Relevance of Consumption Hypotheses in Pakistan: integrating the Micro and Macro Levels of Analysis.

Abstract

This thesis provides new inside in the scanty literature related to consumption hypotheses in Pakistan. The earlier studies were based on aggregate data and none of the study estimated the consumption functions by using the household data. Therefore one the early contribution of my dissertation is; it employed cross sectional data of Pakistani household sector. Another contribution of my dissertation is that it examined three main consumption functions Absoluteincome hypothesis, lifecycle hypothesis and permanent-income hypothesis, using latest available aggregate data set. Third significant contribution of this thesis is that it examines first time the adaptive-expectation model for permanent-income hypothesis in Pakistan. The fourth contribution of the thesis is to use first time the channel of saving, growth, consumption and wealth to explore the life-cycle hypothesis in Pakistan. The time frame(s) for analyses are different, depending on the household and aggregate data. At macro level, annual aggregate data series from 1973 to 2015 has been employed. The data series has been taken from Annual Reports of State Bank of Pakistan, and various issues of Economic Survey of Pakistan. For household level analysis this thesis is based on the cross-sectional data of Pakistan household-integrated-economic-survey (HIES) 2015-16 which had been especially designed for the household-integrated-income and consumption survey (HICS). In the first stage, empirical consumption functions based on the absolute-income hypothesis have been estimated. The results show that the income has significant and positive effect on consumption. The estimated marginal propensity to consume under this hypothesis was 0.78. More specifically, the results of the study strongly support the absolute-income hypothesis for Pakistan and also support the Keynesian assertions i.e. the MPC is less than one, APC is greater than MPC and the elasticity is also less than unity. The lifecycle hypothesis is examined under two different assertions, where one is related to consumption and wealth relationship and the other is based on the saving and income growth relationship, however the results of our studies do not support the lifecycle hypothesis of Modigliani (1989) & Modigliani and Shi (2004). Similarly, two different variants of the permanent-income hypothesis have also been examined at aggregate level. The first variant of permanent-income hypothesis in case of Pakistan is based on the Flavin (1981) modified model and the second variant of permanent-income is examined under the assumptions of adaptive-expectations model. The results based on both of these studies do not provide any evidence in favor of the permanent-income hypothesis at aggregate level. The household level analysis tests the efficacy of absolute-income, lifecycle hypothesis and permanent income hypothesis in case of Pakistan. Where the results of lifecycle hypothesis are against the lifecycle assertion and support the absolute-income hypothesis. The results of permanent income hypothesis rejected the strict version of the Friedman theory; and found strong support for the loose version of permanent-income hypothesis. These results are consistent with the findings of Vakil (1973) who reported the similar findings in case of India. The results of this dissertation show that in case of Pakistan the available data do not support any consumption theory except the absolute-income hypothesis, neither at aggregate level nor at household level, which is a common phenomenon of the developing economies. Few reasons can be offered in support of theses rejections, as it may be due to the primitive society of Pakistan which has low literacy rate and unawareness of primary macroeconomic variables. Second reason is the non-availability or lack of borrowing facilitates, as the financial market is relatively narrow in Pakistan. Third reason may be the religious and cultural aspects of our society where current consumption is not sacrificed over future consumption. In the same way the rejection of the permanent-income hypothesis indicates that the Pakistani people are shortsighted and less informed and they do not plan for their entire life span. The recommendation for policy are very straight forward under the absolute-income hypothesis as if the policy maker wants to increase the welfare and standard of living of the people through increasing the consumption they just need to increase the income of the citizens, without having the difficulty that which part of the income needs to be raised.