This Selected Issues and Statistical Appendix paper presents cross-country regression results that identify investment in physical capital and improvements in institutional quality as having the largest pay-off in terms of increased growth. The paper employs three approaches to forecast inflation in Pakistan. A leading indicator model outperforms a univariate autoregressive moving average model as well as a vector autoregressive model in terms of forecast quality. The paper presents three case studies of Pakistani public sector enterprises that have recently witnessed strong improvements in their financial performance.

Abstract

This Selected Issues and Statistical Appendix paper presents cross-country regression results that identify investment in physical capital and improvements in institutional quality as having the largest pay-off in terms of increased growth. The paper employs three approaches to forecast inflation in Pakistan. A leading indicator model outperforms a univariate autoregressive moving average model as well as a vector autoregressive model in terms of forecast quality. The paper presents three case studies of Pakistani public sector enterprises that have recently witnessed strong improvements in their financial performance.

VII. Trade Integration Between Pakistan and India49

A. Introduction

153. It is natural and beneficial for neighboring countries to trade. Often, transportation costs between neighbors are low, and language similarities reduce communication and transaction barriers. Where cultural affinity leads to similarity in tastes, profitable complementarities can emerge. Trade in turn improves economic efficiency, as it helps countries to exploit their comparative advantages. In addition, it fosters growth through economies of scale, innovation and knowledge spillover, and can aid the broader process of political and social integration.

154. India and Pakistan are two neighbors that hardly trade with each other. Partly, this is due to their history of being overall relatively closed economies, but more importantly past political frictions have influenced their mutual trade regimes. It is also, in part, a regional South Asian phenomenon.

155. This chapter reviews the degree of trade openness of India and Pakistan, and the degree of trade integration between them. It assesses the existing trade policies and tariff and non-tariff trade barriers of the two countries. It provides an outlook of how trade barriers may be dismantled, partly in the advent of the South Asian Free Trade Area (SAFTA), and what the likely economic effects of such trade liberalization would be.

B. Trade Between India and Pakistan

156. Trade between India and Pakistan has been very low. While in 1948, just after India and Pakistan had become independent countries, 60 percent of Pakistan’s exports were destined for India and 17 percent of Indias exports were sent to Pakistan, trade soon dried up as tensions between the neighbors increased. Since 1990, on average 0.9 percent of annual Pakistani exports went to India, while 0.3 percent of Indian exports were shipped to Pakistan (see Figure VII.1).

Figure 1.
Figure 1.

Pakistan and India: Inter-Country Exports, 1990-2003

(in percent of total exports)

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Source: DOTS

157. India and Pakistan are relatively closed economies (see Figure VII.2). Indias trade share of GDP has increased steadily since the early 1990s, but has remained low at 31 percent in 2003, while Pakistan’s trade openness was 38 percent (2003).

Figure 2.
Figure 2.

Trade Opennes, 1990-2003

(in percent of GDP)

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Source: WEO

158. Low trade and trade integration are in part a regional phenomenon. The South Asian average trade openness was 65 percent in 2003 (simple average of Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka), much below the openness of other regions, as, for example, trade openness of ASEAN countries averaged 144 percent in 2003. In addition, South Asia is among the least integrated regions in the world (see Figure VII.3). Intraregional trade in South Asia was below one percent of GDP in 2002, less than the Middle East and North Africa (3.5 percent), and much less than East Asia (26.5 percent).

Figure 3.
Figure 3.

Intraregional Trade, 2002

(in percent of GDP)

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Source: World Bank

C. Tariff and Nontariff Trade Barriers

159. Low levels of trade can in part be explained by the large size of the two countries, but low volumes of trade and low trade integration in India and Pakistan have their roots also in their respective trade systems, as both Indias and Pakistan’s trade regulations are relatively restrictive. According to an index prepared by Fund staff, Indias trade restrictiveness measures 8 (on a scale from 1 to 10), while Pakistan’s index stands at 6. Comparatively high trade restrictiveness is partly a regional feature, with the average of South Asian countries at 5.9, compared to an average of all Asian countries of 4.4.

160. In both countries, tariffs have already been reduced from very high levels (see Figure VII.4). India started its trade liberalization in 1991/92, and between that year and 1997/98 the unweighted average tariff was brought down from 128 percent to 34 percent. Since 2002/03, the general maximum customs duty was reduced from 35 percent to 20 percent (30 percent for agricultural products), and a previously imposed Special Additional Duty (SAdd) of 4 percent was abolished. An exception to Indias trade liberalization efforts are agricultural tariffs, which at an average of 40.1 percent (see Table VII.1) remain substantially higher than non-agricultural tariffs (19.7 percent). In Pakistan, trade liberalization started in the 1980s and continued cautiously through the 1990s. It was reinforced in 1996/97 with a new comprehensive tariff reduction and simplification program. By 2002/03, the basic maximum tariff was reduced to 25 percent, and currently Pakistan operates with a relatively simple, four-rate structure. The average unweighted customs duty is 14.9 percent. Pakistan’s trade liberalization has included the agricultural sector, where the unweighted average tariff (20.5 percent) is only moderately above the non-agricultural tariff average (13.8 percent).

Figure 4.
Figure 4.

Import Duties, 1990-2002

(in percent of goods imports)

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Sources: GFS, WEO, IFS
Table VII.1.

Tariff Structures in India and Pakistan (in percent)

article image
Source: World Bank (2004), Pakistan Central Board of Revenue

In India: 20 percent when excluding agricultural products

161. Both India and Pakistan maintain tariff peaks, lines with basic customs duties exceeding the general maximum rate. In India, 6.9 percent of tariff lines exceed the general maximum customs duty of 30 percent, with rates ranging between 35 percent and 182 percent. In addition, 5.3 percent of Indian tariff lines are subject to specific customs duties, in some cases equivalent to 50-100 percent. Indias tariff peaks are concentrated in the agricultural, automobile, and textiles and garments sectors. Pakistan, which maintains few tariff peaks (1.1 percent of tariff lines), mainly protects the edible oils and automobile industries, and also imposes high tariffs on alcoholic drinks.

162. Overall, despite the efforts to open up for trade, both Indias and Pakistan’s average tariffs remain relatively high. Indias average tariff stands at 22.2 percent, while Pakistan’s is 14.9 percent. This compares with a developing country median of 11.2 percent.50

163. India has moved to protective measures by means of anti-dumping duties. Since 1995, India has initiated 383 anti-dumping cases.51 Currently, India applies anti-dumping duties on a wide range of intermediate materials and inputs, including chemicals and petrochemicals, pharmaceuticals, synthetic fibers and steels products, as well as more recently also on some consumer goods. 47 exporting countries are affected by Indian anti- dumping duties, including China, Taiwan Province of China, the European Union, Korea, Japan and the United States. No cases have been initiated against exporters from Pakistan.

164. Pakistan has only started using anti-dumping measures recently, with the first case having been decided in November 2002. However, Pakistan has recently institutionalized anti-dumping procedures by means of a new anti-dumping law and the strengthening of the Pakistan Tariff Commission. Therefore, increased use of anti-dumping tariffs by Pakistan is possible in the future.

165. Apart from anti-dumping duties, both India and Pakistan employ other types of tariff-like measures. For India, this includes small-scale industry and other exemptions on domestic excise taxes. Moreover, for some products, India applies values other than actual cif prices as the base for ad valorem import duties, which is done mainly as a preventive measure against under-invoicing, but may at times effectively result in higher tariffs. By contrast, Pakistan has in the past employed “regulatory duties” imposed on top of normal customs duties to provide extra protection to particular local industries. However, this policy is being phased out, with no new duties being imposed and only a few still remaining in place.

166. In the area of non-tariff barriers, the most significant hurdle to trade is Pakistan’s restriction of imports from India. While India granted MFN status to Pakistan in the mid-1990s, Pakistan’s imports from India remain restricted to items covered on a positive list. In 1986, Pakistan issued a list of 42 items that were allowed to be imported from India. This list has progressively been extended to 686 items52 (May 2003). In addition, a 2002 Statutory Regulatory Order (SRO) permits the import from India of not locally produced raw materials, required for the production of exports, covering approximately an additional 1000 items. The increase in items allowed for importing notwithstanding, the positive list approach remains an important hurdle to imports from India.

167. Starting in the 1980s, and reinforced in 1996/97, Pakistan progressively dismantled its use of import licensing and other non-tariff trade barriers, few of which are remaining today. Apart from the positive list of imports from India, the main remaining restrictions are largely confined to local content programs in the automobile industry. Under these programs, which have to be phased out under the WTOs Trade Related Investment Measures (TRIMs) agreement, import duty reductions are given to individual firms in return for their commitment to incorporate specified amounts of locally produced inputs in their production. In addition, Pakistan employs health- and safety-related barriers, including a ban on imports of some second-hand household machinery, some used motor vehicles, and some categories of industrial machinery.

168. In India, significant non-tariff barriers remain. While import licensing was largely abandoned in 2001, remaining forms of non-tariff barriers include government-mandated import monopolies, which are in place in the areas of agricultural products and petroleum products. In addition, India maintains tariff rate quotas in the agricultural sector, permitting the import of small quotas at moderate tariffs while applying much higher tariffs on imports in excess of the quota amounts. Besides, India uses technical barriers to trade in the form of technical standards and regulations, administered by the Bureau of Indian Standards, which operates a certification scheme for foreign exporters of products on a list of more than 100 items.53 Moreover, India employs sanitary and phytosanitary rules in the agricultural sector and other health and safety regulations (e.g. in pharmaceuticals), which may in part serve the purpose of discouraging trade.54 For instance, under health and safety concerns, India bans the import of used clothing and second hand household machinery and cars.

169. Efficiency of customs operations can act as a de facto barrier to trade. As delays in clearing goods at customs effectively constitutes an extra cost for traded products relative to domestically produced goods, long delays at customs diminish the competitiveness of imports. Pakistan and India both have above average times of customs clearance. According to the World Bank Investment Climate Surveys, the average number of days to clear imports through Pakistani customs was 17 days (2003), while it took an average of 10 days in India (2000).55 These compare with a average of 4 days for developed countries, and 6 days for East Asia56. Figure VII.5 plots the relationship between imports and customs delays for 38 low- and middle-income countries, confirming the expected negative correlation.

Figure 5.
Figure 5.

Customs Delays and Imports in Low and Middle Income Countries

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Sources: World Bank Investment Climate Surveys, WEO

170. Limited foreign investment can also be part of the explanation for low trade flows. By means of FDI, companies can set up infrastructure to allocate each stage of production to the country with lower cost, leading to trade flows in the production stages. However, substantial FDI flows are more likely between countries that trade extensively, so that the cause and effect relationship is not unidirectional.

171. FDI flows both to India and Pakistan have been rather low (see Figure VII.6). In India, FDI inflows have started to materialize slowly with the economic opening of the 1990s, but levels have remained below 1 percent of GDP. In Pakistan, inflows have fluctuated between 0.6 percent and 1.4 percent of GDP. Low FDI inflows are in part a regional phenomenon, with unweighted country averages in South Asia having gyrated around 1 percent. For comparison, between 1999 and 2003, FDI inflows to Indonesia averaged 2.2 percent of GDP, and were substantially higher for China (3.7 percent). Moreover, the share of FDI flows to Pakistan originating from India is negligible, as is Pakistan’s share in flows to India.

Figure 6.
Figure 6.

Foreign Direct Investment Inflows, 1990-2003

(in percent of GDP)

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Source: WEO

172. Limits in foreign investment can in part be explained by restrictions to equity investment, as measured by the Foreign Ownership Restrictions Index (see Figure VII.7).57 The index measures the share of equity markets that is barred from foreign investment, and shows that, despite substantial opening since the early 1990s, significant restrictions remain.

Figure 7.
Figure 7.

Foreign Ownership Restrictions Index, 1993-2004

Citation: IMF Staff Country Reports 2004, 415; 10.5089/9781451830644.002.A007

Sources: S∓P/IFC, Staff calculations

173. Other restrictions hamper trade between India and Pakistan, in part reflecting the past political tensions. A highly restrictive visa regime continues to curb travel between the countries, thus limiting possibilities to engage in business contacts. Direct air travel between India and Pakistan used to be banned and started to be restored only in late 2003. Similarly, sea and land transportation between the countries is made difficult. Ships plying between Indian and Pakistani ports are obliged to first touch a third-country port before being allowed to land. In addition, India limits the ports and inland customs posts at which imports can be cleared for a number of products it labels as sensitive items. Moreover, the lack of coordination between the railway authorities of both countries and border closure continue to hamper land transportation. In the area of the payments system, current restrictions and underdeveloped financial links act as an impediment to the efficient settlement of payments.

D. Dismantling the Trade Barriers

174. Efforts at boosting regional trade integration have in the past focused on the South Asian Preferential Trade Agreement (SAPTA). Signed in 1993 by Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka, the members of the South Asian Association for Regional Cooperation (SAARC), the SAPTA agreement provided a framework and institutional base for trade liberalization by means of exchange of concessions on tariff and non-tariff barriers. However, SAPTA had only marginal impact on regional trade, as the political will to follow through on trade liberalization was low, and, in the event, restrictions excluded 90 percent of tariff lines, and quotas often rendered even the liberalized lines meaningless.

175. On January 6, 2004, the seven SAARC member countries signed the SAFTA Framework Treaty, which establishes SAFTA effective as of January 2006. The Treaty covers tariff reductions vis-a-vis members, rules of origin, safeguards, institutional structures, and dispute settlement. It also calls for trade facilitation by means of harmonization of standards and customs procedures, and cooperation in transport infrastructure.

176. Under the treaty, members agreed to gradually reduce their import tariffs vis-a-vis other SAFTA members to a range of 0-5 percent. In 2006 and 2007, members must reduce their maximum tariff rates to 20 percent (the non-Least Developed Countries (non-LDCs) India, Pakistan and Sri Lanka) or 30 percent (the Least-Developed Countries (LDCs) Bangladesh, Bhutan, the Maldives and Nepal). In a second stage, tariffs must then be reduced to 0-5 percent. India and Pakistan must implement this second stage by January 1, 2013, Sri Lanka by January 1, 2014, and the LDCs by January 1, 2016.

177. In addition, member countries will be allowed to maintain ‘sensitive lists’ of products that would be exempted from tariff reductions. These lists are currently under negotiation and will be periodically reviewed with a view to reducing the number of items covered.

178. Member countries are also obliged to eliminate all quantitative trade restrictions, except those permitted under WTO rules. This would include the obligation for Pakistan to abolish its positive list for imports from India. No mechanism to deal with non-tariff barriers has so far been specified for SAFTA, other than notification and consideration by the SAARC Committee.

179. The treatys prospective impact remains uncertain. Under the agreement, back-loaded tariff cuts are possible, opening the possibility that tariff reduction takes place slowly and over a prolonged period. On the other hand, unless the high external protection levels for some sectors are dismantled in parallel, steep regional tariff cuts could lead to either substantial trade diversion (see below) and thereby to large welfare losses, or to strong resistance to concessions, which could result in extensive use of the sensitive lists. In addition, the agreement provides for temporary suspension of concessions in the face of balance of payments difficulties, and does not exclude the use of anti-dumping measures between members, creating uncertainty and the potential for disruptions of the trade integration process. Moreover, important elements of the agreement, including the rules of origin, remain to be negotiated. Regardless of the SAFTA process, the broader process of tariff reduction should be continued and the existing distorting protection for certain industries should be phased out.

E. Advantages of Trade Liberalization for Pakistan

180. The potential advantages of trade liberalization for Pakistan are large. Going well beyond the immediate creation of trade flows, the advantages of dismantling tariff- and non-tariff barriers include the potential for boosting productivity and economic growth, and can also extend to promoting regional cooperation in all areas.

181. SAFTA may result in trade creation. The lowering of trade barriers can create trade flows that are increasing economic efficiency, in product groups that are currently being produced in Pakistan mainly due to trade barriers, while the cost for production and transportation from other SAFTA members are at a competitive level. Similarly, the lowering of trade barriers will create new export opportunities for Pakistani products in other SAFTA member countries.

182. As with all regional trade areas (RTAs), SAFTA could also lead to welfare-reducing trade diversion. An RTA may create new and additional trade that is internationally competitive, but may also divert trade from low-cost international sources to a higher-cost source within the RTA. The latter may occur in situations in which trade barriers are initially high to all trading partners, and then lowered for RTA members only. With this, trade may be diverted from outside to within the RTA, which, if production costs in the RTA are higher than elsewhere, would be welfare-reducing.

183. Evidence from existing RTAs is encouraging. 58 In six existing RTAs (Mercado Comun del Sur (MERCOSUR), Andean Pact II, Caribbean Community (CARICOM), ASEAN Free Trade Area (AFTA) and Gulf Co-operation Council (GCC)), five years after signature of the RTA, intra-RTA imports as share of GDP had grown, but simultaneously, extra-RTA imports as share of GDP had similarly increased. While this is no evidence that trade diversion has not occurred, it does not appear to have been a dominating factor at the aggregate level. Nonetheless, the most effective means to avoid trade diversion is to continue the process of dismantling the protection of heavily shielded industries from outside the RTA at the same time as trade within the RTA is liberalized.

184. There is an enormous potential for increasing trade flows between India and Pakistan. Compared to the officially reported trade flows of US$260 million (2003), estimates for the potential under a free trade setting vary. At the upper end, Pakistan’s Ministry of Commerce estimates the potential at US$2.7 billion (informal estimate), and at the lower end Nabi et al. (2001) estimate it at US$750-1000 million. More formally, the potential has been estimated at US$1.85 billion, using a fixed-effects gravity model of bilateral trade based on 2001 data.59 However, part of such flows would certainly be the conversion of existing trade now routed through third countries into direct India-Pakistan trade, as well as the formalization of trade flows currently smuggled across the border.

185. Trade liberalization will unambiguously benefit Pakistani consumers, since product prices fall and consumer choice increases with reduced trade barriers. Importing from India rather than from more distant locations would often imply lower transportation costs. These, as well as lower tariffs, would be largely passed on to consumer prices if there is sufficient competition among suppliers. In addition, in agriculture, price volatility could decline, as at times of unforeseen shortages in local production, access to Indian production could help smooth out price spikes. Pakistani consumers could also gain from liberalized trade in power, as the reliability of power supply could increase. In textiles, lower prices would be expected to have a beneficial income distribution effect, as lower-income households spend a higher share of their income on textiles. For upper income groups, access to high-quality Indian silk products would imply an increase in consumption choices. In engineering goods, consumers may benefit from lower prices for cars, motorcycles and bicycles, as well as access to Indian products.

186. Pakistani producers would lose due to falling consumer prices, but gain due to access to larger markets and possible efficiency boosts. As a main disadvantage for producers, trade liberalization is expected to lower prices of imported products, so that Pakistani firms competing with those imports would face decreasing profit margins. By contrast, trade integration will allow producers to access a much wider market, allowing for greater efficiency in production by exploiting economies of scale in production, thereby boosting productivity. In addition, since freer trade provides entrepreneurs with an incentive to explore new export opportunities and to compete more fiercely with imports, trade liberalization can lead to efficiency gains through learning and innovation. Finally, like consumers, producers would also benefit from increased reliability of power supply that may materialize if trade in power is liberalized.

187. The Pakistani government would gain customs revenue. Increased trade flows that stem from the lifting of import prohibitions from India would lead to additional customs revenue for Pakistan. By contrast, to the extent that existing tariffs are lowered, there would be corresponding losses in customs revenue.60 However, as the existing trade flows are much lower than the potential for such flows (see above), the former effect would almost certainly dominate the latter. In addition, to the extent that trade integration increases the likelihood of the construction of a gas pipeline from Iran or Afghanistan through Pakistan to India, the government could then profit from transit fees, estimated in the order of US$500-700 million.61

188. Trade liberalization can lead to higher economic growth. Gains in total factor productivity achieved by economies of scale and enhanced efficiency (see above) are likely to translate into higher growth of potential output. In addition, a liberalized trade regime is likely to attract FDI, as an economy with larger access to regional markets becomes more attractive to foreign investors. Higher FDI, in turn, may lead to increased technology transfer and thereby to increases in total factor productivity. Moreover, trade liberalization can be seen as part of the broader process of regional integration, which may foster closer relations between India and Pakistan.

F. Conclusions

189. This chapter finds that the potential for expanding trade in the South Asian region and especially between India and Pakistan from its current, abnormally low level is substantial. In order to foster trade, multilateral and bilateral trade restrictions should continue to be dismantled. SAFTA could provide an institutional vehicle for such trade liberalization, but, as shown by the limited progress reached under SAPTA, SAFTA alone cannot guarantee continued advancements, and political determination will be needed to reduce the trade barriers. The potential advantages for Pakistan are large, reaching from lower prices and higher choice for consumers and increased efficiency and market access for producers to higher revenue for the government. Moreover, trade integration can foster economic growth and the broader process of regional integration.

Table 1.

Pakistan: Sectoral Origin of Gross Domestic Product at Constant Prices,

1999/2000-2003/04

article image
Source: Pakistani authorities.
Table 2.

Pakistan: Sectoral Origin of Gross Domestic Product at Current Prices,

1999/2000-2003/04

article image
Source: Pakistani authorities.
Table 3.

Pakistan: Gross Domestic Product—Expenditure Side, 1999/2000-2003/04

article image
Source: Pakistani authorities.
Table 4.

Pakistan: Consumer and Wholesale Price Indices, 1997/98-2003/04

(2000/01 = 100)

article image
Sources: Federal Bureau of Statistics; and Fund staff calculations.

For fiscal year data, refers to the change in in indices at the end of the year.

Table 5.

Pakistan: Domestic Retail Prices of Selected Petroleum Products, 1997/98-2003/04

article image
Source: Ministry of Petroleum and Natural Resources; and Federal Bureau of Statistics.

Annual averages.

MS 87-RON.

Fuel oil prices were deregulated with effect from July 1, 2000.

Table 6.

Pakistan: Natural Gas Prices, 1997-2004 1/

article image
Sources: Ministry of Petroleum and Natural Resources; and Fund staff estimates.

Columns indicate date of price adjustments.

The weights used, based on the 1984/85 consumption pattern, are as follows: fertilizer industry, 0.148; other industries, 0.644; household use, 0.165 (with equal shares for all classes of users); and commercial, 0.043.

Table 7.

Pakistan: Federal Government Fiscal Operations, 1997/98–2003/04

article image
Source: Ministry of Finance and Economic Affairs.

Fiscal year 2003/03 includes $1 billion (PR’s 58 billion) U.S. special grants for debt retirement and also increase in project/other grants.

Accrued payments. Excludes interest expenditure by the military which is included in the defense allocation.

Includes interest and principal payments on military debt; excludes military imports financed by external grants and disbursements.

Includes certain current outlays under the public sector development program.

Table 8.

Pakistan: Provincial Government Fiscal Operations, 1997/98-2003/04

article image
Sources: Ministry of Finance and Economic Affairs; and Fund staff calculations.
Table 9.

Pakistan: Government Debt, 1997/98-2003/04

(In billions of Pakistani rupees, unless otherwise indicated)

article image
Sources: Pakistani authorities; and Fund staff calculations.
Table 10.

Pakistan: External Debt 1997/98-2003/04

article image
Sources: Pakistan authorities; and Fund staff calculations.