An Empirical Investigation of the Relationship Between Trade Liberalization and Tax Revenue in Pakistan

• Author: ATIF ALI JAFFRI, FARHANA TABASSUM and ROOMA ASJED

Abstract

This study examines the empirical relationship between trade liberalization and tax revenue in Pakistan for the period 1982-2013. Estimation results based on ARDL model show that there exists positive relationship between trade liberalization and total tax revenue in Pakistan over the study period. The coefficient of lagged error term (ECMt-1) in short-run model is negative and significant suggesting speed of con- vergence to equilibrium. The coefficient (-0.3119) implies that deviation from the long-term equilibrium is corrected by 31.19% over one year. Sound and stable trade policy along with favourable environment are needed that promotes import of raw material, capital and intermediate goods which enhances trade in the country leading to enhancement of tax collection in Pakistan.

Keywords: Trade liberalization, Tax revenue, ARDL, ECM

1. INTRODUCTION

   Trade liberalization has become an important development policy initiative in many developing countries since the 1980s. Under Structural Adjustment Programs (SAPs) suggested by the World Bank and the IMF countries have been shifting from destructive protective policies to free trade. Trade liberalization is defined as "the total or part elimination of trade barriers such as quotas, import duties, tariffs and non-tariff barriers imposed by governments on imported and exported goods" (Marchant and Snell, 1997).

   During 1980s Pakistan's economy moved towards trade liberalization, deregulation and privatization. In 1995, Pakistan became member of the World Trade Organization (WTO) as a result of the Uruguay Round (UR) of trade negotiations to extract gains from implementation of multilateral trade liberalization. Trade liberalization leads to growth, competition, efficiency, productivity and, hence, development in developing and developed countries (Newman, Rand and Tarp, 2013; Manni and Afzal, 2012; Topalova and Khandelwal, 2011; Geng, 2008; Utkulu and Ozdemir, 2004; Dornbusch, 1992).

   Recently, the subject of trade liberalization and tax revenue has become an important issue both theoretically and empirically. Theoretically, trade liberalization is expected to increase total tax revenue by increasing the share of trade tax revenue through imposing custom and excise duties on both imports and exports (Keen and Ligthart, 2002). In contrast, others argue that under trade liberalization reforms the reduction of trade restriction leads to reduction in trade tax receipts to federal government and thus less proportionate increase in total tax collection (Pritchett and Sethi, 1994).

   The impact of trade liberalization is an empirical question because when trade liberalization reduces import duties and other trade restrictions then there will be revenue loss but if volume of trade increases then tax revenue can increase (Tanzi, 1989; Glenday, 2002; Greenaway, Morgan and Wright, 2002; Suliman, 2005).

   In case of Pakistan, there are a number of factors that influence tax revenue such as exchange rate, openness, per capita income, urbanization, population, inflation, external debt, foreign aid, effective rate of trade taxation, political stability and broad money (Mahmood and Chaudhary, 2013; Chaudhry and Munir, 2010). Studies regarding determinants of tax revenue in Pakistan show that there is lack of such studies which look at the impact of external factors consistent with trade liberalization period.

   According to Pakistan Economic Survey (2013-14), tax to GDP ratio in Pakistan is below 10% even worse than its neighboring countries. The main issues related to tax revenue are the structural problems and low tax base. The historical trend of tax revenue from 1990 to 2012 has been shown in Figure 1.

2. LITERATURE REVIEW

   Basirat et al. (2014) examined the empirical relationship of economic determinants and aggregate tax revenue in Iran by using annual time series data. Findings show that exchange rate, imports, value added by agriculture and industry sector have significant effect on tax collection during 1974-2011 (Basirat et al., 2014).

   Velaj and Prendi (2014) provide the evidence on factors that determine taxes in Albania during 1993-2013. Findings show that inflation, GDP and imports increase tax revenue. Coefficient of GDP indicates that with 1% increase in GDP the taxes grow by 0.62% while unemployment has negative effect on tax revenue. Karagoz (2013) discussed the determinants of tax revenue in Turkey using the time series data for the period 1970-2010. Results show that variables that significantly affect tax revenue include agricultural and industrial sector share, monetization, foreign debt and urbanization. Agriculture share has negative effect while trade openness found to be as insignificant variable among all variables.

   Cage and Gadenne (2012) analyzed the fiscal cost of trade liberalization using a panel data set of 103 developing countries for the period 1945-2006. Trade liberalization leads to lower tax revenue. Revenue can be increased from trade openness by investing in tax capacity because countries which are trapped in high tax capacity have experienced positive effect of trade openness on tax revenue.

   Ghani (2011) focused on both conventional and non-conventional determinants of tax to GDP ratio by using the panel...