Information about Europe Europa
Chapter

4 Palestinian Trade: Performance, Prospects, and Policy

Author(s):
International Monetary Fund
Published Date:
August 2001
Share
  • ShareShare
Information about Europe Europa
Show Summary Details
Author(s)
Geoffrey J. Bannister, Ulric Erickson and von Allmen 

Openness to foreign trade is probably the single most important factor in supporting the growth and development of a small economy like the West Bank and Gaza. One implication of this proposition, which is supported by a large amount of empirical research, is that the future trade regime of the Palestinian Authority (PA) is one of the policy decisions that will have the most far-reaching implications for the Palestinian economy’s prospects to grow and prosper.1

To set the stage for a discussion of the trade policy options for the PA, trade flows between the Palestinian economy and the rest of the world are examined. They are then compared with those predicted by a gravity model to gauge the potential for trade expansion under a new trade regime. When predicted trade flows are analyzed, results similar to other gravity models that analyze Palestinian trade are found. The Palestinian economy trades more with Israel and less with the rest of the world than would be expected. When these results are checked carefully, however, by introducing dummy variables for trade between Israel and the West Bank and Gaza and for trade between the West Bank and Gaza and the rest of the world, we find a more nuanced result: the Palestinian economy does not overtrade with Israel, but does significantly undertrade with the rest of the world. This is taken to mean that there is great potential to increase Palestinian trade with the rest of the world without substantially reducing its trade with Israel.2 Thus, there is potential for an overall increase of trade in the Palestinian economy that could significantly increase welfare and economic growth.

The realization of this potential, however, depends crucially on the type of trade policy implemented by an independent West Bank and Gaza, and the extent to which the current restrictions on trade are eased. Later in the chapter, the current state of Palestinian trade policy is reviewed and then options for the future are examined. We make a case for the PA to adopt a trade regime with a low, uniform tariff across all products, with no quotas or trade monopolies.3 It is further argued that free trade agreements should be limited to major markets and that a plethora of bilateral free trade agreements should be avoided. With respect to the future trade arrangement with Israel, which is likely to remain a key economic partner for the Palestinian economy in the foreseeable future, the chapter underscores the need to minimize transaction costs that the establishment of a customs border between the two entities inevitably would entail.

Trade Performance and Prospects

Some Data on Trade Performance

It is difficult to establish with certainty past Palestinian trade performance because of weak trade statistics. Existing data from the Palestinian Central Bureau of Statistics (PCBS) and IMF staff estimates suggest that exports of goods and services in 1999 (about US$740 million, 18 percent of GDP) were more than 60 percent higher than in 1994, and imports (at US$3.4 billion) were almost 140 percent higher.4 The implicit growth rates are high by any standard. By contrast, if trade performance is examined over a longer period of time—piecing together data from various sources—exports in 1999 were about the same as in 1980 in dollar terms but only a third of the 1980 level when expressed as a percentage of GDP.5 The decline in exports in the early 1990s, and the low level since then, are in large part due to the establishment of the recondite system of permits, inspections, and transportation procedures, as well as border closures, imposed by Israel for its security (see Chapter 3). Palestinian exports are currently low compared with other countries and regions. For example, the value of exports per capita is about half of that in Jordan (Figure 4.1).

Figure 4.1Comparison of Exports and Imports, 19991

Source: IMF, World Economic Outlook; Palestinian Central Bureau of Statistics: and IMF staff estimates.

1Goods and services.

Imports of goods and services currently equal approximately 80 percent of GDP—a high level by international standards but similar to the level recorded for the West Bank and Gaza in 1985 (Table 4.1). Figure 4.1 shows that the average for most regions in the world, as well as for transition and developing country groupings, lies at about 35–50 percent of GDP. The high level of imports and low level of exports is a reflection of the particular circumstances of the West Bank and Gaza, which relies on labor income from Israel and the settlements and, more recently, on substantial inflows of foreign aid disbursements. To some extent, labor income from Israel (20 percent of GDP) substitutes for exports (factor trade versus goods trade) and, together with foreign aid disbursements (13 percent of GDP), supports domestic expenditure significantly in excess of domestic production.

Table 4.1Palestinian Exports and Imports, 1980–99
1980198519871991199419951996199719981999
Exports of goods and services
in millions of U.S. dollars7576821,1631,042456611653661730740
as a percentage of GDP58675349131718171918
Imports of goods and services
in millions of U.S. dollars8948941,3711,5001,4332,0892,5092,7353,1693,398
as a percentage of GDP69886370425869718161
Memorandum items:
Net factor income from abroad
in millions of U.S. dollars148359314715575535469606828903
as a percentage of GDP11351433171513162121
The sum of exports and net factor income
in millions of U.S. dollars9051,0401,4771,7571,0311,1461,1221,2661,5581,643
as a percentage of GDP701036782303231334039
Sources: For 1980–91, World Bank (1993), Volume 2, Tables 1 and 10. For other years, IMF staff estimates based on data from the Palestinian Central Bureau of Statistics.
Sources: For 1980–91, World Bank (1993), Volume 2, Tables 1 and 10. For other years, IMF staff estimates based on data from the Palestinian Central Bureau of Statistics.

Results from the Gravity Model

A gravity model is estimated to examine the potential trade flows for the West Bank and Gaza and to shed some light on the effects of existing tariff and nontariff barriers. The gravity model has been found to be a particularly good predictor of trade flows (Frankel, 1997) and is consistent with theoretical models that explain the pattern of trade based on factor proportions, patterns of demand, and product differentiation.6 In its simplest form, the gravity model predicts bilateral trade flows on the basis of the economic size (GDP) of two countries and the distance between them. Trade is assumed to depend positively on the size of the two economies and negatively on the distance. In addition to GDP in the reporting and partner countries and the distance between them (LGDPPART, LGDPREP, LDISTANCE, respectively in our model), variables for the size of the population, language, trade restrictions, transportation costs, the existence of a border between the trading partners, and membership in regional trade arrangements are also included. Box 4.1 provides sources of data and a list of the variables used in the model, as well as an explanation of the model. Population is included on the premise that poorer countries (in terms of GDP per capita) trade less than richer countries. Given that the model already controls for the level of GDP as a separate variable, population would be expected to be negatively related to trade flows. The language variable is included to cover cultural proximity, assuming countries that speak the same language trade more with each other than countries that do not. Trade restrictions (tariffs, nontariff barriers) and transportation costs lower trade, all else being equal, and the border variable is included to capture the fact that a country tends to trade more with its neighbors than with other countries. One reason is that the infrastructure and transportation networks typically are better integrated between countries that share a border.7 A dummy variable (REPLL) is also included to capture whether the country is landlocked. Direct access to a seaport has been found to be significant in explaining differences in economic growth across countries; landlocked countries tend to grow slower (for example, Gallup, Sachs, and Mellinger, 1999). This variable is relevant to the case of the West Bank and Gaza, since Palestinian exporters and importers today have to use ports in Israel. Finally, dummy variables are included to capture the effects on bilateral trade from belonging to a regional trading arrangement.

Box 4.1.The Gravity Model

The standard gravity equation is given by:

where Tij is the flow of trade between country i and j, Yi and Yj are the GDP of countries i and j, Ni, and Nj are the populations of countries i and j, Dij is the linear distance between countries i and j, Aij includes other factors that influence trade, and eij is a log normally distributed error term. The equation is estimated by taking the logs and expressing it in linear form:

Regionij is a series of dummy variables that take the value of 1 if countries i and j belong to preferential trading arrangements, including AMU, ASEAN, EU, and the GCC. Languageij is a series of dummy variables that take the value of 1 if countries i and j share the same language (English, French, or Arabic), and can be considered a proxy for cultural similarity. Tariffi is the simple average MFN tariff of the reporting country; CIFi is the ratio of the value of imports CIF to FOB of the reporting country, which is a measure of transportation and transaction costs exclusive of tariffs; and REPLLi is a dummy variable that takes the value of 1 if the reporting country is landlocked.

A common problem with the estimation of gravity models is the loss of data points when the data is transformed into logs for those pairs of countries where recorded trade is zero. The assumption is adopted in this case that the underlying value of the log of trade will be a large negative number, but the observed value is zero. This effectively results in censored data on the value of trade.1 Estimation using ordinary least squares (OLS) will result in biased estimators. Accordingly, the gravity equation is estimated using a censored regression model (TOBIT).

The data used include economic variables for, and bilateral trade between, 64 countries, taken from Al-Atrash and Yousef (2000) and augmented to include Israel and the West Bank and Gaza. In addition to Israel and the West Bank and Gaza, the data set consists of 18 Arab countries and 43 other countries that represent over 90 percent of trade with the Middle East. For most of the sample, the data represent the average of values for 1995–97 to smooth out any yearly anomalies. Distance is measured as the direct distance between capitals.

Explanation of Variables
LGDPPARTLog of GDP partner country
LGDPREPLog of GDP reporter country
LPOPPARTLog of population in partner country
LPOPREPLog of population reporter country
LDISTANCELog of distance
AMUArab Monetary Union
ASEANASEAN
EUEuropean Union
GCCGulf Cooperation Council
ARABICLanguage-Arabic
ENGLISHLanguage-English
FRENCHLanguage-French
BORDERBorder
TARPARTTrade restrictiveness of partner country
TARREPTrade restrictiveness of reporter country
CIFPARTCIF partner country
CIFREPCIF reporter country
REPLLLandlocked
WBGSISRWBGS-Israel dummy
WBGSROWWBGS-rest of the world dummy
Sources of data: IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness, and IMF staff estimates.
Sources of data: IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness, and IMF staff estimates.
1See Frankel (1997).

Table 4.2 shows the results using exports and imports respectively as dependent variables.8 Most of the coefficients have the “right” signs and conform well to what would be expected from gravity equations. The adjusted R2 is 0.7, which is respectable. The coefficients on the log GDP of both reporters and partners in trade are positive and significant. The variable for distance is negative and significant, also as expected. Contrary to what would be expected, however, the coefficient on the population size is positive and significant for the reporter when exports are the dependent variable and for the partner when imports are the dependent variable. The dummy variables for regional trading arrangements are not significant, except for the ASEAN region. At first this seems surprising, although it is in line with the findings in other studies (for example, Al-Atrash and Yousef, 2000; and Hamilton and Winters, 1992). One possible interpretation is that the intensity of trade between members of geographic arrangements other than the ASEAN is already well explained by their size, their populations, proximity to each other, cultural similarities, lack of trade barriers, and transport costs.9 The language variables are significant and positive in all runs. The existence of a border is positive and significant, and the existence of high tariff barriers for the reporter and partner country are significant and negative. The transportation costs for reporter and partner are significant in explaining exports, though only partner’s costs are significant for imports. Finally, the dummy variable for landlocked status of the reporting country is significant and negative for both dependent variables.

Table 4.2Results from the Gravity Model Regressions—Base Run
VariableDependent Variable
ImportsExports
CoefficientStandard ErrorsCoefficientStandard Errors
Constant1.990.35*2.390.36*
LGDPPART1.010.02*0.840.03*
LGDPREP0.830.03*0.990.03*
LPOPPART0.070.03*0.020.03
LPOPREP0.020.030.060.03*
LDISTANCE–0.800.04*–0.840.04*
AMU0.140.430.180.44
ASEAN1.680.36*1.680.36*
EU0.230.150.300.15*
GCC–0.210.35–0.170.35
ARABIC0.540.12*0.520.12*
ENGLISH1.130.12*1.200.12*
FRENCH0.900.21*1.030.21*
BORDER0.980.17*0.800.17*
TARPART–0.030.00*–0.010.00*
TARREP–0.020.00*–0.040.00*
CIFPART0.010.00*–0.010.00*
CIFREP0.000.000.010.00*
REPLL–0.430.09*–0.220.09*
Adjusted R20.700.69
Standard Error1.761.79
Observations4,0324,032
Sources: IMF staff estimates based on data from IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness, and IMF Direction of Trade Statistics.

= indicates that the coefficient is significantly different from zero at the 5 percent confidence level.

= indicates that the coefficient is significantly different from zero at the 10 percent confidence level.

Sources: IMF staff estimates based on data from IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness, and IMF Direction of Trade Statistics.

= indicates that the coefficient is significantly different from zero at the 5 percent confidence level.

= indicates that the coefficient is significantly different from zero at the 10 percent confidence level.

Predicted Versus Actual Flows

The gravity model has been used in a number of studies to predict trade flows between Israel and the West Bank and Gaza (World Bank, 1993; Arnon, Spivak, and Weinblatt, 1996; and Arnon, Luski, Spivak, and Weinblatt, 1997). These studies compare actual trade flows with those predicted by the model. They view actual trade patterns as influenced by the many distortions found in the Palestinian and Israeli economies and predict trade patterns as those that would exist if these distortions are removed. By comparing actual and predicted trade patterns they attempt to measure the effects of the distortions on trade.

This approach seems intuitive, but there are some significant problems with using predicted trade flows from the gravity model. First, the predictions do not represent trade patterns that would exist under a free trade equilibrium. They are predictions of what the Palestinian economy’s trade would look like if it behaved, on average, like other economies in the sample, given the level of GDP, distance to trading partners, population, trade restrictions, and other factors that affect trade between the Palestinian economy and its trading partners. Thus, the predictions include the trade restrictions and distortions that can be captured in the model.

Second, since the predictions do not refer to a less distorted trade equilibrium, they do not provide information about possible trade creation and trade diversion effects. Trade creation and trade diversion occur when, as a result of the removal of trade barriers to a subset of a country’s trading partners, that country expands its trade with the rest of the world (trade creation) or diverts existing trade from one supplier to another (trade diversion).10 Since predictions of trade flows from the gravity model do not compare two trade equilibria where trade distortions are removed, and since they do not result from calculations that include parameters for the price elasticity of demand for imports or the substitutability of imports from different sources, they cannot be used to make statements about trade creation and trade diversion.11

Finally, and perhaps most important, looking at the predicted levels of trade from the gravity model does not allow assessment of the statistical significance of the results. It is not possible to attach any level of confidence or probability to the result that the Palestinian economy over- or undertrades with any of its partners from looking at predicted trade flows, since these are point estimates. To look at the statistical significance of bilateral trade flows it is necessary to introduce dummy variables that allow an assessment of the statistical importance of these trade flows in the context of all bilateral trade flows in the sample. This exercise is performed in the next section.

With these caveats in mind, the comparison of actual and predicted trade flows from the model, presented in Table 4.3, can be used to discuss some interesting features of trade of the Palestinian economy. The first result of the model is that the Palestinian economy should trade considerably less than it actually does. Specifically, the model predicts imports to be significantly lower and exports to be slightly higher than is currently the case, a finding that is consistent with results of other studies mentioned above. Three characteristics of the Palestinian economy can explain most of this result. First, the model only considers goods trade; trade in services and factor income are ignored. As mentioned above, Palestinian labor income from Israel and the settlements (factor trade) can be considered a substitute for exports (goods trade), a feature that becomes increasingly important in the presence of impediments to world markets for goods exports. Labor income from Israel, which amounted to US$725 million in 1998, also helps sustain a higher level of Palestinian imports, after controlling for the level of GDP and the other variables in the model.12 A second, and complementary, explanation is the importance of foreign aid. The West Bank and Gaza is one of the largest recipients of foreign aid, both as a percentage of gross national income (GNI) and on a per capita basis.13 In 1998, the total inflow of foreign aid was estimated at US$590 million. After taking into account all the variables in the model, it is clear that foreign aid can help raise the level of imports in an economy that otherwise lacks access to foreign capital.14 Finally, a third, and complimentary, explanation for the level of imports relates to the importance of private transfers for the Palestinian economy. There is no hard data available on the size of transfers from the Palestinian diaspora to the West Bank and Gaza, but they are generally considered to be very substantial. Such transfers allow a higher level of imports than would otherwise be possible.

Table 4.3Actual Versus Predicted Palestinian Trade FlowsIn millions of U.S. dollars)
ActualPredictedDifference
ExportsImportsExportsImportsExportsImports
Partner country
The world358.82,061.2423.51,165.664.8–895.6
Israel338.81,773.3135.6220.9–203.2–1,552.4
Rest of the world20.0288.0288.0944.7268.0656.7
Of which
United States0.323.632.7204.332.4180.8
European Union0.5156.3119.0440.8118.5284.5
Of which
France0.017.817.772.217.754.4
Germany0.013.626.0129.525.9115.8
Italy0.151.822.172.022.020.1
United Kingdom0.218.113.963.413.745.3
Japan0.04.919.580.719.575.9
Egypt0.024.720.228.220.23.4
Jordan16.215.015.614.1–0.6–0.9
Saudi Arabia1.40.07.414.26.014.2
Other1.463.573.4162.472.098.9
Sources: Palestinian Central Bureau of Statistics (PCBS), and IMF staff estimates based on data from PCBS, IMF, International Financial Statistics, IMF, World Economic Outlook database, and the IMF’s database on trade policy restrictiveness. Actual trade data refer to 1998. Goods trade only.
Sources: Palestinian Central Bureau of Statistics (PCBS), and IMF staff estimates based on data from PCBS, IMF, International Financial Statistics, IMF, World Economic Outlook database, and the IMF’s database on trade policy restrictiveness. Actual trade data refer to 1998. Goods trade only.

The second result of the comparison of actual and predicted trade flows is the prediction for bilateral trade flows. The model indicates that the trade relationship of the West Bank and Gaza with Israel is much more intense than would be expected, with actual exports and especially imports considerably higher than predicted values. Indeed, the higher than predicted imports from Israel more than account for the higher than predicted overall imports. The level of imports is almost certainly overstated, though, since a large part of Palestinian imports from other countries are shipped through Israel but are treated as having Israeli origin. (The same applies, but to a more limited extent, to exports.) The level of transshipments through Israel is difficult to measure, however, so it is impossible to tell by how much this trade is overstated in the data, and the reader should keep this caveat in mind when interpreting the results. It is well known, though, that the bulk of Palestinian imports do have Israeli origin.

One interpretation of the model’s result is that the Palestinian and Israeli economies are integrated to a much higher degree than would be expected from countries that have effective borders between them and are independent. In the regressions, a dummy variable is included for the border to capture the many synergies and increased interactions that take place between countries that share a border. Nevertheless, compared with doing business domestically (or in a well functioning customs union) the presence of a border is a barrier to trade. For example, estimates based on gravity models show that Canadian provinces along the U.S. border trade more among themselves than with the neighboring states in the United States, and the same is true for the states within the United States (see for example, McCallum, 1995). The absence of a physical (or customs) border between the West Bank and Gaza and Israel (although the two are adjacent) can help explain some of the apparent overtrade between the two economies.15 In addition, the West Bank and Gaza and Israel share the same currency and the same trade regime. The transportation infrastructure (roads) is also closely integrated, all factors that could explain the higher than predicted bilateral trade.16

The current high level of trade between the two economies may also be due to trade diversion caused by the complex system of permits, security checks, and transportation procedures, which are more cumbersome for Palestinian exports and imports with trading partners other than Israel, and by the lack of direct access to world markets. The transaction costs that these restrictions create, discussed at length in Chapter 3, make trade with the rest of the world very expensive and can be expected to divert some trade toward Israel.17 In addition, despite being on the coast, the West Bank and Gaza is considered landlocked, since it has to rely on Israel’s seaports and airport for exports and imports. Under normal security conditions this should not be much of a problem. Indeed, it would probably make more sense for the West Bank and Gaza to use existing—and very close—port facilities in Israel (in Ashdod and Haifa) rather than build its own. Because of the various restrictions and permit requirements that have been put in place, however, Palestinian traders do not have unimpeded and predictable access to world markets and the Palestinian economy is essentially landlocked. The regression results show that being landlocked significantly reduces overall trade, and the growth regressions in Chapter 2 showed that landlocked economies tend to experience slower growth in per capita GDP.

In sum, the results from the model suggest that there is considerable potential for trade expansion with the rest of the world, both on the import and on the export side, especially with industrial countries such as the United States and the European Union (EU), which is not surprising given their size (Table 4.3). But, there is also scope for more trade with Arab neighbors, especially Egypt. It is interesting to note that the predicted levels of trade with Jordan are very close to the actual levels, suggesting that there is not much scope for trade expansion there, under prevailing trade regimes. The latter is an important qualification—if trade restrictions in Jordan and the West Bank and Gaza were put to zero, the model suggests that exports and imports between the two economies could be about 60 percent higher than they are today. As indicated above, it is difficult to know how much confidence we can have in these results, so the next section attempts to examine the statistical validity of these results by introducing dummy variables into the gravity model equation that control for trade between the Palestinian economy and Israel, and the Palestinian economy and the rest of the world.

Does the West Bank and Gaza Overtrade with Israel?

To examine the issue of whether the West Bank and Gaza overtrades with Israel or undertrades with the rest of the world, the approach used in a number of recent studies is followed and dummy variables are inserted for trade between two or more partners. A positive and significant coefficient on the dummy variable would indicate that the partners in the group trade more with each other than would be expected from the model given their level of GDP, distance, and other explanatory variables. A negative and significant coefficient would indicate the opposite. If the coefficient is not significant (that is, not statistically different from zero), it indicates that the trading partners do not trade with each other more or less than would be expected in the model. 18

The results of the exercise are presented in Table 4.4. A dummy variable was introduced with the value of 1 for trade between the West Bank and Gaza and Israel and zero elsewhere (WBGISR). Another dummy variable was introduced with the value of 1 for trade between the West Bank and Gaza and all other partners except Israel (WBGROW). While the coefficients on WBGISR are positive (suggesting overtrade) they are not statistically significant. Thus, when bilateral trade between the West Bank and Gaza and Israel is examined in the context of bilateral trade of all other countries in the sample, no support can be found for the hypothesis that they overtrade with each other. Put differently, no significant evidence is found that trade between Israel and the West Bank and Gaza is higher than what we might expect on average given their proximity, GDP, population, and other variables. In fact, to the extent that the actual bilateral trade flows are overrecorded, the two economies may actually undertrade with each other. This result contradicts the earlier conclusion from the difference between predicted trade and actual trade.

Table 4.4Results from the Gravity Model Regressions with Dummies for Palestinian Trade with Israel and the Rest of the World
VariableDependent Variable
ImportsExports
CoefficientStandard ErrorsCoefficientStandard Errors
Constant2.230.35*2.630.35*
LGDPPART0.980.02*0.810.03*
LGDPREP0.810.03*0.980.03*
LPOPPART0.080.03*0.040.03
LPOPREP0.030.030.070.03*
LDISTANCE–0.800.04*–0.850.04*
AMU0.080.420.120.43
ASEAN1.660.36*1.660.36*
EU0.220.150.300.15**
GCC–0.380.35–0.350.35
ARABIC0.690.12*0.670.12*
ENGLISH1.120.12*1.200.12*
FRENCH0.800.21*0.930.21*
BORDER0.920.17*0.740.17*
TARPART–0.030.00*–0.020.00*
TARREP–0.020.00*–0.040.00*
CIFPART0.010.00*–0.010.00*
CIFREP–0.010.00*0.010.00*
REPLL–0.270.09*–0.060.09
WBGISR1.451.251.361.27
WBGROW–1.510.17*–1.590.17*
Adjusted R20.710.70
Standard Error1.751.77
Observations4,0324,032
Sources: IMF staff estimates based on data from the PCBS, IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness and IMF Direction of Trade Statistics. Goods trade only.

= indicates that the coefficient is significantly different from zero at the 5 percent confidence level.

= indicates that the coefficient is significantly different from zero at the 10 percent confidence level.

Sources: IMF staff estimates based on data from the PCBS, IMF, International Financial Statistics, IMF, World Economic Outlook database, the IMF’s database on trade policy restrictiveness and IMF Direction of Trade Statistics. Goods trade only.

= indicates that the coefficient is significantly different from zero at the 5 percent confidence level.

= indicates that the coefficient is significantly different from zero at the 10 percent confidence level.

On the other hand, the coefficients on the dummy variable for trade between the West Bank and Gaza and the rest of the world (WBGROW) are negative and statistically significant (as indicated by the asterisk) in all the equations. Thus, there is evidence that the West Bank and Gaza undertrades with the rest of the world, both on imports and exports. According to the magnitudes of the coefficients, Palestinian exports to the rest of the world are almost 80 percent below what would be expected from a country with the characteristics of the Palestinian economy.19 This suggests that there is significant scope for higher trade between the West Bank and Gaza and the rest of the world.

Trade Prospects

The previous sections indicated that there is considerable scope for higher Palestinian exports, especially to the EU and the United States. But what products could the West Bank and Gaza export? And what are the policies needed to support growth in the tradable sector? It is beyond the purpose of this chapter to assess the comparative advantage of the Palestinian economy, and Chapter 2 discusses the factors that are important for long-term economic growth—many of which relate directly to trade (for example, reduced uncertainty and output volatility, a simple, open trade regime, return of infrastructure, legal and regulatory framework, and appropriate macroeconomic policies). We would note, though, that the potential for trade is easily underestimated when making export projections, since the current production structure and export base are usually assumed to remain unchanged, and elasticities are applied to the existing set of goods that a country exports (Romer, 1994). Policy reforms and other significant changes to the Palestinian business environment can release forces that accelerate inventions and the adoption of innovations of new goods and services. Reforms and investments in technology that make trade flows more predictable and less costly (see Chapter 3) might very well set in motion a complete transformation of the goods and services produced and exported by the Palestinian economy, changing its comparative advantage.

The West Bank and Gaza’s small natural resource endowment is not necessarily an important constraint for long-term growth and development, although it can be important in the short run.20 A somewhat surprising, but common, finding in empirical studies is that countries rich in resources tend to grow more slowly than other countries, all else being equal.21

The gravity model focused on merchandise trade, but the West Bank and Gaza might also see important growth in the export of services. For example, tourism has expanded rapidly in recent years, and although the turmoil since late 2000 dealt a severe blow to the tourism industry, once the political situation improves, it should be a sector with considerable potential for growth. Furthermore, recent background studies (see below) for industrial estates argue that the West Bank and Gaza might have a comparative advantage in the provision of some services in the region, for example call centers for technical support and software conversion and customization.

The background studies for the establishment of industrial estates in the West Bank and Gaza suggested activities that could be successful in attracting foreign direct investment (see Chapter 3 for a more detailed discussion of industrial estates).22 The study compared access to world markets, political risk, investment environment (including taxation and investment incentives), human resources, transportation facilities, utilities, telecommunications, and access to capital in the planned industrial estates in the West Bank and Gaza with industrial estates elsewhere in the region. The study found that factor costs are relatively high in the West Bank and Gaza, but productivity was also higher than in Egypt, Jordan, and Turkey. The study concluded that the West Bank and Gaza is not competitive in the region in production using low skilled labor, whereas it is competitive in production using high skilled labor, like engineers. Another conclusion was that, compared with industrial estates in Jordan, Egypt, Turkey, and Dubai, there was great scope for the Palestinian economy to better exploit its closeness and integration with Israeli industry, the most developed in the region. The study also found that the quality of telecommunications is generally good, albeit more costly than elsewhere in the region. The key negative factors for foreign direct investment identified in the study related to political risk, cumbersome export and import procedures that often lead to long delays, land and building costs, and high costs of electricity, water, and telecommunications. Based on these findings, the study recommended that the industrial estates in the West Bank and Gaza target light manufacturing industries that can exploit the ties with Israel (for example, higher-end apparel manufacturing, finished consumer electronics, and electrical appliances), professional services in the information technology sector (for example, software conversion and customization, and technical help desks), and location-based services related to the future cargo operations at Gaza airport (for example, logistics handling and other services related to air transportation).

Trade Policy for an Independent West Bank and Gaza

The future trade policy of the West Bank and Gaza will have profound implications for the Palestinian economy’s potential to generate growth and prosperity. Trade openness can bring benefits through a variety of channels, including: (i) improved resource allocation in line with social marginal costs and benefits; (ii) access to better technologies, inputs, and intermediate goods; (iii) improving the economy so as to better exploit economies of scale; (iv) greater domestic competition; and (v) exposure to favorable externalities, like the transfer of technical knowledge (see Dornbusch (1992) for an overview).

The Current Setting

The West Bank and Gaza is currently in a customs union with Israel, as set out in the Protocol on Economic Relations of 1994 (or the Paris protocol). A customs union was the only trade regime that could satisfy Israel’s insistence on the absence of borders and the Palestinian demands for continued access to Israel’s markets (Kleiman, 1994, p. 355). The external tariff rates and some excise tax rates, such as the purchase tax, as well as quality and safety requirements are all decided by Israel, with a few exceptions. Israel’s (unweighted) average MFN tariff rate is 8.8 percent, and the dispersion of rates is very high (0–250 percent). Most of Israel’s trade goes through its free trade agreements with the United States and the EU, so its effective tariff is very low (around 2.5 percent in 1999).23 The Paris Protocol grants the PA some limited trade policy autonomy, however, on the goods specified on its lists: A1, A2, and B. Since Israel does not have trade (or diplomatic) relations with several countries in the region, Palestinian trade with these countries is also restricted, or severely limited to the above-mentioned lists. The trade provisions and performance under the Paris Protocol will not be reviewed here, as they have been reviewed in detail by others, including Calika (1998), and Kessler (1999). Kessler (1999) argues convincingly that the disappointing trade performance over the past six years is not so much due to the design of the trade provisions of the Paris Protocol itself as to the lack of implementation of these provisions and to the difficult situation on the ground.

Before the turmoil that started in late September 2000, the Palestinians and Israelis appeared ready to agree, as part of the final status negotiations, to replace the customs union with a free trade arrangement. This would require the establishment of a customs border between the two economies. At the time this chapter was written, the future trade regime between the two parties was very much an open question, but it seems probable that it will involve a customs border, and this will be taken as given. It will be argued that the PA should consider applying a low uniform import tariff across all products with exceptions strictly limited to the goods that qualify for duty-free treatment under free trade agreements, but these types of agreements should be few and limited to key markets. The Palestinian economy would also benefit most from a trade regime characterized by the absence of trade quotas and trade monopolies. We would note, though, that a simplified customs union between the West Bank and Gaza and Israel with a more liberal external trade regime and with improved access to world markers for Palestinian traders, would also be an attractive option.

The Case for a Low Uniform Tariff

A nondistortionary trade regime with a low uniform tariff rate is considered more welfare enhancing than regionalism in the form of a customs union or free trade arrangement.24 Indeed, for the West Bank and Gaza, a case can be made that the uniform rate should be zero because since it is a small economy—in the sense that it has to take world prices as given for the goods it imports and produces—the imposition of trade tariffs unambiguously diminishes its welfare (Corden, 1974). Nontariff barriers, such as quotas, are even worse since they have much the same effect as tariffs but without the benefit of tariff revenue.

While only a few countries have adopted a tariff structure with zero tariffs across the board and with hardly any nontariff barriers, some of those that have—Hong Kong and Singapore, for example—are among the most successful economies in the world. The use of tariffs in other countries has been justified on several grounds, but most of them do not hold under closer scrutiny. For example, governments have imposed tariffs to achieve objectives relating to the balance of payments, fixed revenue, income distribution, health, social values, security, or some other objective. But tariffs are not the best policy instruments to achieve any of these objectives, and in practice, high and dispersed tariffs distort the allocation of resources, inhibit the use of foreign technology, and reduce the comparative advantage that a labor-intensive developing country could exploit (Subramanian, Ibrahim, and Torres-Castro (1993); and Subramanian (1994)). In reality, one reason why most countries have tariffs is that once tariffs are in place—for whatever reason—they are very difficult to remove even long after their initial validity (if there was one) has expired. In the following sections, we will discuss the possible role of tariffs for protectionist and revenue purposes in the West Bank and Gaza.

Tariffs for Protectionist Purposes

Tariffs are often used to protect certain industries. The argument might be that some industries are not considered viable yet at world market prices (as in the classic, but discredited, infant industry argument) or are considered to have strong positive externalities (as in the New Trade Theory—see Helpman and Krugman, 1989).25 In practice, however, the application of tariffs is more likely to reflect the influence of vested interest groups. The case for protection of so called infant industries is flawed theoretically and, even if it could be justified, tariffs would not be the most appropriate instrument for such protection—the government should rather use either a temporary production subsidy or intervene at the source of the market failure (Krueger, 1984; and Subramanian, Ibrahim, and Torres-Castro, 1993). Moreover, based on the international experience, there is good reason to be skeptical about any government’s ability to identify the industries that would qualify for protection because of their externalities (to pick the winners). As soon as the government shows that it is inclined to grant special treatment to certain industries, a general pressure for protectionism is certain to arise.

The Palestinian economy operates in a very difficult environment, one that includes the risk of border closures, cumbersome transportation procedures, and costly security-related inspections and restrictions (see Chapter 3). As discussed above, the easing of these obstacles may well set in motion a process of complete transformation of the Palestinian tradable sector, and it is impossible to say now what the Palestinian comparative advantage will be in an environment of significantly fewer restrictions. Targeting industries for trade protection would definitely slow this transformation process. Rather, complementary domestic reforms that facilitate the adjustment of the Palestinian economic structure towards its comparative advantage would be important—for example—improved infrastructure, a strengthened legal and regulatory framework, economic policies that ensure macroeconomic stability, good governance, and further development of the financial sector.

Tariffs for Revenue Purposes

Tariffs are often used in developing countries because they bring important fiscal revenues to the government when the domestic tax institutions are not efficient at raising revenue with less distortionary means, such as VAT or income tax. The PA’s fiscal revenue (excluding foreign grants) was close to 22 percent of GDP in 2000—a respectable level for an economy at this level of development. Revenue came mainly from VAT and, to a lesser extent, income tax, with import duties accounting for only 2.8 percent.26 About 60 percent of all tax revenue is collected by the government of Israel on behalf of the PA, a consequence of the customs union and the sizeable Palestinian imports (80 percent of GDP), most of which come from or through Israel. Abandoning the customs union means that the PA would be responsible for the collection of all its taxes, and it would be realistic to expect tax revenue to decline initially, until the capacity of the PA’s own tax administration was strengthened. Given the uncertainty over future tax revenue, it would be imprudent to lower sharply the customs tariffs (or any other tax) until the revenue base has stabilized. It would be important to preserve the average tariff rate, however, and the PA would be well advised to take the opportunity to replace the current, highly dispersed tariff structure with a low, uniform tariff that applies to all goods.27 In 1999, PA customs revenue amounted to NIS 112 million, equal to less than 1 percent of total imports to the West Bank and Gaza (most imports come from Israel and are not subject to any tariffs). Because of a lack of detailed trade data, it is impossible to determine with certainty the effective tariff rate on goods coming from the rest of the world, but using the data in Table 4.3, the effective tariff rate would be 9 percent. The uniform rate should be set so as to avoid too much of a loss in tariff revenue, but it is more important that it be low, preferably around 5 percent, and not more than 10 percent. If these rates still would entail some revenue loss, it would be preferable that revenue be raised by other measures.28 As explained in the following chapter, the demographic dynamics in the West Bank and Gaza can be expected to have a positive effect on fiscal revenue over the medium term, creating room for adopting a low, uniform tariff.

Avoid a Plethora of Bilateral Free Trade Arrangements

Having an independent trade policy would allow the PA to enter into free trade arrangements with other countries, for example, Jordan, and enter into regional trade arrangements like the Arab Free Trade Area. The benefit to an individual country from joining a preferential trade arrangement in terms of trade, investment, and growth is generally difficult to assess. It depends on whether the agreement would lead to trade creation or diversion for the particular country, and whether it would lead to investment creation or diversion. Often, regional trading arrangements are based more on political aspirations than economic considerations, and the results from the gravity model showed that in most cases, belonging to a regional trade arrangement was not a significant explanation for bilateral trade patterns (except for the ASEAN). The theoretical case against regionalism more generally, as expounded in Bhagwati and Panagariya (1996a and b) for example, is compelling, and empirical evidence presented in Vamvakidis (1999) suggests that investment and growth performance has been distinctly better in countries that have undertaken broad trade liberalization compared with those that have undertaken discriminatory trade liberalization by joining regional trade arrangements. Thus, it would not be desirable to enter into bilateral free trade agreements with a long list of countries, but it would make better sense to have such agreements with key markets. A low uniform tariff on imports from the rest of the world would help minimize the trade diversion effects caused by such preferential trade agreements. One reason why it would be important for the PA to avoid entering into several preferential trade agreements is because overlapping agreements tend to differ from each other, making the trade regime complex and nontransparent. This is particularly the case when it comes to rules of origin requirements, the focus of the next two sections.

Rules of Origin Requirements

Rules of origin define those goods that will qualify for duty-free treatment under a free trade arrangement and usually require that the good be fully grown, produced, or manufactured in the exporting country, or meet one or more of the following three commonly used criteria: (i) a certain percentage of the good’s value added must be produced in the exporting country; (ii) the processing undertaken in the exporting country must change the classification of the good according to a pre-defined list (product tariff shift); or (iii) a specific operation must be undertaken in the exporting country.

Rules of origin are inherently costly instruments and can only cause trade diversion, never trade creation (Krueger, 1995). With many different definitions and procedures for rules of origin, it becomes a cumbersome process for exporters (and importers). Costs from rules of origin requirements arise in at least three different areas. First, the administrative costs of complying with rules of origin are not insignificant. These entail at a minimum the issuance and verification of certificates of origin for every shipment that is presented for clearance at customs. Simply obtaining the necessary documents to verify origin entails costs, which have been estimated at 3–5 percent of the import value in the European Free Trade Area (EFTA).29 If the rules are more complicated these costs can be higher. One way to hold down these costs is to grant approved exporter status to regular exporters who would then not need to provide documentation for each shipment. Second, the rules of origin requirement can be very costly in terms of trade and investment diversion and can undermine efforts for trade diversification. To the extent that the rules favor products from a less efficient producer within the free trade area compared with a more efficient producer outside the area, they will lead to intra-free trade area trade in products that could be sourced more cheaply and efficiently outside the area. In the case of Israel and the West Bank and Gaza, rules of origin would most probably lead to imports from Israel into the West Bank and Gaza that could otherwise be sourced more cheaply from the rest of the world. Such trade diversion is common in other free trade areas, (for example, NAFTA and Mercosur). Finally, perhaps the most convincing argument against rules of origin is that they are inherently discriminatory trade instruments (that is, they tend to be designed explicitly to favor some suppliers over others) and as such they are not only inefficient, but also subject to tremendous pressures in terms of the political economy of protection. 30 Thus, there is a real danger that rules of origin will be used to favor certain producers or products over others, even within the free trade area. In a sense, they can become instruments of tailored protection.31

Rules of Origin Requirements in PA’s Existing Trade Agreements

That rules-of-origin requirements can add complexity to the trade regime is evident from the PA’s own experience with its trade agreements with Canada, the EFTA states, Egypt, the EU, Jordan, Saudi Arabia, and the United States. For example, to qualify for duty-free treatment under the trade agreement with the United States, a product must either be wholly obtained in the West Bank and Gaza or its production cost in the West Bank and Gaza plus Israel must be at least 35 percent of its final value. A different definition is used for textile products. The agreement with the EU establishes rules-of-origin requirements to qualify for duty exemption as a change in tariff classification according to an agreed list. Otherwise, the product must undergo a specific operation in the West Bank and Gaza or in the EU. There is also a limit on the use of materials not originating in the West Bank and Gaza or in the EU. The trade agreement with Egypt defines rules of origin in yet a different way—local (West Bank and Gaza) costs must account for at least 40 percent of production costs for industrial products. The trade agreement with Jordan requires that the product should have at least 35 percent of value added produced locally. In addition to these different definitions, there are procedural differences between the agreements. A Palestinian exporter to the EU must get its certificate of origin stamped by the PA Ministry of Finance, while for exports to the Arab countries, the certificate of origin should be stamped both by the Ministry of Economy and Trade and the Chamber and Commerce. There are also great variations in documentation required to get the origin certified.

The Future Trade Arrangement with Israel

It is an open question what trade regime will eventually be adopted between Israel and the West Bank and Gaza, but it is useful to review some of the implications of moving away from the present customs union arrangement. Some of these apply irrespective of whether Israel and the West Bank and Gaza enter into a free trade agreement or whether trade between them will take place on an MFN basis. The key economic implications are these: trade policy independence for the PA, transaction costs, and fiscal aspects.32 Fiscal aspects are dealt with at greater length in Chapter 5.

Independent Trade Policy

Abandoning the customs union in favor of a free trade arrangement or a MFN-based trade regime means that the PA would have trade policy independence vis-à-vis the rest of the world. The value of such trade policy independence for the PA would seem to depend on: (i) the current and future trade policy in Israel; and (ii) the future trade policy chosen by the PA vis-à-vis third parties.

On the first point, Israel has undertaken extensive trade liberalization in recent years, and overall, the trade regime is quite liberal compared with other countries in the region. Nonetheless, its trade system is still not very transparent and contains important protectionist elements. As mentioned, Israel’s unweighted average MFN tariff is 8.8 percent, and the dispersion is very large with the highest tariff rate at 250 percent. In addition, according to the World Trade Organization (WTO), Israel’s nontariff measures remain an impediment to trade.33 The system of import and export licensing, prohibitions, and quotas remain complex. The purchase tax and mark-up system (TAMA) and protection taxes also raise prices of certain goods considerably. Furthermore, a frequent comment from Palestinians is that the standards and health requirements that Israel places on imports are not compatible with the needs of the Palestinian economy and constitute nontariff barriers. The possibility of adopting a more transparent and liberal trade regime for the West Bank and Gaza is an attractive feature of moving away from the customs union.34 At the same time, if the trend of the past ten years toward a more liberal trade regime in Israel—with a phased reduction in tariffs—were to continue (which seems likely), this would reduce the attractiveness of leaving the customs union. The purchase tax also, was reduced significantly in the summer of 2000, and the plan seems to be to phase it out more or less completely.

On the second point, while with an independent trade policy the PA would be able to adopt a very liberal and transparent trade regime vis-à-vis the rest of the world, it cannot be ruled out that it could come under intense pressure to adopt a more protectionist trade regime. Indeed, in view of the PA’s past inclinations when it comes to direct interference in the economy, it is not a foregone conclusion that a liberal and transparent trade regime would be adopted. The sheer act of lobbying for protectionism is also a typical example of unproductive use of resources in an economy. A way to handle the concerns over the PA’s future trade policy would be to announce at the outset a liberal, MFN-based trade regime and make this commitment credible by joining the WTO as soon as possible.

Transaction Costs

As mentioned, transaction costs (including for transportation) are very high in the West Bank and Gaza. Reducing these transactions costs and making the conduct of foreign trade simpler and more predictable will probably have implications for the performance of the tradable sector that are as profound as the choice of trade regime.

Abandoning the current customs union arrangement in favor of a free trade arrangement or MFN trade relationship requires the establishment of a customs border between the two economies. The customs border is required to check for origin, classification, standards, valuation, and taxes and more generally in order to prevent trade deflection that could arise since the two entities can have different trade policies vis-à-vis third parties. All import taxes (including those on imports from Israel) would have to be collected at these entry/customs stations. In itself, the establishment of a customs border will inevitably add to transaction costs, which in turn will reduce trade volumes. At the same time, the choice of trade regime with Israel might affect the possibility to improve the situation on the ground. A customs border might allow simpler security procedures for Palestinian exports and imports with the rest of the world.

It will be important to make investments in technologies that minimize the additional transactions costs arising from the establishment of borders (see Chapter 3). It should be emphasized that these transaction costs would arise even if all Palestinian exports to Israel were to qualify for duty-free treatment under a free trade arrangement with Israel. In addition, not all trade currently taking place between Israel and West Bank and Gaza (all without import duties) would qualify for duty-free treatment under a free trade arrangement because of high import content, and it would thus be subject to import taxes.

Fiscal Implications

There are important fiscal implications associated with moving away from the customs union with israel. Chapter 5 discusses them in some detail, but a brief summary will be given here. There would be additional expenditures related to an expanded customs bureaucracy, including standards institutes, which would need to be established. The cost of creating a customs authority with customs posts between the two economies could be quite high. Fewer customs posts would reduce the administrative costs somewhat but at the expense of higher transportation costs.

On the revenue side, about 60 percent of the PA’s revenue is currently collected by Israel and transferred under the clearance system. If the PA were to collect all taxes on its own, there most likely would be an initial period of loss in tax revenue, since the Israeli system is relatively more efficient. At the same time, the PA would be able to collect import taxes and purchases tax that are not currently transferred to the PA. Dumas (1999) estimates the amount of tax leakage to be as much as 5 percent of GDP. With the reduction in purchase taxes in Israel in the summer of 2000, the amount of tax leakage could now be somewhat lower. Also on the positive side, abandoning the customs union would give the PA some fiscal policy independence to set tax rates and would reduce the risk of the tax revenue clearance being disrupted.35 Tax revenue (and expenditure) in the West Bank and Gaza has risen rapidly in the last five years, and on current trends, the West Bank and Gaza might soon have the characteristics of a premature welfare state, with tax rates at a level (in percent of GDP) that is not conducive to economic development. The overall tax level in Israel is quite high (33 percent of GDP, compared with 23 percent of GDP in the West Bank and Gaza). In particular, the Israeli indirect tax regime might not be optimal for the West Bank and Gaza, especially the system of purchase tax in combination with TAMA.

Concluding Remarks

The bulk of Palestinian trade today is with Israel, although the exact share is unknown because of data weaknesses. Israel’s large share in Palestinian exports and imports is not in itself evidence of trade distortions, but a good case can be made that the security restrictions and very complex and costly trade and transportation procedures for trade with other countries have tended to favor trade with Israel. The transaction costs that the impediments on the ground have caused can only lead to trade diversion not trade creation. Using a gravity model, however, we find no significant evidence that trade between Israel and the West Bank and Gaza is higher than what might be expected given their proximity, GDP, population, and other variables. The results suggest that Israel will remain a key trading partner to the West Bank and Gaza under almost any type of trade arrangement between the two economies. We find evidence, though, that there is considerable scope for expansion of Palestinian trade with the rest of the world, in particular with the EU and the United States. We take this to mean that a reduction in the very high transaction costs caused by the complex system of permits, security controls, and transportation restrictions could be expected to lead to increased trade with the rest of the world without (a significant) reduction in Palestinian trade with Israel.

Some trade diversification will surely take place as the trade infrastructure improves, including with free access to a seaport and a better functioning airport. Such diversification would take place even under the existing customs union. While there is scope for more trade with countries in the region, the results from the gravity model suggest that the region is unlikely to present a serious alternative to the Israeli market for Palestinian exporters. With respect to the future trade policy of the West Bank and Gaza, this chapter has argued that the PA should adopt an open, nondiscriminatory, and transparent trade regime characterized by the absence of quotas and trade monopolies. The chapter also makes the case for the PA to adopt a low, uniform import tariff rate (5–10 percent) across the board with no exceptions. Further, it does not make much sense for the PA to enter into a long list of bilateral free trade agreements, but rather, it should limit free trade arrangements to countries and regions that represent important markets for Palestinian exporters. Given the volatility and uncertainty of the economic situation in the West Bank and Gaza and its negative effect on growth and investment in the past (see Chapter 2), implementing measures to ensure stability and simplicity of the policy regime should probably be a priority. In this context, the simplicity of the trade regime we propose is a virtue. It should be announced ahead of time and explained with the rationale behind it. It should be bound at the WTO (or at least it should be announced that such binding is being sought, even though this may take a while).

1See, for example, Barro and Sala-ì-Martin (1995), Frankel and Romer (1999), and Sachs and Warner (1995b and 1997). Rodriguez and Rodrik (1999) offer some skeptical observations on the causal link between trade and growth.
2Given that some trade enters the Palestinian economy from the rest of the world through Israel it might appear as if some trade diversion were taking place, but even in this case, direct trade with the rest of the world, as opposed to trade through Israel, might be a more efficient outcome for the Palestinian economy.
3Astrup and Dessus (2000), in a study of trade options for the West Bank and Gaza, argue against a continuation of the customs union with Israel in favor of a most-favored-nation (MFN) trade relationship between the two economies.
4Data from the PCBS show that more than 75 percent of Palestinian goods imports come from Israel while almost all (96 percent) of Palestinian goods exports go to Israel. The dependency is exaggerated, however, since the data include goods from other countries that are imported (exported) via Israel to (from) the West Bank and Gaza. At the same time, to the extent that there is unrecorded trade, it is safe to assume that it is mostly with Israel.
5The data before 1994 come from the World Bank (1993), volume 2, Tables 1 and 10 (based on Israeli Central Bureau of Statistics), and for 1994 and onward they are IMF staff estimates based on data from the PCBS.
6For theoretical discussions of the gravity model, see, for example, Anderson (1979), Helpman and Krugman (1989), Helpman (1987), and Feenstra, Markusen, and Rose (2000). Recent applications of the gravity model to the middle east region include Al-Atrash and Yousef (2000) and Blavy (2000).
7Another is that, irrespective of distance between economic centers, countries that share borders will interact more with each other than those that do not. For example, while the economic centers of the United States and Mexico are quire far apart geographically, and while they have different languages and cultures, the border between them is one of the most economically active areas in the world and includes a significant amount of cross-border regional and local trade.
8We chose to focus on imports and exports separately, rather than the sum of the two, since we expect the effect to be different in each case. Also, when we ran the regression with the sum of exports and imports, it added little value. We focus only on goods trade; no data are available on the direction of trade in services.
9This is consistent with the Krugman-Summers claim that most of trade can be explained by geographic proximity, see Frankel (1997).
10The concept of trade creation and trade diversion was originally developed in the context of the establishment of a customs union (see Viner, 1950). For a good explanation of the concept see Panagariya, 2000.
11On this point see Hamilton and Winters (1992), p. 89.
12At the same time, the role of factor income should not be overstated since the sum of exports and factor income has declined considerably over the past 15 years as a share of GDP (Table 4.1).
13Chapter 5 of the IMF study in Alonso-Gamo, Alier, Baunsgaard, and Erickson von Allmen (1999) showed that, in 1994–96, foreign aid disbursements to the West Bank and Gaza amounted to about US$200 per capita, compared with around US$25 in sub-Saharan Africa. Only Israel received more foreign aid during this period on a per capita basis.
14If foreign aid causes an increase in the price of nontraded goods and services it might also adversely affect exports (the Dutch disease effect), unless the foreign aid also helps raise productivity in the traded sector that offsets the negative effect in competitiveness.
15Nevertheless, trade from the West Bank and Gaza is subject to considerable transactions costs and in that (economic) sense, there is a border with Israel, but it is clear that this border is considerably less restrictive, for Palestinian traders, than the border to the rest of the world, which includes a customs border.
16Rose (2000) finds that countries that share a common currency tend to trade three times more than they would otherwise.
17Furthermore, the absence of banking services in the West Bank and Gaza until 1994 meant that Palestinian merchants had to rely on their Israeli counterparts for formal and informal trade credit facilities, thus creating close ties with the Israeli economy.
18Frankel (1997), Coe and Hoffmaister (1998), Al-Atrash and Yousef (2000), and Subramanian and Tamirisa (2001). As mentioned above, the power of using this approach is that it allows for a statistical evaluation of the results.
19The marginal effect of a dummy variable on the dependent variable in the gravity model equation can be calculated as (in percentage terms): 100*[exp(β)–11, where β is the coefficient on the dummy variable. See Coe and Hoffmaister (1998), footnote 9.
20Currently, the most important natural resource for industrial development in the West Bank and Gaza is probably stones and marbles. Natural gas might become an important resource with the recent findings off the coast of Gaza.
21See, for example, Sachs and Warner (1995a), and Rodriguez and Sachs (1999), for a discussion of why this is the case.
22The study, which is summarized in The Services Group (1999), focused on the export potential from industrial estates in the West Bank and Gaza (which would not include sectors like tourism), and although the projected export potential from these estates was quite small, they are nonetheless suggestive of sectors where the Palestinian economy might see an expansion in exports.
23It is not known with precision what the effective tariff rate is because of the weaknesses in trade data, and because large amounts of imports are exempted. For example, donor-financed imports are exempted from duties as are those imports that qualify for the tax exemptions under the investment promotion law.
24This result is usually for global welfare, but might not necessarily apply for a particular country.
25The only way to justify infant industry protection is in the presence of market failures, but even then, the government should not use trade intervention. Erecting trade barriers usually introduces additional distortions that counteract other benefits of the government policy. For a good discussion, see Corden (1974). Chapter 9.
26Foreign grants equaled roughly 12 percent of GDP in 1999, including for the operations of the United Nations Relief and Works Agency (UNRWA).
27A credible political commitment to a uniform structure also reduces the incentive for vested interest groups to lobby for high tariffs for their own products (Panagariya and Rodrik, 1993). Ideally, it is the effective tariff rate that should be uniform rather than the nominal one.
28To start with, profits from the Palestinian Commercial Services Company (a public investment company) or its replacement, the Palestinian Investment Fund, should go to the PA budget. There is also a case for more judicious use of exemptions.
29Krueger (1995, p.15).
30See Krueger and Krishna (1995), for a discussion of the experience under NAFTA and Yeats (1997), for a discussion of Mercosur.
31The clearest example of this is the negotiation of rules of origin under NAFTA, which are tailored to provide protection for certain industries. For television sets, for example, the rules were designed to protect the only remaining U.S. producer. In textiles, the rule is “fiber forward” meaning that all stages of production from the spinning of the fiber forward must take place in a NAFTA country. The automobile sector also benefited from tight rules of origin, specified in terms of the percentage of value (65 percent) in the final product that had to originate within the free trade area. One observer has qualified the rules of origin for automobiles in the NAFTA as “complex and convoluted… that can only be read to have an intent to restrain import competition for automotive parts and automotive goods from third countries.” Jackson (1997, p. 169).
34It should be pointed out, though, that as regards tariffs, protection is highest for footwear, textile products, and agricultural products, goods that also are produced in the West Bank and Gaza.
35The Paris Protocol allows the PA to have a VAT two percentage points lower than Israel and to adopt a lower purchase tax on goods on the Protocols lists: A1, A2, and B.

    Other Resources Citing This Publication