Our results also match with findings of Romar and Gugerty (1997) who believed that an increase in GDP growth directly increases the incomes of the poor. The co efficient of population growth was highly significant on poverty and had expected theoretical sign implies a positive relation with poverty. Our results also match with the findings of Chaudhary (2005), who concluded that there was a need in the improvements in socio and demographic variables to reduce poverty in the southern areas of Pakistan. 796

10 The coefficients of major, minor crops and livestock had negative relationship with the incidence of poverty. Econometric analysis given in table-4.6 revealed that three variables had significant impact on poverty. So, we cannot reject the alternative hypotheses and concluded that the reason behind was that the share of major crops, minor crops and livestock in GDP was about 25% of the total. The co efficient of per capita income postulated an inverse relation with poverty and significant impacts the incidence of poverty. So we can reject the null hypothesis and conclude that per capita income causes change in the incidence of poverty. Our estimated results also match with findings of Chaudhary, Malik and Ashraf (2006) who explained the macroeconomic determinants of rural poverty in Pakistan. Their results also concluded that per capita income has negative relation with poverty but it has insignificant impact on poverty which is not match with our results. The reason behind is that they took nominal per capita income which has no impact on the incidence of poverty while we take real per capita income which defiantly has significant impact on poverty. 5.0 Conclusions Based on the empirical analysis obtained through the econometric analyses of the time series data, it was concluded that the macro economic variables like GDP growth, per capita income, major and minor crops had the inverse relationship with poverty, the reason behind was that said macro variables contributed towards the enhancement of per capita income of the people leading to poverty reduction. On the other hand the inflation and population growth both were positive correlated with poverty. The increase in population number will reduce the available per capita as denominator and inflation due persistent rising level of prices will reduce the purchasing power of the people leading to rise in poverty level. 5.1 Suggestions and policy implications Following suggestions are given to improve the situation 1. GDP growth rate should be improved on consistent basis. Government has opted growth oriented policies to promote sustained economic growth with the help of fiscal, monetary and trade incentives. 2. The main characteristic of the poor is their large size of family. The people are convinced that increasing number of children is responsible for their poverty. Rather increasing the children they should increase the power to work: they should improve their skill efficiency and knowledge. All this will have the effect of increasing their incomes which may result in the reduction of poverty. Accordingly, government must stimulate efforts at curtailing the growth rate of population. 3. In the agriculture sector the production of major crops miner crops and livestock should be encouraged. If the role of middle man and land tenure system eliminated then the farmers will get better prices of their crops which increase their income and enable them to get ride from the vicious circle of poverty. 4. Real per capita income should be increased and would not be accompanied with worsening the distribution of income. Real per capita income should be increased rather than the nominal income. If the increase in real per capita income is greater 797

11 than the increase in inflation, this will lead to increase the standard of living of the people and poverty will come down. 5. Persistent rising level of prices hurt the consumers; lessen the purchasing power leading to poverty. In this regard, the strict price monitoring system should be established in economy. So in the light of the findings of the study government should bring stability in macroeconomic variables and implement such growth oriented and stabilization policies especially at macro level which will helpful for poverty alleviation 798

12 References Adeyemi, S.L., Gafar T Ijayia and Usman A Raheem (2009), Determinants of Poverty in Sub Sahara Africa, International Multi Disciplinary Journal Ethopia, Vol. 3, No. 2 Agenor, P.R (1998), Poverty Income distribution and Labor market Issues in Sub Sahara Africa, Collaborative Research Project, CR-2-3 Chani, M.I., Zahid Pervaiz, Sajjad Ahmed Jan, Amjad Ali and Amanat R Chaudhary (2011), Poverty Inflation and Economic Growth: Evidence from Pakistan, World Applied Science Journal, 14(7), Chaudhary, I.M., Shahnawaz Malik and Muhammad Ashraf (Winter 2006), Rural Poveerty in Pakistan some related concepts and issues an Empirical Analysis, Pakistan Economic and Social Review, 44(2): Chaudhry, I. S., Malik and Hassan (2009), The Impact of Socioeconomic and Demographic Variables on Poverty: A Village Study, The Lahore Journal of Economics, 14(1), pp Egbe, G and Awogbemi Clement (2011), The Impact of Macroeconomic Policies And Programs on Poverty Problems, Journal of Economics and Sustainable Development, Vol 2, No. 9 Govt of Pakistan (2002), Pakistan Economic Survey , Ministry of Finance Govt of Pakistan Govt of Pakistan (2008), Pakistan Economic Survey , Ministry of Finance Govt of Pakistan Gujarati, D. N (2003), Basic Econometrics, McGraw Hill Education, fourth edition, pp Jamal, Haroon (2006), Does Inequality Matter for Poverty Reduction? Evidence from Pakistan Poverty Trends, The Pakistan development Review, Vol 45, No.3 (Autumn), pp Richard, A.P (2002), Macro Economic Adjustment and the poor: Analytical issues and Cross country evidence, Policy research working paper, Report No. WPS2788 Vol. 1 Romar, M and Mary Kay Gugerty (March 1997), Does Economic Growth Reduce Poverty?, Harvard Institute for International Development Zeman, K., Rashid, K, Khan, M.M and Ahmad, M (2011), Panel Analysis of Growth, Inequality and Poverty: Evidence from SAARC countries, Journal of Yasar University, Vol. 21, No. 6, pp