A Computable General Equilibrium Analysis Of Trade Policy Options For The Palest

TABLE OF CONTENTS

ACKNOWLEDGMENTS .....................................................................................4 LIST OF TABLES ...............................................................................................5 LIST OF FIGURES .............................................................................................6 ABSTRACT ........................................................................................................7 CHAPTER 1 INTRODUCTION........................................................................9 1.1 Background ............................................................................................9 1.2 Description of Study .............................................................................11 1.2.1 1.2.2

1.3

Palestinian Trade Structure and Trade Policy Options ......................................11 CGE Models and SAMs.....................................................................................15

Outline of Study ....................................................................................19

CHAPTER 2 PALESTINIAN TRADE PATTERNS AND TRADE POLICIES ..21 2.1 Introduction...........................................................................................21 2.2 Palestinian Trade Patterns ...................................................................22 2.2.1 2.2.2 2.2.3

2.3

External Merchandise Trade .............................................................................22 Net Transfers from Abroad ................................................................................24 Non-Tariff Barriers and Transaction Costs ........................................................26

Palestinian Trade Policies ....................................................................28

2.3.1 2.3.2 2.3.3

The Paris Protocol and EU Agreements............................................................28 Average Tariff Rates .........................................................................................31 Revenue Clearance System..............................................................................33

CHAPTER 3 REVIEW OF THEORY OF CGE MODELS ...............................34 3.1 Introduction...........................................................................................34 3.2 Computable General Equilibrium Models..............................................35 3.2.1 3.2.2 3.2.3

3.3

Definition and Classification ..............................................................................35 Summary of the Core CGE Model.....................................................................37 Model Construction, Calibration, and Closure ...................................................52

Trade Policy Analysis ...........................................................................56

CHAPTER 4 A PALESTINIAN SOCIAL ACCOUNTING MATRIX..................59 4.1 Introduction...........................................................................................59 4.2 Data Sources........................................................................................60 4.2.1 4.2.2 4.2.3

4.3

Supply and Use Tables .....................................................................................62 Economic Surveys.............................................................................................64 A Palestinian Input-Output Table.......................................................................67

A Palestinian SAM................................................................................71

4.3.1 4.3.2

Level of Aggregation .........................................................................................71 A Macroeconomic Palestinian SAM ..................................................................72

1

CHAPTER 5 A PALESTINIAN CGE MODEL ................................................79 5.1 Introduction...........................................................................................79 5.2 Circular Flow of Income ........................................................................81 5.3 Model Structure (PALMOD)..................................................................83 5.3.1 5.3.2 5.3.3 5.3.4 5.3.5

5.4

Households .......................................................................................................83 Firms .................................................................................................................84 Imports and Exports ..........................................................................................85 Government.......................................................................................................86 Market Clearing Definitions ...............................................................................87

Calibration ............................................................................................90

CHAPTER 6 TRADE POLICY SIMULATIONS ..............................................92 6.1 Introduction...........................................................................................92 6.2 Reduction in Import Tariffs....................................................................94 6.3 Reduction in Labor Demand in Israel..................................................102 6.4 Full Separation ...................................................................................104 6.4.1 6.4.2

‘Peace’ Scenario – Increases in Aid, Labor Demand and Labor Supply .........105 ‘War’ Scenario – Increase in TC and Decrease in LD in Israel........................107

CHAPTER 7 SUMMARY AND CONCLUSIONS..........................................110 7.1 Overview of Study ..............................................................................110 7.2 Conclusions........................................................................................114 7.3 Policy Implications ..............................................................................115 A. APPENDIX ............................................................................................119 A1. Variables ............................................................................................119 A2. Parameters.........................................................................................120 REFERENCES ...............................................................................................121

2

To the memory of my father

3

ACKNOWLEDGMENTS My special thanks to Amos Golan of the American University in Washington DC for making this endeavor as enjoyable as possible, especially under very difficult political circumstances. I would also like to thank Irma Adelman, Daniel Schydlowsky, Mieke Meurs, John Willoughby, Hussein Abdul-Hamid, Tarik Alami, and Ali Bayar for their technical assistance and valuable comments. A very special thanks to my wife, Alin, and my two children, Seto and Vartan, and to the late Ms. Alidz Keshishian, for their continuous support and sacrifices. I am also grateful to my sister, Sylva Balian-Kalbian, for her encouragement and support especially in the final stages of this study. Finally, I would like to express my sincere gratitude to Oded Stark for his guidance over a period of many years, for his altruism, and beyond.

4

LIST OF TABLES Table 1.1 Total value and share of Palestinian exports and imports by country of origin and destination, 1998 ($US million) Table 2.1 Palestinian foreign trade partners for 1998 ($US million) Table 2.2 Total value of Palestinian imports and exports by SITC Rev.3, 1998 ($US million) Table 2.3 Total commitments and disbursements for 1994-2000 ($US million) Table 2.4 Palestinian population in the world, 1998 Table 2.5 Taxes on imports by country of origin, 1998 (%) Table 2.6 Dispersion of import taxes across economic activities, 1998 (%) Table 4.1 An input-output table for the Palestinian economy, 1998 ($US million) Table 4.2 Inter-sectoral transactions, 1998 ($US million) Table 4.3 Direct input-output coefficients, 1998 Table 4.4 Aggregated macroeconomic Palestinian SAM, 1998 ($US million) Table 4.5 Aggregated macroeconomic Palestinian SAM with diagonal entries, 1998 ($US million) Table 4.6 Aggregated macroeconomic Palestinian SAM with indirect taxes, 1998 ($US million) Table 4.7 Aggregated macroeconomic Palestinian SAM with savings and investment, 1998 ($US million) Table 4.8 Disaggregated macroeconomic SAM for the Palestinian economy, 1998 ($US million) Table 6.1 Simulation results

5

LIST OF FIGURES Figure 3.1 Classification of CGE models Figure 3.2 Flowchart of CGE model structure Figure 5.1 Circular flow of income

6

ABSTRACT A Computable General Equilibrium Analysis of Trade Policy Options for the Palestinian Economy The Palestinian economy is a small open economy with poor endowments of natural resources and unmarked national borders. Almost 80% of its foreign trade is conducted with a single partner, Israel. It exports about 15% of its output which in turn finances only about 25% of its imports. The rest of the import bill is financed through wage-income earned in Israel, foreign aid, and remittances from abroad. This dependency on foreign capital inflows has severely retarded the Palestinian export sector. Moreover, the presence of high transaction costs and other non-tariff barriers have dramatically reduced the competitiveness of its export sector. At the Oslo peace accords in 1993, the Palestinian Authority sought to implement a policy of export-led growth with the assistance of both Israel and the international community. The cornerstone of this outward-oriented strategy was the Paris Protocol which formalized the de-facto economic relations between Israel and the West Bank and Gaza Strip that had existed since 1967. Due to political reasons, the Paris Protocol was not implemented and the Palestinian Authority has sought alternative strategies to promote its exports. An important means to achieve this objective is to raise the share of manufactured exports as a source of import finance. This study constructs a static trade–focused computable general equilibrium model to analyze the effects of four policy scenarios on the Palestinian economy. The first scenario is a 50% reduction in transaction costs, and the second is a 50% reduction in Palestinian labor employed in Israel. The remaining two scenarios simulate the effects of a full-separation between Israel and the West Bank and Gaza Strip. One separation scenario – the ‘peace’ scenario – simulates simultaneous increases in foreign aid, labor demand in Israel, and domestic labor supply. The other sce7

nario – the ‘war’ or Intifada scenario – simulates a simultaneous increase in transaction costs and a decrease in labor demand in Israel. The results of these four simulation exercises demonstrate that first, reductions in transaction costs substantially increase the volume of trade and improve the Palestinian terms of trade. Second, the lower dependency of the Palestinian workforce on Israel also increases exports by 8%, but real household income falls since workers who were previously employed in Israel are now forced to work at the lower Palestinian wage rate. And third, simulations of the full-separation scenarios demonstrate that the outcome primarily depends on whether separation is achieved through peaceful negotiations or unilaterally through the Intifada. The simulation results of this last scenario show that the volume of trade is substantially reduced by as much as 56% for exports and 30% for imports.

8

CHAPTER 1

1.1

INTRODUCTION

Background

The Palestinian economy, also known as the West Bank and the Gaza Strip (WBGS), is a small open economy with scarce natural resources and unmarked national borders. It covers an area of approximately 6,000 km2 and has a population of 3.1 million; 1.1 million living in the Gaza Strip and about 2 million in the West Bank. The Oslo peace accord in 1993 was an effort to reach a Final Status agreement between Israel and the Palestinian Authority (PA) for the eventual declaration of a Palestinian State recognized by both Israel and the international community. These agreements have not been fully implemented but both sides are making serious efforts to reach a Final Status agreement. In 1998, the latest year for which National Accounts have been compiled, Palestinian GDP was $4.3 billion and GNP was estimated at $5.2 billion1. Foreign trade has traditionally played an important role in the Palestinian economy, especially vis a vis Israel. Imports directly from Israel account for 77% of total Palestinian imports, or $2.35 billion, and exports to Israel are over 95% of total Palestinian exports, or $704 million. The ratio of trade to GDP is about 65%, reflecting the high dependence of the Palestinian economy on foreign trade, especially with Israel. The Palestinian economy exports 15% of its output, which in turn finances only about 25% of its imports. The rest of the import bill is financed by wage income earned in Israel, foreign aid, and remittances from abroad. This inflow of funds, which is unrelated to the endogenous productive capacity of the Palestinian economy, raises the economy’s wage rate. While producers in the nontradable service sector can shift the resulting higher costs to consumers through higher prices, producers in the tradable manufacturing sector are not able to do 1

Palestinian Central Bureau of Statistics, 1999. National Accounts: Preliminary Results, 1998. Ramallah: West Bank. This difference between GDP and GNP is attributed to Palestinian wage income from Israel, foreign aid, and remittances from abroad.

9

so since the prices of tradables are essentially determined in the world market. The direct effect of this inflow of funds is a resource shift from the tradable manufacturing sector to the non-tradable service sector, causing deindustrialization2. The indirect effect is induced by an increase in spending, which in turn increases aggregate demand and reinforces the price rise in the non-tradable service sector, further contracting the tradable manufacturing sector. Both effects are reflected in a real exchange rate appreciation – an increase in the price of non-tradables relative to the price of tradables – reducing the competitiveness of the tradable export sector. The main symptoms of deindustrialization in the Palestinian economy are the low share of manufacturing in GDP, a booming nontradable construction sector, and a relatively high wage rate as compared to the wage rates in the neighboring economies of Egypt, Jordan, and Lebanon. An important means to halt and reverse the deindustrialization process is to raise the share of manufactured exports as a source of import finance. To this end, trade liberalization could be helpful. Trade liberalization opens up new markets and reduces trade barriers through Preferential Trading Agreements. A Customs Union, for example, would eliminate all tariff and non-tariff barriers between Israel and the WBGS, with a common external tariff against nonmembers. A less restrictive form of trading agreement would be a Free Trade Area where again there would be no trade barriers between the two, but each would be free to set its own tariff rate against non-members. Finally, a complete separation between the WBGS and Israel would require a deeper integration of the Palestinian economy with the rest of the world such as the EuroMediterranean agreement. This study constructs a trade-focused static computable general equilibrium model to quantify the economic effects of four trade policy options: a reduction in transaction costs brought about through a reduction in import tariffs; a de2

This phenomenon is also known as “Dutch Disease” as illustrated in Cordon and Neary, 1982. However, deindustrialization should not be treated as a ‘disease’, especially in developing countries. The movement of resources from tradable to non-tradable sectors can be regarded as an improvement to the expansion of the domestic non- tradable sector.

10

crease in Palestinian labor employed in Israel; increases in labor supply and foreign aid; and an increase in transaction costs and a decrease in labor demand in Israel. The study demonstrates that these policies help to promote export-led growth, albeit with varying effects on sectoral resource allocation and other macroeconomic variables. However, complementary domestic policy reforms are also required to reap the maximal benefits of trade liberalization to achieve export-led growth. In this way, the study provides a basis for quantifying the impact of domestic and regional policy reforms within a broader general equilibrium framework in the future.

1.2

Description of Study

1.2.1 Palestinian Trade Structure and Trade Policy Options

Table 1.1 shows the total value and share of Palestinian trade by country of origin and destination. The traditional importance of trade with Israel is reflected in the high share of Palestinian exports to Israel (96%) and imports from Israel (77%), resulting in a trade deficit of $2.35 billion. This is followed by trade with the EU which constitutes about 9% of total Palestinian imports but absorb only 0.4% of Palestinian exports, resulting in a trade deficit with the EU of $287 million. Exports to Jordan are about 2% of total exports and 2.4% of total imports. Imports from the Asian countries as a group constitute about 4% of total Palestinian imports with no reciprocal exports to these countries. Trade with North America, Canada, and Latin America is again one-sided with a small amount of Palestinian exports going to these countries, about 0.2%, but 1.7% of Palestinian imports originate in this group. Imports from Turkey constitute about 2% of total Palestinian imports, and a small number of imports originate in Eastern European countries. This one sided Palestinian trade structure has resulted in a trade deficit of $2.35 billion in 1998, of which 73% is with Israel. The current trade relations between Israel and the WBGS are based on the Protocol on Economic Relations 11

signed in Paris on April 29, 1994. The Protocol consists of four major elements: first, the abolition of all tariffs and duties between Israel and the WBGS, but Quantitative Restrictions on Palestinian exports to Israel of melons, cucumbers, tomatoes, potatoes, eggs, and poultry. Second, on products classified in Lists A1 and A2 – mostly agricultural and food imports from Jordan, Egypt and other Arab countries – the PA can set its own tariff rates but is bound by quantitative import restrictions, volumes of which are determined by estimates of the consumption needs of the Palestinian economy.

TABLE 1.1 TOTAL VALUE AND SHARE OF PALESTINIAN EXPORTS AND IMPORTS BY COUNTRY OF ORIGIN AND DESTINATION, 1998 ($US 000).

36,636

Share of Exports (%) 0.6

Share of Imports (%) 1.2

Net Trade Balance -32,262

14,580

73,272

2.0

2.4

-58,692

704,214

2,356,916

96.6

77.2

-1,652,702

European Union (EU)4

2,916

290,035

0.4

9.5

-287,119

North America, Canada, and Latin America Turkey

1,458

51,901

0.2

1.7

-50,443

0

61,060

0

2.0

-61,060

Asian Countries

0

131,279

0

4.3

-131,279

Eastern Europe

729

15,265

0.1

0.5

-14,536

Other

729

36,636

0.1

1.2

-35,907

Country

Exports

Imports

4,374

Jordan Israel

Arab Countries3 (except Jordan)

5

Source: Palestinian Central Bureau of Statistics (PCBS), 2000. Foreign Trade Statistics – 1998: Main Results. Ramallah: West Bank. Compiled from Tables 4, 5, 8, 9 and 10.

Third, on products in List B – mostly consisting of basic food items and other goods used in construction, industry, agriculture and government development programs – the PA is free to import any quantity and to set its own tariff rates. For all other products not included in Lists A1, A2 and B, there are no 3

Arab countries include United Arab Emirates, Saudi Arabia, Yemen, Qatar, Morocco, Syria, Iraq, Lebanon, and Egypt. 4 EU countries include Luxemburg, Finland, Ireland, Portugal, Sweden, Denmark, Austria, Greece, Belgium, Netherlands, France, UK, Germany, Spain, and Italy. 5 Asian countries include Vietnam, Philippines, Malaysia, Indonesia, Singapore, Sri-Lanka, Thailand, Japan, India, Hong-Kong, Taiwan, South Korea, and China.

12

quantitative restrictions but Palestinian tariff rates should be at least as high as the Israeli rates. Finally, the PA is free to set its own custom duties on motor vehicles. The current trading arrangement between Israel and the PA includes elements of both a Customs Union and a Free Trade Area. This is expected since national borders between Israel and the WBGS are not yet clearly defined. Moreover, the increasing use of Non-Traditional Non-Tariff Barriers (NTNTBs)6 has rendered the concepts of a “free Trade Area and a Customs “Union” meaningless. Nevertheless, assuming that a Final Status agreement will be reached in the future, the WBGS has certain trade policy options to pursue a strategy of export-led growth. Trade liberalization will increase the competitiveness and productivity of the Palestinian manufacturing sector. A Customs Union with Israel will always be associated with the risk of trade diversion, especially from neighboring Arab countries with lower costs of production. A Free Trade Area, on the other hand, must be accompanied by strict enforcement of rules of origin if one wishes to see the benefits of trade creation. Alternatively, politicians on both sides have been talking of a “complete separation” between Israel and the WBGS. However, the policy options available to the WBGS economy must not be taken at face value. Adherents of the “Natural Trading Partner”7 hypothesis argue that forming a preferential trading agreement is more likely to raise economic welfare if member countries already trade disproportionally with each other. Opponents of the hypothesis claim that the opposite is true: welfare is likely to be higher if member countries trade less with each other. Schiff (1999) demonstrates that neither analysis is entirely correct and that the ‘Natural Trading Partner” hypothesis can be rescued if it is redefined in terms of complementarity or substitutability in the trade relations of member countries rather than in terms of their value of trade. 6

These are barriers associated with closures of the WBGS for security reasons. The “double” NT is appropriate since these barriers restrict Palestinian producers twice – once in the traditional sense, and a second time due to restrictions of movement between Palestinian villages and towns. 7 See Schiff, M., 1999. “Will the natural trading partner please stand up?”. Policy Research Working Paper No. 2161. Washington, D.C.: The World Bank.

13

The WBGS economy is currently facing high levels of unemployment and a rapid process of deindustrialization, the latter primarily caused by Palestinian wage income from Israel (Alonso-Gamo et al, 1999). Trade liberalization can go a long way to halt and reverse this process by shifting resources back into the tradable sector and increasing the share of manufactured exports. A Customs Union with Israel would force the WBGS to lower its trade barriers since Israel has low average tariff rates, especially after its accession to the World Trade Organization. This would generate competition among Palestinian domestic producers and raise productivity in the tradable sector. On the other hand, the low levels of protection may not be compatible with the development needs of the Palestinian economy. The PA receives substantial revenues from existing levels of protection, and this revenue loss must somehow be compensated through, say, an increase in indirect taxes. A Free Trade Area with Israel would be less restrictive where the PA could be tempted to set high tariff rates on final goods and low tariffs on intermediate inputs. In such a situation, rules of origin must be strictly enforced to prevent exporters from penetrating the Palestinian market through Israel. This is especially important since there are no border crossings between Israel and the WBGS. Finally, a complete separation between Israel and the WBGS would require a deeper integration of the Palestinian economy with neighboring Arab countries and Europe. Although this would also require domestic institutional strengthening and reform, it would nevertheless increase the share of manufactured exports through market access improvements. Regardless of which trade policy option we analyze, one must bear in mind that the ability of the WBGS to conduct its own trade policies will ultimately depend on the Final Status agreements reached with Israel.

14

1.2.2 CGE Models and SAMs

A CGE model is a large set of demand and supply functions that cover every market, both for goods and factors of production8. The demand side of commodity markets consists of households, government agents, and firms. Some of these agents are domestic, and others are foreign and therefore capture export demand. Private households are able to buy only as much as their income allows them. They receive this income as they sell labor services to firms, but they also receive a return on any capital investments that they have made. As private households sell their labor and capital services to firms, it enables firms to produce. In addition to buying these primary factors of production, firms also buy intermediate inputs from each other. Commodity purchases by all agents comprise of both imported and domestically produced goods. The government, in addition to participating in the regulatory process, has a number of tax and subsidy instruments available for redistributive uses9. The use of the term general equilibrium corresponds to the well-known Arrow-Debreu model, elaborated in Arrow and Hahn (1971). The Walrasian general equilibrium model provides the necessary framework for appraising the economy-wide effects of policy changes on resource allocation, and for assessing the gainers and losers, policy impacts that are not very well covered in empirical macro models. A closely related class of models, known as structuralist10 models (popularized by Adelman, Lustig, Taylor, Robinson, and others), base their economic analysis on institutions and political economy. They incorporate social classes and market power in their models and utilize an extended functional distribution based on the unique institutional structure of the economy. The actions of economic agents reflect their positions in the economic systems revealed by their 8

Most applied general equilibrium models are numerical analogues of traditional two-sector general equilibrium models developed by Meade, Johnson, Harberger, and others in the late 50’s. 9 For an in depth exposition of the core CGE model, see Dervis et al. 1981. General Equilibrium Models for Development Policy. Cambridge: Cambridge University Press. 10 For simulation exercises with structuralist models see Srinivasan, T.N. and J. Whalley, (eds.), 1986. General Equilibrium Trade Policy Modeling. Cambridge: The MIT Press.

15

principal income source. Neo-classical authors, on the other hand, start from hypotheses of optimizing agents and full employment, “their moves towards incorporating market power come as afterthoughts at best.” (Taylor, 1990). A CGE model is developed by combining multi-sectoral databases with computer software that simulates the functioning and reactions of the economy. The typical database for this purpose is the Social Accounting Matrix (SAM) which shows the transactions that take place between all agents of an economy in some base year. Essentially, it provides a snapshot picture of the economy as of some base year. The data are then applied to the model to replicate that base year in the benchmark general equilibrium11. By replicates, we mean it reproduces the observed economic transactions in the database. The task of calibrating parameters in a CGE mode, so that it successfully replicates the benchmark database, is called benchmark calibration. The computer software12 that simulates the economy does so by assuming that production and consumption activities follow certain mathematical forms. These are production functions exhibiting how technology determines the way in which goods can be produced using factors such as labor and capital, and utility functions showing how different combinations of goods and services generate well-being for consumers.13 After choosing the parameter values, we introduce a policy shock and solve for the new equilibrium. This new equilibrium is known as the counterfactual (or policy replacement) equilibrium. We then compare this new counterfactual equilibrium with the benchmark equilibrium for final policy appraisal. One important characteristic of trade-focused CGE models is the so-called Armington assumption (Armington 1969). In international trade, one often encounters simultaneous import and export of the same good. However, this observation is inconsistent with traditional (Heckscher-Ohlin) trade theory, since 11

The benchmark of a CGE is actually a solution of the model that replicates the observed economic data in some base year. 12 The most popular software is the General Algebraic Modeling System (GAMS). See Brooke et al. 1992. GAMS: A User’s Guide. California: The Scientific Press. 13 The particular production and utility functions adopted can be a crucial determinant of the results of the model.

16

under perfect competition it cannot occur. The problem that this poses for model builders was first solved by Armington and is called the Armington assumption. The problem is solved using the idea of nested functions. Instead of using the production function F(K, L1, L2), we write it as G(K,H(L1, L2)). In other words, we assume that the producer’s problem is separable and can be solved sequentially in two steps. First, the producer decides how the two types of labor should be combined into a single labor input using the function H(L1, L2). The resulting input is called a composite good, which is then combined with capital in the outer function G(K,H). This type of nesting allows us to construct very complex production structures using CES or Cobb-Douglas production functions. In recent years, numerous studies have been conducted on the effects of trade liberalization. Some are single-country models, while others attempt to capture regional multi-country gains associated with trade liberalization.14Most models attempt to quantify the welfare gains/losses associated with trade liberalization and almost all conclude substantial gains.15 The common characteristic of all such models is the adoption of an external closure16 rule, which specifies import supply and export demand for the rest of the world. Broadway and Treddenick (1978) present an especially clear explanation of their external-sector closure procedure. Their focus is on the domestic impacts of variations in Canadian tariff policy. They construct a model with substantial industrial and commodity detail using several functional forms (Cobb-Douglas and CES). An external-balance condition involving the exchange-rate variable closes the system. CGE models have been used to analyze the effects on: macro, industry, regional, labor market, distributional, and environmental variables; by modeling changes in: taxes, public consumption, social security payments, tariffs, environmental policies, terms-of-trade, technology, union behavior, levels of natural resource deposits, and other policy instruments. CGE models can also be con14

See de Melo, J. 1998. “Computable General Equilibrium Models for Trade Policy Analysis in Developing Countries: A Survey.” Journal of Policy Modeling 10(4): 569-603. 15 Several models also incorporate increasing returns and imperfect competition. 16 By closure we mean the macroeconomic adjustment process to equilibrium. See Devarajan, Lewis and Robinson (1994) for a detailed discussion.

17

structed for several countries simultaneously, for a single country, for a specific region within a country, or for a specific sector (Adelman and Taylor, 1997). The type of model depends on the question to be investigated. For example, if regional integration among neighboring countries is to be investigated, then a multi-country CGE model is used. On the other hand, if we are more concerned with income distribution issues, then it is more appropriate to use a singlecountry model with a high level of disaggregation of income groups. The starting point of any CGE model is the SAM – also known as the benchmark or base-year equilibrium solution – which is a detailed representation of the circular flow of income. It is a square matrix with each row/column representing a separate account. The rows represent income (sources of income) and the columns represent expenditures (uses of income).17 Just as total expenditure must equal total income in the circular flow, the column and row sums of each account must balance in the SAM. The accounts are similar to national income accounts and include information available in national accounts and much more.

18

The size of a SAM depends on the availability of data and

the purpose for which it is constructed. A SAM amplifies the input-output model by accommodating more data and presents a comprehensive picture of the economy. Like an input-output matrix, a SAM is based on the double accounting principle. However, it is extended in several directions beyond input-output tables by adding new rows or columns. A row or column is added for each domestic factor of production – land, labor of different types – and for the rest of the world. Thus, a SAM permits a disaggregation of value added, imports, exports, and foreign capital flows. It is constructed to provide a single unified set of accounts on how the market economy operates. It also gives a consistent picture of the flow of funds between institutions or actors in the economy. In sum, a SAM provides a comprehensive and consistent framework to organize economic data, it gives a snapshot picture of 17

See King (1981) for an introductory exposition. A SAM is normally computed from the National Accounts as classified in the UN System of National Accounts: Studies and Methods. Series F, No.2, Revision 3.

18

18

the economy at a glance, it identifies areas where data is missing19, and it permits the study of various institutions and their effects on the economy.

1.3

Outline of Study

Chapter 2 begins with a brief description of the Palestinian economy in 1998 and provides a detailed analysis of the trade agreements with its major trading partners. Trade policy options are then explored as a viable alternative to the current trading arrangements with Israel, the EU, and neighboring Arab countries. Chapter 3 reviews the CGE modeling literature with a comparison of alternative modeling techniques. In recent years, CGE modeling has been extensively used to quantify the global and regional effects of trade liberalization. This chapter also reviews the static CGE models used for trade policy analysis. Chapter 4 examines data sources and data shortcomings for the construction of a Palestinian SAM. To a large extent, policy outcomes depend on the specific functional forms, parameter values, and elasticities20. These data sources are used to compile an initial Input-Output Table for the Palestinian economy, which is then used to construct a Palestinian macroeconomic SAM for 1998. Chapter 5 constructs and calibrates a trade-focused CGE model using the SAM built in the previous chapter, with special emphasis on the “closure” rules adopted in the specification of the external sector. Chapter 6 utilizes this CGE model to conduct four simulation exercises: a 50% reduction in transaction costs; a 50% reduction in Palestinian labor employed in Israel; a simultaneous doubling of foreign aid, a 50% increase in labor employed in Israel, and a 30% increase in domestic labor supply; and a simultaneous increase in transaction 19

In this way, it helps in setting the research agenda. Although in recent years considerable effort has been expended on the construction and mechanics of CGE models, not enough econometric estimation has been carried out to determine the statistical significance of these functional forms. See McKitrick, R.R. 1998.”The Econometric Critique of Computable General Equilibrium Modeling: The Role of Functional Forms.” Economic Modeling 15: 543573.

20

19

costs and decrease in labor demand in Israel. The primary objective of these exercises is to quantify the extent to which deindustrialization in the WBGS can be halted and reversed through export-led growth. Chapter 7 concludes the study by summarizing the degree and extent to which these policies can cure deindustrialization and generate export-led growth. Although market access improvements and lower trade barriers reduce the detrimental effects of deindustrialization, domestic policy reforms must also be implemented at an early stage to reap the full benefits of trade liberalization and to eventually integrate the Palestinian economy into the World Trade Organization. In this way, the study paves the way for the quantification of numerous other domestic and regional policy reforms within a much broader multicountry CGE model in the future.

20

CHAPTER 2

PALESTINIAN TRADE PATTERNS AND TRADE POLICIES

2.1

Introduction

Palestinian trade patterns in the 1967-1993 period were completely determined by Israeli trade policies. All tariffs, purchase taxes, and other requirements applied to Israeli imports were automatically applied to the WBGS. Since 1994, the guiding principle for Palestinian trade policy has been defined in the Protocol on Economic Relations between Israel and the PLO. The Protocol formalizes the de-facto Customs Union that had existed while the WBGS was totally under Israeli control. The main feature of the trade regime stipulated in the Protocol was the provision of free access for Palestinian goods to the Israeli market (and vice versa) and to maintain policies governing imports from third parties under Israeli control. However, the Protocol also provided the PA with some degree of autonomy to set its own trade policies. Specifically, it created three lists of goods (A1, A2, and B)21 for which the PA is free to set its own tariff rates. Tariffs on all other products not included in the lists were to be set at least as high as Israeli tariff rates. In terms of granting the PA some autonomy over it trade policies, the potential of the Paris Protocol has been somewhat limited since Israeli tariffs on the products included in the three lists are already either zero or very low. Moreover, closures of the WBGS and custom clearing delays caused by security checks have resulted in high transaction costs for both Palestinian exporters and importers (Diwan and Sha’aban, 1998). These factors have not helped in creating an independent Palestinian trade regime, and Palestinian trade policies have been identical to Israelis trade policies.

21

List A1 and A2 are mostly agricultural imports from Arab countries but with Quantitative Restrictions which are determined by the consumption needs of the Palestinian economy. List B includes mostly equipment goods which the PA can import without any restrictions as long as they meet Israeli standards and specifications.

21

This chapter examines Palestinian trade patterns and trade policies. It outlines the provisions stipulated in the Paris Protocol as a means of improving the trade relations between the WBGS and Israel, and looks at the trade agreements with the European Union. In recent years, although net transfers from abroad22 have partly offset the effects of large trade deficits with Israel, the increasing use of Non-Tariff- Barriers – such as closures and custom clearing delays – have led to a substantial increase in transaction costs. This chapter also describes the Revenue Clearing System in place between the PA and Israel.

2.2

Palestinian Trade Patterns

2.2.1 External Merchandise Trade Palestinian trading partners can be divided into four groups: Israel; Free Trade Area countries; Arab countries; and the rest of the world. As shown in Table 2.1, the most important trading partner is Israel which is the main source of Palestinian imports. Israel also absorbs almost 89% of Palestinian exports resulting in a trade deficit of $1.65 billion. TABLE 2.1 PALESTINIAN FOREIGN TRADE PATTERNS FOR 1998, ($US million) Country/Region

Israel FTA Countries23 Arab Countries24 RoW TOTAL

Exports

Imports

Share of Total Exports

Share of Total Imports

Net Trade Balance

704 3 21 1 729

2,357 421 197 168 3,053

96.6% 0.4% 3% 0 100%

77.2% 13.8% 3.5% 5.5% 100%

-1,653 -418 -86 -167 -2,324

Source: Palestinian Central Bureau of Statistics, 2000. Foreign Trade Statistics – 1998: Main Results. Ramallah, West Bank.

The next source of imports are those countries which have signed Free Trade agreements with Israel which supply about 14% of Palestinian imports, 22

These are primarily foreign aid, remittances, and wage income from Israel ($1.8 billion). Free Trade Area countries are those countries which have signed Free Trade Agreements with Israel. They include the EU countries, USA, Canada, The Czech and Slovak Republics, Turkey, Hungary, Poland, and Slovenia. 24 Arab countries include Jordan, UAE, Saudi Arabia, Yemen, Qatar, Morocco, Syria, Iraq, Lebanon, and Egypt. 23

22

but only about 0.5% of Palestinian imports are destined to these countries, resulting in a trade deficit of $418 million. Arab countries do not play a major role in Palestinian trade patterns primarily because of the absence of diplomatic relations with Israel. Palestinian trade with Arab countries is only about 3% of total trade, and a large part of this is with only two countries, Jordan and Egypt. These goods enter the WBGS through the various crossing points between Jordan and West Bank on the one hand , and Gaza Strip and Egypt on the other. Finally, almost 6% of imports originate in the rest of the world, which enter the WBGS through Israel.

TABLE 2.2 TOTAL VALUE OF PALESTINIAN IMPORTS AND EXPORTS BY SITC Rev.3, 1998 ($US million). Product Division Food and live animals Beverages and Tobacco Crude materials (inedible) except fuels Mineral fuels, lubricants and related materials Animal and vegetable oils, fats, and waxes Chemicals and related products Manufactured goods classified mainly by material Machinery and Transport equipment Miscellaneous manufactured articles

Total

Exports

Imports

117 47 16 10 5 43 306 40 145 729

580 122 92 519 30 244 732 458 462 3,050

Source: Palestinian Central Bureau of Statistics, 2000. Foreign Trade Statistics – 1998: Main Results. Ramallah: West Bank. Table 13, p48.

This one-sided Palestinian Trade structure has resulted in an overall trade deficit of $2.35 billion in 1998. As Table 2.2 shows, Palestinian exports consist mainly of agricultural goods, stone, clothing, and basic metal products. These exports reflect Palestinian comparative advantage vis-à-vis Israel in the present circumstances, and account for approximately 60% of total Palestinian exports. Palestinian imports consist of food and live animals, mineral fuels and chemicals, manufactured goods, and machinery and equipment. Figures for trade between WBGS and Israel are subject to some degree of uncertainly because not all trade flows are registered due to porous border, especially between Israel and the West Bank. Furthermore, it is not always possible to accurately deter23

mine the extent to which imports from Israel represent genuine Israeli value added or whether Israel merely serves as a transit country for imports from the rest of the world. 2.2.2 Net Transfers from Abroad The Palestinian trade deficit is financed by wage income earned in Israel, foreign aid, and remittances from abroad. About 120,000 Palestinians25, or 22% of the labor force, commute daily to work in Israel. Given an overall median daily net wage of $25 in Israel, these workers received $812 million in 1998. Most of these workers are employed in the construction, agricultural, and service sectors in Israel. Although the daily flow of these workers has sometimes been restricted due to closures of the WBGS, they continue to be a vital source of income. Moreover, Israeli producers have been increasingly dependant on this pool of inexpensive semi-skilled labor. In recent years, the Israeli government has tried to substitute foreign workers from Eastern Europe and the Far East, but the elasticity of substitution between Palestinian and foreign workers is low, about 0.826. The final economic choices on the trade side, ranging from a Customs Union to a Free Trade Area, will affect labor mobility between the WBGS and Israel. Under the existing Customs Union arrangement, approximately one-fourth of the Palestinian labor force commutes daily to jobs in Israel. Palestinians entering Israel for work must pass through border crossings with valid work permits, subject to strict border controls along the perimeter of the Gaza Strip, but less stringent monitoring in the West Bank due to its porous border. Periodic border closures because of security concerns give rise to spikes in unemployment as the large pool of commuting Palestinian workers (with or without permits) is kept for entering Israel. 25

Palestinian Central Bureau of Statistics, 1999. Labor Force Survey: Annual Report: 1998. Ramallah: West Bank. 26 See Balian, O., 1997. “The PAL/FOE Model: The elasticity of substitution between Palestinian and Foreign Labor.” Advanced Econometrics Research Paper, American University, D.C.

24

Another source of finance of the trade deficit has been foreign aid. Since the Oslo agreement, donor countries and multilateral organizations have committed about $4.6 billion towards the assistance of the Palestinian Development Plan. Table 2.3 shows total commitments27 and disbursements for the 19942000 period. Annual average commitments for the 1994-2000 period was $688 million and the annual average disbursement for the same period was $453 million. TABLE 2.3 TOTAL COMMITMENTS AND DISBURSEMENTS FOR 1994-2000 ($US m) Commitments Disbursements

1994 829 504

1995 652 425

1996 744 514

1997 600 553

1998 641 390

1999 685 473

2000 527 163

Source: Ministry of Planning and International Cooperation (MOPIC), 2000. Quarterly Monitoring Report on Donor’s Assistance. Ramallah: West Bank.

The Palestinian Development plan identifies four main sectors for socioeconomic development. These sectors are: Infrastructure and Natural Resource Management; Institutional Capacity Building; Human Resources and Social Development; and Productive Sector Development. For the period from 1994-2000, the Infrastructure sector received $1,023 million (34% of total disbursements), $878 million (29% of total disbursements) was spent on Human Resources and Social Development, Institutional Capacity Building received $785 million (26% of total disbursements), and $306 million (11% of total disbursements) was spent on Productive Sector Development. The largest part of assistance was committed to the following five subsectors: Water and Sanitation ($828 million), Institution Building ($697 million), Education ($456 million), Infrastructure ($367 million), and Health ($359) million). In terms of individual categories, 60% of total committed assistance was allocated to Public Investment ($1.9 billion) and Technical Assistance ($1.045 billion). The other leading categories are Budgetary Assistance ($488), Equipment ($411 million), and Other Assistance ($302 million). Overall, about 80% of 27

Commitments imply donor financial obligations that are allocated to specific programs and/or projects based on signed protocols, memoranda, and agreements. Disbursements mean the total amount of previously committed funds paid out from donors’ headquarters to the implementing agencies or directly to the beneficiary.

25

assistance committed is in the form of grants and the remainder 20% is in the form of loans or guarantees. The Palestinian trade deficit is also financed by remittances from abroad. The 1948 and 1967 conflicts in the region have created a large Palestinian Diaspora abroad, especially in neighboring Arab countries. As Table 2.4 shows, a large number of Palestinians live in Jordan, Lebanon, and Syria. Most Palestinians living abroad have been assimilated in the host country populations but a sizable proportion is still living in refugee camps. Remittances from abroad have been estimated at a monthly average figure of $21 per household28. Given 425,000 households in the WBGS, total remittances for 1998 are calculated at $107 million. This is also the figure used in the Palestinian SAM. TABLE 2.4 PALESTINIAN POPULATION IN THE WORLD, 1998. Country/Area

Population 1,094,000 1,113,000 1,972,000 2,434,000 450,000 487,000 51,000 287,000 40,000 110,000 78,000 6,000 210,000 267,000 8,599,000

Israel Gaza Strip West Bank Jordan Lebanon Syria Egypt Saudi Arabia Kuwait Other Gulf countries Iraq and Libya Other Arab countries The Americas Other countries Total

Source: Palestinian Central Bureau of Statistics, 1999. Population in the Palestinian Territories, 1997-2025. Ramallah: West Bank.

2.2.3 Non-Tariff Barriers and Transaction Costs The existing trade arrangement between Israel and the PA is based on the principle of non-discrimination. This principle eliminates all tariff barriers between the two economies, and stipulates that Palestinian tariff rates on imports 28

Palestinian Central Bureau of Statistics, 1998. Palestinian Expenditure and Consumption Survey (PECS). Ramallah: West Bank. Table 1, p39. Annual remittances received per household is estimated at US $252.

26

from third countries should be as high as Israeli tariff rates29. However, in practice, stringent security checks and closures of the WBGS have given rise to a new class of non-tariff barriers, namely, Non-Traditional Non-Tariff Barriers (NTNTBs)30. The existence of NTNTBs is primarily due to the absence of independent Palestinian trading routes, except for a small volume of trade flows between the West Bank and Jordan, and between the Gaza Strip and Egypt. In such a situation, Israel acts as a transit country for Palestinian imports from third countries and considerably increases transaction costs faced by Palestinian producers. Although these are economies of sale to be gained by importing through established and efficient Israeli ports, these scale economies are more than offset by high transaction costs associated with stringent security checks and custom clearing delays. High transaction costs are also reflected in the fact that trade and transport costs add 35% to the cost of domestic tradable goods, according to the Supply and Use Tables published by the PCBS. This has to be compared with a premium of 10% in the Rest of the Middle East Region31 (GTAP, 1999). Although the source of these high transaction costs cannot be determined with accuracy, a study carried out by the Federation of Palestinian Chambers of Commerce in 1998 concluded that transaction costs on average are 30% higher for Palestinian exporters and importers than for their Israeli counterparts. Other reasons for high transaction costs may be a deficient infrastructure and inefficient administrative procedures.

29

Except for products included in Lists A1, A2, and B in the Paris Protocol, on which the PA is free to set its own tariff rates. 30 The “double” NT is appropriate since theses barriers restrict Palestinian products twice – once in the traditional sense through standards and specifications, and a second time due to closures of the WBGS. For a detailed explanation on the measurement of non-tariff barriers, see Deardorff and Stern, 1997. “Measurement of Non-Tariff Barriers.” OECD Economics Department Working Paper No. 179. Paris: OECD. 31 This region is comprised of Israel, Jordan, Lebanon, Syria, and the Arab Gulf States.

27

2.3

Palestinian Trade Policies

2.3.1 The Paris Protocol and EU Agreements Economic relations between the PA and the State of Israel are governed by the Protocol on Economic Relations which was signed in Paris on April 29, 1994. The Protocol is based on the premise that building a sound economic base for relations between the PA and Israel will increase their interest in achieving peace. Furthermore, the two parties recognize each other’s economic ties with other markets and the need to create a better economic environment for their peoples. The Protocol states that each party views economic matters as one of the key elements in relations between them with a view to enhancing their interests in achieving a “just, lasting, and comprehensive peace”. This peace is to be based on the concept of mutual respect of each other’s economic interest. The sovereignty of the Palestinian side in economic policy and decision making is clearly recognized. The Paris Protocol has several major declared aims with regards to the Palestinian economy. First, the Protocol implicitly recognizes the need to rectify WBGS’s skewed economic relationship with Israel. It aims “to establish a sound economic base for these relations to be governed by principles of mutual respect of each other’s economic interest, reciprocity, equity and fairness”. Second, the Protocol explicitly addresses the need to equip the PA with the legal, institutional and procedural means to manage and develop Palestinian external trade (and other areas) within a Palestinian perspective of the future shape of the economy. It lays the groundwork for strengthening the economic base of the Palestinian side and for exercising its right of economic decision-making in accordance with its own economic development plans and priorities. Third, the Protocol recognizes each party’s distinct economic interests. Recognition of each other’s economic ties with other markets is a significant feature in the case of the Palestinian economy, especially when over 90% of its trade is conducted 28

with Israel. Fourth, the Protocol establishes a free trade area where the flow of goods and services is uninterrupted by political borders. Inherent within the Protocol was the establishment of a semi-Customs Union between the PA and Israel. Finally, these points were re-affirmed in the September 1995 Interim Agreement, in the Protocol concerning Israeli-Palestinian cooperation programs. The Principles of Economic Cooperation emphasize that cooperation is targeted, inter alia, at developing the Palestinian economy’s infrastructure and strengthening capabilities on the Palestinian side for independent and institutional economic decision making processes. With respect to trade agreements between the PA and Israel, the Protocol consists of four major elements. First, the abolition of all tariffs and duties between Israel and the WBGS, but Quantitative Restrictions (QRs) on Palestinian exports to Israel of melons, cucumbers, tomatoes, potatoes, eggs, and poultry. Second, on products classified in Lists A1 and A2 – consisting of agricultural and food imports from Jordan, Egypt and other Arab countries – the PA can set its own tariff rates but is bound by QRs, volumes of which are determined by estimates of the consumption needs of the Palestinian economy. Third, on products on List B – mostly consisting of basic food items and other goods used in construction, industry, agriculture and government development programs – the PA is free to import any quantity and set its own tariff rates. For all other products not included in Lists A1, A2, and B, there are no QRs but Palestinian tariff rates should be at least as high as the Israeli rates. Finally, the PA is free to set its own custom duties on motor vehicles. The agreement with the EU is formally known as “The Euro-Mediterranean Interim Association Agreement in Trade and Cooperation between the EU and the PLO for the Benefit of the WBGS”. It came into effect on July 1, 1997 and encompasses 75 articles. It s explicitly recognized as an interim agreement and that it should be replaced by a Euro-Mediterranean Association Agreement as soon as conditions permit. It establishes the conditions for the progressive liberalization of trade between the WBGS and the EU. It also stipulates a transitional 29

period, not exceeding beyond 31 December 2001, in conformity with the multilateral agreements on trade in goods annexed to the agreement establishing the World Trade Organization. With respect to industrial products, all import tariffs between the EU and the WBGS will be completely eliminated over a five year period. A Joint Committee is to be established to settle any trade disputes that may arise between the two parties, and products originating in the WBGS shall be granted favorable treatment. With respect to agricultural products, taking into account the volume of trade between the EU and the WBGS and of the particular sensitivity of such products, the Joint Committee shall examine product by product, and in an orderly and reciprocal basis, the possibility of granting each other further concessions. On the issue of dumping, if it can be justified by either party on the grounds of causing injury to a specific industry or sector, then the affected party can take retaliatory actions. For example, to protect its infant industries or to protect sectors undergoing restructuring and experiencing serious difficulties, particularly where those difficulties entail severe social hardships. The agreement shall not preclude the maintenance or establishment of Customs Unions or Free Trade Areas or other arrangements with other countries, except insofar as they alter the trade arrangements provided for in the agreement. Moreover, consultations between the EU and the WBGS shall take place within the Joint Committee concerning agreements establishing Customs Unions or Free Trade Areas and, where required, on other major issues related to their respective trade policies with third countries. In particular, in the event of a third country acceding to the EU, such consultation shall take place so as to ensure that account can be taken of the mutual interests of the parties. Finally, the agreement safeguards Intellectual Property Rights, encourages capital mobility, and specifies product standards.

30

2.3.2 Average Tariff Rates The total amount of import taxes that accrues to the PA in 1998 was $128 million, according to figures form the Palestinian Ministry of Finance. This figure also includes revenues from regular tariffs on imports, as well as purchase tax32 levied on imported goods from third parties, and revenues from so-called protection tax, which his marginal. This corresponds to a total average implicit import tax of 16.6% on goods imported from third parties. The combined effect of tariffs and purchase taxes creates a sizable wedge between world market prices and the prices which consumers and producers in the WBGS face. Moreover, the wedge between domestic and world market prices is exacerbated by the fact that the effective rate of collection of value added tax (VAT) is significantly higher on imports from third parties than on domestic production. We do not know the effective VAT collection rate on imports fro Israel. We only know the value of VAT revenue which is cleared by Israeli authorities. This amount, approximately $200 million, corresponds to the net refunding of VAT paid by Palestinian enterprises on imports from Israel. It is assumed that the effective rate of collection of VAT on imports from Israel is the same as that for imports from third parties, that is 8.6% for agricultural goods (fruits and vegetables are exempted) and 12.8% for other goods and services. Furthermore, as not duties are levied on imports from Israel, a preference is obviously given to Israeli products in comparison to imports from third parties. In order to derive implicit import tax rates across countries of origin and across products, we use data on individual transactions in 1998 obtained from the Customs Department of the Palestinian Ministry of Finance. Data in Table 2.5 include information on the country of origin, the type of product identified by 8-digit Harmonized System Codes, the value of the transactions, and the tariffs and purchase tax levied on the good. 32

The purchase tax is considered here as an import tax. Although purchase tax is in theory levied on both imported and domestically produced goods, the delineation of the tax base is such that in practice only goods not produced domestically are subject to the purchase tax. Indeed, the revenue generated by the purchase tax is small – only $200,000 in 1998.

31

TABLE 2.5 TAXES ON IMPORTS BY COUNTRY OF ORIGIN, 1998(%) Region

Tariffs

Purchase Tax

VAT

Total

Arab Countries 11.1 0.5 10.5 23.4 FTA 2.1 11.3 12.2 27.2 Israel 0 0 12.6 12.6 RoW 10.7 11.1 12.5 37.2 Source: Own calculations based on data from the Ministry of Finance and the Palestinian Central Bureau of Statistics (PCBS).

It is interesting to note that although the average tariff rate is lowest for goods originating from Free Trade countries, 2.1%33, the preference given to low tariffs is offset by a high purchase tax, 11.3% on average. This reflects the composition of products imported from these countries, which is characterized by a large share of equipment goods and durable consumer goods where purchase taxes are relatively high. On the other hand, the burden from tariffs is significant on imports from the Arab world, whereas hardly any purchase tax is levied on these products. The total implicit tax on imports from the Arab world is of the same order of magnitude as that for imports from Free Trade countries. We also note that imports from RoW are particularly discriminated. As shown in Table 2.6, there is substantial dispersion of tariffs and purchase taxes across products. The total implicit import tax ranges from virtually zero on dairy products to around 50% on tobacco and beverages, with a standard deviation of 11.5%.

TABLE 2.6 DISPERSION OF IMPORT TAXES ACROSS ECONOMIC ACTIVITIES, 1998 (%) Product Crops Other agriculture Mining Other food products Olive products Vegetable products Meat and dairy products

Average tax on imports (%) 21 10 0 9 0 8 0

33

Share of imports from third parties (%) 14 19 100 35 100 84 100

The average tariff on goods originating from Free Trade countries exceeds zero because not all goods, notably agricultural goods, are included in the free trade agreements. The average tariff on agricultural products reached 22% in Israel in 1999. Average tariffs on dairy products exceeded 95%. The Israeli agricultural sector also benefitted from large domestic support measures as well as considerable NTBs.

32

Animal foods Beverages and tobacco Textiles Clothes Leather products Wood products Paper products Printing Chemical products Rubber Non-metallic products Stones Basic metal products Equipment goods Furniture Other manufacturing

6 49 14 25 17 8 2 8 6 11 17 11 20 29 4 25

9 18 37 47 22 21 61 89 14 37 27 35 35 44 82 26

2.3.3 Revenue Clearance System Most customs clearing of goods destined for the WBGS takes place at Israeli points of entry and is carried out by Israeli customs officials. The accrued amount of import taxes is subsequently transferred to the PA through the socalled Revenue Clearance System (Jawhary and Sha’aban, 1995). Almost 90% of total revenues from import taxes received by the PA comes from the clearance system. Customs declarations of imports into the Gaza Strip through Rafah and Erez crossing points, and into the West Bank via the Allenby Bridge (from Jordan), are overseen by Palestinian officials and revenues are immediately collected by the PA, but the total amounts are only minor. The Revenue Clearance System is based on the principle of ‘final destination’ which implies that only levies on imports from third parties whose official destination is clearly marked ‘West Bank’ or ‘Gaza’ are transferred. On the other hand, purchase taxes collected on imported Israeli goods and on indirect imports (imports from third parties that are ultimately consumed in the WBGS, but imported through an Israeli agent) has until now not been included in the Revenue Clearance System34. 34

As of August 2000, this problem has been partly solved within the framework of the Joint Economic Committee by reaching an agreement, in principle, on the rebate to the PA of the purchase tax collected on some Israeli products exported to the WBGS.

33

CHAPTER 3

REVIEW OF THEORY OF CGE MODELS

3.1

Introduction

In most interesting policy settings, economic theory provides guidelines for justifying if a policy change will be beneficial or not for some economic agent, but often cannot give a definite answer. A CGE model provides a quantitative evaluation of the effects of government policies and presents interactions in the economy in a consistent manner. If something is changed in one part of the economy due to a government policy, then there will be effects on other parts of the economy, and these are automatically taken into account when one computes effects using a general equilibrium model. In recent years, CGE modeling has been used extensively in the analysis of trade policy issues35. These models can be multi-country models to analyze the effects of regional issues, or single country models to analyze the effects of specific trade policies on individual countries. CGE models have also been used to analyze the effects on macro, industry, labor market, distributional, and environmental variables, by modeling changes in taxes, public consumption, social security payments, environmental policies, union behavior, and levels of natural resource deposits. Partial equilibrium analysis treats different sectors and economic agents (households, firms, government, and the ‘rest of the world’) as distinct economic entities and looks at market conditions in each market in isolation. In general equilibrium, we need to incorporate equilibrium links among production structures, incomes of various groups, and the pattern of demand. General equilibrium describes various interrelated markets and indicates the degree of interdependence in a regional economy. Also, a mutual determination of market outcomes (prices and quantities) in many interrelated markets or sectors of a re35

See Francois and Reinert, eds., 1998. Applied Methods of Trade Policy Analysis. Cambridge: Cambridge University Press.

34

gion is emphasized (Hertel, 1985). This requires examination of all the effects of a change in one market on all other markets. This chapter starts with the definition and classification of CGE models used in policy analysis in recent years. The core CGE model is then presented with an emphasis on the production process, commodity markets, and factor markets and incomes. Different functional forms are introduced to show the ‘nested’ structure of production and consumption functions in a typical CGE model. The next section outlines the structure of a typical CGE model and takes a closer look at the elasticity values and the calibration procedure. This section also draws together the common features of ‘closure rules’ adopted in trade – focused CGE models. The chapter is concluded with a review of some of the more prominent trade policy CGE models constructed in the last decade.

3.2

Computable General Equilibrium Models 3.2.1 Definition and Classification

A CGE model is a large set of demand and supply functions that cover every market, both for goods and factors of production36. The demand side of commodity markets consists of households, government agents, and firms. Some of these agents are domestic, and some are foreign and therefore capture export demand. Private households are able to buy only as much as their income allows them. They receive this income as they sell labor services to firms, but they also receive a return on any capital investments that they have made. As private households sell their labor and capital services to firms, it enables them to produce. In addition to buying these primary factors of production, firms also buy intermediate inputs from each other. Commodity purchases by all agents comprise of both imported and domestically produced goods. The gov-

36

Most applied general equilibrium models are numerical analogues of traditional two-sector general equilibrium models developed by Meade, Johnson, Harberger, and others in the late 1950s.

35

ernment, in addition to participating in the regulatory process, has a number of tax and subsidy instruments available for redistributive uses37. Before going into the classification of CGE models, it is worthwhile to consider the family tree of economic models. Traditional macroeconomic models can be considered as a crossbreed between Vector Auto Regressive (VAR) and CGE models. The VAR models have rich statistical content but almost no economic content – basically one tries to find a pattern in the data that is subsequently explained by economic phenomena. In other words, the modeling strategy goes from data to theory38. In CGE models, it is the other way round – one starts with the theoretical model and then finds data that fit the construct. CGE models have a theoretical depth but they take a very liberal view on statistical methodology39. The three different classifications40 of CGE models in the literature are sketched in Figure 3.1. The main classification of CGE models is based on the historical development of CGE modeling and the intended purpose of CGE models. The following two lines of development may be distinguished. The first type of CGE models evolved from multi-sector analysis and the macro models of the 1970s. These macro CGE models became especially popular for policy analysis in developing countries. The second type of CGE models, Walrasian CGE models, evolved from the general equilibrium framework of Walras and the pioneering work of Scarf (1967) on the numerical solution of a Walrasian system. Among researchers working on macro CGE models, it has become common to classify their models based on so-called ‘closure rules’. This second classification is based on the main theoretical differences in the earlier macro models. Often these authors base their choice of closure rules on the debate among economists which dominated economics in the 1960s and 1970s. They 37

For an introductory exposition, see Shoven and Whalley, 1984. “Applied General Equilibrium Models of Taxation and International Trade.” Journal of Economic Literature 22: 1007-1051. 38 In VAR models, one tries to explain a vector of variables as a function of their own lagged values. 39 Traditional econometric models are located somewhere in between, drawing on both statistical methods as well as some economic theory. 40 See Thissen, M. 1998. “A Classification of Empirical CGE Modeling. SOM Research Report 99C01. The Netherlands: University of Groningen.

36

consider the Walrasian model as a specific sub-group of all CGE models, namely those with a neo-classical closure41. FIGURE 3.1 CLASSIFICATION OF CGE MODELS Macro

Other closures

Walrasian

Neo-classical closure

Econometric

SAM based

Source: Thissen, 1998, p4.

Besides this theoretical classification of CGE models, there exists also a third classification based on the technique used to determine the parameters of the model equations. This classification distinguishes between models with parameters based on the calibration technique and models with parameters based on econometric estimation. It should, however, be noted that although most CGE models use some parameters which are econometrically estimated, almost all models are based on the calibration technique. 3.2.2 Summary of the Core CGE Model In the core CGE model42, production creates demand for value-added (primary) factors and goods and services used as intermediate inputs. Interme41

For models used in economies of transition, see Zalai, E., “Computable General Equilibrium Modeling and Applications to Economies in Transition.” Center for Economic Reform and Transformation (CERT). Edinburgh: The Heriot-Watt University. 42 This section closely follows the regional model described in Schreiner at al, 1999. Computable General Equilibrium Modeling for Regional Analysis. Regional Research Institute, West Virginia University.

37

diate inputs consist of both imports and domestically produced goods and services. Demand for value-added factors interacts with available factor supplies to determine factor prices. Margins, such as taxes and transportation costs, increase factor costs to firms, which in turn increase product prices. Factor prices and factor ownership determine personal income, which in turn influences the demand for domestic and imported goods. Equilibrium occurs at prices which equate the demand for goods and services with supplies, and the demand for factors with factor supplies. The specification of the production system in the core CGE model is guided by neo-classical theory. Production is represented as a multi-level or ‘nested’ production process43. The first level of the three-level production process characteristic of most CGE models is represented by a Leontief (fixed inputoutput coefficient) production function with non-substitutability between valueadded (primary) and intermediate inputs. For a single sector, the Leontief production function is represented by:  VA V  Xi = min  i , i   a0i a1i 

(3.1.1)

Where Xi is gross output of sector i, VAi is composite primary input, Vi is composite intermediate input, and a0i and a1i are the respective input-output coefficients for primary and intermediate inputs in sector i. By rearranging terms in equation (3.1.1), we obtain the parameters of the Leontief production function44:

a0i =

VA i Xi

and a1i =

Vi Xi

43

(3.1.2)

To allow for differing elasticities between sets of factors, nested production functional forms are used in a CGE model, with each level containing a different set of factors and their own corresponding elasticities of substitution. That is, the use of nested structures allows for use of both fixed-coefficients and price responsiveness in the Constant Elasticity of Substitution (CES) production function. 44 Note that a0i and a1i can be easily calculated from the data set in the SAM. See section 3.2.3 below.

38

At the second level of production, nesting allows different treatment of intermediate inputs from that of value-added inputs. Substitution between value added inputs – labor, capital, and land – are commonly represented by CobbDouglas or Constant Elasticity of Substitution (CES) production functions45. For simplicity, we adopt the Cobb-Douglas production function to represent the second level of production: L

K

T

VA i = ΓiVALABiαi CAPiαi LANDiαi

(3.1.3

where αLi + αKi + αiT = 1 ⇒ CRTS and LABi, CAPi, LANDi are labor, capital, and land inputs for sector i, coefficient

ΓiVA>0 is total factor efficiency parameter for composite value added inputs, and parameters αiL, αiK, and αiT are the production elasticities corresponding to labor, capital, and land respectively. By substitution and rearranging the terms in equations (3.1.2) and (3.1.3), sectoral gross output, Xi, can be expressed in the Cobb-Douglas forms: L

L

T

Xi = ΓiαLABiαi CAPiαi LANDiαi where

Γiα

(3.1.5)

ΓiVA = a0i

The profit function can be written as πi = (PNi Xi ) - (PL.LAB + PK i .CAPi + PT.LAND)

(3.1.6)

Where πi is profit of sector i, PNi is net output (i.e. gross output less cost of intermediate inputs and indirect taxes), PKi is capital rent (assuming capital is 45

The Cobb-Douglas production function implicitly specifies unitary substitution elasticities, while the CES is a more general case that allows different from unitary elasticities of substitution (See Varian, 1992).

39

fixed by sector), and PT is land rent. Assuming all firms in the sector maximize profits, differentiating equation (3.1.6) with respect to each of the inputs and equating to zero will give us the conditional demands for the three factors:

CAPi = αKi

PNi Xi PK i

(3.1.12)

LABi = αLi

PNi Xi PLi

(3.1.13)

LANDi = αiT

PNi Xi PTi

(3.1.14)

The second level in the production structure is represented by substitution possibilities between domestically produced and imported intermediate inputs. By the so-called Armington assumption46, goods produced in different countries are assumed to be imperfect substitutes, usually represented by a CES function. The Armington specification has three advantages over alternatives for matching the model to data on trade flows. First, it accounts for the large amount of cross-hauling or intra-industry trade, where a country both imports and exports goods of the same product category. Second, it explains the empirical observation that even at a highly disaggregated level, most countries produce goods in all product categories47. And third, it allows for changes in the relative prices of

46

Named after Armington (1969). The Armington assumption distinguishes goods by sector and by country of origin. For example, an automobile produced in Japan is a different good from an automobile produced in Germany – a close but imperfect substitute. 47 In models where goods are not distinguished by country of origin, and produced goods exceed factors of production, countries typically specialize in the production of a limited number of goods.

40

different imported goods, and allows for differing degrees of substitution among domestic and imported goods across different products48. The CES function representing the relationship between the two categories of intermediate inputs can be expressed as: 1 ρv Vji = Γ jiv δ vji VM ji j 



where

ρvj

+

=

(

)

ρv 1 − δ vji VR ji j

 ρ vj 

(3.1.16)

σ vj σ vj

and ρjv>0 is the intermediate input efficiency parameter, 0<δjiv<1 is the share parameter, VMji represents intermediate goods imported by sector i from sector j in the exporting country, VRji is the domestically supplied intermediate good for sector i from sector j, σjv is the elasticity of substitution fro sector j, and ρjv is the substitution parameter49. The following cost minimization problem is used to derive demand functions for domestically produced and imported intermediate inputs:

min

s.t

{VMji, VMji}

PMjVMji + PRjVRji

(

)

1

v Vji = Γjiv δ vji VM ji + 1 − δ vji VR ji  ρ j  

ρ vj

ρ vj

where PMj and PRj are the prices of imported and domestic intermediate inputs from sector j. Solving the first-order conditions and rearranging terms, we obtain

48

Empirical studies indicate that both of these phenomena are found in time-series data (See Shiells, Stern, and Deardorff, 1986). Neither is possible in models that aggregate all imports together, or in models that treat domestic and imported goods as perfect substitutes. 49 The value of σjv depends on the degree of substitutability between the two sources of intermediate inputs. If σjv=∞, then the two are perfect substitutes. If σjv=0, then the two inputs are used in fixed proportions.

41

VR ji  1 − δ vji  PM j    = VM ji  δvji  PR j   

σ vj

(3.1.17)

Rearranging equation (3.1.17) we get

1   v σ   VR j   ji  σ vji =  + 1  VM ji      

−1

(3.1.18)

And the efficiency parameter, Γjiv, is obtained by rearranging equation (3.1.16): Vji

Γjiv =

1

(

)

δ v VM + 1 − δv VR  ji ji ji   ji  ρ vj

ρ vj

(3.1.19)

ρ vj

A third level in the nested production structure may represent substitution among different labor skills, among different ‘classes’ of land, or among different types of capital. A common procedure is to consider the CES form which allows elasticities of substitution to differ among industries but requires the elasticity among any two sub-categories to be the same. Alternatively, sub-categories could be grouped into two parts, such as production labor and ‘all other’ labor, with one elasticity of substitution between the two, and then two different classes of production labor with a different elasticity of substitution. A final remark in the production system is on the net output price, PNi, which is defined as the domestic output price less the unit cost of intermediate inputs and unit value of indirect taxes:

PNi = PXi − ∑ a jiPj − ibt iPXi j

42

(3.1.20)

where PNi is net price of commodity j, PXi is composite domestic output price, aij is the amount of the jth commodity used in the production of one unit of the ith commodity, and ibti is the indirect tax per unit value of output50. Commodity markets comprise of both domestic and export markets. Each sector produces a composite commodity that can be exported or sold domestically. Exports and domestically sold goods are assumed to be differentiated by market, with the relationship between them represented by a Constant Elasticity of Transformation (CET) function. Price ratios and elasticities of transformation determine the levels of output exported and sold domestically. The substitution possibilities are represented by:

[

x Xi = Γi δixEXPiρi

x

+

where ρix =

(

1 − δix

1 ρix ρ x Ri i

)

]

(3.2.1)

σ ix − 1 σ ix

and where Xi is total output of sector i, Γix>0 is the output efficiency parameter, 0<δi<0 is the share parameter, EXPi is sector i’s supply for export markets, and Ri is sector i’s supply to the domestic market. σix is the elasticity of transformation for industry i, and ρix is the output substitution parameter. The value of σix depends on the degree of transformability between the two market outlets. If

σix=∞, then the two are perfectly transformable. If σix=0, then the two markets are not substitutable and we must specify market behavior in each market. Each firm in the sector allocates its output between the domestic and export markets so as to maximize revenues subject to the CET function. Since it is assumed that the production process is the same in each market, revenue maximization may be substituted for profit maximization. Thus, for given domestic and export prices, the problem faced by the firm is to:

50

The net output price is the per unit value of output available to compensate for primary factor use. Under conditions of constant returns to scale, the sum of the marginal value products for all primary factor use should exactly equal the net output price. See equation (3.1.6).

43

max {EXPi , Ri} s.t.

[

x Xi = Γi δixEXPiρi

x

PEi EXPi + PRi Ri +

(

1 − δix

1 ρix ρ x Ri i

)

]

where PEi and PRi are, respectively, prices of exported and domestically sold commodities from sector i. Solving the first-order conditions and rearranging terms, we get:

 1 − δix  PEi  Ri  =  x  EXPi  δi  PRi 

− σ ix

(3.2.2)

Rearranging terms, we get:

1   − x σ   R i   δix =  i  + 1 EXPi   

−1

(3.2.3)

In addition, from equation (3.2.1), we get:

Γix =

Xi

(

)

1

(3.2.4)

δ xEXPρix + 1 − δ x Rρix  ρix i i i   i  Household income available for expenditure is calculated as gross income

minus taxes, savings, and in this example, payments for labor employed by households: HEh = DYh – HSAVh – PL.LHh

(3.2.5)

where HEh is household expenditure, DYh is household disposable income, HSAVh is household saving, PL is the wage rate, and LHh is labor employed di-

44

rectly by households. The subscript h represents household type (low-income, middle-income, or high-income). Consumption by households is nested at two levels. At the first level, households maximize utility from leisure and consumption of composite market commodities, subject to total time (work plus leisure time), household budget constraints, and prices. At the second level, they choose optimal combinations of imported and domestically produced commodities which are perfect substitutes (Armington specification) so as to minimize their cost of purchasing predetermined amounts of commodities. Substitution between these commodity groups is captured in a CES function. We now present a detailed description of these two levels of consumption. The Linear Expenditure System (LES) is the most commonly used demand system in CGE models due, in part, to convention, and because it allows representation of subsistence consumption, in addition to satisfying the above restrictions. In the LES, demand equations are assumed linear in all prices and incomes, and the set of demand functions is express in expenditure form: n

piqi = c i + ∑ aijp j + βi y

(3.2.6)

j =1

where pi is the price of the ith commodity, qi is the quantity of the commodity demanded, ci is the ith intercept, aij are the price parameters, βi is the marginal budget share for the commodity, and y is household income. Empirically, the LES is derived from constrained maximization of the Klein-Rubin (also known as the Stone-Geary) utility function, whose general form is given by:

U = ∑ βi ln(Qi − γi )

where

∑ βi = 1 45

(3.2.7)

and where U is utility level, Qi is the level of commodity consumption, and γi, if positive, is the subsistence level of consumption as perceived by the consumer. Given a fixed level of household income that can be allocated to consumption, HEh, the consumer faces the following constrained maximization problem: n

max {Qih}

U(Qih ) = ∑ βih ln(Qih − γih ) i =1 n

s.t.

HEh − ∑ PiQih = 0 i =1

Solving the first-order conditions of the Lagrangian to the above optimization problem produces the following result: n   PiQih = Piγih + βih HEh − ∑ Pjγ jh  j =1  

(3.2.11)

The first term on the RHS of equation (3.2.11) is the subsistence level of consumption, and the second term is the ‘rest’ of consumption expenditure. The first derivate of equation (3.2.11) with respect to total expenditure, HEh, is the marginal budget share, βih. The LES is obtained by dividing equation (3.2.11) by Pi:

β Qih = γih + ih Pi

n   HEh − ∑ Pjγ jh    j =1

(3.2.12)

To evaluate equation (3.2.12) we need values for γih and βih, prices, and total consumption expenditure data. Since γih cannot be estimated directly from empirical data, and since βih cannot be calculated from a one-period dataset, equa-

46

tion (3.2.12) is often implemented using a simplified version of the Stone-Geary LES given by: Qih = βih

HEh Pi

(3.2.16)

Note that equation (3.2.16) is based on very restrictive and somewhat unrealistic assumptions. It implies that income elasticities of demand are unitary for all commodities. Although the results are not appropriate for dynamic analysis, this assumption does not pose serious problems for comparative static analysis, particularly if expenditure patterns for several household income groups are embodied in the CGE model. The second level of household commodity demand involves determination of the minimum cost combination of domestic and imported commodities. For each commodity i, substitution between the two sources is captured in the following CES function:

Qih = ΓiQ

[

Q δiQQMρihi

+

(

1 − δiQ

where ρiQ =

1 ρiQ ρQ QRih i

)

]

(3.2.17)

σ iQ −1 σ iQ

and where ΓiQ>0 is the household consumption efficiency parameter, 0<δiQ<1 is the share parameter, QMih is household demand for imports, QRih is household demand for domestic goods, σiQ is the elasticity of substitution, and ρiQ is the substitution parameter. Government and capital formation are the two remaining commodity markets. Quantity demanded is assumed exogenous for each of these markets. However, price is endogenous and, hence, expenditure by the government and for capital formation varies with price. Similar to intermediate inputs and house-

47

hold demands, imports and domestically produced goods are imperfect substitutes in meeting the composite commodity demands. Exogenous commodity demand for the government, QGi, and capital formation, QCi, from the two sources is given by the following CES function:

ρ gc QXi = Γigc δigc QXMi i 



where

+

(

1 − δigc

ρigc

=

1 ρigc  ρigc QXRi 

)



(3.2.18)

σ igc − 1 σigc

and where QXi = QGi + QCi . QXMi is quantity imported, and QXRi is quantity produced domestically. All parameters are identified similar to those for equation (3.1.16). Solutions to quantities imported and domestically produced goods are similar to equations (3.1.16) to (3.1.19). Commodity purchases prices (consumer prices) are a composite of domestic and imported prices:

Pi =

PRiRi + PMiMi Ri + Mi

(3.2.19)

The composite purchases price, Pi, is the unit value for household consumption goods, intermediate inputs, and institutional purchases. PRi is domestic purchase price, and PMi is the import price. Ri is the total amount of commodity i produced and consumed, and Mi is the total amount of the imported commodity. Commodity output prices (producer prices) are a composite of domestic and export prices:

PXi =

PRiRi + PEiEXPi Ri + EXPi

48

(3.2.22)

The composite output price, PXi, is the weighted unit value of revenue received from domestic and export sales. PRi is the domestic price, and PEi is the export price. Ri is the domestic quantity, and EXPi is the export quantity. To attain commodity market equilibrium, we equate total commodity supply to total commodity demand. Total commodity demand is the sum of intermediate demand, institutional demand, and export demand. Total commodity supply is the sum of domestic output and imports. Market equilibrium of community i is given by: Xi + Mi = TVi + TQi + QXi + EXPi

(3.2.23)

where Xi is domestic output, Mi is imports, TVi is total composite intermediate input demand, TQi is total composite household demand, QXi is total composite exogenous commodity demand (government plus capital formation), and EXPi is total export demand. Factor markets and factor incomes in CGE models are usually analyzed in a perfect competition setting. Although we have discussed the demand for factors by firms, we also need to discuss the demand for factors by institutions (government and households). In the CGE framework, primary factors are studied from both the short and long run perspectives. In the short run, we assume fixed capital supply and labor mobility, and in the long run both factors are assumed to be mobile. Land is assumed to be fixed in both long and short runs. Equilibrium in factor markets is attained when quantity demanded is equal to quantity supplied. Assuming a homogeneous labor force, labor market equilibrium is expressed as: LSO = LDI + LDE

49

(3.3.1)

where LSO is total initial household labor supply, LDI is total demand for labor, and LDE is exogenous demand for labor. Note that n

LDI = ∑ LABi

and

LDE = LDH + LDG

(3.3.2)

i =1

where LDH is labor directly demanded by households, and LDG is labor demanded by the government. Total labor income, LY, is the sum of the product of labor demanded and the wage rate:   LY = PL ∑ LABi + LDH + LDG  i 

(3.3.4)

where PL is the wage rate, LABi is labor demanded by sector i, LDH is labor demanded directly by households, and LDG is labor demanded by the government. THE capital market is in equilibrium when quantity demanded by each sector, CAPi, is equal to that sector’s initial capital stock, KSOi: CAPi = KSOi

(3.3.6)

Total capital income, KY, is the sum of the product of capital demanded and capital rent:

KY = ∑ PK iCAPi

(3.3.9)

i

where PKi is capital rent and CAPi is the quantity of capital demanded by sector i. Land is assumed perfectly inelastic and the market is in equilibrium when LANDi = TSOi

(3.3.13)

50

where LANDi is land use, and TSOi is the initial quantity of land. Total land income, TY, is the sum of the product of quantity of land and rent:

TY = ∑ PTiLANDi

(3.3.14)

i

where PTi is gross land rent and LANDi is the quantity of land demanded by sector i. The source of enterprise income, ENTY, are gross capital rents:

ENTY = PK.ENTK

(3.3.16)

where PK is capital rent, and ENTK is the initial stock of enterprise capital. Most household income is derived from factor payments. Gross factor payments are subject to government taxes and capital depreciation. It is, thus, total earnings less the applicable deductions that are available for distribution to owners of factors. Other sources of household income include inter-household transfers, government transfers, and net remittances from the rest of the world. Gross household income, GHY, can therefore be represented by:

GHY = NLY + PK.HHK + NTY + HENTY + GOVTH + ROWTH (3.3.20)

51

where NLY is net labor income, PK is capital rent, HHK is capital stock owned by households, NTY is net land income, HENTY is household enterprise income, GOVTH is government transfers to households, and ROWTH is net transfers and remittances from the rest of the world. Disposable household income, DHY, is given by:

DHY = (1 − ht )GHY

(3.3.21)

where ht is the household income tax rate. Household savings, HSAV, is given by:

(3.3.22)

HSAV = hsGHY

where hs is the savings rate.

3.2.3 Model Construction, Calibration, and Closure Model construction depends on the purpose for which the model will be used. Some are multi-country models analyzing a broad range of issues, while others are country specific models designed to analyze specific issues. However, most CGE models follow a common structural pattern as shown in Fig 3.2.

52

FIGURE 3.2 FLOWCHART OF CGE MODEL STRUCTURE

SAM (Benchmark equilibrium)

Replication check

Choice of functional forms

Specification of exogenous elasticity values

Policy shock

Computation of counterfactual equilibrium

Policy appraisal

Further policy appraisal

RESULTS (Policy recommendations)

Source: Shoven and Whalley, 1995.

The starting point of a CGE model is the Social Accounting Matrix (SAM) which provides us with an initial equilibrium, or benchmark equilibrium, against which policy outcomes are compared. The next step is to describe the behavior of economic agents (households, producers, government, rest of the world) using standard functional forms. These functional forms are typically Constant Elasticity of Substitution (CES) or Cobb-Douglas forms depending on the behavioral assumptions of the models. Once these functional forms have been chosen, we need to ‘calibrate’ the model such that the solution of model equations replicates the benchmark equilibrium in the SAM. We then introduce a policy shock and compare the new counterfactual equilibrium with the initial benchmark equilibrium for policy appraisal. The behavior of economic agents in the 53

model is determined by values for elasticities and parameters of the functional forms. The exogenous elasticity values are obtained from previously computed tables for specific sectors – Central Tendency Tables – drawn from previous econometric estimates for each sector. For example, Piggot and Whalley (1985) in their UK tax model use an elasticity of substitution of 0.6 in the agricultural and fishing sectors, 0.85 in the furniture and wood sector, and 0.9 in the textiles sector. In trade models, export and import demand elasticities are of critical importance and can be extracted from the central tendency tables compiled by Stern, Francis, and Schumacher (1976). In the Palestinian CGE model, we need to determine 3 sets of elasticities: a CES production elasticity of substitution between labor and capital inputs (sigmaF); an Armington elasticity of substitution between domestic and imported goods (sigmaA); and elasticities of export demand for each sector (alphaE). These elasticity values are given below: AGRC

MANF

CONS

COMM

TRSP

PRVS

PUBS

sigmaF

3

4

2

2

2

2

2

sigmaA

5

5

2

3

3

3

3

alphaE

5

5

5

5

5

5

5

Source: Own calculations and estimates from various sources.

The CES elasticities of substitution between L and K are not the same in each sector. In the agricultural and manufacturing sectors, they are higher than in other sectors. This reflects the fact that the elasticity of substitution is higher in tradable than in non-tradable sectors since non-tradable sectors are more factor specific, especially the service sector. Similarly, the Armington elasticities of substitution between domestic and imported goods are higher in tradable sectors due to the relative ease with which Israeli and domestic Palestinian goods are substituted. Moreover, the Palestinian economy does not produce a large variety of goods to choose from. Finally, the export elasticities of demand are high and uniform across all sectors due to the very small global share of total

54

Palestinian exports and the availability of a wide range of substitutes from the rest of the world. Another important procedure in CGE model construction is the calibration of the parameters (share and efficiency) of the functional forms. Calibration can be defines as the procedure used to select the parameters of the functional forms in such a way that the model ‘reproduces’ the benchmark equilibrium data set (or Sam). For example, given the simple Leontief input-output form:  VA V  Xi = min  i , i   a0i a1i 

The share parameters are given by:

a0i =

VA i Xi

and

a1i =

Vi Xi

Since we know the values of value-added inputs (VAi) and intermediate inputs (Vi) from the SAM, we can easily obtain the share parameters. Parameters for all the functional forms in the model are calibrated in this way by using data from the SAM. Finally, an important consideration in model construction is the ‘closure rule’ adopted in the model structure. The ‘closure’ of a model determines the theoretical inclination of the model. In mathematical terms, closure specifies the exogenous and endogenous variables. For example, the neo-classical closure rule holds that investment demand is savings drive. In trade models, closure rules specify the behavior of the external sector – i.e. the specification of the export-demand and import-supply functions that the single country is assumed to face in its transactions with the rest of the world. Boadway and Treddenick (1978), in their Canadian CGE model of tariff analysis, describe a closure rule in which the rest of the world is assumed to

55

have import-supply and export-demand functions of constant price elasticity, and a zero trade balance condition. They assume that Canada is a taker of world prices for imported goods and a perfectly elastic foreign import-supply function, although domestic and imported goods follow the Armington specification of [product heterogeneity. Dervis et al (1981) also use the Armington specification along with constant price elasticity export-demand functions, import price taking behavior, and a balance of trade condition. The Deardorff and Stern (1981) multi-country model differs somewhat in that the separate economies are linked through an international price system. Each country’s export-demand function is given by the sum of import demands for the country’s exports across all other countries. However, they also use the Armington specification of product heterogeneity. Although these formulations are somewhat different from each other, they nevertheless possess certain common characteristics with respect to external closure rules. First, a balance of trade condition is always satisfied even through export-demand and import supply elasticities are separately incorporated. Second, although the models are in real terms, there is a nominal exchange rate variable. And finally, they all adopt the Armington assumption with respect to domestic demand, but foreign demand for the country’s exports is assumed to be price responsive.

3.3

Trade Policy Analysis

In recent years, there has been a noticeable surge in studies dealing with trade policy analysis using CGE models. This may be explained by the increased interest in globalization issues for both developing and industrialized countries, and by the ever-increasing involvement of the World Trade Organization in creating a rules based trading environment51. The studies have sought to

51

See Michalopoulos, C. 1998. “The Participation of Developing Countries in the World Trade Organization.” Policy Research Working Paper No. 1906. Washington D.C.: The World Bank.

56

determine the extent to which trade liberalization and trade integration have affected growth, poverty alleviation, and environmental degradation. The recent demonstrations at WTO ministerial meetings are a testament to the increasing importance of these issues. These CGE studies can be classified into two broad groups. One group deals with trade liberalization issues such as reductions in tariff and non-tariff barriers, accession to the WTO, and the effect of exogenous external shocks, while the other group includes studies concerned more with trade integration issues in the form of Preferential Trade Agreements such as Free Trade Areas and Customs Unions. Although these studies adopt different assumptions and different closure rules, the underlying structure of the CGE models used is basically the same. Some are multi-country multi-sectoral trade CGE models dealing with the effect of trade policy changes on a group of countries, while others are single-country models dealing with country-specific issues. Below we report the results of some of the more important CGE trade models developed in the last few years. In the study by Harrison, Rutherford, and Tarr (1997) on trade integration for Chile, the authors use a multi-regional and multi-sectoral static CGE model to analyze different preferential trade agreements for Chile. The structure of the model is very similar to that of the basic model of de Melo and Tarr (1992). Some of their more important results indicate that lowering Chile’s tariffs leads to only small gains since Chile starts with a rather different efficient external regime – uniform tariff rates of 11%. Because of this efficient trade regime, tariff reductions will reduce Chilean welfare through trade diversion, unless Chile can improve its market access to partner countries. The model, however, ignores dynamic gains as the result of importing a greater variety and more technologically advanced range of products. Rattso and Torvic (1998) provide a comprehensive evaluation of CGE modeling of trade liberalization in sub-Saharan Africa. The main areas of controversy are how the labor market functions, and how savings and investment 57

come into balance. They argue that under strict neo-classical assumptions (full employment and savings-driven investment), no model will predict unfavorable consequences of reform. Short-run contraction of output of importable goods leads automatically to an expansion of exportable goods, but this is not a guarantee for future productivity growth. Indeed, the deindustrialization due to trade liberalization may push the region away from badly needed sources of productivity growth. They conclude that most CGE models ignore transitional problems of trade reforms, and that the effects of trade liberalization on the rate of inflation, exchange rates, and interest rates are not easily captured in a CGE framework. The study on the economic effects of an FTA between Tunisia and the EU by Brown, Deardorff, and Stern (1996) uses the Michigan-BDS CGE model to show that the static welfare benefits for Tunisia depend on assumptions on capital mobility. They conclude that the best path for Tunisia would be to reduce its trade barriers multilaterally and reinforce this action with further liberalization of its foreign investment policies52.

52

For trade strategies of other Mediterranean countries see Petri, 1997. “Trade Strategies for the Southern Mediterranean.” OECD Development Center Technical Paper. Geneva: OECD.

58

CHAPTER 4

A PALESTINIAN SOCIAL ACCOUNTING MATRIX

4.1

Introduction

There are no blueprints for constructing a SAM for a specific country. The construction of a SAM depends on the availability of data and the purpose for which it will be used53. It is typically the case that recent data is at a higher level of sectoral aggregation than less recent data. For example, detailed input-output accounts may be available 10 years prior to the chosen base period. To solve this problem, we generally use the most recent data first for broad sectoral aggregates and then estimate more detailed sectoral transactions using shares from less recent data. This process will evolve into stages of data steps progressing from higher levels of aggregation and recent data, to progressive disaggregation with less recent data, the latter being used in share form. While tedious, this approach attempts to make the SAM as timely as possible given the available data. A SAM for trade policy analysis must be constructed in such a way that it incorporates certain assumptions of the model. One of these assumptions is the so-called Armington assumption which treats similar goods from different regions as distinct and imperfectly substitutable goods. To incorporate the Armington assumption in our model, we need to construct a SAM in which imports are classified according to sector of origin (as opposed to sector of destination). For example, the entries for imports in the SAM should tell us different types of goods the Palestinian economy imports from different countries, and not the different types of goods that a specific sector imports from the different countries.

53

The construction of the Palestinian SAM follows closely Pyatt and Round, 1985. Social Accounting Matrices: A Basis for Planning. Washington D.C.: The World Bank.

59

One important advantage of classification by sector of origin is that it accounts for cross-hauling or intra-industry trade54. Another consideration in the construction of SAMs for trade policy analysis is the consolidation of the use and make matrices into a single transaction matrix. There are two possible approaches: to eliminate the commodity account or to eliminate the activity account. The conventional practice is to eliminate the activity account since the resulting transaction matrix preserves the trade accounts in their original form by commodity (see Pyatt, 1998).

4.2

Data Sources

This chapter starts with a description of the various sources of data and its shortcomings. Most of the data used in the construction of the Palestinian SAM for 1998 is obtained from the Supply and Use Tables (SUT) published by the PCBS in 1998. There has been no attempt to estimate data in the SAM but some entries are residuals, such as foreign savings and government savings. For example, given total savings and household savings, foreign savings are the residual difference between total savings and the sum of household and government savings. It is also important to note that the definition of the geographical term Remaining West Bank and Gaza Strip for 1998 differs from the definition adopted in previous years by the PCBS. For 1998, it refers to all of the WBGS excluding those areas of Jerusalem which were annexed by Israel after the 1967 war, whereas in the pre-1998 period it referred to the West Bank excluding the whole of Jerusalem. This difference in definition is taken into consideration when comparing time-series data of this geographical area. Various other publications and sectoral surveys conducted by the PCBS are also used in the compilation of the Palestinian SAM for 1998. These are: Agricultural Statistics, 1997/1998; Bulletin of Consumer Prices; Average Prices for 54

See de Melo and Robinson, 1989. “Product Differentiation and Foreign Trade in CGE Models of Small Economies.” Country Economics Department Policy Research Working Paper No. 144. Washington D.C.: The World Bank.

60

October-December 1998; Construction Statistics; Existing Building Survey, 1998; Expenditure and Consumption Levels: The Final Report, JanuaryDecember 1998; Foreign Trade Statistics - 1998: Main results; The Industrial Survey – 1998: Main Results; The Internal Trade Survey – 1998: Main Results; Labor Force Survey – 1998: Main Findings; National Accounts – 1998: Preliminary Estimates; Producer Price Index – 1997: Preliminary Results; The Services Survey – 1998: Main Results; Transport, Storage, and Communication Survey; The Formal Sector – 1998: Main Results; and The International Standard Classification of All Economic Activities, Third Revision, 1995. These surveys and publications, together with the Supply and Use Tables, are used to compile the Palestinian input-output table. Section 4.2 describes the main sources of data from which the final Palestinian SAM is constructed. These are the Supply (or output) and Use (or input) Tables compiled by the PCBS which include data on value added, intermediate and final demand, indirect and direct taxes, and foreign trade. This section also describes other sources of data published by the PCBS such as the Economic Surveys, and compares them to the figures in the Supply and Use Tables. Using all of these data sources, an initial input-output table is constructed with values for intermediate inputs, value added, indirect taxes, imports, and final demand in seven important sectors. In section 4.3 we provide the rationale for the choice of these seven sectors and the level of aggregation in the SAM55. We note that a SAM includes the input-output matrix and much more, providing us with valuable insights on the inter-institutional transactions in the Palestinian economy. The final section tabulates and describes in detail the disaggregated macroeconomic Palestinian SAM for 1998 with 15 accounts: 2 factors of production; 4 institutions; a combined capital account; 7 production sectors; and the rest of the world (RoW) account.

55

See Lloyd, P.J. 1991. “Aggregation by Industry in General Equilibrium Models with International Trade.” OECD Technical Paper No. 50. Paris: OECD.

61

4.2.1 Supply and Use Tables The Supply and Use Tables (SUT) contain information showing the interrelationships between producers and users, thus providing a framework for checking consistency of data on flows of goods and services obtained from various statistical sources. They are also an appropriate tool for calculating economic data in the national accounts and detecting weaknesses. This is particularly important for the decomposition of values of flows of goods and services into prices and volumes, and for the calculation of integrated set of price and volume measures. The SUT were constructed by the PCBS using data from a wide range of surveys and publications. By “supply” we mean the make matrix in addition to other data, and by “use” we mean the use matrix and other data. The data in the SUT are classified according to the United Nations System of National Accounts (UNSNA) 1993 Rev. 3, using the International Standard Industrial Classification (ISIC). The Supply Table shows the goods and services supplied to the commodities account by the various activities in the Palestinian economy. It also includes data on imports and import tariffs, and shows the entries in the commodities column in the SAM, which are the make matrix, imports, and tariffs. The sum of this column is aggregate supply. The Use Table shows aggregate demand, which is divided into intermediate demand by the activities and final demand for household consumption, investment, government consumption, and exports. It is the row sum of the commodities account, or aggregate demand. The Use Table also includes data on value added, indirect taxes, and gross operating profits. It is the column sum of the activities account in the SAM, or total production cost56. Unlike in the use and make matrices, activities in the SUT are always represented by column entries and commodities by row entries. The Palestinian 56

Note that although similar, the supply and use tables are not the same as the use and make matrices in a SAM.

62

SUT for 1998 contains 97 activities and 98 commodities grouped under six major sectors: agriculture and fishing; mining and quarrying; manufacturing, electricity, water, and gas; construction; and services. Although in the SUT these sectors are classified according to ISIC at the 4-digit level, we have grouped them into seven sectors for analytical purposes. For example, the manufacturing sector consists of 22 sub-sectors but we have aggregated them into one large manufacturing sector. Similarly, the service sector consists of 11 separate sectors. The reason for this aggregation is that we are more concerned with the quantification of the effects of trade liberalization on deindustrialization. The SUT are also closely related to the standard input-output tables for a given country. Although there was no ‘standard’ input-output table for the Palestinian economy, one could easily construct one from the information provided in the SUT, which is shown in later sections of this chapter. In the SUT, activities are always represented by column entries and commodities by row entries. The inter-industry transaction matrix in the input-output table (or the use matrix in the SAM) can therefore be directly extracted from the SUT. However, to construct the make matrix, we need to transpose the ‘supply’ table in the SUT and include the corresponding entries for imports and tariffs in the input-output table. Once the input-output table is constructed, it can then be used to construct the SAM which also includes the inter-institutional transactions in the economy. The data in the SAM are classified according to ISIC under 16 main sectors and over 90 sub-sectors. However, some sectors have been consolidated and other sectors have been aggregated into single sectors. For example, the fishing sector has been consolidated with the agricultural sector, and the 11 service sectors have been consolidated into one giant service sector. Trade data are extracted from the SUT and are reconciled with data from Foreign Trade – 1998: Main Results, March 2000.

63

4.2.2 Economic Surveys In accordance with the Master Plan for developing official statistics, the Palestinian National Accounts have high priority and are being developed following the standards set down in the System of National Accounts (SNA, 1993). The National Accounts provide a systematic framework to check the quality and coverage of data gathered by means of economic statistical sources. The SNA delineates the most recent internationally accepted standards and recommendations for the establishment of a coherent set of macroeconomic accounts. It provides a set of concepts, definitions, and classifications within a broad accounting framework designed for purposes of policy making, decision taking, and economic analysis. Most of the data in the compilation of the Palestinian National Accounts are obtained from the Economic Surveys conducted by the PCBS over the 1995-2000 period. These surveys cover a wide range of economic activities classified under the International Standard Industrial Classification (ISIC) system. For activities not covered by the Economic Surveys, the figures are compiled using other statistical surveys conducted by the PCBS, as well as administrative records such as accounts of local governments in the WBGS. In very few instances, figures are also based on estimates made by other parties such as the Israeli Central Bureau of Statistics (ICBS). In addition, numbers from international organizations were used as a source of data on Central Government. The Agricultural Survey (category A in ISIC) includes both plant and livestock production. Volume data is estimated from cultivated area and productivity. The value of output is calculated by multiplying these quantities by the annual average prices - the latter were available by type of crop from the Ministry of Agriculture. As for intermediate inputs in the agricultural sector, Ministry if Agriculture officials made estimates based on information provided at the local level. Production inputs consists of fertilizers, water, pesticides, veterinary medicines, animal feeds, purchased chicks and calves, and other miscellaneous inputs including seeds and seedlings. To compile gross value added, intermediate 64

consumption is subtracted from output. Compensation of employees is based on the findings of the Labor Force Survey conducted by the PCBS for the relevant reference period. The total compensation of agricultural workers is then computed as the number of agricultural employees multiplied by the average daily wage multiplied by the number of working days during the year. Gross operating surplus was obtained as a residual given gross value added, compensation of employees and indirect taxes on agricultural production. Estimates for output from fisheries were reached with the cooperation of the Fishermen’s Association in Gaza and the Ministry of Agriculture. Both quantities of fish catches and prices were available. However, data on intermediate consumption had not been collected; therefore, the ratio of intermediate consumption to output in agriculture was applied to fisheries. This method is likely to lead to a slight bias in intermediate consumption in fisheries, and thus, to value added. Note that in the input-output table and the SAM, the fisheries sector was consolidated with the agricultural sector (i.e. category A in ISIC includes category B). Under the Mining & Quarrying (category C in ISIC) and Manufacturing (category D in ISIC) classifications it is important to note that stone quarrying is classified under category C, whereas stone cutting and shaping is classified under category D. Data on thee two sectors, in addition to Water, Electricity & Gas, are obtained from the Industrial Survey conducted by the PCBS. Gross value added in these sectors is again defined as the difference between output and intermediate consumption, while gross operating surplus is estimated as a residual, subtracting compensation of employees, and indirect production taxes, from gross value added. The Construction sector (category F in ISIC) includes all activities related to construction and maintenance of fixed structures. It encompasses site preparation, building of complete or parts of construction including civil engineering structures, and building completion and installation. It also includes rental machinery and equipment for construction. Like many other developing countries, 65

the Palestinian construction sector is divided into formal and informal sectors, with the latter making a sizable contribution to output. For the formal sector, data was obtained from the Construction Contractor’s Survey for 1998. Data for the informal sector was obtained from the Existing Buildings Survey for 1998. This survey also provided data on current maintenance expenditures on buildings in the Palestinian territories. These expenditures by owners and tenants represent output value for the informal sector. As for intermediate consumption in the informal sector, data provided by the Existing Buildings Survey was used to obtain information on the costs of materials used for construction. The ratio of these inputs to the output of building construction and capital maintenance, excluding the payments to registered contractors, was used to add up the intermediate consumption of the informal sector. With respect to the estimation of employee compensation for the informal sector, the findings of the Labor Force Survey conducted by the PCBS were used. Since this survey provides information on employee compensations in the whole construction sector, the value of employee compensations obtained from the Construction Contractor’s Survey was subtracted. The Service sector in the Palestinian economy consists of 11 sub-sectors which are grouped together under this heading for analytical purposes. These are: Wholesale and Retail Trade (category G); Transport, Storage, and Communication (I); Financial Intermediation (J); Real Estate, Renting and Other Business Services (division 70 in ISIC); Community, Social, and other Personal Services (O); Hotels and Restaurants (H); Education (M); Health and Social Work (N); Production of Government Service; Private Non-Profit Institutions Serving Households; and Private Households with Employed Persons. We have deliberately aggregated all these sectors under one heading to represent the non-tradable sector.

66

4.2.3 A Palestinian Input-Output Table An input-output table depicts the industrial structure of the economy and provides a view of interdependence caused by the passage of intermediate commodities between different activities. Inter-industry or input-output analysis emphasizes the idea that the economy should be viewed as a complete system of interdependent sectors. Individual industries supply produced inputs to each other, they compete for sales in domestic markets, and they compete in international trade. The implications of industrial interdependence, be it via forward linkages or backward linkages, are often crucial to the understanding of the effects of changes in economic circumstances both on particular industries and on the economy as a whole. Input-output tables can be used to analyze the structural effects of shocks to the economy, such as: a change in final demand for a good; a substitution of imports for domestic output; changes in tariffs and domestic taxes; and changes in the wage rate in various sectors. Input-output tables are also used for forecasting the effects on demand – for the products of an individual industry or sector – of a change in final demand (e.g. exports). For this purpose, input-output models are used in large-scale computer generated mathematical models of the economy. In sum, input-output analysis entails the description and analysis of the interdependence which exists amongst industries or sectors of the economy, and has two aims: (i) to measure and describe interindustry transactions, and (ii) to assist in the forecasting of the direct and indirect effects of exogenous policy changes on individual industries or sectors. The first step in the construction of an input-output table consists of classifying all the firms (or all of the activities of various firms) to one industry or another. Industries are usually defined in terms of homogeneity of product (e.g. wooden furniture, ice-cream, etc) but sometimes they are defined according to homogeneity of process (e.g. mining, metal fabrication, etc). Within an inputoutput table, a flow table is a rectangular array of data arranged in rows and columns containing information on the industrial composition of final demand and also the purchases and sales of intermediate goods and services between 67

industries. The inter-industry flow table for any given year attempts to record the flows occurring between industries or sectors, the sales of each industry or sector to the various components of final demand, and the amounts spent by each industry on wages, imported raw materials, and taxes. Each row of the flow table shows the output sold by each industry or sector listed down the left-hand side of the table, to each industry or sector listed across the top of the table. It follows that each column shows the purchases made by the industry or sector listed across the top of the table for the industries or sectors down the left-hand side of the table. We also have a column which shows the industry composition of final products produced in a given year. This is given under the heading ‘Final Demand’ and includes production which is exported. Table 4.1 is an input-output table for the Palestinian economy for 1998. The entries in the flow table showing the inter-industry transactions are extracted directly form the Supply and Use Tables (SUT). Final demand components are also extracted from the SUT. TABLE 4.1 AN INPUT-OUTPUT TABLE FOR THE PALESTINIAN ECONOMY, 1998 ($ US million) Sectors

Final Demand

SubTotal

Total

C

I

G

X

357

69

0

153

990

2306

430

0

524

5459

36

411 2199 188

68

1108

0

38

1402

13

58

1365

156

1

0

10

1532

14

40

33

0

1

452

19

225

67

160 558

258 855

32

215

2

1662 785

1

2

3

4

5

6

7

1. Agr

107

114

0

189

0

0

1

2. Manf

169

788

829

118

73

84

138

3. Const

0

10

113

7

0

22

4. Comm

124

895

3

263

9

5. Transp

5

45

4

45

7

6. PrivS

55

53

20

119

7. PubS

0

0

0

0

0

15

0

15

13

-5

761

1

Sub -Total

460

1905

969

741

108

373

340

4896

4013

1668

976

729

Val added

304

563

425

689

183

976

412

3552

Ind.Taxes

0

36

7

30

9

15

0

97

Tariffs

26

595

0

8

17

34

4

684

Imports

200

2360

1

64

135

264

29

3053

990

5459

1402

1532

452

1662

785

12,282

Total

12,282

Note: C=Consumption expenditure I= Investment expenditure G= Government expenditure X= Exports

GDP = C + I + G + (X-M) where M are imports.

As a matter of definition, the ‘value of production’ or ‘value added’ by a sector is equal to the total value sold – i.e. ‘value of output’ – less the cost of raw materials and intermediate inputs used up in the production process. For exam68

ple, the Agricultural sector column shows the outlays on intermediate inputs (raw materials) and other factors of production (K and L). Total agricultural output is $990 million of which $411 million is sold to other sectors, $357 million for consumption, $69 million is in the form of investment demand, and $153 million exported. The Manufacturing sector is a relatively large sector with a total value of output of $5,459 million. It primarily supplies to the Construction sector ($829 million) with a substantial amount ($788 million) used up in its own production process. This gives us total intermediate demand for manufactured goods of $2,199 million. The rest of manufacturing output is distributed among the various components of final demand with a large chunk ($2,306) for private consumption. Exports of manufactured goods constitute about 70% of total Palestinian exports, and $430 million is in the form of investment demand. The manufacturing sector purchases $1,905 million worth of intermediate inputs from other sectors, with a relatively large amount ($788 million) for own consumption. The total value of output of the construction sector is $1,402 million of which a large chunk ($1,108 million) is in the form of investment. This sector purchases primarily from the manufacturing sector ($829 million) and $115 million is for own consumption. The input-output flows for the remaining sectors are shown in Table 4.1. The Palestinian Service sector is a relatively large sector and is aggregated into one ‘giant’ sector with a total value of output of $2,727 million. Table 4.2 below is a disaggregation of the 11 sub-sectors which make up the service sector. These sub-sectors are wholesale and retail trade (G); hotels and restaurants (H); transport, storage , and communications (I); financial intermediation (J); real estate, renting, and business activities (K); public administration, defense, and compulsory social security (L); education (M); heath and social work (N); other community, social, and personal service activities (O); private households with employed persons (P); and extraterritorial organizations (Q). Most of the data in the service sector was obtained from the Services Survey and the In-

69

ternal Trade Survey conducted by the PCBS in 1997-98, and were carefully checked with data obtained from the Supply and Use Tables. Going back to Table 4.1, a large part of intermediate demand for services is for own consumption ($493 million), final demand by consumers is $1141 million, and $976 million is in the form of government consumption. On the expenditure side, the service sector purchases $1182 million worth of goods and services from other sectors, with relatively large amounts from the manufacturing sector ($388 million) and for own consumption ($493 million). It also pays $391 million in indirect taxes, and $1156 million is paid to labor and capital. TABLE 4.2 INTER-SECTORAL TRANSACTIONS, 1998 ($US million) Sectors

A+ B

Agr M&Q Manf EWG Const W Trade Hotels Trsp Fin Int

105 0 169 17 0 9 0 0 0 14 0 0 25 0

0 0 9 0 0 0 0 1 0 2 0 0 0 0

114 68 567 23 0 0 2 20 6 19 0 0 0 1

338

12

821

R Estate

Pub S Educ Health Other TOTAL

C

D

E

F

G

H

I

J

K

L

M

N

O

Tot.

0 0 11 67 1 0 0 0 0 0 0 0 0 0

0 0 829 9 114 0 0 4 4 6 0 0 0 0

165 0 63 27 4 2 7 43 6 67 0 1 0 2

26 0 41 3 1 0 0 1 8 4 0 0 0 0

0 0 74 1 0 5 0 7 7 12 0 0 0 0

0 0 4 2 0 0 3 5 109 12 0 0 0 0

0 0 5 1 0 0 0 2 0 5 0 0 0 0

1 0 138 9 35 27 31 4 2 38 0 4 0 14

0 0 10 3 9 0 1 2 0 4 0 0 0 0

0 0 49 3 12 0 1 3 0 8 0 0 0 1

0 0 6 2 1 0 1 2 0 5 0 0 0 0

409 68 1972 164 176 42 44 129 144 196 0 5 25 18

78

966

386

82

105

136

14

336

28

77

17

3397

Table 4.3 shows the input-output coefficients or ‘technical’ coefficients. These are the direct57 requirement coefficients for estimating the input requirements for any given output of a sector. These input requirements are calculated by dividing the purchases of inputs of each sector by total sectoral output. We can also calculate GDP from the input-output Table 4.1 using the simple income/expenditure identity: GDP = C + I + G + (X − M) = 4013 + 1668 + 976 + (729 − 3053 )

= $4,333million 57

Direct here refers to the fact that these are dollar values of the purchases which forms in one sector actually make from firms in other sectors.

70

TABLE 4.3 DIRECT INPUT-OUTPUT COEFFICIENTS, 1998. 1. Agr 2. Manf 3. Const 4. Comm 5. Trsp 6. Pvt S 7. Pub S

1 0.140 0.221 0 0.162 0.007 0.072 0

2 0.046 0.319 0.004 0.363 0.018 0.021 0

3

4 0.132 0.083 0.005 0.184 0.031 0.083 0

0 0.595 0.081 0.002 0.003 0.014 0

5

6

0 0.251 0 0.031 0.024 0.065 0

0 0.062 0.016 0.010 0.010 0.167 0.011

7 0.001 0.184 0.048 0.077 0.053 0.089 0

Source: Own calculations using GAMS in Palestinian CGE model.

4.3

A Palestinian SAM

4.3.1 Level of Aggregation In practice, two major considerations enter the choice of the aggregation level: the need to accurately capture the main discriminatory features involved in the policy issues under discussion; and the limits of data availability, especially in developing countries. The levels of aggregation in a SAM are also influenced by the orientation of the model - whether it is designed as an issue-specific model or whether a general purpose capability is intended. Among issue specific models, dimensions tend to be smaller. On the demand side, the level of aggregation is very much a function of the focus of the modeling effort. If the main issues are efficiency issues, it may be quite acceptable to aggregate households into one single consumer group, as we have done in the Palestinian SAM. If, however, distributional issues are at the heart of the analysis, then detail on the demand side becomes more important. Generally speaking, applied models are not very detailed on the demand side. In initial model development, a highly aggregated data set may be used to simplify model simulations. For first model runs, where only initial broad indications of results are desired, an intermediate level of aggregation can be used. Only when the model is sure that all development problems in designing the modeling system have been resolved will more detailed calculations be made. All the above considerations were taken into account in constructing the Palestinian SAM. The level of aggregation in our SAM is primarily influenced by 71

the need to analyze deindustrialization effects. Our primary objective is to show how trade liberalization can shift resources from the tradable into the non- tradable sector. We are primarily concerned with resource allocation and efficiency issues. Hence, households are aggregated into one consumer group. The seven production sectors in the Palestinian SAM generate most of the output. As mentioned above, the service sector is an aggregation of 11 separate service sectors. Indirect taxes have also been separated from direct taxes since the former is extracted from the production process. The combined capital account provides the balance between savings and investment, and the government account shows total government revenue and expenditure. Foreign trade is also at a high level of aggregation since almost 80% of trade is conducted with Israel.

4.3.2 A Macroeconomic Palestinian SAM Table 4.4 is an aggregated SAM for 199858. Starting from the production account (4), the row sum is total sales of $7,386 million of which $6,657 million is sold to institutions59 (households, enterprises, and government), and $729 million are exported. The column sum of $7,386 million shows total expenditure of the various production sectors of which $3,552 million is total value added (labor and capital), $781 million accrues to institutions (in the form of taxes), and $3,053 million are imports of all types of goods. The factor account (1) includes both capital and labor which receives a total income of $4,363 million ($3,552 million from the production process just mentioned), and $812 million from wage income earned abroad (primarily form Israel). This income of $4,364 million accrues to institutions (households and enterprises) in column 1.

58

Note that the empty cells in the SAM Tables 4.4 – 4.8 indicate that the entries are not economically meaningful and therefore are not relevant for analytical purposes. 59 It must be noted that ‘institutions’ are defined rather loosely in this context. Although firms are also institutions, they are nevertheless separated into production activities for analytical purposes.

72

TABLE 4.4 AGGREGATED MACROECONOMIC PALESTINIAN SAM, 1998 ($US million). 1 1. Factors 2. Institutions 3. Surplus/Deficit 4. Production 5. RoW 6. TOTAL

4364

4364

2

3

132 -1015 6657 5774

0

4 3552 781

3053 7386

5 812 497 1015 729

6 4364 5774 0 7386 3053

3053

Institutional income is derived from value added ($4,364 million), direct taxes ($132 million) and indirect taxes ($781 million including import tariffs), and $497 million of foreign capital flows (remittances of $107 million received by Palestinian residents and foreign aid of $390 million), giving us a total institutional income of $5,774 million in row 2. Institutional expenditure is shown in column 2 of which $6,657 million are paid to the production sector for consumption of goods and services, and income taxes of $132 million. Note that total institutional expenditure exceeds total institutional income by $1,015 million, which is the deficit of the institutional account. Looking now at the rest of the world (RoW) account (5), we see that the row sum of $3,053 million consists of total imports and the column sum represents foreign exchange earnings (i.e. payments from abroad). We note that the RoW’s total income exceeds its total expenditure by $1,015 million, which is exactly equal to the institutional deficit of $1,015 million. In Table 4.5, there is one modification. This is cell (4, 4) which represents the inter-sectoral transactions. Note that the corresponding row and column total are also increased by the amount of the diagonal entries. The diagonal entry 2, 2 implies that institutions make transactions between themselves in the form of transfer payments, direct taxes, and distributed profits. This entry represents the total of transactions among households, enterprises, and the government.

73

TABLE 4.5 AGGREGATED MACROECONOMIC PALESTINIAN SAM WITH DIAGONAL ENTRIES, 1998 ($US million). 1 2 3 1. Factors 2. Institutions 3. Surplus/Deficit 4. Production 5. RoW 6. TOTAL

4364

132 -1015 6657

4364

5774

4 3552 781

5 812 497 1015 729

4896 3053 12282

0

6 4364 5774 0 12282 3053

3053

The diagonal entry 4, 4 represents total transactions among different production sectors since one sector’s output is another sector’s input. By including diagonal entries, we have changed the meaning of the totals. In Table 4.6, we separate indirect taxes from the institutional account to distinguish them from direct taxes. Direct taxes are extracted directly from the TABLE 4.6 AGGREGATED MACROECONOMIC PALESTINIAN SAM WITH INDIRECT TAXES, 1998 ($US million). 1 2abc 1. Factors 2abc. Institutions 2d. Indirect taxes 3. Surplus/Deficit 4. Production 5. RoW 6. TOTAL

4364

132

2d

3

4 3552

781

5 812 497

781 -1015 6657 4364

5774

781

0

4896 3053 12282

1015 729

6 4364 5774 781 0 12282 3053

3053

income stream and therefore constitute a transfer between institutions (i.e. from households and enterprises, to the government). Indirect taxes are levied on intermediate purchases of inputs (both domestic and imported) by producers in the production process, and on purchases of final goods and services by consumers (including imports). The $6,657 million purchased by households is not what the consumers pay. We must add indirect taxes to this to obtain the actual payment. Similarly, the $4,896 million is not what the purchasers of intermediate inputs pay. We need to add indirect taxes. Therefore, we separate indirect taxes from the institutional account. Hence, the entry of $781 million is brought down form cell (2abc, 4) to (2d, 4) since it directly accrues to the indirect tax account. 74

Note that the total of indirect taxes extracted from the institutional account in row 2d is included in cell (2abc, 2d) since it accrues to the government (one of the Palestinian institutions). Consequently, the row/column totals are the same as before. In effect, by separating indirect taxes, we have allocated indirect taxes to the relevant types of expenditure. So far, we have not distinguished between consumption expenditure and investment expenditure. The entry of $6,657 million in cell (4, 2abc) in Table 4.6 represents purchases of both consumption and investment goods and services. In Table 4.7, we extract the investment goods in rows 4 and 5 from column 2abc

TABLE 4.7 AGGREGATED MACROECONOMIC PALESTINIAN SAM WITH SAVINGS AND INVESTMENT, 1998 ($US million). 1 2abc 2d 1. Factors 2abc. Institutions 2d. Indirect taxes 3. Combined capital 4. Production 5. RoW 6. TOTAL

4364

132

3

4 3552

781

5 812 497

781 653 4989 4364

5774

1668 781

1668

4896 3053 12282

1015 729

6 4364 5774 781 1668 12282 3053

3053

and enter them in column 3 which is renamed ‘combined capital’ account. The total of all these changes in column 3 is investment expenditure of $1,668 million. Expenditure in column 2abc is reduced by this amount. The balance of $1,015 million, which formerly appeared on line 3 of Table 4.6, is altered accordingly by the amount spent on investment: -1,015 + 1668 = $653 million. This is now the difference between income and consumption (rather than the difference between income and total expenditure). Therefore, the $653 million is the saving of domestic institutions. Total savings are equal to $1,668 million (domestic sources of $653 million and foreign sources of $1,015 million) which finance total investment. In Table 4.8, we disaggregate factors into labor and capital, and institutions into households and government. Total labor income is $2,759 million, of which $812 million is derived from wage income earned in Israel, and $1,947 75

million from value added. This income accrues to households in column 1a. Total capital income is $1,605 million which is paid by the production sector to households. Total household income of $4,471 million is derived from payments by labor ($2,759 million), capital ($1,605 million), and remittances from abroad ($107 million). Total household expenditure of $4,471 million in column 2a consists of direct taxes ($132 million) paid to the government, domestic household savings ($326 million), and consumption expenditure ($4,013 million). Total government revenue ($1,303 million) is derived from direct taxes ($132 million) paid by households, indirect taxes ($97 million), import tariffs ($684 million), and foreign aid ($390 million). The combined capital account shows the equality between total savings and total investment. Total savings are derived from domestic household savings ($326 million), government savings ($327 million), and foreign savings ($1,015 million). Total investment expenditure is $1,668 million of which $69 million is investment expenditure on agricultural goods, $430 million on manufactured goods, $1,108 million on construction (mainly housing), $1 million on communications, $33 million on transportation, $432 million on private services, and there is a $5 million disinvestment on public services. The negative sign in cell (4g, 3) in Table 4.8 can be translated into expenditure by the public sector. In other words, the economy has disinvested in the public sector. Total sales ($12,282 million) are shown by the row totals of accounts 4a, 4b, 4c, 4d, 4e, 4f, 4g, which is derived from household consumption expenditure, intermediate demand, and imports. Finally, the RoW account shows foreign exchange earnings (column total) and foreign exchange payments (row total). The column total of the RoW ($ 3,053 million) is distributed to factor payments from Israel to Palestinian workers ($812 million), remittances from abroad ($107 million), foreign aid ($390 million), surplus of the RoW ($1,015 million), and exports ($729 million = 153 + 524 + 38 + 10 + 1 + 2 + 1) of final goods and services.

76

1605

327

258

855

13

4e. Trsp

4f. PrvS

4g. PubS

1303

761

215

0

0

0

77

1668

-5

32

33

1

1108

430

0

55

5

124

0

169

107

990

156

4d. Com

4471

68

4c. Con

0

684

69

6. Total

2306

4b. Manf

0

200

357

4a. Agr

5. Row

326

3. CAP

97

684

26

97

4a. Agr

2d. Tarf

132

3. CAP

0

2759

1605

2d. Tarf

2c. iTax

2b. Gov

2a. HH

2759

2c. iTax

174

2b. Gov

1b. K

2a. HH 130

1b. K

1a. L

1a. L

5459

2360

0

53

45

895

10

788

114

595

36

328

235

4b. Manf

1402

1

0

20

4

3

113

829

0

0

7

245

180

4c. Con

1532

64

0

119

45

263

7

118

189

8

30

214

475

4d. Com

452

135

0

19

7

9

0

73

0

17

9

80

103

4e. Trsp

1662

264

15

225

14

13

22

84

0

34

15

546

430

4f. PrvS

785

29

0

67

40

58

36

138

1

4

0

18

394

4g. PubS

3053

1

2

1

10

38

524

153

1015

390

107

812

5. RoW

TABLE 4.8 DISAGGREGATED MACROECONOMIC SAM FOR THE PALESTINIAN ECONOMY, 1998 ($ US million)

3053

785

1662

452

1532

1402

5459

990

1668

684

97

1303

4471

1605

2759

Total

The circular flow of income is still maintained in the SAM. The core of the matrix in Table 4.8 is still the circular process of demand, production, and income. Most of the other entries are required by the existence of the RoW. The circular flow is also ‘dual’ in nature – one can move round a circle clockwise or anti-clockwise. If we move in the anti-clockwise direction, corresponding to the order just given, we are implicitly following the flow of money: factors of production receive money from production, and production receives money from institutions. We may think of the other direction as the supply of goods and services: factors of production to the production process, production to institutions, and institutions to factors of production.

78

CHAPTER 5

A PALESTINIAN CGE MODEL

5.1

Introduction

Equilibrium in a CGE model is ‘general’ in the sense that all markets are in equilibrium simultaneously. The modeler has to consider two sets of decisions: microeconomic constraints which are determined by individual economic agents (households, government), and macroeconomic or system-wide constrains which are outside the control of individual agents. The mechanisms through which these system-wide constrains are determined are known as ‘closures’ of the model. Although a number of CGE models have been developed for Egypt60, Tunisia, Morocco, and Turkey, there have been very few models for Jordan, Israel, and the West Bank and Gaza Strip. The reason for this was the lack of accurate data and input-output tables. However, in 1998, the Palestinian Central Bureau of Statistics (PCBS) compiled the first detailed Supply and Use Tables (SUT) with highly disaggregated data on a number of sectors. This has greatly improved the prospects of modelers to build CGE models. The only CGE model on the Palestinian economy is by Astrup and Dessus (2000) in which they attempt to find some orders of magnitude for Palestinian trade policies. Their model is a static trade-focused model with 31 industries to analyze welfare and resource allocation effects. Their main conclusion is that a new trade regime, which departs from the existing Customs Union arrangement with Israel and which re-balances trade flows with the rest of the world, is likely to generate specialization gains, improvements in household welfare, and reductions in transaction costs. Their analysis suggests that a Free Trade Area is not necessarily the only option since the cost of implementing the rules of origin will significantly offset the benefits of trade liberalization. Another option would 60

See Lofgren, H., 1993. “Egypt’s Experience with CGE Modeling: A Critical Review.” American University of Cairo Economics Department Working Paper. Cairo.

79

be a complete or ‘full’ separation between the WBGS and Israel, which is one of the trade policy options analyzed in this study. Production at the micro level in the Palestinian CGE model, PALMOD, is modeled using CES production functions for seven sectors representing substitution possibilities between value-added factors of production. Producers in each of theses seven sectors minimize total costs subject to their output constraints, and constant return-to-scale is assumed. Household income is derived from labor, capital, and remittances from abroad. Household demand is obtained by maximizing a Cobb-Douglas utility function subject to the budget constraint of the representative household. The government acts just like any other economic agent in the model by maximizing a Cobb-Douglas utility function subject to its tax revenue constraint. Both savings and investment are endogenous, and investment demand is a fixed share of savings. Domestic and imported goods are imperfect substitutes represented by a CES Armington aggregation function, and exports are represented by an export-demand function for the rest of the world with a constant elasticity of demand for exports. In considering the macro system wide-constraints, real constraints are represented by equilibria in commodity and factor markets, and nominal constraints are represented by two macro balances: the savings-investment balance, and the trade balance. For factor markets, factor supplies are exogenously fixed and factor prices adjust to equate factor demand with factor supplies. For the investment-savings balance, total investment demand is equal to total savings, which in turn is the sum of household saving, government saving, and foreign saving. The model assumes a fixed level of foreign savings and total investment is the endogenous sum of the separate savings components. In the trade balance constraint, since foreign savings are fixed, the equilibrating variable is the exchange rate. Equilibrium is achieved by changing the relative price of tradable and non-tradable goods61.

61

An illustrative example is given in Devarajan, S., J. Lewis, and S. Robinson, 1994. Getting the Model Right: A General Equilibrium Approach to Adjustment Policy. (Draft Manuscript).

80

This chapter begins with a description of the structure of the Palestinian CGE model, PALMOD, using the circular flow of income diagram. Section 5.3 derives the behavioral equations for households, firms, the rest of the world, and the government. It also describes the macro system-wide market clearing equations for external and internal balance. Section 5.4 describes the calibration process such that the model equations replicate the benchmark equilibrium. The Appendix lists the variables (exogenous and endogenous) and the parameters of the Palestinian CGE model.

5.2

Circular Flow of Income

Figure 5.1 describes the circular flow of income between major economic agents and markets in the Palestinian economy. The model consists of seven major ‘blocks’ – producers, households, commodity markets, factor markets, government, investment and savings balance, and the RoW. Starting with producers, there are seven production activities which sell their output to the domestic and export markets. They make payments for factor services intermediate inputs used-up in the production process. In the domestic product market, the components of demand are household consumption, government consumption, investment demand, and intermediate input demand. The supply side is obtained from sales of imports from the RoW. Demand for factors of production is satisfied by household supply of factors that receive income payments for their factor services. Equilibrium in factor markets is achieved through factor price adjustments which equate quantities supplied and demanded. Household income from producers is spent on consumption and taxes, and the remainder is saved. The government in this model acts just like any other economic agent by spending this tax revenue on consumption. It also receives transfers from the RoW, and saves. Total investment demand is financed by total savings by households, government, and foreign savings. To the extent that spending on imports ex-

81

ceeds exports and other revenues from the rest of the world, foreign savings (the current account deficit) is positive.

FIGURE 5.1 CIRCULAR FLOW OF INCOME

Domestic Household Savings

FACTORS Factor Costs

HOUSEHOLDS

PRODUCERS

GOVT.

Intermediate Inputs Private Consumption

Domestic Sales

Govt. Savings

Taxes

Wages

Transfers

Govt. Consumption

PRODUCTS Investment Imports Foreign Transfers Exports

RoW Foreign Savings

82

SAV/INV

5.3

Model Structure (PALMOD) 5.3.1 Households

There is one representative household that maximizes a Cobb-Douglas utility function of the form: h

U = ∏ Ciαi

Max {Ci}

i

s.t.

Y = ∑ PiCi (1 − tc i ) i

where Ci an Pi are the consumption and price of good i, αih is the Cobb-Douglas exponent, and tci is the tax rate on consumption goods. Forming the Lagrangian of the above optimization problem and setting the partial derivatives with respect to Ci and λ to zero, we obtain the first-orderconditions for utility maximization, given by:

(1 + tc i )PiCi = αhi Y Given household savings, HHSAV, the above first-order-condition can be rewritten as:

(1 + tc i )PiCi = αhiCBUD

(5.1)

where CBUD is the consumer’s budget and is given by:

CBUD = −(1 − ty )Y − HHSAV

and ty is the income tax rate. Note that the budget constraint of the household is already included in equation (5.1).

83

5.3.2 Firms There are i sectors and each sector is represented by a perfectly competitive firm with Constant Elasticity of Substitution (CES) production functions. The objective of the firm is to minimize total cost (TC) subject to the output constraint. This optimization problem is given by: min { Ki, Li } s.t.

XDi =

TC = PKK i + PLL i

[

F ai γFiK iαi

+

(

1 − γFi

1 αFi αF Li i

)

]

where XDi is total domestic output, ai is the CES production function shift parameter, γiF is the share parameter, and αiF is the exponent. Forming the Lagrangian to the above optimization problem, we obtain the first-order conditions: F

γFi (1 + tl i )PLK iσ i

−1

(

)

F

= 1 − γFi (1 + tk i )PKLσi i

−1

(5.2)

The production constraint is given by:

XDi =

[

F ai γFiK iαi

+

(

1 − γFi

1 αFi αF Li i

)

]

(5.3)

Finally, since under perfect competition profits are zero, the zero-profit condition is given by:

PDi XDi = (1 + tk i )PKK i + (1 + tl i )PLLi + ∑ a ji XDiPj

(5.4)

j

where the LHS represents total revenue, the first two terms on the RHS are value added costs, and the last term is the cost of intermediate inputs.

84

5.3.3 Imports and Exports Good Xi is a composite of domestic goods sold domestically (XXDi) and imports Mi). This aggregation of the composite good is represented by an Armington CES function given by: Xi =

[

A aA i γiAMiαi

+

(

1 − γiA

1 α iA α A i XXDi

]

)

The objective of the firm is to minimize total cost subject to the above production constraint. Forming the Lagrangian and setting the partial derivatives with respect to Mi and XXDi to zero, we obtain the first-order-conditions: A

γiAPDiMiαi

−1

(

)

A

= 1 − γiA PMi XXDiαi

−1

(5.5)

where PDi is the price of good XDi, PMi is the price of imported goods, and XXDi is the domestic good sold in the domestic market. aAi is the Armington CES shift parameter, γiA is the share parameter, and αiA is the exponent. The production constraint of the firm is given by: Xi =

[

A aA i γiAMiαi

+

(

1 − γiA

1 α iA α A i XXDi

)

]

(5.6)

and the zero-profit condition for the Armington composite good is given by: PiXi = PMiMi + PDiXXDi

(5.7)

Finally, the export function is given by:  PWEZ i  Ei = EZi    PWE i 

αEi

(5.8)

Where Ei is the export good, EZi is the initial level of exports, PWEZi is the initial world price of exports, and αiE is the elasticity of export demand.

85

5.3.4 Government The government is treated just like any other economic agent maximizing its utility subject to its budget constraint. The optimization problem of the government is given by: max {CGi, KG, LG}

CG

KG

U = CGiαi KGα LGα

LG

TAXR = ∑ PiCGi + PK.KG + PL.LG

s.t.

where

∑ αCG + αKG + αLG = 1 ⇒ CRTS

and where CGi is government consumption of good i, KG and LG are government uses of capital and labor respectively, and TAXR is total government revenue. Forming the Lagrangian of the above optimization problem and setting the partial derivatives to zero, we obtain the first-order conditions: PiCGi = αiCGTAXR

(5.9)

PL.LG = αLGTAXR

(5.10)

PK.KG = αKGTAXR

(5.11)

where αiCG is the Cobb-Douglas power for the government utility, and αLG and

αKG are the Cobb-Douglas power for the government utility for labor and capital respectively. TAXR is total government revenue, which is given by:

TAXR = ∑ tk iPK.K i + ∑ tliPL.Li + ∑ tc iPiCi + i

i

i

∑ tmiER.PWMZiMi + ty.Y + ER.FORAID

(5.12)

i

where the first three terms on the RHS are tax revenues on capital, labor, and consumption goods respectively, the fourth term is import tariff revenue, and tyY is income tax revenue. FORAID is foreign aid, and ER is the exchange rate.

86

5.3.5 Market Clearing Definitions • Labor market equilibrium is given by:

∑ Li + LG + LISRAEL = LS

(5.13)

i

where LISRAEL is Palestinian labor demand in Israel, and LS is the exogenous labor supply. Note that LS is assumed to be endogenous such that the wage rate is fixed to allow for employment effects. This assumption of the labor supply ensures that a reduction in Palestinian labor demand in Israel is not automatically absorbed into the Palestinian economy.

• Capital market equilibrium is given by:

∑ K i = KS

(5.14)

i

where KS is the exogenous capital supply.

• Goods market equilibrium is given by: Xi = Ci + CGi + Ii + ∑ aij XDj

(5.15)

j

where Ci is household consumption of good i, CGi is government consumption of good i, Ii is total investment demand, and the last term is demand for intermediate goods.

• Domestic goods market equilibrium is given by: XDi = XXDi + Ei

(5.16)

where XDi is total domestic output, XXDi is domestic output sold only in the domestic market, and Ei are exports of good i.

87

Both investment and savings are endogenous, and investment demand from different sectors is a fixed share of total savings (Cobb-Douglas) which is given by the following optimization problem:

max{Ii }

I

U = ∏ Iiαi i

s.t.

TOTSAV = ∑ PiIi i

where TOTSAV is total savings, Ii is total investment, and αiI is the CobbDouglas power for investment demand from savings. Solving the above optimization problem gives us: PiIi = αIiTOTSAV

(5.17)

• Finally, external market equilibrium (the trade balance) is given by: ΣPWMZiMi = ΣPWEiEi + FORAID + REMIT + PLISRAEL.LISRAEL + FORSAV

(5.18)

Where the LHS is value of imports, the first term on the RHS is the value of exports, FORAID is foreign aid, REMIT are remittances from abroad (accruing to households), PLISRAEL is the wage rate in Israel, LISRAEL is Palestinian labor demand in Israel, and FORSAV is foreign saving which is equal to the deficit of Palestinian domestic institutions with the rest of the world. For a complete specification of the model, we define the following variables:

• CPI = ∑ (1 + tc i )Piαhi

(5.19)

i

where CPI is the consumer price index. Note that the CPI is calibrated to the LHS. The labor supply is no longer exogenously fixed while the wage rate is fixed. In other words, we are assuming a fixed real wage (CPI indexed) and any reduction in labor demand in Israel would have no direct impact on the domestic labor market.

88

• Y = PK.KS + PL(ΣLi + LG) + ER.PLISRAEL.LISRAEL + ER.REMIT

(5.20)

where Y is household income.

• CBUD = (1 – ty)Y – HHSAV

(5.21)

where CBUD is the household consumption budget, ty is income tax, and HHSAV is household savings given by: HHSAV = mps(1 – ty)Y

(5.22)

where mps is the marginal propensity to save.

• Government savings, GOVSAV, is the residual difference between government revenue and expenditure, given by: GOVSAV = TAXR – ΣPiCGi

(5.23)

• Total savings, TOTSAV, is the sum of household savings (HHSAV), government savings (GOVSAV), and foreign savings (FORSAV): TOTSAV = HHSAV + GOVSAV + ER.FORSAV

(5.24)

where ER is the exchange rate.

• Domestic price of imports, PMi, is given by: PMi = PWMZi.ER(1+tmi)

(5.25)

Where PWMZi is the world price of imports, and tmi are import tariffs. 89

• The world price of exports, PWEi, is given by:

PWE i =

PDi ER

(5.26)

where PDi is the price of the domestic good XDi. • Finally, the objective function is given by : TRICK = 1

(5.27)

where TRICK is an artificial variable. The final PALMOD model consists of a total of 27 equations and 34 variables. The Walras Law allows us to drop 1 equation leaving us with 26 equations. We also have 7 exogenous variables and a fixed exchange rate leaving us with 26 endogenous variables. We thus have a fully determined model where the number of equations is equal to the number of endogenous variables.

5.4

Calibration

Calibration is the process of calculating the parameters of the model such that the model equations ‘replicate’ the benchmark equilibrium (SAM). However, not all parameters are calculated in this way. Some are extracted directly from the Central Tendency Tables which provide elasticity values for a wide range of sectors. In the Palestinian CGE model (PALMOD), elasticities of substitution are extracted from these tables whereas other parameters are calibrated using data provided in the SAM. Note that the order of calibrating the parameters is very important. For example, suppose we wish to calibrate

90

Xi = XXDi + Mi + TRMi where Xi is composite goods supply, XXDi is the supply of goods to the domestic market, Mi are imports, and TRMi is tariff revenue on imports of good i. We first need to determine the value of all the variables on the RHS of the above equation. The values of some of these variables can be extracted directly from the SAM, while others have to be calibrated before this operation.

91

CHAPTER 6

TRADE POLICY SIMULATIONS

6.1

Introduction

Trade policy analysis in practice has to deal with the inherent characteristics and distortions of the economy under examination. For the Palestinian economy, these distortions are further compounded by three considerations that the policy analyst must take into account. The first has to do with the nature of the trade agreement between the WBGS and Israel. The second has to deal with the very definition of ‘import tariffs’, and the third consideration is the absence of national borders and the presence of Non-Traditional Non-Tariff Barriers. These issues were discussed extensively in Chapter 2, but are useful to summarize them here before conducting the simulations. The trade agreement between Israel and the WBGS is based on the Paris Protocol which stipulates four major provisions. First, some Palestinian exports to Israel62 are bound by Voluntary Export Restraints (VERs). Second, on imports of products in Lists A1 and A263, the PA can set its own tariff rates but is bound by Quantitative Restrictions (QRs). Third, on products in List B64, the PA is free to set its own tariffs and can import any quantity it desires. On all other products not included in Lists A1, A2 and B, the PA can import any quantity but tariff rates should be at least as high as Israeli rates. The second consideration that one has to take into account is that over 90% of exports and 75% of imports are with one trading partner, Israel. Although there are no import tariff barriers in the traditional sense between the WBGS and Israel, there are other taxes such as value-added (VAT) and purchase taxes. For example, there is a 13% VAT on imports from Israel. Similarly, there is a 12.2% VAT, 11.3% purchase tax, and a 2.1% import tax from countries 62

These are tomatoes, cucumbers, potatoes, melons, eggs, and poultry. Consisting of agricultural and food imports from Jordan, Egypt, and other Arab countries. 64 Consisting of basic food items and other goods used in construction, industry, agriculture, and development programs. 63

92

which have signed Free Trade Agreements with Israel, and a 12.5% VAT, 11.1% purchase tax, and a 10.7% import tax from the rest of the world65. In other words, ‘import tariffs’ in our model are a composite of value-added taxes, purchase taxes, and import taxes in the traditional sense. Finally, the absence of national borders and the presence of NonTraditional Non-Tariff Barriers between the WBGS and Israel lead to abnormally high transaction costs for both importers and exporters. Moreover, the porous nature of borders, especially between the West Bank and Israel, makes an accurate determination of the volume of trade very difficult between these two trading partners. In conducting trade policy simulations for the Palestinian economy, one has to take into account these considerations carefully; otherwise it would be difficult to classify the trade agreements under one heading or another. For example, the Paris Protocol indicates that the trade agreement between Israel and the WBGS is a quasi-Customs Union with elements of a Free Trade Agreement. If the modeler is not too carful, the presence of abnormally high transaction costs does not yield economically meaningful results. Although some CGE modelers have explicitly taken into account transaction costs (Lofgren, 2001), our model treats transaction costs as part of import tariffs. Hence, a reduction in the cost of trade can be implicitly translated into a reduction in transaction costs. This chapter uses the CGE model constructed in Chapter 5 to conduct several trade policy simulations to analyze Palestinian trade policy options. Section 6.2 analyzes the effects of reductions in import tariffs on both macro and micro economic aggregates and resource allocation. Section 6.3 simulates reductions in demand for Palestinian labor in Israel to determine the degree of dependency on the Israeli economy, and section 6.4 simulates two ‘separation’ scenarios; the first simulation is a ‘peace’ scenario in which Israel and the PA reach a Permanent Status settlement represented by simultaneous increases in foreign aid, labor demand in Israel, and domestic labor supply. The second sim65

See Table 2.5 in Chapter 2.

93

ulation is a ‘war’ scenario in which either the PA or Israel (or both) declare a unilateral separation. This last scenario, to some extent, reflects the current state of ‘war’ and is simulated by simultaneously increasing transaction costs (because of border closures) and decreasing demand for workers employed in Israel.

6.2

Reduction in Import Tariffs

This section simulates a 50% reduction in import tariffs to reflect a reduction in transaction costs. It has been estimated by the Palestinian Chambers of Commerce that transaction costs add approximately 30% to the price of imports. This is because of poor infrastructure, custom clearing delays, and tighter border closures of the WBGS due to Israeli security concerns. Hence, a reduction in transaction costs is translated into a proportionate reduction in the price of imports. Column S1 of Table 6.1 describes the effects of this simulation on both micro and macro economic variables. A reduction in transaction costs decreases the price of imports and generates competition between domestic and imported goods. This decreases the price of domestic goods. Since composite goods are an aggregation of domestic and imported goods, domestic goods supplied to the domestic market also falls, and producers begin to supply more to export markets since the terms of trade has improved. Moreover, since the trade deficit (or foreign savings) is assumed to be fixed, government saving falls. Household consumption also increases because of the price fall, but government revenue is lower due to lower revenues. Total exports increase by 51% and total imports increase by 12%. Exports of agricultural and manufacturing goods increase by 46% and 53% respectively, exports of construction and commerce goods increase by 54% and 48% respectively, and exports of transportation goods, private services, and public services increase by 47%, 39%, and 55% respectively, although the total volume of exports of these last three sectors is relatively small at $5.8 million. This substan-

94

tial increase in the volume of trade is the result of improvements in the Palestinian terms of trade caused by the relative decline in the price of imports. Imports of agricultural goods are virtually unchanged at $201 million with a substantial increase in imports of manufacturing goods (12%) and construction goods (164%), although the latter is at a relatively low figure of $2.6 million. Imports of commerce goods increase by 55% from $64 million to $99 million, and imports of transportation goods increase by 32%. Finally, there is a small increase in imports of private services (5%) and a relatively large decrease of 19% in imports of public services. These changes brought about by a decrease in import tariffs explain the prohibitive effects of high transaction costs that have traditionally hindered Palestinian trade. However, since agricultural goods face relatively low tariff rates, the volume of agricultural imports is virtually unchanged. Both household income and household savings increase by 7.4%, while government saving decreases by 18.6%. The increase in household income and savings is caused by the fall in prices, which in turn increase real income of households. The consumption budget of households also increases by 7.4% and tax revenues fall by 18%. This fall in tax revenues must somehow be compensated for by increasing the direct tax rate or consumption taxes. The price of labor is unchanged since it is a numeraire price against which all other prices in the model are measured and is assumed to be fixed. The price of capital, on the other hand, increases by 5.2% since there is an associated movement of resources from the non-tradable service sectors to the tradable sectors, especially into manufacturing and agriculture which experience substantial increases in exports. Demand for consumer goods increases from $4,013 million to $4,308 million distributed as follows: 8.6% increase in manufactured goods; 7.5% increase in construction; 7.2% increase in commerce; 6.3% increase in transportation; 7.1% public services; 5.5% agricultural goods; and a 5.3% increases in consumption of private services. This increase in demand is caused by the fall in prices and the increase in real income as import prices fall. 95

TABLE 6.1 SIMULATION RESULTS Variable EXPORTS (E) $US million

Benchmark 729

S1 1105

S2 787

S3 630

S4 318

153 524 38 10 1 2 1

225 801 58 15 1.5 2.8 1.5

167 565 40 11 1 2 1

124 440 33 8 0.8 1.5 0.9

59 183 11 4 0.3 0.9 0.3

3053

3427

2743

3750

2129

Agriculture Manufacturing Construction Commerce Transportation Private services Public services

200 2360 1 64 135 264 29

201 2646 2.6 99 178 277 24

176 2133 1 60 119 228 26

258 2817 1.1 73 170 387 44

269 1375 0.8 64 144 270 8

Labor supply (LS) $US million Capital Supply (KS) $US million Household income (Y) $US million Household savings (HHSAV) $US million Gov savings (GIVSAV) $US million Foreign savings (FORSAV) $US million Total savings (TOTSAV) $US million

2759 1605 4471 326 327 1015 1668

2919 1605 4800 350 266 1015 1727

2226 1605 3897 284 306 1015 1616

3586 1605 5387 392 463 1015 1847

1865 1605 3268 238 258 1015 1298

Agriculture Manufacturing Construction Commerce Transportation Private services Public services

IMPORTS (M) $ US million

Note on Scenarios: S1 = 50% reduction in import tariffs S2 = 50% reduction in labor demand in Israel S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand in Israel; and 30% increase in labor supply S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel

96

TABLE 6.1 SIMULATION RESULTS (contd…) Variable Demand for consumer goods (C) $US m

Benchmark 4013

S1 4308

S2 3497

S3 4839

S4 2922

357 2306 68 156 258 855 13

376 2505 73 167 274 899 14

312 2001 59 137 225 752 11

428 2809 81 186 310 1009 16

285 1573 49 120 204 681 10

8448

8912

8231

9068

7577

764 2468 1394 1430 291 1349 752

894 2709 1408 1513 376 1355 657

756 2460 1383 1404 271 1252 705

764 2387 1423 1446 321 1671 1056

481 2556 1334 1341 213 1044 608

1605 174 328 245 214 80 546 18 1947 130 235 180 475 103 430 394

1605 190 330 237 211 98 525 14 2108 166 289 193 518 139 457 346

1605 178 343 249 218 77 522 18 1820 122 219 173 458 94 386 368

1605 158 279 235 196 81 634 22 2370 147 267 200 501 121 577 557

1605 128 414 262 241 68 473 19 1460 64 174 147 408 67 285 315

Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Domestic output (XD) $ US m Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Capital input (K) $ US million Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Labor input (L) $US million Agriculture Manufacturing Construction Commerce Transportation Private services Public services Note on Scenarios:

S1 = 50% reduction in import tariffs. S2 = 50% reduction in labor demand in Israel. S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand; and 30% increase in labor supply. S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel.

97

TABLE 6.1 SIMULATION RESULTS (contd…) Variable Investment goods (I) $US million

Benchmark 1668

S1 1824

S2 1668

S3 1812

S4 1668

69 430 1108 1 33 32 -5

69 430 1108 1 148 32 36

69 430 1108 1 33 32 -5

85 430 1108 1 33 134 21

69 430 1108 1 33 32 -5

876

790

916

1374

783

0 0 0 0 0 215 761

0 0 0 0 0 172 618

0 0 0 0 0 203 713

0 0 0 0 0 298 1076

0 0 0 0 0 185 598

1303 812 107 390 0 0

1060 812 107 390 0 5.2

1220 406 107 390 0 -2.9

1846 1218 107 780 0 7.1

1026 406 107 390 0 -12.6

0 0 0 0 0 0 0 0 0 0 0 0

1.8 -1.2 -0.2 0.1 1.0 2.1 0.3

-0.3 0.4 -0.3 -0.5 -0.2 -0.9 -0.1

0.6 -1.1 0.7 1.2 0.3 2.1 0.2

-8.5 5.5 1.0 -4.6 -7.8 -8.3 0.3

7.40 7.40 -18.6 7.40 -18.6

-12.8 -12.8 -6.30 -12.8 -6.30

20.5 20.5 42.0 20.5 42.0

-27.0 -27.0 -21.2 -27.0 -21.2

Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Government demand (CG) $ US m Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Tax revenue (TAXR) $US million Labor demand in Israel $US million Remittances (REMIT) $Us million Foreign aid (FORAID) $US million % change in price of labor (PL100) % change in price of capital (PK100) % change in price of goods (P100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in income (Y100) % change in hh saving (HHSAV100) % change in Gov saving (GOVSAV100) % change in consp. Budget (CBUD100) % change in tax revenue (TAXR100) Note on Scenarios:

S1 = 50% reduction in import tariffs. S2 = 50% reduction in labor demand in Israel. S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand; and 30% increase in labor supply. S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel.

98

TABLE 6.1 SIMULATION RESULTS (contd…) Variable

Benchmark

S1

S2

S3

S4

0 0 0 0 0 0 0 0 0

0 0 0 0 0 -20.1 -18.8 5.8 0

0 0 0 0 0 -5.4 -6.2 -19.3 0

0 0 0 0 0 39 42 30 0

0 0 0 0 0 -14.1 -21.5 -32.4 0

0 0 0 0 0 0 0

5.5 8.6 7.5 7.2 6.3 5.3 7.1

-12.6 -13 -12.6 -12.4 -12.7 -12.1 -12.8

19.8 21.8 19.7 19.1 20.1 18 20.3

-20.1 -31.8 -27.6 -23.4 -20.1 -20.4 -27.1

0 0 0 0 0 0 0

9.5 0.5 -3.3 -1.4 22 -3.9 -20.6

2.7 4.6 1.7 2.3 -3.4 -4.7 -0.8

-9.2 -15 -4.1 -8.6 1.7 16 22.3

-26.2 26.4 6.7 12.5 -15.4 -13.3 4.5

0 0 0 0 0 0 0

27.4 23.1 7 9.1 35.1 6.2 -12.2

-5.9 -6.9 -4.1 -3.6 -9.0 -10.1 -6.5

12.7 13.6 10.8 5.6 17.4 34.1 41.3

-50.1 -26.2 -18.4 -14.1 -35.3 -33.7 -20.2

% change in gov demand for goods (CG100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in labor supply (LS100) % change in capital supply KS(100) % change in hh consumption (C100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in capital input (K100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in labor input (L100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services Note on Scenarios: S1 = 50% reduction in import tariffs. S2 = 50% reduction in labor demand in Israel. S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand; and 30% increase in labor supply. S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel.

99

TABLE 6.1 SIMULATION RESULTS (contd…) Variable

Benchmark

S1

S2

S3

S4

0 0 0 0 0 0 0

0 0 0 0 34.8 0 -8.18

0 0 0 0 0 0 0

36.1 0 0 0 0 45.8 0

0 0 0 0 0 0 0

0 0 0 0 0 0 0

17.1 9.8 1.0 5.8 29.3 0.5 -12.5

-1.1 -0.3 -0.8 -1.8 -6.5 -7.2 -6.3

0 -3.3 2.1 1.8 10.4 23.9 40.5

-37.1 3.6 -4.3 -6.3 -27 -22.6 -19.2

0 0 0 0 0 0 0

9.7 -1.8 -0.5 5.5 29.3 0.4 -12.6

-3.7 -2.5 -0.9 -1.9 -6.5 -7.2 -6.3

4.8 0.2 2.6 1.2 10.5 24 40.5

-31.1 22.1 -2.5 -5.8 -26.8 -22.5 -19.1

0 0 0 0 0 0 0

7.2 6.6 0.3 8.6 30.3 1.3 -12.9

-6 -6.8 -1 -2.2 -8.4 -8.4 -6.4

11.3 11.7 2.6 2.1 16 28 41

-14.3 -17.8 -2.6 -5.5 -15.4 -18.1 -21.8

% change in investment demand (I100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in domestic output (XD100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in domestic output supplied to the domestic market (XXD100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in composite good supply (X100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services Note on Scenarios: S1 = 50% reduction in import tariffs. S2 = 50% reduction in labor demand in Israel. S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand; and 30% increase in labor supply. S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel.

100

TABLE 6.1 SIMULATION RESULTS (contd…) Variable

Benchmark

S1

S2

S3

S4

0 0 0 0 0 0 0

0.5 12 164 55 32 5 -19

-12 -9.6 -3.6 -6.6 -11.6 -13.5 -9.5

29.3 19.4 9.2 13.7 25.6 46.6 52.4

34.4 -41.8 -23.1 -0.5 6.5 2.2 -73

0 0 0 0 0 0 0 1146 1146

46 53 54 48 47 39 55 1146 1230

9.5 7.9 6.9 8.7 9.5 12.5 6.1 1146 998

-19 -16 -14.5 -17.5 -19.2 -24.4 -12.7 1146 1381

-14.3 -17.8 -2.6 -5.5 -15.4 -18.1 -21.8 1146 832

0

7.4

-13

20.5

-27

% change in imports (M100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

% change in exports (E100) Agriculture Manufacturing Construction Commerce Transportation Private services Public services

Initial household utility level (UZ) New utility level (UNEW) Per capita utility level % change in utility (U100) Note on Scenarios:

S1 = 50% reduction in import tariffs. S2 = 50% reduction in labor demand in Israel. S3 = ‘Peace’ scenario: doubling of foreign aid; 50% increase in labor demand; and 30% increase in labor supply. S4 = ‘War’ scenario: 50% increase in transaction costs; and 50% reduction in labor demand in Israel.

In terms of resource allocation, there is considerable inter-sectoral resource mobility because of the reduction in import tariffs. The agricultural and manufacturing sectors experience an inflow of capital of 9.5% and 0.5% respectively, and the transportation sector exhibits a 22% capital inflow. The public services sector experiences the most severe capital outflow of 20.6%. The agricultural and manufacturing sectors experience the largest labor flows of 27.4% and 23.1% respectively. The construction, commerce, and transportation sectors experience labor flows of 7%, 9.1%, and 35.1% respectively, while the public services sector experiences a relatively large labor outflow of 12%. These changes in resource allocation are clearly reflected in domestic output and composite good supply. Domestic output of the agricultural sector increases by 17.1%, and domestic output supplied to the domestic market (as opposed for exports) increases by 9.7%. The composite good supply (domestic output supplied to the domestic market plus imports) of agricultural goods in101

creases by 7.2%. The manufacturing sector experiences a 9.8% increase in domestic output and a 1.8% decrease in output supplied to the domestic market since manufactured exports have increased by 53%. The composite good supply of manufactured goods increases by 6.6% since imports increase by only 12% but exports increase by over 50%. Finally, government consumption of private and public services fall by 20.1% and 18.8% respectively, which is the result of lower import tariff revenues. Prices in the model are determined endogenously and therefore the reduction in import tariffs is reflected in the prices of the composite good. The manufacturing sector experiences the largest drop in prices of 1.2%, followed by a price reduction of 0.2% in the construction sector. The price fall in the manufacturing sector improves the competitiveness of this sector and thereby increases exports. Prices in the agricultural sector increase by 1.8%, and prices in the private services sector increase by 2.1% and there is a marginal price increase of 0.3% in the public service sector. Finally, the level of household utility increases by 7.4%.

6.3

Reduction in Labor Demand in Israel

Palestinian wage income from Israel has traditionally played an important role in the Palestinian economy. In 1998, about 120,000 workers (or 25% of the total Palestinian labor force) were commuting daily to jobs in Israel. This wage income amounted to over $800 million, mostly through employment in the Israeli agricultural and service sectors. Although in recent years the number of Palestinian workers in Israel has fallen sharply due to border closures of the WBGS, it still remains an important source of Palestinian income. This section simulates a 50% decrease in labor income from Israel to analyze the degree of sensitivity of changes in this important source of Palestinian income. The results of this simulation exercise are reported in Table 6.1 under column S2. Given a flexible Palestinian labor supply, a decrease in labor de102

mand in Israel will decrease labor inputs in the domestic economy, and domestic output falls. The price of capital also falls by 2.9% which in turn pushes down the domestic price level. Moreover, since composite goods are an aggregation of domestic and imported goods, domestic goods supplied to the domestic market decreases and imports fall. Household consumption falls because of the fall in household income, and household savings are lower. Government savings are also lower to maintain the same level of foreign savings. Total exports increase by 8% while total imports fall by 10%. Exports of agricultural and manufactured goods increase by 9.5% and 7.9% respectively, exports of construction goods and commerce increase by 6.9% and 8.7% respectively, and exports of transportation goods, private services, and public services increase by 9.5%, 12.5%, and 6.1% respectively, although the total volume of exports of these last three sectors is relatively low at $4 million. Imports of agricultural goods fall by 12%, with a substantial decrease in imports of manufactured goods (9.6%). Imports of commerce goods decrease by 6.6%, and imports of transportation goods fall by 11.6%. Finally, imports of private services decrease by 13.5% and there is a relatively large decline of 9.5% in imports of public services. Both household income and household savings decrease by 12.8%, while government savings decrease by 6.3%. The consumption budget of households also decreases by 13% and tax revenues fall by 6%. The price of labor is unchanged, while the price of capital decreases by 2.9%. Demand for consumer goods decreases from $4,013 million to $3,497 million distributed as follows: 13.2% decrease in manufactured goods, 12.6% decrease in construction, 12.4% decrease in commerce, 12.7% decrease in transportation, 12.8% decrease in public services, 12.6% decrease in agricultural goods, and 12.1% decrease in consumption of private services. In terms of resource allocation, there is considerable inter-sectoral resource mobility as a result of the reduction in labor demand in Israel. The agricultural and manufacturing sectors experience an inflow of capital of 2.7% and 103

4.6% receptively. The commerce sector also experiences a capital inflow of 1.7%. Since the total supply of capital is exogenously fixed, this inflow is at the expense of the remaining sectors. The private services sector experiences the most severe capital outflow of 4.7%, followed by the transportation sector (3.4%). The transportation and private services sectors experience about a 1% capital outflow. Since labor supply is assumed flexible, all sectors experience substantial labor outflows ranging from 3.6% in the commerce sector to 10.1% in the private sector. These changes in resource allocation result in high levels of unemployment and are clearly reflected in changes in domestic output and composite good supplies. Domestic output of the agricultural sector decreases by 1.1% and domestic output supplied to the domestic market (as opposed to exports) increases by 15%. The composite goods supply (domestic output supplied to the domestic market plus imports) of agricultural goods decreases by 6%. The manufacturing sector experiences a 0.3% decrease in domestic output and a 2.5% decrease in output supplied to the domestic market. The composite good supply of manufactured goods decreases by 6.8% since imports have fallen by 9.6%. All other sectors exhibit declines in domestic output. Prices of composite goods in all sectors decrease by less than 1% with a 0.4% increase in the price of manufactured goods. Finally, the level of household utility decreases by 6.3%.

6.4

Full Separation

This section simulates a full-separation scenario between the WBGS and Israel. The idea of full-separation has gained considerable support in the face of the political stalemate between the PA and the Israeli government. However, the full-separation scenario is in turn separated into two possibilities. One is fullseparation in which both parties reach a mutually acceptable agreement in creating a Palestinian State bordering Israel, Egypt, and Jordan (‘peace scenario),

104

and the other possibility is a situation in which either Israel or the PA unilaterally declare separation ( ‘war’ scenario). 6.4.1 ‘Peace’ Scenario – Increases in Aid, Labor Demand and Labor Supply In the mutually agreed upon full–separation scenario, it is assumed that donor commitments and disbursements to the PA will be doubled and there world be a 50% increase in the demand for Palestinian workers in Israel. Moreover, the resolution of the question of the ‘Right of Return’ is expected to increase the labor supply by 30% as some Palestinian refugees return to the West Bank and the Gaza Strip. The results of this simulation exercise are reported in column S3 of Table 6.1. These exogenous shocks have the effect of increasing the composite goods supply. The increase in labor supply increases domestic output sold in the domestic market, and the price level is virtually unchanged. This increase in composite goods supply is met by an increase in imports since household consumption is now larger. Both government savings and household savings also increase to accommodate a higher level of total savings. Government revenue is also substantially higher due to the larger volume of imports. Total exports fall by 13.6% while there is a 23% increase in total imports. The largest decrease in exports is in the private services sector (24.4%), followed by the agricultural and transportation sectors (19% each). Exports of manufactured goods fall by 16% while exports of the commerce sector fall by 17.5%. The construction sector experiences a 14.5% decreases in its exports, and exports of public services fall by 12.7%. It is important to note that while the percentage changes are large – especially in the construction and services sectors – changes in exported volumes are relatively small. On the import side, the services sectors experience a relatively larger increase in imports than the manufacturing sector. Although total imports increase by 23%, the distribution of this change is not uniform across sectors. The public services sector experiences the largest increases in imports at 52.4%, followed by private services at 46.6%, and imports transportation goods increase by 105

25.6%. Agricultural imports also increase by 29.3%, followed by a 19.4% increase in imports of manufactured goods, which translates into $2,817 million of manufactured imports. Finally, commerce and transportation imports increase by 13.7% and 25.6% respectively. Household income and household savings increase by 20.5%, while government saving increases by 42%. The consumption budget of households also increases by 21%, and government revenue increases by 42%. The price of capital increases by 7.5%, while the price of labor is unchanged since it is fixed and is the numeraire in the model. Demand for consumer goods increases from $4,013 million to $4,839 million. In terms of the sectoral distribution of this increase, each sector experiences about a 20% increase in demand. In terms of resource allocation, capital inputs in the agricultural, manufacturing, construction, and commerce sectors decrease by 9.2%, 15%, 4.1%, and 8.6% respectively, while there is a capital outflow of 1.7%, 16%, and 22.3% in the transportation, private and public services sectors respectively. Since we have increased the labor supply in our simulation exercise, there is a labor inflow of 12.7%, 13.6%, and 10.8% in the agricultural, manufacturing, and construction sectors. The commerce sector experiences a relatively small labor inflow of 5.6%, and 17.4% and 34.1% labor inflows into the transportation and private services sectors respectively. Finally, the labor input in the public services sector increases by about 42%. These changes in resource allocation are not reflected in changes in domestic output as clearly as they were evidenced in the previous scenarios since the increase in demand for labor by Israel is associated with increases in foreign aid and labor supply. Domestic output of agricultural goods is virtually unchanged at $764 million, while the manufacturing sector experiences a 3.3% decrease in output. The construction and commerce sectors experience relatively small increases of 2.1% and 1.1% respectively, while domestic output in the transportation sector increases by over 10%. The two sectors that experi-

106

ence the highest increase in output are private services (23.9%) and public services (40.5%). Domestic output supplied to the domestic market also increases across all sectors with the smallest increase in manufacturing (0.2%), and the largest in public services (40.5%). This is followed by a 24% increase in private services and a 10.5% increase in transportation. The agricultural sector, construction, and commerce sectors experience relatively small increases in domestic output supplied to the domestic market of 4.8%, 2.6%, and 1.2% respectively. Composite goods supplies also increase since there are substantial increases in imports. The largest percentage increase is in public services at 41%, followed by a 28% increases in private services. The agricultural and manufacturing sectors experience an 11% increase, while composite supply in the construction and commerce sectors increase by about 2.5%. Finally, the increase in transportation composite goods is 16%. Prices of composite goods increase by relatively small amounts, and there is a 1.1% decrease in the price of manufactured goods. The largest price rise is in the private services sector (2.1%), followed by a 1.2% increase in the commerce sector. The agricultural and construction sectors experience relatively small price hikes of about 0.6%, while price increases in the transportation and public service sectors are a marginal 0.3%. Finally, household utility increases by an impressive 20.5% in this scenario. 6.4.2 ‘War’ Scenario – Increase in TC and Decrease in LD in Israel. This scenario, to some extent, represents the current state of affairs between Israel and the PA. Since September 2000, the political situation in the WBGS has sharply deteriorated and both parties have been in an unofficial state of ‘war’. The Palestinian Intifada or uprising has been a direct result of the failure of the Oslo peace accords. Israel, on its part, has considerably tightened its borders and imposed strict closures of the WBGS which has had a negative effect on the Palestinian economy. This scenario is simulated by assuming that Pales107

tinian transaction costs (TC) in the WBGS increase by 50%, and labor demand (LD) for Palestinian workers in Israel falls by 50%. The effect of an increase in transaction costs is to increase the price of imports, and domestic output supplied to the domestic market decreases because of a relatively lower domestic price level. The composite goods supply also decreases because of the decrease in imports. The fall in labor demand in Israel decreases the relative price of capital which partly explains the decrease in the domestic price level. The fall in household income causes a reduction in household consumption. Exports fall because of the deterioration in the terms of trade. Both household savings and government savings are lower, government revenue declines, and unemployment is higher. The results of this simulation exercise, column S4 in Table 6.1, show that the volume of exports falls by 56% and imports decreases by 30%. The largest decrease in exports is in public services (21.8%) followed by decreases of 18.1% and 117.8% for private services and manufactured goods respectively. Agricultural exports fall by 14.3%, and construction and commerce exports fall by 2.6% and 5.5% respectively. Total imports fall by 30% with the largest decline in public services (73%), followed by manufacturing (41.8%). Imports of agricultural goods increase by 34.4%, while construction and commerce imports decrease by 23.1% and 0.5% respectively. Private services imports and transportation good imports increase by relatively small amounts of 2.2% and 6.5% respectively. Household income and household savings decrease by relatively large amounts, 27% each, while government saving decreases by about 21%. Household consumption decreases by 27%, and government tax revenue falls by 20%. The price of capital decreases by an impressive 12.6%, while the wage rate is fixed since it is the numeraire in our model. Total demand for consumer goods decreases from $4,013 million to $2,922 million, with the largest drop in manufacturing (31.8%) since imports of manufactured goods constitute about 77% of total Palestinian imports. In terms of resource allocation, there is a capi108

tal inflow from the agricultural, transportation, and private service sectors, into the manufacturing, construction, and commerce sectors. Since we have reduced the demand for Palestinian labor in Israel, the resulting outflows are clearly evident in all sectors causing high rates of unemployment. The largest labor outflow is in the agricultural sector (50.1%), followed by transportation (35.3%) and private services (33.7%). Manufacturing and public service sectors exhibit about 20% of labor outflows each, and there is a 14% labor outflow from the commerce sector. Total domestic output falls from $8,448 million to $7,577 million, or about 10%, with the largest decrease in agriculture (37.1%), followed by transportation (27%). Again, this is an expected outcome since transaction costs are now higher and Palestinian workers who were previously employed in Israel are now unemployed. These decreases in domestic output are also reflected in decreases in domestic outpour supplied to the domestic market (as opposed to export markets) since the volume of trade is now lower. One important outcome of this ‘war’ scenario is that prices decrease by relatively large amounts. The manufacturing sector, on the other hand, experiences a price rise of 5.5%. All other sectors exhibit price reductions in the 4% - 9% range. Finally, household utility declines by a dramatic 27% in this ‘war’ scenario.

109

CHAPTER 7

SUMMARY AND CONCLUSIONS

7.1

Overview of Study

The objective of this study was to analyze trade policy options for the Palestinian Authority (PA) to halt the process of deindustrialization and promote exports. The current trading arrangement between Israel and the West Bank and Gaza Strip (WBGS) is largely determined by the Paris Protocol, which is a quasi-Customs Union with elements of both a Free Trade Area and Customs Union. Over 95% of exports and 75% of imports are with Israel. Moreover, Palestinian wage income earned in Israel has been an important source of revenue. However, this flow of funds, coupled with the inflow of foreign aid and remittances, has increased the economy-wide wage rate. Producers in non-tradable sectors have been able to increase prices to maintain their levels of profit, but producers in the tradable sectors have not been able to do so since prices in these sectors are determined in world markets. This capital inflow has therefore led to a shift of resources from tradable to non-tradable sectors causing deindustrialization. The symptoms of deindustrialization in the Palestinian economy are the high share of non-tradable sectors in GDP and a relatively high wage rate as compared to the wage rates in the neighboring economies of Jordan, Egypt, and Lebanon. Although there are no import tariffs in the traditional sense between the WBGS and Israel, there are other taxes such as value added (VAT) and purchases taxes. Moreover, the continued closures of the WBGS due to security concerns have resulted in abnormally high transaction costs for Palestinian exporters and importers alike. This has further reduced the competitiveness of the export sector and accelerated the process of deindustrialization. This study analyzed four trade policy simulations by constructing a tradefocused static CGE model. Chapter 2 provided an overview of the Palestinian trade structure and current policies, with a special emphasis on the trading ar110

rangements with Israel, the European Union, and neighboring Arab countries. It also described the Revenue Clearance System, which was put in place in the Oslo Accords in 1994 to facilitate the transfer of tariff and other revenues accruing to the PA. The Revenue Clearance System is an important source of revenue since the PA does not deal directly with its trading partners and all trade is channeled through Israel due to the absence of Palestinian national borders and independent trading facilities. Chapter 3 was a review of the theory of CGE models especially within trade policy analysis. A CGE model is a set of behavioral equations and identities that describe the workings and behaviors of economic agents (households, firms, government) and encompasses fundamental economic relationships such as market equilibrium and external balance. The core CGE model was introduced to illustrate the construction and mechanics of a typical CGE model, and to show the method through which the model equations are calibrated to replicate a benchmark equilibrium. Chapter 4 constructed this benchmark data set or Social Accounting Matrix (SAM) for 1998 to be used in the simulations of the Palestinian CGE model. The main source of data for the SAM was the Supply and Use Tables (SUT) compiled by the Palestinian Central Bureau of Statistics (PCBS) with technical assistance provided by various international organizations. This chapter also constructed an initial Input/Output (I/O) table, which was subsequently incorporated into the SAM. The level of sectoral disaggregation was determined by the purpose for which the CGE model was to be utilized. The SAM, therefore, contained 15 accounts where the rows of each account represent incomes, and the columns represent expenditures. These accounts are: 2 factors of production (L and K); 1 household; 1 government account with separate accounts for indirect taxes and tariffs; a capital account to demonstrate the equality between investment and savings; 7 production accounts – agriculture, manufacturing, construction, commerce, transportation, private and public services; and 1 account for the rest of the world (RoW) to satisfy the equality between foreign exchange 111

earnings and foreign exchange payments. The SAM was balanced with the sum of each column (expenditure) equal to the sum of the corresponding row (income), and thus satisfies the fundamental law of economics of equality between total expenditures and total income. Chapter 5 constructed the Palestinian CGE model – PALMOD – which was used in the next chapter for policy simulations. PALMOD is a static CGE model represented by 27 blocks of non-linear equations. Production was modeled using CES production functions for the 7 sectors. Household demand for goods and services was modeled using a Cobb-Douglas utility function subject to the budget constraint of the representative household. Both savings and investment were endogenous, and investment demand was a fixed share of savings. Domestic and imported goods were represented by a CES Armington aggregation function, and exports were represented by a constant elasticity of export demand. In considering the macro system-wide constraints (or ‘closures’) of the model, real constraints were represented by equilibrium in commodity and factor markets, and nominal constraints were represented by the investment-savings balance and external equilibrium. The model assumed a fixed level of foreign savings and the equilibrating variable was the exchange rate. The model was calibrated using data provided in the SAM and it was shown to satisfy the Walras Law and was homogenous of degree zero (since only relative prices matter). Several replication checks were conducted to make sure that the model generated a feasible and optimal solution. Chapter 6 conducted four simulation exercises to compare and contrast the effects of policy options available to the PA. The first simulation demonstrated the effects of trade liberalization through a 50% reduction in import tariffs. This simulation was also helpful in demonstrating the prohibitive effects of high transaction costs facing Palestinian traders. The results of this simulation were large increases in the volume of trade and a reduction in the prices of the composite goods. However, a 50% reduction in import tariffs caused an 18.6% 112

decline in government revenue, indicating the importance of import tariffs as a source of much needed government revenue. In this simulation, the adverse effects of deindustrialization were partly offset. The second simulation showed the importance and effects of a 50% reduction in Palestinian wage income from Israel. Although this capital inflow has traditionally been a vital source of income for Palestinian workers, it has caused deindustrialization by shifting resources from tradable into non-tradable sectors. The results of this simulation exercise were an increase in exports by a relatively small amount, and a decrease in imports. It was concluded that this policy was helpful in offsetting deindustrialization effects. The third simulation dealt with a scenario in which Israel and the PA reach a Final Status Agreement with the blessing of the international community. (‘Peace’ scenario). This scenario was simulated by simultaneously changing three policy instruments: doubling of foreign aid, increasing the demand for Palestinian labor in Israel by 50%, and increasing the supply of labor by 30% as Palestinian refugees return to Palestinian controlled areas. (‘Right of Return). The effects of this simulation were an increase in exports and a substantial increase in imports. To the extent that this increase in imports originated in countries other than Israel, there was a 20% improvement in the level of household utility. However, this scenario reinforced the adverse effects of deindustrialization. The fourth and final simulation dealt with a ‘war’ scenario in which transaction costs increased by as much as 50%, and demand for Palestinian labor in Israel fell by 50%. Both changes in these simulation variables were justified in light of the tight closures of the WBGS. The simulation results of this scenario demonstrated that the volume of trade fell substantially and there was a noticeable decrease in the domestic price level. Domestic output levels also fell and there was a 27% decline in household utility levels.

113

7.2

Conclusions

The effects of the four simulation exercises on export promotion are not the same. A reduction on transaction costs brought about by reductions in tariff rates substantially increases the volume of trade. Agricultural exports increase by 46% and manufacturing exports increase by 53%. There is also a clear resource shift from non-tradable to tradable sectors, especially into agriculture and manufacturing. The reduction in import prices has reduced the price of composite goods and improved the terms of trade. However, since foreign savings are fixed at $1,015 million, and government revenues are now lower, government savings are also lower. This is reflected in a reduction in government consumption. We can therefore conclude that this policy of trade liberalization partly offsets the adverse effects of deindustrialization and promotes exports. Moreover, there is a 7.4% increase in household utility because of higher income levels. The effects of a 50% reduction on Palestinian wage-income from Israel are to increase exports by 8%, and decrease imports by 10%. The reduction in imports is caused by the reduction in household incomes since Palestinian workers who were previously employed in Israel are now forced to work at the lower Palestinian wage rate. This ‘excess’ supply of labor is now absorbed into domestic production activities and the price of capital falls by 2.9%. Resource shifts are not as evident as in the tariff reduction scenario and there is a slight improvement in the terms of trade as prices of composite goods decrease. As before, given a fixed level of foreign savings, government savings decrease by 6.3%, which is reflected in a small decrease in government consumption. The general conclusion from this simulation exercise is that a 50% reduction in labor demand in Israel causes a slight improvement in the terms of trade as prices of composite goods decrease. To the extent that this price decrease can be passed on to consumers, this policy partly offsets the adverse effects of deindustrialization and promotes exports. The third scenario simulates three policy variables simultaneously: a doubling of foreign aid, a 50% increase in labor demand in Israel, and a 30% in114

crease in labor supply as Palestinian refugees return to areas controlled by the PA. We see that in this simulation, exports decrease by 13.5% and imports increase by as much as 23%. This is an expected result since capital inflows have increased substantially. Imports increase because household income has increased from $4,471 million to $5,387 million, and there is a 20% increase in household consumption. Moreover, there is considerable movement of resources from tradable to non-tradable sectors with marginal increases in prices of composite goods, leaving the terms of trade virtually unchanged. The results of this simulation exercise suggest that although deindustrialization effects are reinforced, the level of household utility increases by an impressive 20.5%. The fourth scenario simulates a 50% increase in transaction costs and a 50% reduction in labor demand in Israel. The effects of this simulation dramatically reduce the volume of trade causing a 27% decline in household utility levels. This decline in income also reduces domestic consumption and output. The price of composite goods also falls and the price of capital falls by 12.6%. Labor supply is also reduced substantially by 33% causing high levels of unemployment. The movement of resources from tradable to non-tradable sectors is not as evident as before since there is a 26% capital outflow from the agricultural sector and a 26.4% capital inflow into the manufacturing sector. However, both the transportation and private services sectors exhibit large capital outflows (15.4% and 13.3% respectively). The decline in household income causes a 27% fall in household utility.

7.3

Policy Implications

The circumstances under which the PA conducts trade and other policies are not the same as in other developing countries. We have seen that the trade agreement between Israel and the WBGS is largely based on the Paris Protocol, which is bound by Quantitative Restrictions (QRs) and Voluntary Export Restraints (VERs). Moreover, the absence of national borders and the presence of 115

Non-Traditional Non-Tariff Barriers - such as border closures due to security concerns – make it even more difficult to derive clear-cut policy implications and recommendations. Nevertheless, the PA has some autonomy over the conduct of its fiscal and trade policies. One important policy implication is derived from the results of the trade liberalization simulation (S1). We have seen the prohibitive effects of high transaction costs on both importers and exporters. The reduction in import tariffs resulted in a decrease of tax revenues, which must be somehow compensated to achieve a revue-neutral tariff reform66. Since a large proportion of intermediate inputs are imported, the PA could set a low tariff rate on intermediate goods and a relatively high tariff rate on imports of final goods. In addition, the PA could increase the value added (VAT) tax from its current level of 17%. Although this policy may not be politically feasible under the current economic hardships, an increase in VAT on non-tradable goods could reinforce the resource shifts into tradable sectors. The extent to which this policy will promote export-led growth will depend on reductions in transaction costs and improvements in market access. Another policy implication is the ability of the PA to absorb large numbers of Palestinian refugees and workers who were previously employed in Israel. It was shown that the reduction in wage income from Israel reduced imports by 10% and increased exports by 8%. The reduction in imports was explained by the fall in real household incomes since workers are now forced to work at the lower Palestinian wage rate. Although the price of labor is fixed, the price of capital decreased by 2.9%. This caused a substantial increase in the rate of unemployment. The ability of the PA to increase labor productivity through training programs and capital investments will determine the extent to which it can promote a strategy of export-led growth.

66

See Dahl, H., S. Devarajan, and S. van Wignbergen, 1986. “Revenue-Neutral Tariff Reform: Theory and An Application to Cameroon.” Country Policy Department Discussion Paper No. 1986-25. Washington D.C.: The World Bank.

116

The third simulation dealt with a scenario in which there is a Final Status Agreement. It was shown that increases in foreign aid, labor demand in Israel, and labor supply, reinforced the adverse effects of deindustrialization. An important policy implication of this scenario is the ability of the PA to improve market access. One way of achieving this objective is through membership in the World Trade Organization (WTO)67. The WTO is defined as a single institutional framework comprising of commitments on a wide choice of policy instruments affecting trade in goods and services, and providing for protection in intellectual property rights. These commitments are necessary for the enforcement of obligations, for dispute settlement, and for the monitoring of trade policies to increase transparency and credibility. The WTO also embodies the Most Favored Nation (MFN) principle which grants the same concession to countries with different bargaining powers in trade negotiations. Once accession is fully implemented, products can also be accorded ‘favorable treatment’ under the Generalized System of Preferences (GSP), which grants preferential treatment to the exports of least developed countries. Although WTO accession improves market access and lowers trade barriers, industrialized countries have been adopting ‘hidden’ trade barriers in the form of tariff escalation, countervailing duties, and antidumping policies – which have severely restricted the exports of developing countries. Moreover, a newly admitted country may face fiercer competition from countries with which it had no contact before. Finally, the ‘war’ scenario demonstrated that the volume of trade fell substantially. This was an expected outcome since the increase in transaction costs retards the Palestinian export sector. Although this may not be a ‘wise’ policy decision, it can nevertheless be imposed on the PA if the current Intifada continues. In recent weeks, there has been considerable debate on ‘fortifying’ the Green Line68. One implication of this policy choice may be to legitimize the de67

For the current state of market access conditions see Croome, J., 1998. “The Present Outlook for Trade Negotiations in the WTO.” Policy Research Working Paper No. 1992. Washington D.C.: The World Bank. For a developing country’s market access conditions, see OECD, 1997. 68 This is the name given to the pre-1967 borders of the West Bank with Israel.

117

marcation lines of a future Palestinian State. However, to reap the full benefits of a complete separation, alterative trading routes and channels must be established. This study has shown that market access improvements and lower trade barriers can reduce the detrimental effects of deindustrialization and promote exports only partially. However, two reservations need to be taken into consideration when looking at the results of the model. The first is the assumption of ‘full’ employment, and the second is the ‘open’ economy assumption. Since we have adopted a neo-classical model, the assumption of ‘full’ employment is ever present in our model. A more realistic assumption is to adopt alternative ‘closures’ which incorporate less-than-full-employment equilibria69. Secondly, the assumption of an ‘open’ economy is against the current spirit of repeated closures of the WBGS. Moreover, domestic policy reforms must also be implemented at an early stage to reap the full benefits of trade liberalization, and to eventually integrate the Palestinian economy into the world trading system. In this way, the study will pave the way for the computation of numerous other domestic and regional policy reforms in the future using a multi-country Computable General Equilibrium model.

69

For a discussion of alternative ‘closures’ see Dutt, A.K., 1990. Growth, Distribution, and Uneven Development. Cambridge : Cambridge University Press.

118

A.

APPENDIX

A1.

Variables

PMi

domestic price of import good

Ei

export good

Mi

import good

Xi

composite good supply (Armington aggregator)

XXDi

supply of domestic goods to the domestic market

Pi

price of composite good

PWEi

world price of exports

ER

exchange rate

PL

return to labor (numeraire)

PLISRAEL return to Palestinian labor in Israel (exogenous) PK

return to capital

PDi

price of good XD

LS

labor endowment

KS

capital endowment (exogenous)

Y

household income

HHSAV

household savings

GOVSAV government savings FORSAV

foreign savings (exogenous)

Ci

demand for consumer goods

CBUD

household consumption budget

XDi

domestic output

Ki

capital input

Li

labor input

Ii

investment goods

LG

government use of labor

KG

government use of capital 119

CGi

government demand for goods

TAXR

tax revenue (total government revenue including FORAID)

LISAREL

labor demand in Israel (exogenous)

REMIT

remittances (exogenous)

FORAID

foreign aid (exogenous)

TRICK

artificial objective variable

CPI

consumer price index. A2.

Parameters

γiA

Armington function share parameter

aAi

Armington function shift parameter

αiA

Armington function exponent

σiA

Armington elasticity of substitution

αiE

export demand elasticity

ty

income tax rate

tmi

import tariff

tci

tax rate on consumption goods

tli

tax rate on labor use

tki

tax rate on capital use

αiCG

Cobb-Douglas power for the government utility

αLG

Cobb-Douglas power for the government utility (labor)

αKG

Cobb-Douglas power for government utility (capital)

αiH

Cobb-Douglas power for the household utility on good i

αiI

Cobb-Douglas power for investment demand (from savings)

ai

efficiency parameter in the CES production function

αiF

CES power parameter in the production function

γiF

CES share parameter in the production function on K

σiF

CES elasticity of substitution between K and L

aij

technical ( input-output) coefficients

mps

marginal propensity to save. 120

REFERENCES Adelman, I. and J. E. Taylor, 1997. Village Economies. New York: Cambridge University Press. Alonso-Gamo and P. M. Alier, 1999. West Bank and Gaza Strip: Economic Developments in the Five Years Since Oslo. Middle East Department. Washington D.C.: International Monetary Fund. Astrup, C. and S. Dessus, 2000. “Trade Options for the Palestinian Economy: Some Orders of Magnitude”. MENA Discussion Paper. Washington D.C.: The World Bank. Armington, P., 1969. “A Theory of Demand for Products Distinguished by Place of Production.” IMF Staff Papers 16: 159-177. Arrow, K. and F. Hahn, 1971. General Competitive Analysis. San Francisco: Holdenday. Balian, O., 1997. “The Elasticity of Substitution between Palestinian and Foreign Workers: The PAL/FOE Model.” Advanced Econometrics Research Paper. American University, Washington D.C. Boadway, R. W. and J. Treddenick, 1978. “A General Equilibrium Computation of the Effects of the Canadian Tariff Structure.” Canadian Journal of Economics 11: 424-426. Brooke, A., D. Kendrick and A. Meeraus, 1992. GAMS: A User’s Guide. California: The Scientific Press. Brown D., Deardorff and R. Stern, 1996. “Some Economic Effects of the Free Trade Agreement between Tunisia and the European Union.” University of Michigan Department of Economics Working Paper No. 96-06. Corden, W. M. and P. J. Neary, 1982. “Booming Sector and Deindustrialization in a Small Open Economy.” Economic Journal 92:825-848. Croome, J., 1998. “The Present Outlook for Trade Negotiations in the WTO.” Policy Research Working Paper No. 1992. Washington D.C.: The World Bank. 121

Dahl, H., S. Devarajan, and S. van Wignbergen, 1986. “Revenue-Neutral Tariff Reform: Theory and an Application to Cameroon.” Country Policy Department Discussion Paper No. 1986-25. Washington D.C.: The World Bank. de Melo, J., 1998. “Computable General Equilibrium Models for Trade Policy Analysis in Developing Countries: A Survey.” Journal of Policy Modeling 10(4): 569-603. de Melo, J. and S. Robinson, 1989. “Product Differentiation and Foreign Trade in CGE Models of Small Economies.” Country Economics Department Policy Research Working Paper No. 144. Washington D.C.: The World Bank. de Melo, J. and D. Tarr, 1992. General Equilibrium Analysis of US Foreign Trade Policy. Cambridge: The MIT Press. Deardorff, A. V. and M. Stern, 1981. “A Disaggregated Model of World Production and Trade: An Estimate of the Impact of the Tokyo Round.” Journal of Policy Modeling 3: 127 – 152. Deardorff, A. V. and M. Stern, 1997. “Measurement of Non-Tariff Barriers.” OECD Economics Department Working Paper No. 179. Paris: OECD. Dervis, K., J. de Melo, and S. Robinson, 1981. General Equilibrium Models of Development Policy. Cambridge: Cambridge University Press. Devarajan, S. and J. Lewis, 1990. “Policy Lessons from Trade-Focused Two Sector Models.” Journal of Policy Modeling 12(4):625-657. Devarajan, S., J. Lewis, and S. Robinson, 1994. Getting the Model Right: A General Equilibrium Approach to Adjustment Policy. (Draft Manuscript). Diwan, I. and R. Sha’aban, eds., 1998. Development Under Adversity: The Palestinian Economy in Transition. Washington D.C.: The World Bank. Dutt, A. K., 1990. Growth, Distribution, and Uneven Development. Cambridge: Cambridge University Press. Francois, J. and K. Reinert, eds., 1998. Applied Methods for Trade Policy Analysis. Cambridge: Cambridge University Press. 122

Harrison, G. W., T. F. Rutherford, and D. G. Tarr, 1997. “Trade Policy Options for Chile: A Quantitative Evaluation.” Policy Research Working Paper No. 1783. Washington D.C.: The World Bank. Hertel, T. W. ed., 1996. Global Trade Analysis Project (GTAP): Modeling and Applications. Cambridge : Cambridge University Press. Hertel, T. W., 1985. “Partial vs General Equilibrium Analysis and Choice of Functional Form: Implications for Policy Modeling.” Journal of Policy Modeling 7(2):281-303. Jawhary, M. and R. Sha’aban, 1995. “The Palestinian-Israeli Trade Arrangement: Searching for Fair Revenue-Sharing.” Palestine Economic Policy Research Institute (MAS). Ramallah: West Bank. Kehoe, P. J. and T. J. Kehoe, 1994. “A Primer on Static Applied General Equilibrium Models.” Federal Reserve Bank of Minneapolis Quarterly Review Vol. 18. King, B., 1981. “What is a SAM? A Layman’s Guide to Social Accounting Matrices.” World Bank Staff Working Paper No. 463. Washington D.C.: The World Bank. Lloyd, P. J., 1991. “Aggregation by Industry in General Equilibrium Models with International Trade.” OECD Technical Paper No. 50. Paris: OECD. Lofgren, H., 1993. “Egypt’s Experience form CGE Modeling: A Critical Review.” American University of Cairo Economics Department Working Paper No. 9411. Cairo: American University. McKitrick, R. R., 1998. “The Econometric Critique of Computable General Equilibrium Modeling: The Role of Functional Forms.” Economic Modeling 15:543-573. Michalopoulos, C., 1998. “The Participation of Developing countries in the World Trade Organization”. Policy Research Working Paper No. 1906. Washington D.C.: The World Bank. Ministry of Planning and International Cooperation (MOPIC), 2000. Quarterly Monitoring Report on Donor’s Assistance. Ramallah: West Bank. 123

OECD, 1997. “Market Access for Least Developed Countries: Where Are The Obstacles?” Paper presented at the High Level Meeting on Integrated Initiatives for Least Developed Country’s Trade Development. Geneva: OECD. Palestinian Central Bureau of Statistics (PCBS), 1998. Expenditure and Consumption Results: The Final Report (January-December, 1998). Ramallah: West Bank. _____, 1999. Labor Force Survey: Annual Report - 1998. Ramallah: West Bank. _____, 1999. The Internal Trade Survey – 1998: Main Results. Ramallah: West Bank. _____, 1999. Existing Building Survey – 1998. Ramallah: West Bank. _____, 1999. Services Survey – 1998: Main Results. Ramallah: West Bank. _____, 2000. Foreign Trade Statistics – 1998: Main Results. Ramallah: West Bank. _____, 1999. Agricultural Statistics, 1997/1988. Ramallah: West Bank. _____, 1999. The Industrial Survey – 1998: Main Results. Ramallah: West Bank. _____, 1999. National Accounts: Preliminary Results, 1998. Ramallah: West Bank. _____, 1999. Palestinian Supply and Use Tables. Ramallah: West Bank. _____, 1999. Labor Force Survey: Annual Report – 1998. Ramallah: West Bank. _____, 1998. Palestinian Expenditure and Consumption Survey (PECS). Ramallah: West Bank. Rattso, J. and R. Torvic, 1998. “Economic Openness, Trade Restrictions, and External Shocks: Modeling Short-Run Effects in Sub-Saharan Africa.” Journal of Economic Modeling 15:257-286. Petri, P. A., 1997. “Trade Strategies for the Southern Mediterranean.” OECD Development Center Technical Paper. Geneva: OECD.

124

Pyatt, G., 1998. “A SAM Approach to Modeling.” Journal of Policy Modeling 10(3):321-352. Pyatt, G. and J. Round, 1985. Social Accounting Matrices: A Basis for Planning. Washington D.C.: The World Bank. Robinson, S., 1986. “Multisectoral Models of Developing Countries: A Survey.” Giannini Foundation of Agricultural Economics Working Paper No. 401. Berkley: University of California. Rutherford, T. F., E. Rutstrom, and D. Tarr, 1993. “Morocco’s Free Trade Agreement with the European Community.” Policy Research Working Paper No. 1173. Washington D.C.: The World Bank. Scarf, H. E., 1967. “The Approximation of Fixed Points of a Continuous Mapping.” SIAM Journal of Applied Mathematics 15: 1328-1343. Shiells, C. R., R. M. Stern, and A. V. Deardorff, 1986. “Estimates of the Elasticities of Substitution between Imports and Home Goods for the United States.” Weltwirtschaftliches Archiv 122: 497-519. Schiff, M., 1999. “Will the Real Natural Trading Partner Please Stand Up?” Development Research Department Working Paper. Washington D.C.: The World Bank. Schreiner, D. F. and D. Marcouiller, 1999. Computable General Equilibrium Modeling for Regional Analysis. Regional Research Institute. West Virginia University. Shoven, J. and J. Whalley, 1984. “Applied General Equilibrium Models of Taxation and International Trade.” Journal of Economic Literature 22:10071051. Shoven, J. and J. Whalley, 1992. Applying General Equilibrium. Cambridge: Cambridge University Press. Srinivasan, T.N. and J. Whalley, eds., 1986. General Equilibrium Trade Policy Modeling. Cambridge: The MIT Press. Stern, R. M, J. Francis, and B. Schumacher, 1976. Price Elasticities in International Trade: An Annotated Bibliography. London: Macmillan. 125

Taylor, L., 1990. Socially Relevant Policy Analysis. Cambridge: The MIT Press. Thissen, M., 1998. “Two Decades of CGE Modeling Lessons from Models of Egypt.” SOM Research Report 99CO2. The Netherlands: University of Groningen. UNCTAD, 1993. Selected Statistical Series on the Balance of Payments, Foreign Trade, Population, Labor Force, and Employment in the Occupied Palestinian Territories (West Bank and Gaza Strip), 1968-1987. Geneva: UNCTAD. _____, 1998. Palestinian Merchandise Trade in the 1990s: Opportunities and Challenges. New York: The United Nations. _____, 1998. The Palestinian Economy and Prospects for Regional Integration. New York: The United Nations. United Nations, 1968. System of National Accounts: Studies and Methods. Series F, No. 2, Revision 3. New York: The United Nations. United Nations Special Coordinator for the Occupied Territories (UNSCO), 1998. Special Report on the West Bank and Gaza Strip Private Economy. Jerusalem. _____, 1999. “Economic and Social Conditions in the West Bank and Gaza Strip.” Quarterly Report. Jerusalem. Varian, H., 1992. Microeconomic Analysis. Third Edition. New York: Norton. Zalai, E., 1998. “Computable General Equilibrium Modeling and Application to Economies in Transition.” Center for Economic Reform and Transformation (CERT). Edinburgh: The Heriot-Watt University.

126