86255 Pakistan: Reinvigorating the Trade Agenda José Guilherme Reis and Daria Taglioni Policy Paper Series on Pakistan PK 15/12 March 2013 _______________________________________________ This publication is a product of the South Asia Poverty Reduction and Economic Management Unit. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions in Pakistan and around the world. Policy Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at daria.taglioni@graduateinstitute.ch and jreis@worldbank.org. Abstract This paper reviews Pakistan’s recent trade performance, its trade policy and trade costs. Different dimensions of trade performance—growth and orientation, diversification and sophistication—are assessed, complemented by an in-depth analysis of export dynamics in the period 2001-10 using firm-level data. An econometric exercise is also performed to identify the impact of tariffs, exchange rates, fixed costs to export, foreign demand, and preferential trade policy on the ability of firms to increase their exports. The analysis of Pakistan’s trade policy includes tariffs, effective protection and trade restrictiveness estimates, as well as an assessment of the role of preferential trade agreements in the context of regional integration. Finally, the main characteristics of trade facilitation and logistics are analyzed, covering the capacity, performance, quality of services and degree of integration of the logistics system. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development / World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Pakistan: Reinvigorating the Trade Agenda José Guilherme Reis and Daria Taglioni This paper was prepared by José Guilherme Reis and Daria Taglioni, with contributions from Nihal Pitigala and Maros Ivanic (Trade Policy), Charles Kunaka (Trade Facilitation), Megha Mukim (Export Performance) and Florencia Piacci (Research Assistance). The authors would like to thank valuable comments and suggestions from José R. Lopez Calix, Eric Manes, Thomas Farole and Daniel Reyes. The team also consulted with authorities from the Ministry of Commerce, PITAD and Central Bank. Pakistan: Reinvigorating the Trade Agenda Table of Contents Executive Summary 4 Background 6 Overview of Trade Performance, Trends and Outcomes 8 Trade Policy Objectives, Tools for Implementation and Effectiveness 24 Policy Recommendations 54 References 60 Annexure 63 Figures Figure 1 Openness to Trade 6 Figure 2 Export Growth: 1980-2010 8 Figure 3 Pakistan’s Exports, Predicted vs. Actual Flows (2007-09) 11 Figure 4 Herfindahl Index of Export Product and Market Concentration 11 Figure 5 Intensive and Extensive Margin in Products and Markets 12 Figure 6 Technological Content of Exports 13 Figure 7 Export Sophistication (EXPY), 2002-10 13 Figure 8 Export Sophistication 14 Figure 9 Services Trade Restriction Index (STRI) 15 Figure 10 Pakistan: STRI by Sector 15 Figure 11 Export of Service, Pakistan and Peer Countries 16 Figure 12 Service Export Sophistication (EXPY) 16 Figure 13 Total FDI Inflows (US$ Million) 18 Figure 14 FDI Inflows by Sector (US$ Million) 18 Figure 15 Concentration of Exports Across Exporters 18 Figure 16 Entry and Exit Rates in Pakistan and in Peer Countries 19 Figure 17 Annual Export Growth Generated by Firm Churning (Firm Extensive Margin) 19 Figure 18 Annual Export Growth Generated by Firm Entry and Exit from Specific 19 Export Markets (Country Extensive Margin) Figure 19 Annual Export Growth Generated by Firm Entry and Exit into/from New 20 Sectors and New Products (Sector and Product Extensive Margin) Figure 20 The Intensive Margin of Exports, Annual Export Growth 21 Figure 21 Survival Rates After 1 Year of Export Activity 21 Figure 22 The Margins of Export, Annual Export Growth 22 Figure 23 Estimated OTRI (Except for Pakistan, Baseline Year 2007) 26 Figure 24 Pakistan’s Tariff Reform, MFN Unweighted Average (1982-09) 28 Figure 25 Distribution of Tariff Rates, by Frequency, including Regulatory Duties 29 (2009-10) Figure 26 Distribution of effective tariff rates by frequency, including regulatory 29 duties (2009-2010) Figure 27 Decomposition of statutory and applied tariff rate variation 35 Figure 28 Effective Rates of Protection 35 1 Pakistan: Reinvigorating the Trade Agenda Figure 29 Pakistan’s exports enjoying preferential, non-preferential or MFN-zero 38 tariff in largest twenty world export markets (percentage shares) Figure 30 Change in Pakistan’s total imports (% 2009 total imports) 41 Figure 31 Tariff revenue losses (US$ million) 42 Figure 32 Welfare gains (US$ million) 42 Figure 33 Pakistan’s predicted vs. actual exports 45 Figure 34 India’s predicted vs. actual exports 45 Figure 35 Pakistan Logistic Performance 48 Figure 36 Logistics Performance 49 Figure 37 Transport Prices in Pakistan and Other Regions 50 Figure 38 World Air Services Network Structure: 2011 52 Tables Table 1 Composition and Growth of Exports by Sector 9 Table 2 Export Share in Selected Destination Markets 9 Table 3 Pakistan: Commercial Services Exports 16 Table 4 FDI Inflows (% of Gross Fixed Capital Formation) 17 Table 5 Survival Rates by Cohort 21 Table 6 Gini coefficients and average statutory and applied tariffs by sector 30 Table 7 Pakistan’s official tariff schedule by stages of processing (2010) 31 Table 8 The Statutory and Effective Duty Regime based on Customs Transactions 34 (2009-2010) Table 9 Sector-wise distribution of key SROs based on Customs Transactions data 34 (2009-2010) Table 10 Preferential Trade Regime for Pakistan’s Exports to Larges 20 Export 38 Markets Table 11 List of Pakistan’s multilateral and preferential trade agreements 40 Table 12 Pakistan: Changes in bilateral imports from main partners as a result of 41 hypothetical full tariff liberalization in current PTAs Table 13 Changes in tariff revenue and welfare as a result of proposed PTAs 43 Table 14 Provisions in Pakistan’s PTAs with SAFTA partners and China 43 Table 15 Pakistan Bilateral Investment Agreements 44 Table 16 Major Non-Tariff Impediments to Pakistan-India Trade 47 2 Pakistan: Reinvigorating the Trade Agenda Boxes Box 1 Sophistication of Exports in Pakistan 14 Box 2 Overall Trade Restrictiveness Index: OTRI 25 Box 3 Recent import Tariff and Tax Reforms 30 Box 4 Exception to MFN Regime: SROs 565(I)/2006, SRO 567 (I)/2006 and SRO 33 575(I)/2006 Box 5 Protectionism in the Automotive Sector 37 3 Pakistan: Reinvigorating the Trade Agenda Executive Summary 1. Pakistan has many challenges ahead in order to take full advantage of integration into to the global economy. The need for urgent reform in openness and trade policy is underscored by Pakistan’s new Framework for Economic Growth (FEG) issued in mid -2011. Indeed, in the last fifteen years, Pakistan’s trade policy has been erratic; closely following a cycle defined by major departure from and return to protectionist policies. This has been compounded by significant deficiencies in the overall business environment that regulates the economic governance of the trade regime and in the provision of infrastructure services that negatively affect the competitiveness of firms in Pakistan. 2. This paper reviews Pakistan’s recent trade performance, its trade policy and trade costs. Different dimensions of trade performance—growth and orientation, diversification and sophistication—are assessed, complemented by an in-depth analysis of export dynamics in the 2001-10 period using firm-level data. An econometric exercise is also performed to identify the impact of tariffs, exchange rates, fixed costs to export, foreign demand, and preferential trade policy on the ability of firms to increase their exports. The analysis of Pakistan’s trade policy includes tariffs, effective protection and trade restrictiveness estimates, as well as an assessment of the role of preferential trade agreements in the context of regional integration. Finally, the main characteristics of trade facilitation and logistics are analyzed, covering the capacity, performance, quality of services and degree of integration of the logistics system. 3. Main findings can be summarized as follows: i) Pakistan’s recent trade performance is stagnating, as indicated by a decrease in its trade-to-GDP ratio over the last decade and low levels of sophistication of exports; ii) its export bundle is relatively diversified in terms of products, but the geographic concentration of export markets makes the country vulnerable to adverse shocks in developed countries; iii) there are positive signs of export dynamism, with substantial churning of firms. Yet, dynamism has been decreasing, as shown by declining trends in terms of entry and exit of firms; iv) high trade costs and the anti-export bias reflected in the complexity of the tariff regime are both important obstacles for export growth; and v) the complexity of the trade regime is reflected in high effective rates of protection and high and rising overall trade restrictiveness index. As a result, Pakistan remains as one of the most protected economies of the world. 4. Results for preferential trade agreements are mixed, with (vi) shallow Preferential Trade Agreements (PTAs) having impact only at deepening current exports to existing markets; vii) the impact of PTAs on trade creation and trade diversion has been limited, but there is significant untapped potential for expanding its trade relations; viii) India and Pakistan would benefit greatly by a normalization of their trade relations, and the fact that the two neighbors have recently taken concrete steps towards closer economic relations— with a vision to enhance peace and stability in the region—is a very positive change. Our estimates suggest that Pakistani exports to India are 40% below its potential and recent analyses incorporating dynamic supply chain developments find even higher impacts on trade growth; and ix) Pakistan’s overall logistics performance lags below that of India and Bangladesh as well as below the global average of all countries. 4 Pakistan: Reinvigorating the Trade Agenda 5. The country’s trade competitiveness would benefit from reforms that aim at bringing new firms and new products to existing (and new) foreign markets. To achieve this objective, creating a level playing field for firms by promoting intra-industry competition and improving the business environment should be policy priorities. Reducing the anti-export bias associated with the current levels of tariff protection is critical in this regard. In addition, as shown in the empirical analysis conducted, accelerating the implementation of deep preferential trade agreements and reducing trade costs can also play a decisive role for the country’s integration to the world economy. A number of policy recommendations are presented, in particular: 1. Rationalize and liberalize the tariff regime by moving initially to a three-band structure (25%, 10%, and 0%) with the aim of moving toward uniformity within 3 to 5 years—this should be accompanied by elimination of all exemptions, concessions (other than free zones and free trade agreements) and remaining regulatory duties and excise duties (other than strictly for health and safety reasons). In addition to creating a more transparent regime, the elimination of concessions will go a long way toward offsetting any fiscal losses from tariff reductions; 2. Accelerating the implementation of deep preferential trade agreements and signing of other deep agreements. Significant trade creation effects are likely to originate from agreements with China and Malaysia, due to the deeper nature of these agreements and their coverage beyond market access issues. 3. Opening up border with India and facilitating deep forms of trade integration to benefit from high growth rates of neighbors. With the introduction of MFN status with India and the recently signed bilateral agreements, significant efforts will be required to implement measures to reduce the cost of trading across borders by improving infrastructure, policies and procedures. This includes, foremost, removing impediments at the border and along trade routes, such as a transit protocol to link each to neighboring countries, associated infrastructure, a one-stop border post at the Wagah-Attari border, inland container depots, and online payment schemes for traders. Furthermore, given the challenges of integrating border communities, more localized initiatives should be considered, such as border bazaars and other measures to facilitate cross-border trade. 4. Upgrading the quality and reliability of logistics services, through acceleration of reforms such as a unified customs system across the country; to improve ports performance, through complementary measures (including capacity building and modernization); to solidify legal framework for the trucking industry, and through the development of land transport links to Pakistan’s neighbors—a crucial step to develop the country as a regional hub for logistics and to address domestic connectivity challenges. 5 Pakistan: Reinvigorating the Trade Agenda Background 6. Pakistan’s recent trade performance is one of stagnation, as indicated by a very small increase in its trade-to-GDP ratio over the last decade. At the same time peer countries have leapfrogged with high growth rates.1 Pakistan’s position below the predicted line indicates that it ‘under-trades’ compared to smaller countries at comparable levels of per-capita income.2 This is the case for all other populous countries as shown in Figure 1. What is striking, however, is that Pakistan’s average trade-to-GDP ratio for the period 2008-10 was only 2.6 percentage points higher than what it was a decade earlier (1998-00), in contrast to the shares of its peers. In 1998-00, Pakistan’s trade-to-GDP ratio was barely lower than China’s, and much higher than India’s. Ten years on, shares of both China and India almost doubled while Pakistan’s remained stagnant and increased only from 31.5 to 34.1 percent. Furthermore, both countries show highest rates of growth. Figure 1 Openness to Trade Average 1998-2000 Average 2008-2010 250 250 MYS 200 200 Trade to GDP (%), 2008-2010 MYS VNM 150 150 THA THA 100 100 VNM EGY CHN 50 50 BGD IND TUR CHN EGY TUR BGD PAK IRN PAK IND USA BRA USA BRA 0 0 6 7 8 9 10 11 6 7 8 9 10 11 Log of GDP per capita (PPP, av. 1998-2000) Log of GDP per capita (PPP, av. 2008-2010) Source: Authors’ calculations based on COMTRADE database 7. Despite the recent stagnation of Pakistan’s trade performance and the anemic world demand due to international economic crisis, many opportunities exist for Pakistan to take advantage of the global economy. World trade has undergone a dramatic transformation over the recent decade, and is now increasingly characterized by global fragmentation of production. This has resulted in greater trade flows and an increased variety of traded goods and services (World Bank, 2010). On one hand, the evolving regional trade landscape, dominated by emerging economies such as China and India is likely to exert more competitive pressure on firms in other developing economies, such as Pakistan. On the other hand, the competitive pressure will constitute a stimulus to improve productivity. Moreover, these countries also represent a new engine of growth, and are geographically proximate to 1 The trade-to-GDP ratio is one of the most basic indicators of openness to foreign trade and economic integration. It weighs the combined importance of exports and imports of goods and services in an economy. The ratio gives an indication of the dependence of domestic producers on foreign demand and of domestic consumers and producers on foreign supply. There is a concave relationship between trade openness and per capita income: countries tend to trade more as incomes rise, but at a decreasing rate. 2 An OLS regression (using cross-country data from WDI) of trade/GDP on the log of per capita income (PPP current $), its squared value, cost of exporting (using a sub-indicator from Doing Business 2008-10) and population, gives a predicted trade share of 65 percent. The actual trade share of less than 32 percent is therefore extremely low. 6 Pakistan: Reinvigorating the Trade Agenda Pakistan, offering greater market opportunities than the debt-ridden and geographically remote industrialized countries, which currently dominate the trade relations of Pakistan. 8. Yet, for Pakistan there are many challenges ahead. In the last seven years, Pakistan’s trade policy has been defined by major reversal from the previous trade liberalization and regulatory simplification program. With a limited, but effective return to more policies and indigenization programs, Pakistan’s trade policy has become increasingly complex and driven by vested interests. In addition, there are significant deficiencies in the overall business environment that affect its economic governance and shortcoming in the provision of infrastructure services that negatively affect the competitiveness of firms in Pakistan. 9. There is a broad recognition that a new strategy is urgently required in order for Pakistan to revert course, reinvigorate the trade liberalization agenda and support faster and sustainable growth. Pakistan’s Strategic Trade Policy Framework 2009-12 provides a coherent agenda to achieve these objectives by developing a platform that would move Pakistan away from low value-added and resource-driven activities toward higher value- added activities. The recent FEG (Planning Commission 2011) also recognizes the critical role played by trade for self-sustained growth, rightly pointing out that no success story in growth accelerations has featured exclusive inward-looking strategies. 7 Pakistan: Reinvigorating the Trade Agenda Overview of Trade Performance, Trends and Outcomes Openness, Export Levels, Growth, and Market Shares 10. Pakistan has grown in the last 30 years above world’s exports growth but much below the accumulated growth of exports of its South Asian peers. Specifically, Pakistan’s exports grew 10-fold in 30 years, much less than the 25-fold observed for the rest of South Asia. During this period, the most intense growth of Pakistan’s exports was observed at the end of 1980s, at the beginning of 1990s and mid 2000s. The end-2000s, however, is associated with a strong deceleration of export growth, following the drop in demand associated with the current industrialized countries’ global debt and economic crisis. Figure 2 Export Growth, 1980-2010 (1980=100) 3,000 2,500 2,000 1,500 1,000 500 0 1984 1999 1980 1981 1982 1983 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Pakistan World Low income South Asia Source: World Development Indicators (WDI) 11. From a sectoral point of view, Pakistan’s exports are dominated by labor -intensive light manufacturing—like textiles, clothing, footwear and agri-foods. The ‘textiles, apparel, leather and footwear’ sector’s share in total exports in 2008-10 was 65.6 percent. The second most important export sector was the agri-food sector with slightly more than 12.4 percent of total exports. Despite the geo-political and security difficulties posed by the war on terror, all sectors of Pakistani exports appear to have seen a healthy growth during the period 1998- 2010. The majority of sectors experienced double-digit annual export growth, with extractive industries, metals, and chemicals growing above 20 percentage points annually over the last decade. During 2007-09, years coinciding with the financial crisis, however, the manufacturing sectors, especially the textiles and apparel industries saw a drop in demand and annual growth was much slower in all sectors except machinery, electronics and transportation equipment, possibly partly sustained by the vitality of the electronics sector globally. 8 Pakistan: Reinvigorating the Trade Agenda Table 1 Composition and Growth of Exports by Sector Value (US$ mill) % total exports RCA CAGR* 98-00 08-10 98-00 08-10 98-00 08-10 Textiles, apparel, leather, footwear 18,400 34,800 79.7 65.6 9.8 12.3 6.6 (HS 41-42, 50-65) Agriculture, meat and dairy, seafood 1,946 6,568 8.4 12.4 1.8 2.9 12.9 (HS 1-10, 12-14) Extractive industries 440 4,347 1.9 8.2 0.2 0.4 25.7 (HS 25-27, 68-71) Other industries 967 2,193 4.2 4.1 0.6 0.7 8.5 (HS 37, 43, 49, 66-67, 90-97) Food, beverages, tobacco, wood 677 1,550 2.9 2.9 0.4 0.5 8.6 (HS 11, 15-24, 44-48) Chemicals, plastics, rubber 224 1,446 1.0 2.7 0.1 0.2 20.5 (HS 28-36, 38-40) Iron, steel, and other metals 213 1,371 0.9 2.6 0.1 0.3 20.5 (HS 26, 72-83) Machinery, electronics, transp. eq. 203 724 0.9 1.4 0.0 0.0 13.6 (HS 84-89) *Compound average growth rate 1998-2000 v. 2008-2010; RCA: Revealed Comparative Advantage. Source: COMTRADE 12. Geographically, the European Union (EU27) and the United States (US) represent the most important destinations of Pakistan’s exports. In 2008-10, the US and European markets absorbed 20.5 percent and 31.4 percent of Pakistan’s total exports, respectively. The main markets in the European Union are United Kingdom (7.1%), Germany (6.5%), France (4.4%) and Italy (4.1%). China is the third most important destination, with 7.5 percent of total exports. Turkey, Korea and United Arab Emirates have recently become important destinations. Table 2 Export Shares in Selected Destination Markets % total exports CAGR* 1998-2000 2008-10 EU27 33.3 31.4 8.1 United States 25.9 20.5 6.2 China 5.5 7.5 12.1 Turkey 0.7 3.7 28.1 Korea, Rep. 2.7 2.7 8.7 United Arab Emirates 0.0 2.7 - Rest of the World 32.0 31.6 -2.1 Source: COMTRADE * Compound annual growth rate 1998-2000 v. 2008-2010 Diversification 13. Diversification of exports across markets and products is important for a country like Pakistan. It reduces the vulnerability of the country’s export portfolio to partner-specific shocks and to extreme volatility in export prices. The recent global crisis has highlighted the critical importance of diversification in reducing the risks of volatility in global demand and explains why this topic is rising in the agenda of low-and middle-income countries. 9 Pakistan: Reinvigorating the Trade Agenda 14. Overall, Pakistan’s export bundle is mildly diversified in terms of products, but not geographically. The emphasis on industrialized countries export markets makes the country too vulnerable to adverse shocks in these countries. In terms of geographical specialization, Pakistan’s exports are more diversified than most of its peer countries with the exception of China and Turkey (Figure 4)—but not so in terms of markets (see below).3 15. Pakistan’s exports are still very concentrated geographically, in the past decade the country started diversifying, a path followed as well by all the peers. In a decade, the prominence of US and European markets in Pakistan’s exports has reduced by about 7 percentage points (from 59.2 percent in 1998-2000 to 51.9 percent in 2008-2010). At the same time the rest of the world has become more important. There is a small increase in exports to the emerging bloc of Brazil, Russia, India and China (BRIC). UAE, Korea, and Turkey have also become important destinations 16. Deepening this recent trend is key as Pakistan appears to be still under exporting with large and fast growing economies of the 21st century. Pakistan’s exports to Japan, Brazil, Russia, and of course India with which ties are fraught with political relations, are below what could be expected, as can be seen in Figure 3, based on a gravity model of trade4 that depicts countries with which Pakistan under performs above the 45-degree line. Among all trading partners, India is the country with greater potential, followed by Japan. In South Asia, Pakistan’s exports to Sri Lanka have grown in positive response to a bilateral trade agreement in place. Pakistan trades heavily with members of the Gulf Cooperation Council (GCC) which is explained by proximity and trade complementarities as well as from the fact that these serve as re-export bases for Pakistan’s trade with a number of third countries. 3 The Herfindahl index measures concentration of export value. It can be computed either at the market or at the product level. It is computed as the sum of squared shares of each product (market) in total export. A country with a perfectly diversified export portfolio will have an index close to zero, whereas a country which exports only one export (market) will have a value of 1 (least diversified). 4 We regress mirrored bilateral exports (2007-09) on the following bilateral characteristics with trading partners: distance, contiguity, common language, colony, common colonial power, as well as log of GDP, log of GDP per capita, etc. We also incorporate three innovations. First, a measure of remoteness is computed by summing distances weighted by the share of GDP of the destination in world GDP. This is to take note of the fact that relative distances matter greatly, alongside absolute distances. Second, we control for zero trade flows with the use of Heckman sample selection correction method. When observations with non-existent bilateral trade are dropped, as OLS does, our dependent variable is not really measuring bilateral trade, but one contingent on a relationship existing. An important variable left out of the model therefore is the probability of being included in the sample, i.e. having a non-zero trade flow. To the extent that the probability of selection is correlated with GDP or distance, this has the potential to bias OLS estimates. Third, we address heterogeneity of firms, following Helpman, Melitz and Rubinstein (2008), by controlling for firm heterogeneity without using firm-level data utilizing the fact that the features of marginal exporters can be inferred from the export destinations reached. With these steps, the gravity results of whether a country over trades or under trades with particular partners are better grounded on trade theory. 10 Pakistan: Reinvigorating the Trade Agenda Figure 3 Pakistan's Export Partners, 2007-09 Pakistan’s Exports, Predicted vs. Actual Flows (2007-09) PREDICTED VS. ACTUAL (GRAVITY) 16 USA IND JPN CHN GBR FRA DEU RUS ITA BRA KOR AUS THA BGD SAU TUR 12 VNM MEX CAN MYS ARE EGY ZAF HKG SGP KWT LKA NPL QAT OMN 8 8 12 16 Log of Actual Exports, 2007-09 Note: Pakistan has bilateral agreements with China, Sri Lanka, and Malaysia (circled) Source: Authors’ calculation based on COMTRADE database Figure 4 Herfindahl Index of Export Product and Market Concentration Herfindahl Index - Markets Herfindahl Index - Products (HS6) 0.096 0.064 BGD 0.188 BGD 0.036 0.076 0.008 CHN 0.158 CHN 0.005 0.041 0.054 EGY 0.068 EGY 0.104 0.047 0.029 IND 0.079 IND 0.025 0.080 0.033 MYS 0.120 MYS 0.032 0.058 0.017 PAK 0.087 PAK 0.016 0.039 0.007 TUR 0.089 TUR 0.007 0 .05 .1 .15 .2 0 .02 .04 .06 .08 .1 0810 9800 0810 9800 Source: Authors’ calculations based on COMTRADE data 17. In terms of products Pakistan is mildly diversified, more concentrated than China and Turkey, but less concentrated than other middle-low income countries. The value of its Herfindahl index is lower than 0.02. More worrisome, however, is that in almost a decade, 2000-08, there was an increase in product concentration in many countries, including China, Malaysia, Bangladesh and India, whereas Pakistan’s index did not decrease but slightly increased. 18. Overall, Pakistan is broadly moving toward new products and markets; but the pace in terms of new products is slower than those of potential competitors. The definition of the intensive margin (IM) and the extensive margin (EM) of exports follows Hummels and 11 Pakistan: Reinvigorating the Trade Agenda Klenow (2005). Its computation reveals that, in terms of products, Pakistan’s share of exports in products that the rest of the world also exports (e.g. IM) has declined slightly over the last ten years, but it appears to be moving towards new exports that are economically significant (e.g. EM) (Figure 5).5 Countries like Vietnam, India and Indonesia, however, have managed to increase their share of export in goods which the rest of the world produces too (IM) as well as the breadth of their export portfolio relative to all exportable products (EM). In terms of market destinations, Pakistan’s export share in countries it currently exports to (IM) has reduced slightly, but it has increased its reach to markets that cumulatively are larger relative to the world in 2000 than in 2010 (EM). This is consistent with the relatively higher market diversification found above. In contrast, India increased its existing share of exports to existing markets, but has not added major new markets to its portfolio of destinations. Indonesia and Vietnam have done both. Pakistani products that appear as new in 2010, compared to 2000 are petroleum oils, semi milled or wholly milled rice, Portland cement, gold, copper, cement clinkers, polyethylene, collages and undenatured ethyl alcohol. Figure 5 Intensive and Extensive Margin in Products and Markets Products Markets 1.5 1.5 IND IND IDN IDN IDN Intensive Margin 1 IDN Intensive Margin 1 IND IND VNM VNM .5 .5 VNM PAK PAK VNM PAK PAK 0 0 60 70 80 90 100 45.2 45.4 45.6 45.8 46 46.2 Extensive Margin Extensive Margin 2000 2010 2000 2010 Source: Authors’ calculations based on COMTRADE data Quality and Sophistication 19. The technological content of exports in Pakistan is low. High-tech exports constituted less than 2 percent in 2010, a share broadly unchanged in the past 25 years. This is extremely low for a country which has good educational and research institutions, and a large population. For example, Vietnam’s share of high-tech exports went from 0.7 percent to 3.8 percent of total in the same period whereas India’s increased from 2.8 percent to 6.2 percent. Moreover, while the majority of the country’s exports (around 54 percent) was classified as low-tech already in 1985,6 this share further increased over time, to reach close to 75 percent in 20067 (Figure 6). Some evidence of structural change appears after 2007, but the growth in technological content of exports remains small. 5 The IM in products indicates how big is Pakistan in what it exports, and the EM measures how globally important (often called ‘economically strategic’) is what it exports. In openness terms , a country is expected to move from being a ‘big fish in a small pond’ to ‘a small fish in a big pond.’ 6 Lall, S. (2000). “The technological structure and performance of developing country manufactured exports, 1985-98�. Oxford Development Studies. Vol. 28(3). pp. 337–369. 7 World Bank calculations from COMTRADE data 12 Pakistan: Reinvigorating the Trade Agenda Figure 6 Technological Content of Exports 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Low tech Primary Resource based Medium tech High tech Source: Authors’ calculations based on COMTRADE data 20. Furthermore, the low level of export sophistication in Pakistan is also likely to weigh negatively on the country’s future economic growth. Pakistan’s increase in export sophistication is modest. In 1986, the sophistication of its export basket was higher than that of Vietnam. In 20 years, Vietnam has not only caught up, but surpassed Pakistan’s achievement in upgrading exports. Measuring export sophistication (EXPY) over time gives an important indication of the relative growth in sophistication of the export basket and the degree to which this is impacting growth of per capita income. By observing the movement of EXPY indices over time, we cannot definitively say if rising export sophistication pulled up per capita incomes, or countries moved into production of more sophisticated exports after average incomes rose. However, Hausmann and Klinger (2007) have shown for a large pool of countries that export sophistication at present is a good predictor of economic growth in the future. Felipe (2010) estimates that a 10 percent increase in EXPY at the beginning of period raises growth by about half a percentage point on average. Also in terms of sophistication, in the past two decades Pakistan export basket has not undergone as stark an improvement as its Asian peers (See Box 1 for further details). Figure 7 Export Sophistication (EXPY), 2002-10 9.8 2010 2008 2006 9.6 2004 2004 2002 2006 2002 2010 2008 Log EXPY 9.4 2004 2006 2008 2010 2002 2010 2008 2004 2006 2002 9.2 2004 2002 2010 2006 2008 9 7.5 8 8.5 9 9.5 Log GDP per capita, PPP (constant 2005 international USD) PAK IND MYS CHN VNM Source: Authors’ calculations based on COMTRADE data 13 Pakistan: Reinvigorating the Trade Agenda Box 1 Sophistication of Exports in Pakistan Hausmann, Hwang and Rodrik (2006) show that the level of ‘sophistication’ of products matters for economic growth. Countries that have a more sophisticated export basket, proxied by a measure named EXPY, enjoy accelerated subsequent growth while those with less sophisticated export baskets tend to lag behind—in essence, countries’ growth acceleration becomes reliant, inter alia, on what they export. In a sample of 100 developing countries that the World Bank has classified as either low or middle income, Pakistan lies below the ‘average’ regression line implying that its export basket is ‘poorer’ than it should be, given its income per-capita. This is based on one of the measures of export sophistication (EXPY) which assesses the export basket of a country in relationship to the income of countries that produce similar products, weighted by the share of those exports in the national total. Developing new products is much more important for countries below the line. In contrast, countries above the line can expect to see growth from existing products. This includes not only China and India, but also Thailand and Turkey that stand out for producing goods that are deemed sophisticated and therefore notable for countries at comparable levels of economic development. Also in terms of sophistication, in the past two decades Pakistan export basket has not undergone as stark an improvement as its Asian peers (see above). China has made the biggest leaps in upgrading both the income gains and level of sophistication of its export baskets, although starting already from high basis (Figure 7). The assembly role of China in many international value chains has certainly contributed to the spectacular performance of this country. The leap that China demonstrated is only possible as the productive base increasingly mimics rich countries by producing what they produce. With data accounting for disaggregated production, findings would be more nuanced. Figure 8: Export Export inDeveloping Sophisticationin Sophistication Countries Developing Countries 9000 THA TUR EXPY 2008 (Current US$) CHN IND EGY 6000 IDN VNM IRN PAK 3000 BGD 0 5 6.5 8 9.5 Log of GDP Per Capita (PPP) Source: Word Development Indicators - COMTRADE 14 Pakistan: Reinvigorating the Trade Agenda Services and Direct Foreign Investment 21. Restrictions to trade in services in Pakistan are lower than in most comparator countries; however, financial and professional services are affected by relatively higher restrictions. According to the World Bank’s Services Trade restriction Index (STRI), Pakistan’s restrictions to trade in services are fewer than any comparison country with the exception of Turkey (Figure 9) and far below neighboring India. This positive result masks some differences at the sectoral level: trade in services in some sectors like retail, telecommunications and transportation seem to have very few restrictions while services in financial and professional services present higher restrictions to trade. Figures Services and Trade Restrictions Index Figure 9: Services and Trade Restrictions Index Figure 10: Pakistan: STRI by Sector India Financial Iran Egypt Professional Thailand Bangladesh Transportation Viet Nam China Telecom Pakistan Retail Turkey 0 20 40 60 80 0 10 20 30 40 50 60 Source: World Bank’s Services Trade Restrictions Database (2012) 22. Trade in services, although picked up only partially in official statistics, has acquired some momentum, possibly because it can override tangible, physical constraints faced by the goods sector. Increasingly, services trade has also become an important source of export diversification8. In terms of services exports, Pakistan shows some growth, but less than its peer countries: Bangladesh, China, Egypt, India, Malaysia, and Turkey. Outside Bangladesh, Pakistan has the lowest share of services in GDP among its peers. Between 2000 and 2007, exports of commercial services grew at a compound annual rate of nearly 15 percent, which is just below the average rate of peer countries. Transport and information communication technologies services are, respectively, the most important types of services exported. 8 The World Bank Group Trade Strategy 2011-21: Leveraging Trade For Development And Inclusive Growth 15 Pakistan: Reinvigorating the Trade Agenda Figures Export of Services and Service Export Sophistication Figure 11: Export of Services, Pakistan & Peers Services Figure 12: Service EXPYSophistication Export 18000 Trade in Services 16000 40 14000 35 12000 Bangladesh 30 Bangladesh Percentage of GDP 10000 Egypt, Arab Rep. China EXPY 25 Egypt, Arab Rep. 8000 India 20 India Malaysia 6000 15 Malaysia Pakistan 4000 10 Turkey Turkey 5 2000 Pakistan 0 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2001 2002 2003 2004 2005 2006 2007 2008 2009 Source: World Development Indicators Source: Mishra, Lundstrom and Anand (2011) Table 3 Pakistan: Commercial Services Exports Growth rate 2000 2007 2008 2009 2010 00-07 07-10 Commercial Services (US million) 1,284 2,224 2,531 2,548 2,792 8.2 7.9 (% commercial services exports) Transportation 65.4 48.0 48.5 45.3 47.7 4.9 7.7 Other business services 10.6 18.8 19.8 19.0 21.5 25.2 12.7 Travel 6.3 12.4 12.5 10.7 10.9 27.8 3.3 Communication 14.8 5.8 3.6 11.1 8.7 -7.6 23.7 Computer and Information 1.7 5.7 7.4 7.1 6.9 41.8 15.3 Financial 0.8 3.0 2.2 4.0 1.8 46.3 -9.3 Insurance 0.4 1.6 2.8 1.8 1.6 48.4 6.9 Construction 0.0 3.0 1.7 0.7 0.7 - -32.8 Personal, cultural, recreational 0.0 0.0 0.1 0.1 0.1 - 58.7 Royalties and license fees 0.0 1.7 1.5 0.2 0.1 - -56.7 Source: World Development Indicators 23. In order to have a detailed look at the services export trend, we present the export sophistication of services using the data and methodology suggested by Mishra, Lundstrom and Anand (2011)9. Services have gained momentum in explaining economic growth, according to the authors, since new technologies have made them more tradable. As seen in Figure 12, it is clear that since 2003 Pakistan has had a positive trend on services sophistication. This positive trend is similar to the one observed in all peer countries. 24. Until the financial crisis, 2008-09, Pakistan had had a steady rising inflow of FDI, as a percentage of gross fixed capital formation. Inflows were relatively sizeable compared to other Asian economies (Table 4). FDI has mainly come into the largely non-tradable sectors of banking and telecom. Yet, its potential for technology transfer may help upgrading the efficiency of these two important backbone services thereby giving a boost to the country’s 9Mishra, Lundstrom and Anand. 2011. “Service Export Sophistication and Economic Growth�. World Bank Policy Research Paper 5606. March 2011. World Bank: Washington D.C. 16 Pakistan: Reinvigorating the Trade Agenda overall productivity and exports10. According to the World Bank Group’s latest report on FDI regulations, Investing Across Borders 2010, 27 out of a total of 33 sectors are fully open to foreign capital participation in Pakistan. In the financial services sector, the Banking Companies Ordinance allows a maximum of 49 percent foreign ownership of Pakistani banks, while foreign capital participation in local insurance companies is allowed up to a 51 percent share. Table 4 FDI Inflows: (% of Gross Fixed Capital Formation) 2005 2006 2007 2008 2009 2010 Bangladesh 5.5 4.6 3.9 5.2 3.3 3.7 China 12.9 11.2 11.7 9.5 5.0 6.9 Egypt, Arab Rep. 33.4 49.9 42.6 26.2 18.8 15.7 India 3.0 6.8 6.2 11.0 8.3 4.7 Indonesia 12.3 5.6 6.4 6.6 2.9 6.1 Iran, Islamic Rep. 6.1 2.8 2.3 - - - Pakistan 11.5 16.3 18.6 16.2 8.7 8.3 Thailand 15.8 16.3 17.4 11.4 7.6 12.3 Turkey 9.9 17.1 15.6 13.4 8.1 6.6 Vietnam 11.2 11.8 24.7 30.4 22.7 21.1 Source: World Development Indicators 25. FDI inflows dropped by almost two thirds between 2007 and 2010, mainly due to a sharp contraction of FDI inflows to services sectors (mainly transport, communications and finance) and to a lesser extent by lower inflows to manufacturing sectors (mainly food and beverages). FDI inflows increased significantly after 2004 and reached a peak in 2007 at US$ 5,590 million or 3.9% of GDP (Figure 13). However, FDI inflows decreased every year after 2007 to reach 2,018 million or 1.1% of GDP in 2010. The tertiary sector experienced a decline of 84% in FDI inflows between 2007 and 2010, while FDI inflows to the secondary sector declined by 62%. Three sub-sectors were responsible for about 80% of the drop in FDI inflows between 2007 and 2010: ‘transport, storage and communications’ (US$ 1,782 million), ‘finance’ (US$ 830 million), and ‘food and beverages’ (US$ 456 million). The decline in FDI inflows has reached alarming levels: In 2012, FDI inflows fell to 700 million. This declining trend indicates the aversion of foreign private investors towards Pakistan due to perceived country risks, overall insecurity and the weak macroeconomic situation.11 10 Pack, Howard and Kamal Saggi .2006. “The Case for Industrial Policy: A Critical Survey.� World Bank Policy Research Working Paper 3839. February 2006. World Bank: Washington D.C. 11 Paradoxically, Pakistan offers a very open environment for foreign investment. Foreign investors can hold 100% equity in all industrial sectors, with the exception of arms and ammunition, high explosives, radioactive substances, and currency printing and minting. Official approval is not required for new manufacturing investments, although the establishment of new units for manufacturing alcohol is banned, except for industrial alcohol. There are minimum FDI levels of US$300,000 in the social, agriculture and infrastructure sectors and of US$150,000 in the services sector, while all other sectors do not have a minimum investment threshold. In addition, Pakistan allows full repatriation of capital, profits and dividends in all sectors (EIU 2011). General investments incentives, including liberal tariff and tax concessions, do not discriminate between foreign and domestic investors. The investment policy lists openness, liberal tax concessions and legal protection as major inducements to attract foreign investment. Additionally, Pakistan puts no restrictions on employing foreigners, and foreign companies may appoint foreign citizens as chief executives. Although companies that want to employ foreigners must first seek permission from the government’s Board of Investment, this is merely a formality and usually takes no more than 2 to 3 weeks (EIU 2011). 17 Pakistan: Reinvigorating the Trade Agenda Figures FDI Inflows Figure 13: Total FDI Inflows (US$ million) Figure 14: FDI Inflows by Sector (US$ million) 6,000 5,000 5,000 4,000 4,000 3,000 3,000 2,000 2,000 1,000 1,000 0 0 2007 2008 2009 2010 Primary Secondary Tertiary Source: World Development Indicators and ITC Export Dynamism at the Firm Level 26. As in most other countries worldwide, Pakistan’s exports are concentrated in the hands of a few large exporters. The bottom 80% of exporters accounted for only 5.3 percent of total export value in 2010, down from 7.7 percent in 2003. The concentration has not changed much over time as can be seen in Figure 15, which plots the percentile of exports against the percentile of exporters for different years. Concentration across Pakistani exporters is almost unchanged between 2003 and 2010. For the analysis of firm-level data we use data from a customs transactions database for the period 2001-10. Figure 15 Concentration of Exports Across Exporters Source: Pakistan’s Customs Data 27. Export entry and exit rates are lower than in peer countries but new entrants more than offset the loss in exports by exiting firms. Each year, anywhere between 16 and 37 percent of firms exit and another 19 to 37 percent enter. While at first sight there seems to be significant destruction and creation of exporters in the data, entry and exit rates are lower than in other middle-low income countries (Figure 16). Moreover, the net effect on the 18 Pakistan: Reinvigorating the Trade Agenda number of exporters seems to be marginal. This suggests that the number of exporters is not expanding significantly over time. Yet, new exporters seem more effective than sunset exporters. This is reflected through the decomposition of annual export growth by firms churning (Extensive Margin) that is presented graphically in Figure 17 and Figure 18.12 Firm churning, i.e. frequent entries and exits, seems to be associated with a positive selection effect, whereby new entrants more than offsets the loss in exports by exiting firms. Firm export growth due to entry of new firms into the exporting market in the period 2002-10 was on average from 1 to 2 percent higher than the loss in growth due to firms that stopped exporting (see black marker in Figure 17). Indeed, Pakistani firms are quite dynamic along this margin, as seen by significant entries and exits over the years going from 2002 to 2010. Figure 16 Entry and Exit Rates in Pakistan and in Peer Countries Exporter Entry Rate Exporter Exit Rate 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0 0 2004 2005 2006 2007 2008 2009 2004 2005 2006 2007 2008 2009 PAK BGR COL ECU PER ZAF PAK BGR COL ECU PER ZAF 13 Source: Customs data for Pakistan, Bulgaria, Colombia, Ecuador, Peru, South Africa Figures Annual Export Growth Figure 18: Generated by Firm Entry/Exit from Figure 17: Generated by Firm Churning Specific Export Market (Firm Extensive Margin) (Country Extensive Margin) 15 15 10 10 5 5 0 0 -5 -5 -10 -10 Country Entry Country Exit Net Country Firm Entry Firm Exit Net Firm Source: Pakistan Customs Data 12 This decomposition is carried out using the methodology in Bricongne et al. (2011) and applied to Pakistan (see World Bank 2012). 13 The list of countries available is much more limited in case of firm-level data comparisons. Among the existing countries we selected some middle income countries and mid-sized countries—Colombia, Ecuador, Peru, South Africa and Serbia 19 Pakistan: Reinvigorating the Trade Agenda 28. Churning along the destination-country margin also generated positive export growth. By contrast, there is a more anaemic performance in terms of sectoral and product differentiation. Again, taking 2002-03 as the base year over the 2002-10 period, existing firms were instrumental in exporting to new countries, and even while there was evidence of exits, the net effect on the export growth rate was positive (Figure 18). Sectoral and product14 churning over the 2002-10 period, was overall not very significant, and it seems that on average exits outweighed entries (Figure 19). Albeit small, the negative contribution of net export growth from entry and exit of firms from specific products and sectors should not be underestimated. Usually one should expect a higher positive contribution to export growth along the sectoral and product margins than along the firm margin. As Arkolakis and Muendler (2011) show, this is the case because the fixed costs of entry into export is higher than the fixed costs of expanding the product range due to economies of scope at play. Hence the overall interpretation of Figure 19 is that the sector and product margin of growth are clearly underperforming. Figure 19 Annual Export Growth Generated by Firm Entry/Exit Into/From New Sectors Into/From New Products (Sector Extensive Margin) (Product Extensive Margin) 15 40 10 30 5 20 0 10 -5 Sector Exit Product Exit 0 -10 Sector Entry Product Entry -10 -15 Net Sector Net Product -20 -20 -25 -30 Source: Pakistan Customs Data 29. In terms of export volumes (Intensive Margin-IM), performance has been irregular. Some years of very good performance have combined with a few years of low export growth, possibly reflecting developments in the macro-economic environment and in foreign demand (Figure 20). The rates of survival in Pakistan are not low, at least compared to a group of middle-low income countries, for which customs data at the firm level are also available (Figure 21). Table 5 summarizes the rates of survival by cohort. For instance, in 2003, 5,435 exporters start exporting. By 2004, only 57% of the original cohort manages to survive, i.e. 43% of firms exit the market and no longer export. After 7 years, i.e. by 2010, 24% of the original cohort still continues to export.15 Although these figures seem low, a comparison of survival rates, after one year of export activity of Pakistani exporters with exporters from peer countries, suggests that Pakistan is more effective in keeping exporting firms in the market—at least in the short run (Figure 21). 14We use the two level-digit classification for sectors and six-digit for products. 15It should be noted that data relative to 2002, the starting year of our dataset, do not refer to new exporters as in the following years, but to all firms active in exporting at that date. This is due to the fact that we cannot distinguish between new entrants and incumbent exporters for the first year of data. 20 Pakistan: Reinvigorating the Trade Agenda Figures Export Intensive Margins and Survival Rates Figure 20: Intensive Margin of Exports Figure 21: Survival Rates after 1 Year Export Annual Export Growth Activity 1-Year Exporter Survival Rate 60 0.7 40 0.6 20 0.5 0 0.4 -20 0.3 -40 0.2 0.1 0 2004 2005 2006 2007 2008 Intensive Positive Intensive Negative Net Intensive PAK BGR COL ECU PER ZAF Source: Customs data for Pakistan, Bulgaria, Colombia, Ecuador, Peru, South Africa Table 5 Survival Rates by Cohort Year 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total 2002 14,747 14,747 2003 63% 5,435 14,701 2004 49% 57% 4,397 14,730 2005 42% 44% 55% 3,633 14,565 2006 36% 36% 41% 54% 3,628 14,696 2007 33% 32% 36% 42% 66% 3,250 15,076 2008 30% 28% 32% 36% 53% 57% 3,057 15,358 2009 28% 26% 29% 32% 46% 45% 56% 2,982 15,676 2010 26% 24% 27% 29% 41% 40% 44% 58% 2,846 15,925 Note: The top of each row provides the count of new exporters entering the market in the given year. The last column provides the sum total of exporters in each year. 30. Dynamism was significantly reduced after 2004 and has still not recovered. This emerges clearly by the decomposition analysis of export growth in intensive and firm, country, sector and product extensive margins. The correlation across export margins is even more clearly visible in Figure 22, which summarizes in one single figure the information in Figure 17 to Figure 20. The arrows emphasize the declining trends in entries and exit for various extensive margins. On average, fewer firms entered and exited the export market since 2004. Similarly incumbent exporters became more conservative over time in terms of export markets, sector and product experimentation. In short, entries and exits by new firms, existing firms in new countries, sectors and products, all fell over the second half of the past decade. 31. Many important changes took place in Pakistan over the past decade, which may explain these trends in firms’ export growth. The fact that export growth decomposed by entries and exits at the firm, sector and product level have all declined over time is an indication of a decrease in dynamism of exports in Pakistan, and possibly a lack of incentives for the latter to innovate. An econometric analysis by means of OLS of the drivers of export growth at the firm level helps quantifying the importance of selected determinants. 21 Pakistan: Reinvigorating the Trade Agenda Figure 22 The Margins of Export, Annual Export Growth 50 40 30 20 10 0 -10 -20 -30 Firm Entry Firm Exit Country Country Exit Sector Entry Sector Exit Product Product Exit Intensive Intensive Entry Entry Positive Negative 2002-2003 2003-2004 2004-2005 2005-2006 2006-2007 2007-2008 2008-2009 2009-2010 Source: Pakistan Customs Data 32. Results show that diversification is good for export growth. Diversification, be it in terms of firms, products or destinations is positively correlated to export growth, with an effect that is sizeable and significant. Pakistan has been quite successful in increasing its exports by having more firms starting exporting and by having more exporters entering new markets. The performance in terms of helping firms breaking into new sectors and products has however been less satisfactory. 33. Export performance at the firm level is severely constrained by the fixed costs associated with entering in the exporting activity. Econometric analysis on firm level export growth shows, for example, that a 1% change in the amount of impediments at the border implies a 16.7% change in a firm’s export growth.16 We measure impediments at the border as the change in the total number of days it takes to fulfill the customs procedures in Pakistan and in the destination country. We assume that a reduction in the number of days needed for clearing merchandise at the customs is correlated with a more general effort of a country to simplify import procedures and thereby reduce the fixed costs of entry to the market. Hence we assume that this measure is a good indicator of the impact of fixed costs on export growth. 34. Trade impediments first and foremost constrain dynamism and experimentation, a vital determinant of countries export performance over time. For example, trade impediments at the border have a sizeable and statistically negative effect on all the extensive margins of trade, i.e. they constrain the ability of new firms to start exporting, the ability of existing exporters to enter new markets and their ability to diversify their product offer or to expand to new sectors. Arguably, burdens to export are distortive, as they impede resources to flow to the most productive firms, sectors and products. 35. The complexity of the trade tariff structure in the destination market is also a very important obstacle to export growth for Pakistani firms. A 1% increase in tariff ‘complexity’ 16For further references on the econometric assessment of the determinants of firm export growth, refer to World Bank (forthcoming). 22 Pakistan: Reinvigorating the Trade Agenda in the destination country leads to a 13.2% decrease in export growth by Pakistani firms. We measure tariff complexity as the standard deviation at the (HS-4 digit) sector level of individual HS-6 digit MFN and applied tariffs. 36. Finally, findings show that fluctuations in export growth are linked to developments in the global demand. Largely determined by external and market conditions, demand can explain an important share of fluctuations and cyclical patterns in the export performance. Amidst very uncertain and volatile conditions for global demand, Pakistan can maximize its chances of exporting by eliminating barriers to trade and by using trade policy to create the conditions that foster firm dynamism, experimentation and quality upgrading of their products. In the following sections we will assess how trade policy can help improving firms’ export performance. Namely, we will discuss more in detail the importance of a streamlined and transparent tariff regime; the effectiveness of Pakistan’s preferential trade strategy; and the state of the country’s infrastructure as a facilitator for trade. 23 Pakistan: Reinvigorating the Trade Agenda Trade Policy Objectives, Tools for Implementation and Effectiveness 37. In the last fifteen years, Pakistan’s trade patterns have been characterized by a pendulum-like trajectory of trade liberalization. Starting in the mid-1990s, Pakistan embarked on a major trade liberalization program, which included reductions to the level of trade tariffs, simplifications of the overall tariff structure, and the abolition of nearly all remaining quantitative restrictions. Over this period of time, trade policy in Pakistan had favored multilateral and unilateral trade liberalization and aimed at reducing government intervention. Meanwhile, a relatively favorable macro-economic environment supported the process of trade policy reform and export expansion. 38. In the mid-2000s the pendulum started swinging back. New import substitution policies started being implemented, with the pace accelerating since the beginning of the global crisis in 2008.17 Current trade policies have become increasingly oriented to provide extra protection to the processing margins of determined local products and producers. The Planning Commission of the Government of Pakistan (2011) reports that out of the 906 manufactured products that have been granted tariff exemptions, 91% benefit a single local monopoly producer, and 5% two producers, while smaller producers, in particular, are excluded from the exemption benefits. 39. Two major factors explain such acceleration. On the one hand, protection increased as part of a series of extraordinary measures adopted by the Government to counter a severe terms of trade shock and stop massive losses in international reserves (which fell from 15 weeks of imports in November 2007 to less than 5 weeks of imports in September 2008). Such losses did not prevent the sharp depreciation of the rupee. On the other hand, increased protectionism encompassed intensive lobbying reflected in a significant number of SROs introduced since 2006 by the Government, well before the crisis started, to benefit certain firms in both the agricultural and industrial sectors. This explains why, even after the balance of payments crisis ended, those SROs remained. There is no reliable estimate of how much these SROs represent the fiscal losses, but proxy estimates are around 1 percent of GDP. 40. The reversal in trade policy has converted Pakistan into the sixth most protected economy in the world, thus risking further deterioration in its already poor performance in terms of openness. The Overall Trade Restrictiveness Index (OTRI) suggests that Pakistan’s overall restrictiveness has increased over time. On the OTRI scale, which ranges from an index of 0 in Singapore and Hong Kong to 16 in Iran, Pakistan reached a value of 9.9 in 2010, up from 9.0 in 2004 (see Box 2 for further details as to the construction of the index; the full results for 2004 and 2007 can be found in World Bank (forthcoming). While the OTRI of Pakistan is not the highest in the South Asia region, as it is surpassed by the restrictiveness of India, the value of 9.9 places Pakistan in the 88th percentile, making it among the group of countries with more restrictive trade policies. We illustrate this in Figure 23, which depicts the distribution of observed OTRI for all countries in the world and marks the position of Pakistan (dark blue) and some of its main competitors (light blue). Overall trade restrictiveness developments in Pakistan are possibly associated with the increasing 17See Planning Commission Government of Pakistan (2011) “Pakistan: Framework for Economic Growth�, pag.74. 24 Pakistan: Reinvigorating the Trade Agenda complexity of trade policy in Pakistan and to the reversal of some of the early progress towards reforms. Box 2 Overall Trade Restrictiveness Index (OTRI) Measuring the restrictiveness of a trade policy is often a complex task, especially given the multifarious nature of tariff and trade regimes in countries. Standard indicators of restrictiveness do not capture widespread exemptions affecting imports and other policy barriers and are not well grounded in theory (Kee et. al. 2008). The Overall Trade Restrictiveness Index (OTRI) quantifies the uniform tariff that, if imposed on home imports instead of the existing, heterogeneous structure of protection, would leave aggregate imports at their current level. It is a rigorous way to calculate and compare weighted average tariffs of countries, with weights reflecting the importance of each good in total imports and the responsiveness of the import of each good with respect to the relative tariff. Its theoretical foundation, developed in Anderson and Neary (1994, 1996, and 2003) was first applied empirically by Kee et al. (2008 and 2009). Following (Kee et. al. 2008), the authors estimate the OTRI for Pakistan and other countries using the latest GTAP database (version 8, base year of 2007) and the current GTAP model (Hertel, 1997), using applied tariffs and non-tariff barriers. The GTAP database provides a more comprehensive estimate of non-tariff barriers compared to the measures available in the UNCTAD TRAINS database, the latter of which are utilized in the calculation of OTRI published in the World Bank Trade Indicators data set. Results are illustrated in Figure 23, which depicts the distribution of observed OTRI for all countries in the world and marks the position of Pakistan and some of its main competitors 41. Pakistan should consider that its prospects of negotiating improved market access with key trade partners may be negatively affected by maintaining its higher barriers. Countries that maintain higher protection are likely to face high trade barriers on their export bundles (Kee et al. 2009). This may be explained by reciprocity in multilateral and bilateral trade agreements: what you get in terms of market access depends on what you are ready to give up in terms of protection at home. The following section will investigate the features of Pakistan’s domestic market protection, thereby assessing the determinants of the high levels of OTRI for Pakistan. 25 Pakistan: Reinvigorating the Trade Agenda Figure 23 Estimated OTRI (except for Pakistan, baseline year 2007) Overall Trade Restrictiveness Index (GTAP) IRN ZWE IND MAR TUN VEN PAK ETH VNM BGD NGA COL EGY RUS KHM SEN MDG TZA BLR MWI LKA UGA NIC ECU BRA ZMB AZE PAN LAO KAZ PER MOZ ALB GTM ZAF ARG THA CHN KOR UKR PRY MYS PHL URY IDN BOL CRI NOR AUS MEX MUS JPN NZL TUR ARM HRV CAN KGZ CHE USA GEO CHL CYP GBR MLT PRT LTU NLD BWA ESP ITA DEU GRC FIN ROU FRA DNK LVA BEL EST POL BGR SWE SVK IRL HUN CZE SVN AUT LUX SGP HKG 0 5 10 15 20 Ivanic and Pitigala (2011) 26 Pakistan: Reinvigorating the Trade Agenda Domestic Market Protection 42. The guiding principles of trade policy in Pakistan are not fully clear, as different programmatic documents seem to have divergent objectives. On one hand, the Planning Commission (2011) recommends that the way forward should include the re-establishment of the unilateral trade liberalization program, regardless of ongoing multilateral, regional or bilateral trade negotiations; the immediate abolishment of the present system of distortive regulatory duties that interfere with the tariff structure; the maintenance of a neutral real exchange rate policy; the re-establishment of normal trade relations with India, including the border opening; and the acceleration of the maximum potential possible of already signed free trade agreements, notably with China and Malaysia. On the other hand, a few recent publications including the ‘Strategic Trade Policy Framework 2009-12’ by the Ministry of Commerce, and the websites of the Ministry of Industry, of the National Tariff Commission (NTC) and of the Engineering Development Board (EDB) suggest that protection of domestic industries is an objective of Pakistan’s trade policy. For example, the website of the NTC indicates that its objectives are to set tariff measures or other forms of assistance for providing protection to the indigenous industry, after having ascertained that there is a cost disadvantage; and to deal with any other matters relating to protection or assistance to indigenous industry that the federal government may refer to the Commission.18 The criteria for granting protection suggest that an infant industry approach and the cascading principle—whereby relatively low tariffs are set for upstream industries and higher tariffs for downstream industries—are used in granting protection. However, no objective criteria are set to determine the period in which an industry is to be considered ‘infant’ and therefore eligibility for protection and acceptable levels of protection are unclear. Finally, it is acknowledged that, since protection is a cost for the consumer, it should not be excessive. In addition, no further criteria are specified for determining when protection is to be considered excessive. 43. Being unclear how the Government of Pakistan will reconcile the objective of protecting the domestic industry with ad-hoc interventions while also aiming at streamlining its trade policy and pursuing greater openness, it is worth recalling that ‘customized’ domestic protection entails important costs and risks. A first risk is that public support will be distortive, going to industries with high costs and low productivity at the expense of more efficient industries. Second, the very existence of schemes that grant protection increases the scope for smuggling and under-invoicing. The cascading principle, evoked by many governmental documents, makes this likely especially for final goods. Third, domestic market protection achieved through tariff protection generates an anti-export bias. Since Pakistan cannot influence the world market prices, exporting firms do not have any benefit from domestic tariff policies. By contrast, producers for the domestic market enjoy a price advantage over foreign competitors. This leads to market distortions whereby the incentive for producers is to focus on serving the domestic market, unless exports are similarly subsidized. Field interviews confirmed the existence of some anti-export bias in the strategic decisions of some successful exporters. Fourth, protection increases the complexity of the tariff regime and the associated information costs for firms. 18 http://www.ntc.gov.pk/tProtevent.asp 27 Pakistan: Reinvigorating the Trade Agenda 44. The existence of conflicting objectives in Pakistan’s trade policy and the ensuing approach of pursuing very precisely targeted protection measures have given rise to an increase in tariff complexity. Since July 2006, applied levels of protection for many products have diverged from the statutory customs duty rates due to the imposition of various exemptions and/or regulatory duties. Regulatory duties and special duties are issued in Statutory Regulatory Orders (SROs) by the Federal Bureau of Revenue (FBR) and, though published by notification in the official Gazette, are not included in the tariff schedule nor consolidated in any published format, creating a lack of transparency in the trade regime and to a distorted and uncertain regime of incentives. Between 2007 and 2009, protection levels were increased for selected products through regulatory duties, while others were made low or zero through concessions to facilitate selected firms and industries. By 2008, some 379 products were subjected to regulatory duties on top of custom duties. In 2011, signaling a reversal of this policy, Pakistan removed the bulk of the regulatory duties, including all such duties on edible items, leaving regulatory duties on 60 PCT codes in place, including sanitary ceramics, tableware, and other selected manufacturing products, in addition to tobacco (see Box 3). The increases in c.i.f. prices on these products range from 15% to 50%, the latter mainly on automobiles. At the same time, tariffs were reduced on many inputs not produced by Pakistan, which, combined with a system of escalated tariffs, creates a very heterogeneous system of effective protection. 45. The complexity of the tariff regime makes unweighted average MFN tariffs a poor benchmark for assessing Pakistani trade policy. MFN unweighted averages have been decreasing over time (Figure 24)—the simple average tariff is now 14.9 percent. However, these developments are unable to account for the increasingly complex structure of the tariff regime. Virtually all tariffs (99.4 percent) are ad valorem and there are currently 16 bands (up from 14 in 2009-10), ranging from zero to 100 percent (up from a maximum 50 percent in 2009-10), with the greatest frequency around the 5 percent slab, which accounts for approximately 40 percent of trade (Figure 25). The automotive sector attracts the highest tariff levels, as high as 100 percent (up from 60 percent in 2009-10) for a few products (13 at HS 6-digit level). However, the presence of regulatory duties and the plethora of exemptions and partial exemptions suggest that the unweighted MFN tariff is an imperfect indicator of Pakistan’s trade openness—such concessions have been further expanded in the most recent Budgets in 2011-12 and 2012-13, further adding to the complexity of the regime (see box 3). Figure 24 Pakistan’s Tariff Reform, MFN Unweighted Average (1982-2009) 100 80 60 40 20 0 1991 2007 1982 1983 1984 1985 1986 1987 1988 1989 1990 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2008 2009 Pakistan Developing Countries Selected competitors Source: Author’s Own Calculations 28 Pakistan: Reinvigorating the Trade Agenda 46. A significant deviation of applied tariff rates from official rates reveals another important facet of the complexity of trade policy. Marginal deviations from statutory rates due to duty drawback and/or concessions due to free trade and other arrangements are to be expected and are featured in many countries’ tariff policy. However, a significant contrast between MFN ad valorem rates and effective duties are observed in Pakistan. Using available customs data at the transactions level for 2009-10 at the most disaggregated available level (8-digit), World Bank (2012) shows that Pakistan applied some 40 different rates. While changes were introduced in the 2011-12 and 2012-13 Budgets, it is expected that the nature and extent of these changes is unlikely to have had a significant impact on the foregoing analysis—while many regulatory duties were removed, additional concessions were introduced and some tariff rates increased, further increasing the complexity of the overall regime (see Box 3). Over 50 percent of the effective tariffs are either 5 percent or less in ad valorem terms, with more bands appearing at the upper end of the scale, creating a higher level of dispersion. In addition, at least 2 percent of transactions appear to levy a rate between zero and less than 2 percent ad-valorem, tantamount to ‘nuisance taxes’. Nuisance taxes, while adding to the cost of business, increase the administrative burden of collecting them, often outweighing any benefit to the government from the marginal revenues (Pitigala and Hoppe, 2010). As can be expected, imports tend to be higher for low tariff products (Figure 25). Thus tariff levels are likely to have a significant impact on imports. Figures Distribution of Tariff Rates and Effective Tariff Rates (2009-10) Figure 25: Distribution of Tariff Rates, by Frequency, Figure 26: Distribution of Effective Tariff Rates by including Regulatory Duties Frequency, including Regulatory Duties Source: Pursell et all. (2011) Source: Author’s Calculations 47. From a sector perspective, services sectors face the most dispersion—although at low levels of tariff—while the footwear/headgear sector posts the most uniform tariff. The measure of dispersion is given by the Gini coefficient computed at the HS 8-digit level. The higher the index, the more dispersed the tariff schedule is within a sector. Applied duty rates are marginally more dispersed than statutory tariffs, but the ranking of sectors is broadly the same (Table 6). 29 Pakistan: Reinvigorating the Trade Agenda Box 3 Recent Import Tariff and Tax Reforms Since the initial analysis using the 2009-10 trade and revenue data, the Government of Pakistan has introduced a number of changes to the import regime, through amendments to the relevant SROs, which were announced in the annual Finance Bills for 2011-12 and 2012-13. These include changes to the MFN customs duty schedule, various concessions and exemptions, as well as regulatory duties and other indirect taxes imposed on imports. MFN Customs Duties—While statutory duties on selected items were decreased, resulting in a lower simple average tariff of 14.9 percent, the number of MFN duty slabs was increased from 14 to 16, with maximum rate increase from 50 percent to 100 percent. Other notable changes to tariffs included the reduction of duties and taxes on hybrid electric vehicles and their batteries by 25 percent in 2012-13. Efforts were also made to rationalize the tariff codes for several sectors and to realign the schedule with HS-2012 nomenclature. Regulatory Duties—The majority of regulatory duties (through SRO 482) were abolished in 2011-12, from 397 items to 60 items, including the removal of regulatory duties on all edible items. While the abolition of most regulatory duties, on consumer goods represents an important step in liberalizing and rationalizing the trade regime and reducing the incidence of tariff escalation, regulatory duties still remain for a for a selected group of consumer goods, from tableware and kitchenware to sanitary ware to automobiles, as well as tobacco products. The initial 397 items accounted for 10 percent of the number of trade transactions in 2009-10 and the duties collected accounted for a small 0.5 percent of total trade revenues. Duty Concessions and Exemptions—Duty concessions and exemptions on the import of raw materials and intermediate inputs used in domestic industries (through SRO 565, 567, and 575) continue to be an important industrial policy mechanism and are wrought with increasing complexity. Over the past two years, a number of concessions and exemptions have been introduced (beyond those relating to trade agreements) to benefit local industries, including pharmaceuticals; air conditioners, evaporators and condensers; printing and stationary; chemicals; glass; oil and gas exploration; car audio manufacturing; aircraft; cement; and fabricated metals; among others. In 2012-13, a diversified list of raw materials, parts and sub-components were made eligible for full exemption (which were previously at a concessionary rate of 5 or 10 percent), while concessions on several other items were removed — including selected inputs to car air conditioners, microwave ovens, audiovisual systems, and mobile phones. Source: PWC Tax Memorandum 2011 and 2012. Table 6 Gini Coefficients and Average Statutory and Applied Tariffs by Sector Gini Coefficients Average un-weighted duty rate Number of lines Statutory Applied Statutory Applied Services 0.96 0.95 0.4% 0.4% 36 Raw hides, skins, leather, & furs 0.55 0.69 10.7% 6.2% 80 Vegetable products 0.51 0.55 11.7% 9.3% 309 Animal & animal products 0.50 0.51 11.7% 6.5% 116 Mineral products 0.46 0.51 7.7% 10.7% 157 Machinery / electrical 0.44 0.51 12.1% 8.0% 1188 Miscellaneous 0.39 0.48 14.9% 20.5% 481 Chemicals & allied industries 0.38 0.47 8.8% 6.9% 1039 Transportation 0.36 0.45 34.5% 9.5% 201 30 Pakistan: Reinvigorating the Trade Agenda Gini Coefficients Average un-weighted duty rate Number of lines Statutory Applied Statutory Applied Metals 0.36 0.45 13.6% 11.0% 696 Wood & wood products 0.33 0.42 14.9% 15.0% 258 Foodstuffs 0.32 0.40 27.7% 12.2% 196 Plastics / rubbers 0.31 0.40 16.7% 13.2% 300 Stone / glass 0.29 0.38 20.4% 19.0% 212 Textiles 0.22 0.38 18.8% 14.1% 798 Footwear / headgear 0.06 0.14 22.9% 19.8% 51 Source: WB Staff estimates 48. Meanwhile, the official customs schedule also suggests a systematic escalation of tariffs by stages of processing. Raw materials and intermediate inputs attract low tariffs. Tariffs on consumer goods are higher, twice as high in many cases. This pattern is more prominent in consumer durable goods such as the automotive and air conditioning sectors where inputs range from 0 to 10 percent and final goods as high as 60 percent, yielding relatively high levels of effective protection (Table 7). Table 7 Pakistan’s Official Tariff Schedule by Stages of Processing (2010) Consumer goods Intermediate goods Raw materials Mean 20.4 10.8 9.5 Std. dev 11.5 8.0 9.6 Source: WB Staff estimates 31 Pakistan: Reinvigorating the Trade Agenda A Deeper Look into Regulatory Duties and Partial Exemptions as Distortions to Trade 49. The dispersion in tariff rates is further associated with additional duties and exemptions. Until 2006, many products were subject to local content requirements. However, in 2006, local content requirements had to be removed to align with Pakistan’s WTO commitments. In order to maintain a certain degree of protection of the domestic market, Pakistan substituted such local content requirements with a complex mix of regulatory duties and partial exemptions for inputs. While the recent abolition of most regulatory duties on consumer goods represents an important step in liberalizing and rationalizing the trade regime, reducing the incidence of tariff escalation on consumer goods, regulatory duties still remain for a selected group of consumer goods, from tableware and kitchenware to sanitary ware to automobiles, as well as tobacco products. In case of automobiles, regulatory duties may be as high as 50 percent, while special duties may be up to 35 percent. While additional special duties are not applied on other products, it begs the policy rationale of isolating the narrow range of manufacturing products to retain regulatory duties, systematically escalating the effective tariff regime. Retaining regulatory duties on selected consumer goods and not others is symptomatic of the selective distortion created by Pakistan’s trade policy. Furthermore, retaining regulatory duties on excisable products, such as tobacco, adds to the administrative burden as a dual tax. 50. Exemptions and partial exemptions provided for industries under SRO regimes are a key source of deviation from MFN rates. Analysis of the level and structure of Pakistan’s tariffs is greatly complicated by large numbers of exemptions and partial exemptions which are announced separately through SROs and do not affect the customs duty rate shown in the customs duty column of the tariff schedule. Additional SROs further specify whether specific products are exempted from sales and other domestic taxes, as well as rules and ordinances affecting imports. While these are now provided on the Federal Bureau of Revenue’s website, the separate SROs make it difficult to discern the applicable taxes and other measures imposed on individual tariff items, which may be covered under multiple SROs with different exemptions and criteria.19 These create distortions, as in most cases they are exemptions for inputs and are confined to specified firms or groups of firms. By confining regulations to a selected group of firms and by barring other importers, in particular commercial importers, these exemptions act as a de facto licensing scheme (Purcell et al. 2011). 51. The most prominent exemptions and partial exemptions applicable to the industrial sector are found under three SROs: 565(1)2006, 567(1) 2007 and 575(1) 2007, (see Box 4), accounting for 23 percent of Pakistan’s imports in 2009-10 (Table 8). On average, firms or industries under these three SRO provisions received concessions up to about 11 percentage points from the statutory rates, applied non-uniformly across industries (see Box 4). The exemptions under SRO (565, 567 and 575) cover several sectors, including chemicals, automotive, rubber and plastic industries as well as the agricultural sector. Concerning their distribution across sectors, 44 percent of all SROs providing exemptions were on machinery and appliances sector (Table 9). In addition, an extensive scheme of exemptions in the automotive sector, which discriminates by type of market (Original Equipment Manufacturer 19 http://www.cbr.gov.pk/SROsImportA.aspx. 32 Pakistan: Reinvigorating the Trade Agenda (OEM) versus after sales parts), is implemented under SRO 656 (1) 2006, auto vendors under SRO 655 (1) 2006 and SRO 693 (1) 2006. The recent budget in 2011-12 introduced amendments to SRO 565(1) that increased the number of concessions and exemptions (beyond those relating to trade agreements) to benefit local industries, including pharmaceuticals, printing and stationary, chemicals, glass, oil and gas exploration, car audio manufacturing, fabricated metals etc. among others. 52. A common feature of all SROs is that their within sector incidence varies considerably, increasing the inefficiency of the incentive structure. More generally, the complex tariff system in place does not generate easily detectable patterns of effective protection. For example, optical photography and related products which have high technological and capital intensity content enjoy low mean tariffs and many exemptions (30 percent of the total), proving low effective protection for a large range of products. In contrast, footwear and head gear has virtually no exemptions and high tariffs, leading to high effective rates of protection. Box 4 Exception to MFN Regime: SROs 565(I), 567(I) SRO 575(I)[2006] The Ministry of Finance/CBR issues these as Statutory Regulation Orders (SROs), which are approved by the Economic Coordination Committee of Cabinet. Most SROs provide exemptions or partial exemptions from normal tariffs, but others provide for increased tariffs subject to at least one special condition announced in the SRO. The EDB administered SRO’s mostly providing exemptions for inputs and are confined to specified firms or groups of firms. They are not available to other importers, in particular commercial importers. The three key SRO administered for the objective indigenization or domestic orientation for industry development are SRO 565(I)/2006, SRO 567 (I)/2006 and SRO 575(I)/2006. SRO 565(I)/2006 (‘Survey based’) provides a long (45 page) list of 154 domestically manufactured products. For each of these products there is an associated list of inputs (raw materials, sub- components, components, sub-assemblies, and assemblies) which can be imported at specified low Customs duty rates—mostly zero, 5% or 10%. SRO 567 (I)/2006 (Non-survey based) gives another long (28 pages) list of products for which the normal statutory CDs have been reduced, mainly to zero or 5%, but others (e.g. various textile and garment products) to 3%, 6.5% and 9%. Many of the listed products are organized by industry e.g. Sl. 4 consists of 33 inputs for the poultry sector, and there are similar groups of surgical sector, textile and clothing, and pharmaceutical products. SRO 575(I)/2006 similarly give a long (about 40 pages) list of machines and ‘capital goods’ for which the normal CD is reduced, nearly all to either zero or 5%. As for the other SROs, the machines listed in SRO 575 are mostly listed by user industry e.g. machines used in horticulture and floriculture. While many of these machines are specialized and would probably not have uses outside the specified user industry, others (e.g. gen-sets) certainly have alternative uses. Source: Purcell et al., 2011. 33 Pakistan: Reinvigorating the Trade Agenda Table 8 Statutory & Effective Duty Regime based on Customs Transactions (2009-10) Share of Share of Statutory Duty Effective Duty Imports Transactions Mean Std. Dev. Mean Std. Dev. Min Max Total 1.00 1.00 0.15 0.13 0.04 0.11 0.00 1.86 Regulatory 0.10 0.10 0.21 0.12 0.07 0.06 0.00 2.96 Duties* SROs 565, 567 0.23 0.08 0.14 0.08 0.03 0.02 0.00 0.3 and 575 FTA related 0.06 0.04 0.15 0.10 0.06 0.08 transactions Other SROs 0.08 0.02 0.00 0.97 Other, under 0.57 0.76 0.14 1.60 Statutory Duty Note:* The 2011-12budgetary changes on regulatory duty are not included. Statutory duty is the ad valorem rates of regulatory duty (mean and Std. Dev.), whereas Effective Duty is the percentage additional duty on the statutory rate, in this case by about 7 percent on average. Source: Pakistan’s Custom data Table 9 Sector-wise Distribution of Key SROs based on Customs Transactions data (2009-10) Auto Auto HS SRO SRO SRO 60 Section heading Vendo Codes Range 565 567 575 SRO r 655 656 Live Animals; Animal Products 01 to 05 0 0 0 3 0 Vegetable Products 06 to 14 0 22 0 14 0 Fats And Oils 15 5 2 1 4 0 Prepared Foodstuffs; 16-24 0 7 2 11 0 Mineral Products 25-27 4 13 12 31 3 Products Of The Chemical Or Allied Industries 28-38 72 113 75 374 19 Rubber And Plastics 39-40 43 32 100 148 17 Raw Hides And Skins, Leather, Furskins And Articles 41-46 1 2 4 41 0 Pulp Of Wood Or Of Other Fibrous Cellulosic Material; 47-49 15 18 20 53 2 Textiles And Textile Articles 50-63 15 83 42 340 0 Footwear, Headgear, Umbrellas, Sun Umbrellas, 64-67 0 3 3 25 0 Articles Of Stone, Plaster, Cement, Asbestos, Mica 68-70 9 8 63 49 5 Natural Or Cultured Pearls, Precious Or Semi-Precious 71 1 5 0 16 0 Base Metals And Articles Of Base Metal 72-83 120 80 243 354 32 Machinery And Mechanical Appliances; 84-85 124 138 543 653 79 Vehicles, Aircraft, Vessels And Associated 86-89 0 16 22 89 72 Optical, Photographic, Cinematographic, Measuring, 90-92 17 26 62 105 6 Arms And Ammunition; Parts And Accessories 93 1 2 0 9 0 Thereof Miscellaneous Manufactured Articles 94-95 0 2 36 41 2 Works Of Art, Collectors' Pieces And Antiques 97-98 0 0 0 3 0 Source: WB staff estimates based on Custom data 34 Pakistan: Reinvigorating the Trade Agenda 53. As much as 11% of all HS 8-digit tariff exemptions are likely to be driven by demand from individual firms or group of firms, reflecting that the process of tariff setting is, to a great extent, driven by vested interests. We compute the absolute difference between the statutory and collected tariff rates and decompose the resulting changes into exemptions applied uniformly to all products in HS section headings, HS chapters, HS 4-digit headings, HS 6-digit products and HS 8-digit products, respectively. They further distinguish exemptions, within HS 8-digit categories that are granted due to preferential trade agreements. The remaining changes, i.e. changes within HS 8-digit product categories that are not justified by preferential trade agreements are considered as exemptions granted following specific firm needs. The result of the decomposition provided in Figure 27 suggest that about 19 percent of all rate variation is determined at the HS section level, while additional 21 percent of variation is determined at the chapter level and 25 percent at the heading level. A much smaller share of total variation (17 percent) is caused by differentiated tariffs below the heading level. Finally, about 18 percent of applied tariff variability is caused by the use of exemptions, of which 11 percent are detected at the firm level. Figures Decomposition of Tariffs and Effective Rates of Protection Figure 27: Decomposition of Statutory & Applied Figure 28: Effective Rates of Protection Tariff rate Variation Effective Rates of Protection Section Exemption 450% Heading by firm Beverages & 400% 19% 11% Tobacco Effective Rate of Protection Exemption 350% by country 7% 300% HS8 250% 8% 200% Leather Products HS6 150% Vegetable Oils Motor Vehicles 9% 100% Chapter HS4 50% 25% Mineral Products 21% 0% -50% 0% 20% 40% 60% 80% 100% Value Added Source: Pursell et all. (2011) Source: Author’s Calculations 54. The high degree of tariff escalation and complexity of the tariff structure is reflected in the effective rates of protection (ERP). By taking into account protection on both outputs and inputs, ERPs provide a good representation of tariff-generated transfers to producers than nominal rates of protection (NRPs), which are based on protection of outputs only. Protection on services industries is mostly negative. Imports of most basic raw materials are relatively free of import restrictions. Products at an intermediate stage of production have moderate ERPs. By contrast, import substituting industries such as beverages, tobacco, motor vehicles and accessories, and vegetable oils and fats yield very high effective protection rates (see Annexure 1). These results reinforce the earlier assertion about the escalatory tariff structure in Pakistan and provide evidence of the instrumental role of exemptions and regulatory duties to achieve indigenization objectives in a number of manufacturing activities, and in the automotive sector in particular. 55. The pattern of protection in place in Pakistan discourages the production of high value-added products. Effective protection is typically exaggerated with low value added 35 Pakistan: Reinvigorating the Trade Agenda activities such that even quite low tariffs on final goods can lead to increased returns. Figure 28 illustrates this assertion. Sectors with high effective rates of protection tend to have low value-added or conversely there is a higher incentive to produce low-value added goods. Since sectors with higher effective protection tend to be domestic-oriented this demonstrates the inherent bias against the export competing sectors. 56. Industry and product level ERP measures further help illustrate the protective effects of the existing tariff and incentive regime. Using survey based input-output data, provided by the EDB and by Pakistan’s Input Output Coefficient Organization (IOCO) of the FBR sectoral ERPs20, on selected domestic-oriented sectors show that refrigerators, air-conditioners and deep freezers have effective protection rates between 65 percent and 150 percent, whereas cotton yarn has about 5 percent. This suggests that the ‘indigenization programs’—which target refrigerators, air-conditioners and deep freezers, in particular—are a source of economic inefficiency. 57. Since the global crisis, a few measures have been taken that may increase the protection of the domestic market. As a result of the first generation reforms in early 2000s, the use of quantitative restrictions, as protective devices, has been substantially reduced in Pakistan. However, Pakistan has continued to apply a number a number of non-tariff measures. The most prominent is the EDB-administered ‘Indigenization Program’ with its discretionary control over new entrants to the industry, since all potential entrants need to negotiate with the EDB and agree on an individual indigenization program and how that would be phased-in over time. The administration of this often competing dual SRO/CGO (Customs General Order) regime discriminates against the import of domestically-available materials, amounting to a de facto local content requirement. In addition, according to the Global Trade Alert, other NTBs include bail out/state aid measures, export subsidy measures, export taxes and restrictions, import bans, sanitary and phytosanitary measures and trade defense measures such as anti-dumping measures, countervailing duties and safeguard measures. 20 Both EDB and the IOCO carry out firm-level survey-based estimates input-output coefficient for their respective mandates: The former to implement the Indigenization Program, and the latter for establishing coefficients to implement duty-drawback and other temporary importation schemes. These rates are reviewed and revised periodically. 36 Pakistan: Reinvigorating the Trade Agenda Box 5 Protectionism in Automation Sector Starting in 1988 motor cars, trucks and other vehicles were among the industries which were subjected to ‘indigenization’ programs in Pakistan. One of the purposes of these programs was to attract investment by foreign firms under which they received tariff and non-tariff protection in the domestic market in return for commitments to provide specified levels of ‘local content’ in their domestic operations, preferably by supporting local producers of parts and components. These policies succeeded in attracting investment in Pakistan by a number of multinational firms (Suzuki, Honda, Toyota etc.) which are now supplied with parts and components by large number of local firms. As a consequence of these programs, the automobile industry in Pakistan operates under a policy framework that is characterized by prohibitively high custom duties on the imports of finished cars (ranging from 50% to 100%), very high custom tariffs (50%) on most components and auto parts produced in Pakistan, low tariffs (mostly between zero and 10%) on raw materials and components that are used by vendors approved and registered by EDB, a ban on the import of second hand cars except for a few exceptions, and a set of rules that effectively block new firms from entering Pakistan unless they come with the intention to produce parts locally. The complex system that regulates the automobile industry in Pakistan produces several inefficiencies that include: low capacity utilization levels in the industry (43% in 2010), sub-par quality vis-à-vis international standards, creation of economic rents for firms producing large volumes, fewer incentives to respond to consumers demand for new varieties, virtual monopolization of some specialized market segments by producers, and large and economically inefficient variations in effective protection levels between different car models, processes and components. More significantly, car prices in Pakistan are higher than they would in the absence of the present protection system. The differences between domestic and world prices vary between models and over time. Pursell at. al. (2011) find that, after adjusting for sales taxes and importer and dealer margins, domestic ex-factory prices for an inexpensive type of car (796 cc Suzuki Mehran) exceeded Indian prices by about 30% in 2004 and 32% in 2011 and that local prices for an expensive model (Honda City) exceeded Indian prices by 27% in 2004 but were very similar in 2011. Source: Pursell, Khan and Gulzar (2011) 37 Pakistan: Reinvigorating the Trade Agenda Market Access to Foreign Markets and Preferential Trade Agreements 58. Pakistan’s exports benefit from important preferences in the largest export markets, but preferential tariff margins are quite low. While fewer than 42% of Pakistan’s imports receive any preferences, 84% of Pakistan’s exports go to countries granting preferences to Pakistan. This notwithstanding, the preferential tariff margins are quite low. Only 0.3% of Pakistan’s exports to the top 20 importers receive a preference margin of 10% or above, mainly in the sectors of garments and commodities, while 30% of exports have a preference margin between 0.1 and 2.5%. Table 10 provides an overview of the trade regime enjoyed by Pakistan in the largest twenty export markets while Figure gives a graphical description of the share of Pakistan’s exports to each of these markets enjoying preferential, non - preferential or MFN zero-tariff regime. 59. These figures reflect more general trends. Despite the flourishing of preferential trade agreements in recent years, 83% of world merchandise trade still takes place on a non- discriminatory, most favored nation basis (Carpenter and Lendle, 2010). This is firstly because half of world trade is already subject to zero-tariffs, on a most favored nation basis. Secondly, preferential trade agreements tend to exempt high MFN-tariff items from preferential treatment and continue to trade these products at the MFN rate (WTO, 2011). Finally, large preferences tend to fall on very elastic goods, further reducing the efficacy of preferences (Fugazza and Nicita, 2010). In conclusion, the scope for exchanging preferential market access is unlikely to be extensive for Pakistan. Tab. Fig. Table 10: Preferential Trade Regime for Figure 29: Pakistan’s exports enjoying preferential, non - Pakistan’s Exports to Largest twenty world preferential or MFN-zero tariff in largest twenty world export markets export markets (% shares) Importer Exporter Regime Reciprocal? EUR PAK EU_GSP no MFN zero Non-preferential trade Preferential trade USA PAK US_GSP no ARE PAK MFN MFN 100% CHN PAK FTA_CHN yes 90% HKG PAK MFN MFN KOR PAK PSA_KOR yes 80% TUR PAK GSP_TUR no 70% JPN PAK GSP_JPN no IND PAK FTA_IND yes 60% CAN PAK GSP_CAN no 50% AUS PAK GSP_AUS no 40% TWN PAK MFN MFN MEX PAK MFN MFN 30% RUS PAK PREF_RUS yes 20% THA PAK MFN MFN 10% MYS PAK PSA_MYS yes BRA PAK MFN MFN 0% CAN ARE KOR JPN CHE HKG CHN IDN RUS SGP MYS EUR TUR TWN IND AUS BRA USA MEX THA CHE PAK GSP_special_CHE no IDN PAK MFN MFN SGP PAK MFN MFN Source: Market Access Map, Market Analysis Tools, International Trade Centre, www.intracen.org/marketanalysis 38 Pakistan: Reinvigorating the Trade Agenda 60. Yet, Pakistan is increasingly pursuing a strategy of economic integration through preferential trade agreements. Up to the mid-2000’s Pakistan was member of multilateral preferential trade agreements only. Since 2006, it has instead been pursuing a policy of regional and bilateral trade integration. The country is, since that year, a member of the South Asian Free Trade Agreement. Moreover, since 2005, Pakistan has engaged in bilateral trade agreements with China, Iran, Malaysia, Mauritius and Sri Lanka and in 2010 in a Transit Trade Agreement with Afghanistan (Government of Pakistan, 2011) 61. This shift reflects more general global trends in trade policy. In the last two decades, the number of preferential trade agreements has increased four-fold, to around 300 agreements in 2011.21 It also reflects public awareness that Pakistan under trades with several large and fast growing economies from the region, including India and China (Planning Commission Government of Pakistan, 2011). 62. The overall economic and welfare effects of full tariff liberalization in the context of preferential trade agreements contracted by Pakistan are limited. This is what emerges from simulation analysis using the World Bank SMART (System of Market Analysis and Restrictions of Trade)—a partial equilibrium tool that is part of the World Bank—UNCTAD WITS (World Integrated Trade Solution) platform. A caveat of the model on which the simulations are based is that it can only account for effects on the intensive margin of trade. In other words, it is unable to quantify changes in terms of new trade relationships, i.e. new products’ exports or new markets. 63. The increase in Pakistan’s imports of currently traded products as a result of implementation of a corner solution, i.e. the hypothetical establishment of a zero-tariff regime in the framework of currently signed PTAs, is modest and ranges from 0.3 percent to 2.7 percent of total imports (Figure 30). The biggest impact would result from the full implementation of a zero-tariff regime under the PTA with China. Imports would increase by 2.7 percent with respect to the baseline (total imports in 2009). Slashing tariffs in the framework of PTAs with Malaysia and India would have a smaller effect, which is similar for the two agreements. In both cases, bilateral imports would increase by less than 0.5%. The impact of implementing full liberalization with all SAFTA countries would increase imports by 0.33%. India accounts for 94% of that increase. It is to be expected, however, that tariff reduction would also increase the product range of imported goods for which barriers are currently prohibitive. We will quantify the impact in terms of overall trade potential in further section of the paper. 21 World Trade Organization. 2011. “World Trade Report 2011�. Geneva. 39 Pakistan: Reinvigorating the Trade Agenda Table 11 List of Pakistan Multilateral and Preferential Trade Agreements Export Entry Imports Type of s Agreement Current Partners into Main provisions 2010 agreement 2010 force (%) (%) MULTILATERAL AGREEMENTS Protocol on Bangladesh; Brazil; Chile; Egypt; 1973 Partial - 11.44 4.54 Trade Israel; Korea, Republic of; Scope Negotiations Mexico; Pakistan; Paraguay; Agreement Peru; Philippines; Serbia; Tunisia; Turkey; Uruguay WTO 153 WTO Members 1995 Multilatera - 88.30 96.16 l PREFERENTIAL AGREEMENTS Economic Afghanistan; Azerbaijan; Iran, 1992 Partial Cooperation on 11.95 3.39 Cooperation Islamic Republic of; Kazakhstan; Scope projects and Organization Kyrgyz Republic; Pakistan; Agreement programs of mutual Tajikistan; Turkey; Turkmenistan; interest Uzbekistan GSTP Algeria; Argentina; Bangladesh; 1989 Partial Tariff concessions 16.10 24.02 Benin; Venezuela; Bolivia; Brazil; Scope Cameroon; Chile; Colombia; Agreement Cuba; Ecuador; Egypt; Macedonia; Ghana; Guinea; Guyana; India; Indonesia; Iran, Islamic Republic of; Iraq; Korea, Democratic People's Republic of; Korea, Libya; Malaysia; Mexico; Morocco; Mozambique; Myanmar; Nicaragua; Nigeria; Peru; Philippines; Singapore; Sri Lanka; Sudan; Tanzania; Thailand; Trinidad and Tobago; Tunisia; Vietnam; Zimbabwe Pakistan-Sri Sri Lanka 2005 PTA Tariff concessions 1.35 0.14 Lanka SAFTA Bangladesh; Bhutan; India; 2006 PTA Tariff concessions 5.69 4.50 Maldives; Nepal; Pakistan; Sri Lanka Pakistan-Iran Iran 2006 PTA Tariff concessions Pakistan- China 2007 PTA Preferential market 6.55 13.99 China access through tariff concessions and mutual recognition arrangements in services Pakistan- Malaysia 2008 PTA Tariff concessions in 0.67 5.48 Malaysia goods and services + market access provisions. Mutual recognition arrangements in the field of education and services and chapter on investment. Source: WTO 40 Pakistan: Reinvigorating the Trade Agenda Figure 30 Change in Pakistan’s total imports (% 2009 total imports) Source: COMTRADE data and authors’ simulations using SMART 64. Turning to bilateral trade, while imports from partners in the PTAs would increase according to our simulations, imports from some East Asian countries, the EU-27 and the US would be affected somewhat negatively by hypothetical full tariff liberalization with China (Table 12). Imports from all partners in current PTAs would increase by double digits if those PTAs were to lead to full tariff liberalization. Meanwhile, full tariff liberalization with China would have sizeable negative implications for Pakistan’s trade partners. The largest negative impact would be felt by Japan, which would suffer a 2.6 percent of loss in terms of imports value if the PTA with China would lead to full tariff liberalization. Imports from a handful of other East Asian countries (Korea, Thailand and Indonesia) as well as those from the EU-27 and the US would also be affected negatively by the hypothetical implementation of a zero- tariff regime with China. Imports from the rest of Pakistan’s main trade partners woul d be marginally impacted (less than 0.5%). Full tariff liberalization in the framework of the PTAs with Malaysia, India or SAFTA would have less sizeable effects. Table 12 Pakistan: Changes in Bilateral Imports PTAs Imports Partner 2009 (US$ China Malaysia India SAFTA All PTAs million) China 3,777 28.5 -0.3 -0.3 -0.3 28.0 Saudi Arabia 3,494 -0.1 -0.2 -0.1 -0.1 -0.3 UAE 3,349 -0.3 -0.2 -0.1 -0.1 -0.7 Kuwait 1,804 0.0 -0.3 0.0 -0.1 -0.4 United States 1,760 -1.4 -0.1 -0.1 -0.1 -1.6 Japan 1,289 -2.6 -0.2 -0.1 -0.1 -3.0 India 1,080 -1.0 -0.1 12.6 12.6 11.6 Iran, Islamic Rep. 955 -0.1 0.0 -0.4 -0.4 -0.5 Korea, Rep. 627 -2.5 -0.3 -0.4 -0.5 -3.4 Thailand 600 -2.5 -0.7 -0.1 -0.2 -3.5 Australia 535 -0.2 0.0 0.0 0.0 -0.3 Indonesia 506 -1.6 -0.4 -0.2 -0.4 -2.4 Malaysia 496 -2.0 36.2 -0.1 -0.2 34.2 Singapore 492 -1.2 -0.2 -0.1 -0.1 -1.5 Canada 403 -0.7 0.0 0.0 0.0 -0.7 EU-27 5,087 -1.8 -0.2 -0.2 -0.2 -2.2 Germany 1,270 -2.1 -0.2 -0.2 -0.2 -2.5 41 Pakistan: Reinvigorating the Trade Agenda PTAs Imports Partner 2009 (US$ million) China Malaysia India SAFTA All PTAs United Kingdom 780 -1.7 -0.1 -0.1 -0.1 -1.9 Italy 677 -2.1 -0.2 -0.2 -0.2 -2.5 Finland 418 -2.6 0.0 0.0 0.0 -2.7 France 396 -1.6 -0.1 -0.2 -0.2 -1.9 Note: Change in Bilateral Imports from main partners, as a result of hypothetical full tariff liberalization in current PTAs Source: COMTRADE data and authors’ simulations using SMART 65. The fiscal impact of hypothetical zero-tariff FTAs is low with India, Malaysia and SAFTA, and high with China, according to our simulations using SMART. The fiscal impact of a zero-tariff PTA with Malaysia would be equal to about US$ 70 million or 2.4% of tariff revenues in 2009 (Figure 31). A zero-tariff PTA with India is about US$ 65 million or 2.2% of tariff revenues in 2009. Signing a PTA with the rest of the SAFTA countries would result in an additional loss of revenue of US$ 9 million or 0.2% of total imports, taking the total to US$ 74 million (2.6% of revenues). The reduction of tariffs resulting from the implementation of a zero-tariff PTA with China would have by far the biggest impact on revenues and trade. Tariff revenue would go down by US$ 525 million or 18.2%. Including all the previous scenarios (China, Malaysia and SAFTA) as a free trade partner increases the total loss in tariff revenues by $ 663 million or 23% of tariff revenues. Figure 32 provides the welfare gain estimates net of tariff revenue loss. Whilst these are small compared to the tariff revenue losses, they are underestimated. Indeed they do not consider the associated increase in productivity, variety and quality of imported inputs. Hence, full tariff liberalization would not only reduce losses but also generate a gain in welfare above the estimated figures, with the only cost of a shift away from tariffs in the composition of tax revenues. Figures Tariff Revenue Losses and Welfare Gains Figure 31. Tariff Revenue Losses (US$ million) Figure 32. Welfare gains (US$ million) 3,500 120 111 2,885 2,820 2,815 2,811 3,000 100 90 2,360 2,500 2,222 80 2,000 60 1,500 1,000 40 17 500 20 8 7 0 0 Baseline China India Malaysia SAFTA All China India Malaysia SAFTA All Source: COMTRADE data and authors’ simulations using SMART 42 Pakistan: Reinvigorating the Trade Agenda Table 13 Changes in Tariff Revenue and Welfare as a Result of Proposed PTAs Tariff revenue change Tariff revenue change Welfare gains New weighted (US$ million) (%) (US$ million) tariff China -525 -18.2 90.3 8.3 Malaysia -70 -2.4 16.7 13.7 India -65 -2.2 7.3 13.7 SAFTA -74 -2.6 8.2 13.6 All current PTAs -663 -23.0 111.3 7.6 Source: COMTRADE data and authors’ simulations using SMART 66. The preferential trade agreement with China is the most ambitious agreement in scope signed by Pakistan. This is so, because granting market access to this trade partner would cause the biggest economic and welfare impact among the recently signed preferential trade agreements. This agreement goes beyond market access in goods. Table 14 compares the different structure of provisions regulating Pakistan’s PTA with SAFTA (a shallow agreement) and the agreement recently signed with China (the deepest of all of Pakistan’s PTAs). While the bilateral agreement with SAFTA only deals with tariffs for agricultural and industrial goods, the agreement between Pakistan and China covers non-tariff measures, services and regulatory matters, including, among others, provisions regulating the mutual recognition of services suppliers and other technical barriers to trade, international property rights and the establishment of Chinese investment zones in Pakistan. The preferential trade agreement between Pakistan and Malaysia is also deep in nature, and includes investment provisions in the fields of computer and I.T. related services, Islamic Banking, Islamic Insurance and mutual recognition arrangements for accreditation of education institution and academic programs. All remaining PTAs are shallow in nature, i.e. they primarily cover issues of market access. Table 14 Provisions in Pakistan’s PTAs with SAFTA partners and China SAFTA Pakistan – China FTA Legally Legally Provision Provision enforceable? enforceable? FTA Industrial 1 yes 1 yes FTA Agriculture 1 yes 1 yes Customs 1 1 yes Export Taxes 0 1 yes SPS 0 1 yes TBT 1 1 yes AD 0 1 CVM 0 1 State Aid 0 1 Source: WTO (2011) 67. Deep agreements allow creating a framework of trade and investment rules enabling to address the emergence of supply chain production as a prominent mode of 21st century economic integration. Other tools have been increasingly used, globally and in Pakistan, which are complementary to PTAs in enabling this. These include bilateral investment treaties (BITs), and double taxation treaties (DTTs). Generally speaking, investment and trade are regulated by distinct treaties because they focus on different but complementary objectives (Di Mascio and Pawelyn, 2008). Trade agreements seek to increase trading opportunities and investment agreements seek to protect and promote foreign investment. 43 Pakistan: Reinvigorating the Trade Agenda BITs have been an important vehicle that capital-abundant countries have used to guarantee investment protection (Adlung and Molinuevo, 2008) and that labor-abundant counterparts have used to attract off-shored manufacturing jobs and factories (Baldwin, 2010). 68. As of 2011, according to the ICSID website, Pakistan had signed 47 such agreements (Table 15)22. These agreements establish disciplines that govern interactions between private foreign investors and host governments. In the case of Pakistan, most of the bilateral investment agreements are aimed at avoiding double taxation and at facilitating trade and investment. While Pakistan’s older bilateral investment agreements were with developed countries, the more recent agreements are increasingly South-South in nature. This parallels developments in trade phases, which from North-North developed into North-South and now increasingly extent to South-South deals. According to Bergstrand and Egger (2011), the likelihood that a country-pair establishes BITs and PTAs is higher: the larger and the more similar in income level the two countries are and the closer geographically the two countries are. Nevertheless, while the likelihood of signing PTAs is higher if the country shares a border and a language, for BITs the opposite is true. Countries that do not share a border or a language are more likely to sign investment agreements. Finally BITs tend to be more frequent between skilled labor abundant countries. Table 15 Pakistan Bilateral Investment Agreements Entry into Entry into Country Signature Country Signature force force Germany 1959 1962 Denmark 1996 Romania 1978 1978 Indonesia 1996 Sweden 1981 1981 Syria 1996 France 1983 1984 Tunisia 1996 Kuwait 1983 Belarus 1997 Korea 1988 1990 Italy 1997 2001 Netherlands 1988 1989 Mauritius 1997 China 1989 1990 Oman 1997 Uzbekistan 1992 Sri Lanka 1997 Spain 1994 1996 Australia 1998 1998 Turkmenistan 1994 Belgium-Luxembourg 1998 United Kingdom 1994 1994 Japan 1998 2002 Azerbaijan 1995 Czech Republic 1999 Bangladesh 1995 1997 Philippines 1999 Iran 1995 1998 Qatar 1999 Kyrgyz Republic 1995 Egypt 2000 Malaysia 1995 Bosnia and Herzegovina 2001 Portugal 1995 1996 Lebanon 2001 2003 Romania 1995 1996 Morocco 2001 Singapore 1995 Kazakhstan 2003 Switzerland 1995 1996 Cambodia 2004 Turkey 1995 1997 Lao 2004 United Arab Emirates 1995 Tajikistan 2004 Source: http://icsid.worldbank.org/ICSID/FrontServlet 22An alternative source of information on BITs is UNCTAD: http://www.unctad.org/Templates/WebFlyer.asp?intItemID=3150&lang=1 44 Pakistan: Reinvigorating the Trade Agenda 69. PTAs signed by Pakistan to date have generated trade growth but the gains are of secondary importance. Signing PTAs has neither boosted Pakistan’s exports nor created trade diversion effects on Pakistan’s imports from third countries. By contrast, PTAs between third countries have had noticeable trade diversion effects for Pakistan’s exports. Pursell et al. (2011) notes that the complex system of regulatory duties set in place to protect domestic producers of import competing goods may indeed be a source of trade diversion costs. Since countries having signed PTAs with Pakistan are exempt from regulatory duties foreseen in Pakistan’s tariff system, this may ultimately lead to protect its own high cost producers and high cost producers in the preferential supplying country, leading to trade diversion costs for Pakistan itself. To test for trade effects of PTAs we use a standard tool of analysis: the gravity model. Our specification reflects indications from numerous empirical and theoretical contributions that have allowed over the years to generate consensus around the key features that a well specified gravity model should have. In particular, the chosen specification controls for all unobservable exporter and importer characteristics and for any idiosyncratic factor related to the year, the bilateral relationship, and the individual nation. It also corrects for non-linearity, potential selection bias, heterogeneity and zeroes in the data. 70. Interestingly Pakistan appears to have a great untapped potential for expanding its trade relations, both on export and import side. The gravity estimations suggest that while the elasticity of Pakistan’s exports to tariff changes is in line with the world average, Pakistan’s trade seems to be more elastic to trade frictions than the world average. The elasticity of Pakistan’s trade to distance—a common proxy for trade costs—is about 30% higher than the world average. Sharing a border (contiguity) usually facilitates trade. In the case of Pakistan, sharing a border appears to represent an additional substantial burden to trade. Not only sharing a border appears to be a trade cost for Pakistan’s exports, but the overall effect is so important that it doubles the cost of distance. In other words, exporting to two equidistant locations, one in a neighboring country and one in a non-contiguous market, implies twice the cost if exports go to neighboring country. These insights are confirmed if we compare actual vs. predicted flows. Exports to India are 40% below its predicted potential (Figure 33). Similarly, Pakistan is the key missing market for India, once more confirming that both countries would gain greatly by a normalization of their trade relations. Figures Predicted vs. Actual Exports: Comparison of Pakistan & India Figure 33: Pakistan Figure 34: India Pakistan: Predicted vs Actual Exports 30 20 10 0 0 5 10 15 20 25 Log of Actual Exports, 608 Source: Authors’ Calculation based on COMTRADE data 45 Pakistan: Reinvigorating the Trade Agenda 71. Looking ahead, more significant trade creation effects may originate from agreements with China and Malaysia, due to the deeper nature of these agreements and their coverage beyond market access issues. Our analysis (for more details, see World Bank 2012) has shown that in the period 2005-10 and for a sample of 181 countries worldwide, deep PTAs had positive trade creation effects. By contrast, shallower PTAs had a statistically not significant or weakly significant overall effect. 72. Given this background, the way forward in terms of preferential trade agreements should be to scale up efforts to benefit from the geographical advantage of being in the highest growth region of the world. Pakistan is situated in a very dynamic region. The many domestic impediments and trade barriers have thus far prevented the country from benefitting fully from the exceptional growth in the region. Three main areas need to be highlighted for policy intervention. 73. Accelerating the implementation of deep preferential trade agreements and signing of other deep agreements. Significant trade creation effects are likely to originate from agreements with China and Malaysia, due to the deeper nature of these agreements and their coverage beyond market access issues. By contrast regional PTAs that are limited to market access are unlikely to bring benefits. This is the case because of the fast paced and deep nature of integration agreements in Asia. 74. Opening up the border with India and facilitating deep forms of trade integration, so to benefit from the high economic growth rates of the neighbor. Exports to India are 40% below the predicted (in current values this is equivalent to about US$ 140 million in static terms). Similarly, Pakistan is the key missing market for India, once more confirming that both countries would gain greatly by a normalization of their trade relations. These however should aim at deep forms of trade integration and not be limited to market access. A recent study (see De, Raihan and Ghani, 2012) shows that the welfare gains resulting from granting MFN access to India would increase dramatically if supported by improved connectivity and trade facilitation measures. According to these authors, granting MFN status to India plus enhancement of bilateral trade facilitation would lead to a staggering 202 percent increase in exports against 0.4 percent under the MFN only scenario. Pakistan’s total exports would rise by 1.82 percent under the former scenario compared to only 0.17 percent under the later scenario. Similarly, the bilateral trade facilitation together with the MFN would lead to higher dynamic gains. In 2012, under the MFN plus trade facilitation scenario, the real GDP would be 0.21 percent higher than the business as usual scenario whereas by 2030 it would be 0.60 percent higher. Without the trade facilitation component, however, the rise in real GDP would be 0.06 and 0.18 percentage points lower, respectively. Similar patterns are observed in the case of Pakistan’s total imports, exports and aggregate consumption. 75. De et al. point to a number of non-tariff barriers to trade between Pakistan and India that, if not addressed, will continue to hinder bilateral trade in spite of MFN access. Border regions are particularly challenged. In Pakistan, per capita income is lower than average in the border areas of rural Sindh, as are the border states of Rajasthan and Jammu & Kashmir in India. While recently signed agreements will address some of these issues, such as mutual recognition and streamlined visas, a number of other barriers remain to be addressed (see Table 16), particularly restrictions on major trade routes and at the borders, as well as cumbersome payment systems that cause delays and increased transaction costs. 46 Pakistan: Reinvigorating the Trade Agenda Table 16 Major Non-Tariff Impediments to Pakistan-India Trade Category Major Impediments Stringent visa regimes, Trade distorting subsidies, Overland transportation Non-tariff barriers limitation, Air travel restriction, Sea transportation restriction , Transit restriction, Port of call restriction Cumbersome payment systems, Restrictive official foreign exchange allocation, Regulations concerning terms of trade for import payments, Non-acceptance of Finance measures letter of credit, High commission of foreign banks offering letter of credit, Lack of bank branches License with no specific ex-ante criteria, License for selected importers, Sanitary Technical barriers to and phytosanitary measures, Marking requirements, Labeling requirements, trade and quality Testing, inspection and quarantine requirements, Pre-shipment control measures inspection/certificate acquisition Source: De et al. 76. Finally, for enhancing domestic connectivity, and thereby export opportunities, one of the overriding objectives should also be to develop the country as a regional hub for trade and logistics which serves Afghanistan and Central Asia. In this regard, developing the logistics links to Pakistan's neighbors should be the priority. Not only it would reinforce Pakistan's role as the main transit country for landlocked Afghanistan, but would also allow Pakistan to connect better to the rest of the region. Ratifying key international conventions and other regional trade facilitation instruments recently negotiated or under negotiation will be the key. This point is further discussed in detail in the context of trade facilitation and logistics. 47 Pakistan: Reinvigorating the Trade Agenda Trade Facilitation and Logistics 77. A good logistics system can play a decisive role for Pakistan to attain three objectives that have been identified as critical in this paper. The first one is to encourage current and new exporters to have access to foreign markets by lowering logistics costs. The second objective is to specifically bring new firms and new products to the markets, as diversification towards new firms and products remains a challenge for Pakistan. The third objective, closely linked to the previous one, is to foster integration of domestic markets. By helping reducing trade costs, upgrading the quality of services, improving connectivity with foreign markets and moving upstream in the supply chain, a sound trade facilitation and logistics system is of great importance for the country. 78. Pakistan has a core transport and logistics system and environment that support trade in various types of commodities of varying cost, time and reliability sensitivities, ranging from agricultural products to electrical and other manufactured products. However, based on the 2010 Logistics Performance Index (LPI) data, Pakistan’s overall logistics performance was below that of India and Bangladesh as well as below the global average of all countries (Figure 35). Generally, the South Asia neighborhood as a region lagged behind most regions, only surpassing Sub-Saharan Africa. South Asia fared in line with its per capita income, performing marginally better than the low income country group (Figure 35). In the main, the LPI data suggest that weaknesses in Pakistan’s performance are mainly in the areas of infrastructure, customs and logistics competence. Figure 35 Pakistan Logistics Performance Pakistan compared to the Logistics Performance of Pakistan logistics performance compared to the India global average Source: World Bank, Logistics Performance Index 2010 48 Pakistan: Reinvigorating the Trade Agenda Figure 36 Logistics Performance South Asia Logistics Performance Pakistan performance relative to South Asia Source: World Bank, Logistics Performance Index 2010 79. Connectivity in Pakistan features great geographical disparities. Outside the system connecting specialized markets in Karachi, Lahore, Islamabad, Sialkot and Peshawar, which is relatively well functioning, trade logistics services are generally of a poor quality, especially when compared to other countries of a similar level of income. The national logistics hub in Pakistan is located in the south of the country. As one moves north and west of the country and away from the core national corridor, access to high quality infrastructure and services reduces. In relatively lagging regions, plans to grow exports in niche products such as fresh produce for instance, would require measures that better connect remote areas to main export gateway. Given the nature of some of the prospective export commodities, which have high volume to weight ratios and the distances involved, it would appear the key to improving domestic connectivity lies in developing multi-modal systems based on rail and consolidation of trade volumes. The Main Components of the Logistics System—Current Situation 80. Railways—Presently Pakistan Railways plays a relatively limited role in the country’s freight transport despite having a network that extends over much of the country. Pakistan Railways has almost eight thousand kilometers of broad gauge routes linking all the major centers, and also interconnects to the Iranian network. Although the railway market share continues to decline, there has been a leveling off in the volume of freight traffic. 81. Despite the downward trend in traffic carried by rail, there is evidence that Pakistan railways has been increasing the average length of haul—meaning it has been moving towards distances where rail has a comparative advantage over road transport (World Bank, 2012). The pricing policy of Railways is designed to make it attractive to different types of commodities, officially depending on their ability to bear the cost since Pakistan Railway rates are based on the principle of ‘what each type of traffic can bear’. Simply, railway tariffs are based not so much on cost of providing the service but on broader objectives to allow trade in various commodities. Different types of cargo are charged different amounts even if they may weigh the same and are transported over equivalent distances. However, the downside of this approach is that the railways are perennially making losses and have to be 49 Pakistan: Reinvigorating the Trade Agenda subsidized by the government. As a result, the availability of rolling stock and general maintenance of the system has been deteriorating in recent years. 82. Roads and Trucking—Road transport is the dominant mode of overland transport in Pakistan, accounting for more than 80% of cargo volumes shipped. Despite the poor condition of parts of the road network, Pakistan has the lowest road freight transport rates in the world (Figure 37). This is partly due to the structure of the industry. The majority of trucks on the roads are operated informally as small fleets. In 2007 there were as many as 209,000 registered trucks, more than two thirds of them rigid trucks. Generally, the vehicles are old and highly fuel inefficient. As a result, the low rates are obtained by maintaining old fleets and by overloading the vehicles. Competition and high operating costs significantly reduce margins for the industry and hamper efforts to modernize the fleet. Figure 37 Transport Prices in Pakistan and other Countries 8 Average road transport price, US 7 6 7 cents/tonne-km 5 6 4 5 5 3 4 3.5 2 1 2 0 RSA-Zambia Togo-Burkina China France Brazil Pakistan USA Faso Country/Corridor Source: Authors’ elaboration with data from Teravaninthorn and Raballand (2008) 83. However, in recent years there have been some changes in the sector with the emergence of an increasing number of formal companies running large fleets, in excess of 50 vehicles. The largest fleet operator is the state controlled National Logistics Cell (NLC) which owns more than 1500 trucks and previously had a monopoly on Afghanistan transit traffic. In addition to NLC, other large fleet operators are petroleum companies such as Shell and the dry port operators especially Sialkot and Faisalabad. The Sialkot Dry Port Trust operates more than 70 trucks. These larger operators all provide modern services including tracking of vehicles and shipments. 84. In the recent past there has also been a change in the organization of the independent truckers. Previously, they were contracted for individual shipments with rates negotiated directly with the shippers or their brokers. Backhaul shipments were obtained from large number of brokers located in major cities. More recently, some of the larger road haulage companies have developed relationships with independent truck operators in which they contract for their service on a regular basis as a supplement to their own fleet. These transport companies introduce certain performance criteria and in terms of quality and reliability of the vehicles, the skills of the drivers and the route, which in return provide a continuing source of business. 85. In 2007 the Government of Pakistan drafted a trucking policy which is now being implemented. The policy is designed to complement the NCTIP and serve as the basis for 50 Pakistan: Reinvigorating the Trade Agenda modernizing the trucking sector. It contains several progressive provisions which, if realized would contribute to a higher quality of service from the trucking sector. Some of the key provisions of the policy are:  Enhancing access to finance: The policy seeks to make it possible for trucking concerns to access financing from financial institutions. This is being pursued by designating the sector as an ‘industry’ which under Pakistani law then enables trucking concerns to borrow from banks at less than commercial rates. Limited access to finance has often been identified as an impediment to modernizing and replacing the current fleet of predominantly old trucks. The policy provides incentives for this to happen.  Motor vehicle registration: The registration of motor vehicles in Pakistan is handled by the provinces. As a result, it is usually difficult to obtain current information on the vehicle fleet. The policy proposes to establish a central depository for motor vehicles for nationwide maintenance of data.  Vehicle worthiness testing and certification: The policy provides regular tests of fitness of vehicles and their road worthiness certification. This will be accompanied by training and capacity building of staff involved in the tests. 86. While some of these initiatives may raise trucking prices, they would also raise the quality, especially reliability of logistics services. Reliability is a critical performance attribute for logistics services, particularly of sophisticated supply chains. In Pakistan some market segmentation should therefore be expected, with more sophisticated chains paying more for greater reliability and quality of logistics service while other chains continue to rely on low cost but also lower quality services. 87. Air transport—Air transport is the second most important mode of transport for international shipments in Pakistan, but is mostly covered by international carriers. Of the almost thirty air carriers serving the country, only three belong locally from Pakistan. Five international carriers provide 40 weekly scheduled all-cargo services in and out of Pakistan. Karachi, Lahore and Sialkot are the only cities with access to scheduled dedicated cargo services. There are many more exports than imports, most of them passing through the main airports of Islamabad, Lahore and Karachi. Using the total turnaround cost methodology, Pakistani airports strike as one of the most expensive in terms of landing charges. 88. Pakistan’s role in the worldwide airline network structure is far from central, and might present only limited opportunities for transit traffic. Pakistan sits astride main air traffic routes connecting Europe and the Middle East with Southeast Asia and Australasia (Figure 38). However, given the preeminence of other hubs like Dubai and Bangkok that compete in the same markets, it seems unlikely for Pakistan to become a credible connecting hub in the short run, especially for directional long haul connections between Europe and Asia. On the other hand, its geographical location might favor short and medium haul connections in South and Central Asia and the Middle East. 51 Pakistan: Reinvigorating the Trade Agenda Figure 38 World Air Services Network Structure (2011) Source: World Bank Staff Estimates based on DIIO Data 89. Ports—More than 90% of Pakistan’s international trade is transported by sea. Most of the trade traffic is through Karachi where there are two container terminals, the Karachi International Container Terminal (KICT) and Port Qasim. Of the two, KICT handles the most volumes. All major shipping lines provide services to either Karachi or Port Qasim. However, connections to Europe are indirect, with hubbing mostly through Salalah in Oman. 90. It is increasingly apparent that across the developing world cargo stays long periods of time in ports. In Pakistan, port authorities allow for up to 5 days of free storage time in the premises of the port. Recent statistics suggest that, on average, the cargo dwell time in Karachi is 7 days. This average is much higher than the 2-3 days prevalent in most efficient ports in the world. Beyond assessing average duration of stationing in the port, it would be valuable to understand the distribution of the actual dwell terms to get a sense of the reliability of the system. However, the necessary data are not readily available. 91. Customs—Pakistan Customs has been implementing a modernization program for the past several years. Reforms already introduced include the use of the Harmonized Code, creation of a Single Administrative Document (Goods Declaration or GD), and electronic submission and processing of declarations at the seaports and the major land borders. These improvements have allowed a significant reduction in clearance times and an increase in the collection of duties and taxes. However, at present there are two different IT systems for submitting the goods declaration, Pakistan Customs Computerized System (PaCCS) and Web Based One Customs. The rate of physical inspections has been reduced but remains higher than would apply with an effective system of risk management. More important, the decision regarding inspections is often based on informal payments. The typical clearance times for cargoes with proper documentation are two days for imports and one for exports. The procedures for handling transit cargoes have been simplified but the time required for issuing transit permits and the procedures regarding inspection of transit cargo are neither transparent nor efficient. Pakistan Customs has a Risk Management Unit which has developed some risk profiles and is supposed to select the level of inspection applied for individual shipments. However, both PACCS and One Customs perform limited functions 52 Pakistan: Reinvigorating the Trade Agenda and neither has a risk management module (PACCS system uses random selection of containers). Still, the random selection can be overridden by an officer thereby creating significant opportunities for informal payments. 92. Presently, Pakistan does not have a formal system of Authorized Economic Operators, although informal recognition is given to preferred traders who receive expedited clearance. Modern X-ray scanning has been introduced but only at QICT and only for exports to the US as part of the C-TPAT program. 53 Pakistan: Reinvigorating the Trade Agenda Policy Recommendations 93. In this paper, a thorough analysis of Pakistan’s recent trade performance was conducted, with focus on its trade policy and trade costs. In addition to assessing different dimensions of trade performance—growth and orientation, diversification and sophistication —an in-depth analysis of export dynamics in the 2001-11 period, using firm-level data, was performed. It also includes an analysis of Pakistan’s trade policy including tariffs, effective protection and trade restrictiveness estimates, as well as an assessment of the role of preferential trade agreements in the context of regional integration. Finally, the main characteristics of trade facilitation and logistics were analyzed, covering the capacity, performance, quality of services and degree of integration of the logistics system. 94. Pakistan’s recent trade performance stagnated in recent years, as indicated by a moderate decrease in its trade-to-GDP ratio over the last decade, in stark contrast to peer countries that have leapfrogged with high growth rates in the last two decades. The firm-level analysis shows that, despite positive signs of exports dynamism, with substantial churning of firms, Pakistan has not been very successful in terms of helping firms break into sectors and products to be able to deliver higher export growth. In addition, the performance has become poorer after 2004, well before the global financial crisis, and has still not recovered. 95. Amidst very uncertain and volatile conditions for global demand, Pakistan can maximize its chances of increasing domestic competitiveness and hence exports by creating the conditions that foster firm dynamism, experimentation and quality upgrading of their products and by eliminating domestic impediments and barriers to trade. 96. Diversification is critical to enhance export performance. Diversification, be it in terms of firms, products or destinations, is positively correlated to export growth, with an effect that is quantitatively sizeable and statistically significant. To achieve the objectives of more dynamic and diversified export base, Pakistan needs to take substantial measures to develop a competitive business environment and eliminate barriers to export. This will have a clear impact in terms of lowering the entry costs for exporters and firms to venture in new products and sectors by creating a level playing field for firms, promoting intra-industry competition, and creating a more transparent business environment. It is also likely to attract foreign direct investment that is crucial for both acquiring technological know-how and much needed capital. 97. In particular, the following policy actions analyzed in this paper are likely to stimulate diversification and increase of exports: 98. Promoting intra-industry competition and enabling environment conducive for exports requires first and foremost the abolition of highly complex trade regulations, further refining the tariff structure, and the elimination of remaining regulatory duties. Pakistan’s complex import tariff and tax regime creates uncertainty, reduces transparency, and reduces firm dynamism. First, the selective protection and concessions on inputs is a source of economic inefficiency because valuable resources are being diverted to less productive sectors that are protected from otherwise low international prices, and away from Pakistan’s true comparative advantage industries. Second, administering the concessions which operates as a de facto import licensing regime involves transaction costs to the firms that hope to benefit the government bodies that administer it and the EDB. The transaction costs 54 Pakistan: Reinvigorating the Trade Agenda incurred by firms reduce the value to them of participating in the system. Because the information and transaction costs will tend to be lower relative to their output, the system very probably benefits larger established firms and discriminates against small and medium enterprises. Meanwhile ‘customized’ domestic protection entails important costs and risks, including the risk that protected firms have a disincentive to become efficient and often rely on continued protection to remain in business, rendering a loss to consumers and reducing economic efficiency. 99. High duty rates and a regulatory duty imposed only on imports was and remains an extremely inefficient way of cutting imports of ‘luxury’ products and is also a counterproductive method of generating government revenue. The very existence of schemes granting protection to selected sectors and products increases the scope for smuggling and under-invoicing. The cascading principle, evoked by many governmental documents, makes this likely especially for final goods. Third, domestic market protection through tariff protection generates an anti-export bias. Since Pakistan cannot influence world market prices, exporting firms do not have any benefit from domestic tariff policies. By contrast, producers for the domestic market enjoy a price advantage over foreign competitors. This leads to market distortions whereby the incentive for producers is to focus on serving the domestic market rather than exports. We therefore recommend the following phased rationalization and liberalization of the import tax regime:  In the short-term, a simplification of the tariff regime toward a three-band structure. This includes an immediate reduction of the top level to 25% and the move to a transparent, three-band structure (25%, 10%, 0%), with the aim of moving over the medium term (within three to five years) to a uniform 10% tariff rate. This latter stage should be accompanied by the removal of all regulatory duties and elimination for concessions outside of trade agreements and free zones. This would have important positive effects in the long run, with the transitory effects of adjustment in terms of output and employment limited to some sectors and a relatively small impact on tariff revenues. An exercise conducted for this paper shows that a reform that puts a ceiling for tariffs at 25 percent but retains all other features of the current tariff regime, including exemptions, would cause a 3.7 percent reduction in tariff revenues, but only 1.8 percent on total import tax revenues, and a 0.6 percent decline in import prices. As a result of reform, higher value-added export-oriented sectors such as Chemicals and Chemical Products, and Machinery and Equipment are likely to benefit, even in the short run. Meanwhile, many lower value-added, import- competing sectors—such as Processed Foods, Rubber and Plastic Goods, and Other Manufacturing–would adjust production downward in the short run, as would many service sectors that temporarily lose out as consumption adjusts to take advantage of lower prices, particularly in traded goods.  In the medium-term, move to uniform tariff rate that is applied across-the-board. Moving to a uniform 10 percent tariff and eliminating all exemptions, as recommended over the medium-term, would imply a 79.2 percent increase in tariff revenues and a 36.4 percent increase in total import tax revenue, while increasing moderately the price of imports (by 1.4 percent). In the medium run, it is expected that the full effects of reform will start to emerge, as economic agents are better able to respond to the new incentive environment, resulting in the shift of resources from 55 Pakistan: Reinvigorating the Trade Agenda less productive to more productive economic activities and, more importantly, Pakistan is able to attract new capital to its more competitive sectors. 100. This paper also brought supporting evidence that Pakistan could achieve great advances in trade performance by pursuing greater regional integration through deep agreements that look beyond market access and improvement of the country’s over all logistics system. 101. In regards to market access, accelerating the implementation of deep preferential trade agreements and signing of other agreements in that vein are key initiatives. Significant trade creation effects are likely to originate from the agreements with China and Malaysia, due to the deeper nature of these agreements and their coverage beyond market access issues. By contrast regional PTAs that are limited to market access are likely not to bring benefits. This is the case because of the fast paced and deep nature of integration agreements in Asia. 102. Pakistan has to scale up efforts to benefit from the geographical advantage of being in the highest growth region of the world. Pakistan is situated in a very dynamic region. The many domestic impediments and trade barriers have thus far prevented the country from benefitting fully from the exceptional growth in the region. Opening up the border with India and facilitating deep forms of trade integration is necessary to benefit from the high growth rates of the neighbor. Exports to India are 40% below their predicted potential. Similarly, Pakistan is the key missing market for India, once more confirming that both countries would gain greatly by a normalization of their trade relations. These however should aim at deep forms of trade integration and not be limited to market access. Given that normalized trade is only now being established, the focus should be expediting measures to facilitate trade, building on the agreements that have recently been signed on mutual recognition and visas, by improving infrastructure, institutions, services, policies, procedures, and market- oriented regulatory systems. Some specific measures that should be considered for further assessment, potentially on a joint basis with India include the following:  Removing impediments at the border and along trade routes. A number of key impediments remain that raise the time and cost of bilateral trade. To facilitate the movement of goods and people across the border, specific measures should be considered for further analysis include the potential for transit agreements to link Pakistan to Bangladesh and Nepal and India to Afghanistan (similar to that which was negotiated between Pakistan and Afghanistan) and associated infrastructure to support new trade routes, a one-stop border post at the Wagah-Attari border, inland container depots on either side of the border, and online payment schemes.  Further integrating border communities. Given populations along the border regions are among the poorest, the particular challenges of integrating border communities, more localized initiatives to target border populations may include border bazaars and other measures to encourage cross-border trade—something that is already happening to some degree between India and Bangladesh and can be replicated on the Pakistan-India border. 56 Pakistan: Reinvigorating the Trade Agenda 103. Finally, weaknesses in the logistics system and great geographical disparities also restrain Pakistan’s competitiveness and exports. By helping reducing trade c osts, upgrading the quality of services, improving connectivity with foreign markets and moving upstream in the supply chain, a sound trade facilitation and logistics system is of great importance for the country. Pakistan has a core transport and logistics system and environment that support trade in various types of commodities of varying cost, time and reliability sensitivities, ranging from agricultural products to electrical and other manufactured products. However, based on the 2010 Logistics Performance Index data, Pakistan’s overall logistics performance was below that of India and Bangladesh as well as below the global average of all countries. Pakistan’s performance main weaknesses are in the areas of infrastructure, cus toms and logistics competence. Moreover, outside the core north-south corridor, trade logistics services are generally of a poor quality. In order to upgrade the quality and reliability of the logistics services, five areas of action are suggested:  Overall, improving port performance is critical to lowering logistics costs. The stated average is much higher than the 2-3 days that is prevalent in most efficient ports in the world. If trade in time sensitive products is to grow, then this time will have to be reduced. Improvement of port operations and customs procedures is key to reach this objective. In particular, it would be beneficial from a trade logistics perspective if a unified and comprehensive customs system was introduced and operational across the country. This would enable the private sector to develop and invest in appropriate interfaces to the system. Presently this is difficult and inefficient as there are duplications. Critical but still lacking is the introduction of the following: an effective risk management system, a formal Authorized Economic Operators regime, and an expedited regime for transit shipments. These will allow more cargo to be cleared inland thereby reducing bottlenecks at the seaports and land borders and increasing the effectiveness of the clearance procedures and allowing for greater cooperation between shippers and customs officials. As Pakistan is developing trade links to the Central Asian Republics in the north, these systems can be strategic.  Further improving regulation of the trucking industry is another crucial step. There are already a number of initiatives underway to address the problems of the road haulage sector. Overall, the trucking policy seems comprehensive and well founded. It is at par with policies in other middle income countries and when fully implemented should have a significant effect on the quality of trucking services in Pakistan. Given the presence of numerous small players in the logistics market in Pakistan, a solid legal framework for the responsibilities and liabilities, including those of truckers, is paramount. Presently, this is a significant gap in the legal framework in the country especially where it concerns liability for goods as they move along the chain. Truck operators for instance only carry third-party liability. This is one of the reasons why the quality of freight forwarding is critical. Freight forwarders, most of who also serve as clearing agents are the most important intermediary in trade logistics in any country. It is imperative that capacity building programs are implemented for this sector.  For enhancing domestic connectivity, and thereby export opportunities, one of the overriding objectives of the Government of Pakistan is to develop the country as a regional hub for logistics. In this regard, developing the land transport links to 57 Pakistan: Reinvigorating the Trade Agenda Pakistan’s neighbors, including India, should be the priority, since in doing so this allows developing a multi-modal system which connects the more remote north and north-west areas of the country to the well established national logistics hub located in the south. Not only it would reinforce Pakistan’s role as the main transit country for landlocked Afghanistan, but would also allow Pakistan to connect better to the rest of the region. Exploiting this potential requires first of all changing the approach in the country, from viewing herself as a transit country for landlocked Afghanistan to relying on transit rights through the same country for access to and from the Central Asian Republics in particular. It further requires that Pakistan takes additional steps, especially with respect to ratifying key international conventions and other regional trade facilitation instruments. 23 Pakistan has already been working for some time on negotiating transport agreements with several of its neighbors to the north and west. The highest priority of these agreements has been the one with Afghanistan, whose revised version entered into force in mid-2011. Others are being negotiated with the Central Asian Republics.  Addressing domestic connectivity disparities to create trade opportunities especially in fresh produces requires a holistic approach. Exploring dry ports and railway expansion within a logistics cluster based approach can lead to the expansion of exports of new products, specially fruits and vegetables. Small scale production, spread over large areas increases logistics costs as there are limited opportunities to exploit economies of scale. There is therefore a need for strategically located infrastructure for consolidation, grading, processing, packing, cold stores to facilitate such trade by reducing cost. Secondly, these products though time sensitive, have large volumes and low values. This makes the more readily available mode of transport, which is air, unsuited to expanding their trade. Air transport is suited to high value products that are time sensitive. The challenge is how to develop an efficient land transport system that can connect producers in the northern part of Pakistan to the export gateways in the south so as to minimize time and cost. Prospects for this in the Pakistani context lie in efficient road and rail services and strategically located cargo consolidation facilities. The Sialkot Dry Port experience clearly suggests this can be done with significant impacts on efficiency. The Sialkot area has three major industrial complexes in Pakistan, namely sports goods, surgical instruments and tannery industries. A common logistics system was established in 1984 designed around a dry port. The dry port facility was built by 52 reputed exporters to enable them to build an efficient logistics system with Customs Clearance services on site and access to air, road and rail transport services. The services are of high quality in terms of transit time, safety and reliability. In fact Sialkot is one of the most sophisticated and efficient logistics operations in Pakistan. The experience demonstrates the potential of mutually beneficial collaboration between exporters at the same location. 23In 1992 UNESACP identified several international legal instruments that South Asia countries were to ratify to improve trade facilitation: Road Traffic 1968; Road Signs and Signals 1968; Customs Convention on the International Transport of Goods under Cover of TIR Carnets (TIR Convention) 1975; Customs Convention on the Temporary Importation of Commercial Road Vehicles 1956; Customs Convention on Containers 1972; International Convention on the Harmonization of Frontier Controls of Goods 1982; Convention on the Contract for the International Carriage of Goods by Road (CMR) 1956; and Maritime Trade and Transit Facilitation Convention 1965. 58 Pakistan: Reinvigorating the Trade Agenda  Finally, improving coordination is essential. The majority of Pakistan’s logistics providers are small-scale enterprises offering a limited range of services and competing on cost with relatively little consideration for reliability or value. This has two significant consequences: firstly, services are generally of a poor quality and secondly, there is a high level of fragmentation in services, meaning there are numerous parties involved in the supply chain. The fragmentation of services is at the policy, institutional and operational levels. The coordination of trade facilitation and logistics in Pakistan is not easy, mainly due to institutional fragmentation. Each mode of transport falls under a different ministry: Roads are under the Ministry of Communications, while there is also a Ministry of Railways, Ministry of Ports and Shipping and Ministry of Defense which has a say on civil aviation matters. This multiplicity of public agencies makes it a challenge to coordinate actions and engender an integrated core logistics system. However, any coordinating body would need sufficient authority to see through critical reforms, ideally with very high level support in government. 59 Pakistan: Reinvigorating the Trade Agenda References Adlung, R. and M. Molinuevo, 2008. “Bilaterlism in Services Trade. Is there Fire behind the (BIT) Smoke?� Journal of International Economic Law. Vol. 11(2). pp:365-409. Anderson, J.E. and J.P.Neary. 1994. “Measuring the Restrictiveness of Trade Policy�. World Bank Economic Review. Vol. 8(2). pp:151-169. Anderson, J.E. and J Peter Neary. 1996. “A New Approach to Evaluating Trade Policy�. Review of Economic Studies. Vol. 63(1). pp:107-125. Anderson, J.E. and J Peter Neary. 2003. “The Mercantilist index of trade policy�. International Economic Review. Vol. 44. pp:627-649. Antràs, P and R Staiger. 2008."Offshoring and the Role of Trade Agreements". CEPR Discussion Papers 6966. C.E.P.R. Discussion Papers. Arkolakis, C. & M.A. Muendler, 2010. “The Extensive Margin of Exporting Products: A Firm- level Analysis�. NBER Working Papers 16641. National Bureau of Economic Research, Inc Baldwin R. 2011. “21st century regionalism: Filling the gap between 21st century trade and 20th century trade governance�. CEPR Policy Insight No. 56. CPER. Bergstrand, J. H. and P. Egger. 2011."What Determines BITs?". CESifo Working Paper Series 3514. CESifo Group Munich. Bricongne, J. C., Fontagne, L., Gaulier, G., Taglioni, D. & Vicard V. 2011.“Firms and the Global Crisis: French Exports in the Turmoil�. Forthcoming in Journal of International Economics. Carpenter, T and A Lendle. 2010. “How Preferential is World Trade?�. CTEI Working Paper 2010-32. Geneva: The Graduate Institute Centre for Trade and Economic Integration. Carrere C. 2006. “Revisiting the effects of regional trade agreements on trade flows with proper specification of the gravity model". European Economic Review. Vol. 50(2). pp:223- 247. Chaffour J.P., J.C. Maur. 2011. “Preferential Trade Agreement Policies for Development: A Handbook�. The World Bank: Washington DC. Damuri, Y.R. 2009. “How preferential are preferential trade agreements? Analysis of product exclusions in PTAs�. Swiss National Centre of Competence in Research, Working Paper 2009/30. Davis, S. and J.C. Haltiwanger. 1992. “Gross Job Creation, Gross Job Destruction and Employment Reallocation�. Quarterly Journal of Economics. Vol.107(3). pp: 819-63. De, P., S. Raihan and E. Ghani. 2012. “What does MFN mean for India and Pakistan? Is MFN a Panacea?� World Bank Mimeo Di Mascio N and J Pawelyn. 2008. “Non-discrimination in trade and investment treaties: Worlds apart of Two Sides of the Same Coin?� American Journal of International Law. Vol. 102(1). pp: 48-89. Dollar, D. and A. Kraay. 2004.; 60 Pakistan: Reinvigorating the Trade Agenda Dollar, D. and A. Kraay. 2002.; Fernandez, R and J Portes. 1998."Returns to Regionalism: An Analysis of Nontraditional Gains from Regional Trade Agreements". World Bank Economic Review. Vol. 12(2). pp:197-220. EIU. 2011. “Pakistan Country Commerce Report 2011�. New York: The Economist Intelligence Unit. Felipe, J. 2010. “Inclusive Growth, Full Employment, and Structural Change: Implications and .Policies for Developing Asia�. Asian Development Bank. Freund C. and E. Ornelas. 2010. "Regional Trade Agreements". Annual Review of Economics Vol. 2(1). pp:139-166, 09. Fugazza, M. and A. Nicita. 2010. “The Value of Preferential Market Access�. UNCTAD Blue series on Policy Issues in International Trade and Commodities. Hausmann, R., Hwang, J., and Rodrik, D. 2006. “What you exports matters�. NBER Working Papers 11905. National Bureau of Economic Research, Inc. Helpman, E., Melitz, M., and Y. Rubinstein. 2008. “Estimating Trade Flows: Trading Partners and Trading Volumes.� Quarterly Journal of Economics, Vol. CXXIII(2). Horn, H., P. Mavroidis and A. Sapir. 2010. "Beyond the WTO? An anatomy of EU and US preferential trade agreements�. The World Economy. Vol.33(11). pp::1565-1588. Hummels, D. and P. Klenow. 2005. “The Variety and Quality of a Nation’s Exports.� American Economic Review, Vol. 95(3). Kee, H.L., A. Nicita and M. Olarreaga. 2008. “Import Demand Elasticities and Trade Distortions�. The Review of Economics and Statistics. Vol 90(4). pp:666-682. Kee, H.L., A. Nicita and M. Olarreaga. 2009. “Estimating Trade Restrictiveness Indices�. The Economic Journal. Vol. 119. pp: 172-199. Lall, S. 2000. “The technological structure and performance of developing country manufactured exports, 1985-98�. Oxford Development Studies. Vol. 28(3). pp: 337–369. Lawrence, R.Z. 1996. “Regionalism, Multilateralism and Deeper integration�. Brookings Institution. Ministry of Commerce. 2009. “Strategic Trade Policy Framework 2009-12�. Government of Pakistan. Mishra, L. and R. Anand. 2011. “Service Export Sophistication and Economic Growth�. World Bank Policy Research Paper 5606. World Bank: Washington D.C. Orefice, G and N. Rocha. 2011. “Deep integration and production networks: an empirical analysis�. WTO Working Paper. Pack, H. and K. Saggi. 2006. “The case for industrial policy: a critical survey.� World Bank Policy Research Working Paper 3839. World Bank: Washington D.C. Pitigala, N. and M. Hoppe. "The Impact of Multiple-Taxation on Competitiveness in Nigeria". Africa Policy Note Series No. 16. World Bank: Washington DC. 61 Pakistan: Reinvigorating the Trade Agenda Planning Commission. 2011. “Framework for Economic Growth�. Government of Pakistan. Pursell G., A. Kahn and S. Gulzar. 2011. “Pakistan’s Trade Policies: Future Directions�. International Growth Centre Working Paper 11/0361. London School of Economics, London: UK Santos-Silva J.M.C. and S. Tenreyro. 2006. “The Log of Gravity�. Review of Economics and Statistics. Vol.88(4). pp: 641-658. Tinbergen, J. 1962. “Shaping the World Economy�. The 20th Century Fund, New York. Viner. 1950. “The Customs Union Issue�. Carnegie Endowment for Peace. Winters, L. A., N. McCulloch, and A. McKay. 2004. "Trade Liberalization and Poverty: The Evidence So Far". The Journal of Economic Literature. Vol.42(1). pp:72-115. World Trade Organization. 2011. "World Trade Report 2011- The WTO and Preferential Trade Agreements: From Co-existence to Coherence". WTO: Geneva. World Bank. 2010. “Connecting to Compete—The Logistics Performance Index and its Indicators�. The World Bank: Washington DC. World Bank. 2012. “Trade competitiveness in Pakistan�. The World Bank: Washington DC. 62 Pakistan: Reinvigorating the Trade Agenda Annexure 1: Effective Rates of Protection NPR EPR Paddy 10.0 10.5 Wheat 10.0 10.7 Cereal grains nec 4.0 0.5 Vegetables, fruit, nuts 12.2 14.5 Oil seeds 2.7 2.3 Sugar cane, sugar beet 5.0 2.9 Plant-based fibers 0.0 -4.0 Crops nec 4.8 3.6 Bovine cattle, sheep and goats, horses 2.1 -3.3 Animal products nec 7.8 8.7 Raw milk 0.0 -5.0 Wool, silk-worm cocoons 3.3 -0.8 Forestry 6.1 5.5 Fishing 8.8 9.3 Coal 3.3 1.5 Oil 7.5 8.7 Gas 0.0 -1.6 Minerals nec 6.6 5.8 Bovine meat products 6.5 5.3 Meat products nec 21.8 33.3 Vegetable oils and fats 11.3 151.8 Dairy products 23.1 63.7 Processed rice 10.0 17.6 Sugar 10.7 17.0 Food products nec 16.4 121.5 Beverages and tobacco products 57.4 390.0 Textiles 16.4 57.1 Wearing apparel 24.1 66.7 Leather products 15.7 190.1 Wood products 15.2 19.3 Paper products, publishing 15.8 21.4 Petroleum, coal products 8.0 24.6 Chemical, rubber, plastic products 9.9 11.5 Mineral products nec 19.3 26.6 Ferrous metals 10.3 11.6 Metals nec 6.9 4.2 Metal products 17.8 47.0 Motor vehicles and parts 38.7 115.0 Transport equipment nec 14.6 17.3 Electronic equipment 9.2 10.1 Machinery and equipment nec 10.1 11.3 Manufactures nec 15.3 39.7 Electricity 0.0 -9.0 Gas manufacture, distribution 5.0 4.2 Water 0.0 -2.0 Construction 0.0 -10.0 Trade 0.0 -1.1 Transport nec 0.0 -7.9 Water transport 0.0 -2.8 Air transport 0.0 -3.1 63 Pakistan: Reinvigorating the Trade Agenda NPR EPR Communication 0.0 -2.2 Financial services nec 0.0 -3.2 Insurance 0.0 -0.6 Business services nec 0.0 -3.5 Recreational and other services 0.0 -5.2 Public Administration, Defense, Education, 0.0 -2.5 Health Dwellings 0.0 -2.2 Sources: Authors calculation using Pakistan’s Tariff Schedule (2009 -10) and GTAP8 database 64 Pakistan: Reinvigorating the Trade Agenda Annexure 2: Trade Policy Summary Matrix Trade Policy Objective Short Term (1 year) Medium Term Encourage export Simplification of the tariff regime toward a 3-slab structure (0%, In the medium-term, a move toward a uniform 10% diversification, through the 10%, 25%), continued abolition of the present system of distortive tariff regime is recommended. promotion of intra-industry regulatory duties and highly complex trade regulations. This would competition have important positive effects in the long run, with the transitory effects of adjustment in terms of output and employment limited to some sectors and a relatively small impact on tariff revenues. Pursuing greater regional Opening up the border with India and facilitating deep forms of Accelerating the implementation of deep integration through deep trade integration would be mutually beneficial. Potential measures preferential trade agreements and signing of other agreements that look beyond include transit agreements, one-stop border posts, and other agreements in that vein are key initiatives. market access measures to promote both transit and cross-border trade. Upgrade the quality and Developing the land transport links to Pakistan’s neighbors, Modernizing regulation of the trucking industry is reliability of logistics services, including India, should be the priority. Ratifying a few main important to enhancing the quality of logistics contributing to reduce trade international conventions and other regional trade facilitation services. Roads carry the most amount of domestic costs instruments recently negotiated or under negotiation will be the traffic. It is imperative to implement capacity key. This is feasible in the short term based in particular on the TIR building programs for this sector, including fleet (Transports Internationaux Routiers) and other instruments which modernization. A phased approach is recommended Pakistan is recommended to join. to minimize the potential for disruptive upheavals in Improvement of port operations and customs procedures is the road transport markets. paramount, especially if trade in time-sensitive products is to Addressing domestic connectivity disparities is grow. Recent measures by customs to introduce a unified and important to create opportunities for trade in new comprehensive customs system should be complemented with products. Revamped rail services linked to dry ports reforms in: an effective risk management system, a formal AEO can serve as the core for provincial logistics clusters regime, and an expedited regime for transit shipments. to support exports in new products, specially fruits and vegetables. The Sialkot Dry Port experience can be used as a model for the future. 65