Poverty, inequality and trade facilitation in low- and middle-income countries
D. Empirical results
Tables 6 to 9 present the GMM regression of the poverty rate and poverty gap index (measured at the poverty line of $1.25 and $2 PPP/day), per capita GDP and the Gini index on the trade facilitation variables and other explanatory variables. The results from the OLS regression are shown in the annex.
Several points should be noted. First, the instruments are the first lagged difference of trade variables and other explanatory variables. Since the number of observations is not large, the second-order lagged differences cannot be used as instruments. The Sargan test of over-identifying restrictions is performed and reported in the tables 6 to 9. The null hypothesis that over-identifying restrictions are valid is not rejected in all the regressions.
Second, the Arellano-Bond test for zero autocorrelation of the first-order and second-order in first-differenced errors was performed. The P-value of the test in all the regressions was above 0.1, indicating that the null hypothesis of no autocorrelation was not rejected.
Third, in each regression only one variable of trade facilitation was used in order to avoid the multi-collinearity problem. As indicated in table 3, the trade facilitation variables are strongly correlated. Thus, for each outcome, there are four models with different measures of trade facilitation.
As mentioned above, there are only 52 observations for which data are available both on poverty measures and the logistics performance indexes of trade facilitation. There are no panel data on the logistics performance indexes of trade facilitation. However, the OLS regression of outcomes was also tried on the logistics performance indexes. The results were very similar to the OLS results using the time and documents for imports and exports as the trade facilitation measures.32
Table 6 shows the association between the trade facilitation variables and the poverty rate at the poverty line of $1.25 PPP per day. Except for the variable “documents for exports”, all the trade facilitation variables are statistically significant at the 5 per cent level. Countries requiring a large number of documents for imports and more time for imports and exports are more likely to have a higher poverty rate. One additional document for imports can be associated with a 0.77 percentage point increase in the poverty rate.
One additional day in the time needed for exports and imports might increase the poverty rate by 0.49 and 0.47 percentage points, respectively. The sign of trade facilitation variables in GMM regressions is the same as in the OLS regression. Improvement in trade facilitation by reducing the number of documents and times for exports and imports is also negatively associated with the poverty gap.
Although the over-identification test is not rejected, the exogeneity of the GMM-type instruments cannot be fully convincing. Thus, the estimate of trade facilitation variables in the GMM regression could be explained as an association between trade facilitation and outcomes instead of a causal effect of trade facilitation.
Table 7 presents the regressions of the poverty rate at the poverty line of $2 (PPP) per day. All the trade facilitation variables are statistically significant and have the same sign as the regression in table 6. The point estimates in table 7 are larger than in table 6, since the poverty rate measured at the poverty line of $2 (PPP) a day is higher than at the poverty line of $1.25 (PPP) a day.
Table 8 shows a negative relation between per capita GDP and the number of documents and days needed for exports and imports. The OLS regression shown in the annex also shows a negative association. An additional document for exports and imports is associated with a reduction in per capita GDP by the equivalent to 2.9 per cent and 1.5 per cent of per capita GDP, respectively. It should be noted that the average number of documents required for exporting and importing a commodity is 7.3 and 8.3, respectively. It
32 The signs of the logistics performance indexes and the signs of the time and documents to export and import in regression of outcomes are opposite, since the higher value of the logistics performance indexes means improvement in trade facilitation, while the higher value time and documents to export and import means depreciation in trade facilitation.
Table 6. GMM regression of poverty rate at poverty line of $1.25 per day (PPP) ExplanatoryPoverty rate (%)Poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents for exporting0.3570.107 (number)(0.483)(0.220) Documents for importing0.773***0.297** (number)(0.272)(0.122) Time taken to export (days) 0.494***0.162*** (0.108)(0.044) Time taken to import (days)0.474***0.163*** (0.128)(0.052) Population density (people/km2 )-0.367**-0.2470.2760.493-0.174**-0.1210.0420.126 (0.173)(0.196)(0.266)(0.349)(0.079)(0.088)(0.108)(0.142) 2005Base omitted 2006-0.995*-0.670-0.254-0.404 -0.549**-0.425-0.309-0.349 (0.526)(0.580) (0.729) (0.837)(0.240) (0.260) (0.295) (0.341) 2007 -1.701***-1.186*-0.362-0.184 -0.826*** -0.635** -0.387-0.307 (0.559)(0.630) (0.816) (0.977)(0.255) (0.282) (0.331) (0.398) 2008 -1.838***-1.261* 0.2600.646-0.649**-0.441 0.031 0.197 (0.566)(0.644)(0.904)(1.130)(0.258) (0.289) (0.366) (0.460) 2009 -1.314**-0.6700.4960.839-0.421-0.181 0.162 0.309 (0.669) (0.739)(0.974)(1.196)(0.305) (0.331) (0.394) (0.487)
Constant46.884***31.070-25.994-48.479 21.187***14.247-3.204-12.072 (16.998)(19.498)(26.958)(36.063) (7.744) (8.743)(10.920)(14.680) Observations224224224224224224224224 Sargan test of over-identifying8.1293.9572.4274.1776.40312.361.8583.208 restrictions:
×
2 statistic and0.9750.8610.9990.9970.6020.4980.9990.999 P-value Source:Estimation based on the World Bank database. Heteroskedasticity-robust standard errors are shown in parentheses. *** p<0.01, ** p<0.05, * p<0.1.Table 6. (continued) ExplanatoryPoverty rate (%)Poverty gap index (%) variables Model 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4
Table 7. GMM regression of poverty at poverty line of $2 per day (PPP) ExplanatoryPoverty rate (%)Poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents for exporting1.342**0.370 (number) (0.669)(0.332) Documents for importing1.154*** 0.553*** (number)(0.374)(0.188) Time taken to export (days)0.696*** 0.332*** (0.150)(0.074) Time taken to import (days) 0.748*** 0.336*** (0.189)(0.090) Population density (people/km2 )-0.437*-0.257 0.473 0.896*-0.262** -0.1730.1740.346 (0.240) (0.270)(0.371)(0.515)(0.119) (0.135)(0.182)(0.247) 2005Base omitted 2006 -1.170-0.734-0.119-0.220-0.747** -0.523-0.248 -0.326 (0.729)(0.798)(1.015)(1.237)(0.362) (0.401) (0.498)(0.593) 2007 -2.368***-1.651*-0.5160.026-1.285*** -0.928**-0.389 -0.209 (0.774)(0.865)(1.137)(1.444)(0.384) (0.435) (0.558)(0.692) 2008 -3.403*** -2.633*** -0.5400.462-1.392*** -1.001** -0.0010.356 (0.784)(0.886)(1.260)(1.670)(0.389) (0.445) (0.618)(0.800) 2009-2.340**-1.739* -0.2360.721-0.952** -0.554 0.1850.514 (0.926)(1.015)(1.356)(1.767)(0.460) (0.510) (0.665)(0.847)
Table 7. (continued) ExplanatoryPoverty rate (%)Poverty gap index (%) variables Model 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Constant 58.772**41.025-38.146-83.826 33.501*** 22.653*-14.898-33.154 (23.537)(26.796)(37.540) (53.305)(11.683)(13.460)(18.424) (25.548) Observations224224224224224224224224 Sargan test of over-identifying1.3582.7478.2842.6125.6932.4502.1751.897 restrictions:
×
2 statistic and0.9990.9490.8740.9990.6810.9640.99991.000 P-value Source:Estimation based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.Table 8. GMM regression of log of GDP per capita and export volume, PPP (Constant 2005 international $) ExplanatoryLog of GDP per capitaLog of export volume variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents for exporting-0.029***-0.051*** (number)(0.009)(0.012) Documents for importing-0.015***-0.034*** (number)(0.005)(0.008) Time taken to export (days)-0.009***-0.022*** (0.002)(0.005) Time taken to import (days)-0.011***-0.018*** (0.002)(0.004) Population density (people/km2 ) 0.018*** 0.015*** 0.006-0.002 0.013*** 0.009*-0.013-0.017* (0.003) (0.003)(0.004)(0.006)(0.004)(0.005)(0.008)(0.010) 2005Base omitted 2006 0.017* 0.012 0.004 0.005 0.023* 0.019 0.018 0.023 (0.009) (0.010)(0.012)(0.015)(0.014)(0.014)(0.020)(0.022) 2007 0.069*** 0.062*** 0.046*** 0.036** 0.083*** 0.073*** 0.052** 0.045* (0.010) (0.011)(0.014)(0.018)(0.015)(0.016)(0.022)(0.026) 2008 0.103*** 0.095*** 0.066*** 0.049** 0.114*** 0.103*** 0.046* 0.037 (0.010) (0.011)(0.015)(0.021)(0.015)(0.016)(0.027)(0.032) 2009 0.047*** 0.044*** 0.020 0.005-0.012-0.015-0.060**-0.068** (0.012) (0.013)(0.016)(0.022)(0.017)(0.018)(0.028)(0.033)
Table 8. (continued) ExplanatoryLog of GDP per capitaLog of export volume variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Constant 7.058*** 7.210*** 8.173*** 8.943***22.203*** 22.479*** 24.799***25.158*** (0.287)(0.315) (0.428)(0.630)(0.412)(0.464)(0.841)(1.019) Observations222222222222198198198198 Sargan test of over-identifying6.3702.7718.3944.5878.18516.3912.015.546 restrictions:
×
2 statistic and0.6050.9470.4950.9170.4150.2280.6050.986 P-value Source:Estimation based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.means that the elasticity of the per capita GDP with regard to the number of documents required for exporting and importing is around 0.21 per cent and 0.22 per cent, respectively.
Similarly, the increase in the time taken to export and to import is negatively correlated with per capita GDP.
To examine whether the export is the channel through which the trade facilitation affects the GDP, we run regression of log of export values on trade facilitation. The trade facilitation variables are negative and very significant. Improvement in trade facilitation would help countries significantly increase exportation. The point estimates are larger than those in the regression of GDP on trade facilitation.
An increase in the number of documents and time required for exports and imports is associated with a small decrease in per capita GDP but a relatively large increase in inequality. It means that income distribution can be worsened by increasing the number of documents and time required for exports and imports. Table 9 shows this relationship. If the number of documents needed for imports increases by one, the Gini index can increase by 0.4 percentage points. An additional day in the time taken to export and import is associated with 0.22 and 0.25 percentage point increases in the Gini index, respectively.
E. Conclusion
Since trade facilitation can help to boost economic growth, it can also help poverty and inequality reduction. This chapter attempts to examine the effect of trade facilitation on poverty, GDP, exports and income inequality in low- and middle-income countries. Trade facilitation is measured by the number of documents and the number of days needed for exports and imports. The findings show that improvement in trade facilitation is positively correlated with exports and per capita GDP, and negatively correlated with poverty and inequality. More specifically, deterioration in trade facilitation – which is measured by the increase in the number of documents required and days taken for exporting and importing a good – can reduce per capita GDP, albeit to a small amount. Countries requiring a larger number of documents and more time for imports and exports tend to have higher levels of poverty (measured by the headcount and poverty gap index) and inequality (measured by the Gini index) than other countries.
It should be noted that although this study is aimed at estimating the causal effect of trade facilitation on GDP and poverty in developing countries, using instrumental variable regressions, caution is advised in the interpretation of the causal effect as the exogeneity of GMM-type instruments is not fully convincing. Another limitation is the small number of observations used in this study, which does not allow for estimation of the heterogeneous effects of trade facilitation. The extent to which trade facilitation affects GDP, poverty and inequality in a country depends on the structure of the economy; therefore, it can vary across different countries. While estimating the heterogeneous effects of trade facilitation is beyond the scope of this study, it is an important aspect for future studies.
Table 9. GMM regression of Gini index
Explanatory variables Model 1 Model 2 Model 3 Model 4 Documents for exporting (number) 0.444
(0.320)
Documents for importing (number) 0.400**
(0.179)
Time taken to export (days) 0.217***
(0.059)
Time taken to import (days) 0.245***
(0.071) Sum of documents for exporting
and importing
Sum of time taken to export and import
Population density (people per km2 -0.174 -0.107 0.128 0.333
of land area) (0.127) (0.138) (0.154) (0.209)
2005 Base
omitted
2006 0.288 0.456 0.621 0.598
(0.362) (0.386) (0.405) (0.474)
2007 0.022 0.259 0.596 0.826
(0.386) (0.421) (0.455) (0.558)
2008 -0.571 -0.320 0.358 0.748
(0.391) (0.431) (0.507) (0.649)
2009 -0.764* -0.553 -0.038 0.279
(0.462) (0.496) (0.548) (0.688)
Constant 55.683*** 49.077*** 24.544 3.517
(12.211) (13.392) (15.204) (21.008)
Observations 217 217 217 217
Sargan test of over-identifying 11.47 6.868 12.35 8.708
restrictions:
×
2 statistic and 0.648 0.961 0.499 0.892 P-valueSource: Estimation based on the World Bank’s database.
Heteroskedasticity-robust standard errors in parentheses.
*** p<0.01, ** p<0.05, * p<0.1.
Annex
Annex table 1. Definition of trade facilitation measures
Trade facilitation Detailed definition
measure
Documents required for All documents required per shipment to export goods are exporting (number) recorded. It is assumed that the contract has already been agreed upon and signed by both parties. Documents required for clearance by government ministries, customs authorities, port and container terminal authorities, health and technical control agencies and banks are taken into account. Since payment is by letter of credit, all documents required by banks for the issuance or securing of a letter of credit are also taken into account. Documents that are renewed annually and that do not require renewal per shipment (for example, an annual tax clearance certificate) are not included.
Time taken to export Time is recorded in calendar days. The time calculation for (days) a procedure starts from the moment it is initiated and runs until it is completed. If a procedure can be accelerated for an additional cost, the fastest legal procedure is chosen. It is assumed that neither the exporter nor the importer wastes time and that each commits to completing each remaining procedure without delay.
Procedures that can be completed in parallel are measured as simultaneous. The waiting time between procedures – for example, during unloading of the cargo – is included in the measure.
Documents required All documents required per shipment to import goods are for importing (number) recorded. It is assumed that the contract has already been agreed upon and signed by both parties. Documents required for clearance by government ministries, customs authorities, port and container terminal authorities, health and technical control agencies, and banks are taken into account. Since payment is by letter of credit, all documents required by banks for the issuance or securing of a letter of credit are also taken into account. Documents that are renewed annually and that do not require renewal per shipment (for example, an annual tax clearance certificate) are not included.
Time taken to import Time is recorded in calendar days. The time calculation for (days) a procedure starts from the moment it is initiated and runs until it is completed. If a procedure can be accelerated for an additional cost, the fastest legal procedure is chosen. It is assumed that neither the exporter nor the importer wastes time and that each commits to completing each remaining procedure without delay.
Procedures that can be completed in parallel are measured as simultaneous. The waiting time between procedures – for
example, during unloading of the cargo – is included in the measure.
Logistics performance Logistics professionals’ perception of the efficiency of country’s index: Efficiency of customs clearance processes (i.e., speed, simplicity and customs clearance predictability of formalities), on a rating ranging from 1 (very low) process (1 = low to to 5 (very high). Scores are averaged across all respondents.
5 = high)
Logistics performance Logistics professionals’ perception of country’s quality of trade index: Quality of trade and transport related infrastructure (e.g., ports, railroads, roads and transport-related and information technology), on a rating ranging from 1 (very infrastructure low) to 5 (very high). Scores are averaged across all (1 = low to 5 = high) respondents.
Logistics performance Logistics professionals’ perception of the ease of arranging index: Ease of arranging competitively priced shipments to a country, on a rating ranging competitively priced from 1 (very difficult) to 5 (very easy). Scores are averaged shipments (1 = low to across all respondents.
5 = high)
Logistics performance Logistics professionals’ perception of country’s overall level of index: Competence competence and quality of logistics services (e.g., transport and quality of logistics operators, customs brokers), on a rating ranging from 1 (very services (1 = low to low) to 5 (very high). Scores are averaged across all
5 = high) respondents.
Logistics performance Logistics Performance Index overall score reflects perceptions index: Overall of a country’s logistics based on efficiency of customs clearance (1 = low to 5 = high) process, quality of trade- and transport-related infrastructure, ease of arranging competitively priced shipments, quality of logistics services, ability to track and trace consignments, and frequency with which shipments reach the consignee within the scheduled time. The index ranges from 1 to 5, with a higher score representing better performance.
Logistics performance Logistics professionals’ perception of how often the shipments index: Frequency with to assessed country reach the consignee within the scheduled which shipments reach or expected delivery time, on a rating ranging from 1 (hardly consignee within ever) to 5 (nearly always). Scores are averaged across all scheduled or expected respondents.
time (1 = low to 5 = high)
Logistics performance Logistics professionals’ perception of the ability to track and index: Ability to track trace consignments when shipping to the country, on a rating and trace consignments ranging from 1 (very low) to 5 (very high). Scores are averaged (1 = low to 5 = high) across all respondents.
Source: World Bank database. Available at http://data.worldbank.org/.
Annex table 1. (continued)
Trade facilitation Detailed definition
measure
Annex table 2. Summary statistics of variables
Variables Obs. Mean Std. Dev. Min. Max.
Documents required for 225 7.209 1.972 3 14
exporting (number)
Documents required for 225 8.471 2.644 3 20
importing (number)
Time taken to export (days) 225 26.236 15.972 9 102
Time taken to import (days) 225 30.831 17.543 9 101
Population density 224 95.54 135.00 3.20 1 142.29
(people per km2 of land area)
East Asia and the Pacific 225 0.098 0.298 0 1
Europe and Central Asia 225 0.302 0.460 0 1
Latin America and Caribbean 225 0.347 0.477 0 1
Middle East and North Africa 225 0.053 0.225 0 1
South Asia 225 0.044 0.207 0 1
Sub-Saharan Africa 225 0.156 0.363 0 1
Poverty rate at poverty line 225 13.571 18.948 087.72
of $1.25 a day (PPP)
Poverty rate at poverty line 225 25.265 26.490 0.05 95.15
of $2 a day (PPP)
Poverty gap at poverty line 225 5.009 8.447 0 52.76
of $1.25 a day (PPP)
Poverty gap at poverty line 225 10.454 13.558 0.01 67.58
of $2 a day (PPP)
GDP per capita, PPP 223 6 444.17 4 153.85 284.20 21 026.04
(constant 2005 international $)
Gini index 218 42.593 9.277 26.44 67.4
Source: Author’s estimation, based on the World Bank database.
Annex table 3. List of low- and middle-income countries
Afghanistan Egypt Mauritania Syrian Arab Republic
Albania El Salvador Mauritius São Tomé and Principe
Algeria Eritrea Mexico Tajikistan
Dem. Rep. of Korea Malawi St. Lucia
Djibouti Malaysia St. Vincent and Grenadines
Dominica Maldives Sudan
Dominican Republic Mali Suriname
Ecuador Marshall Islands Swaziland
Annex table 4. OLS regression of poverty indexes at poverty line of $1.25 per day (PPP) ExplanatoryDependent variable is poverty rate (%)Dependent variable is poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents required for0.915**0.341 exporting (number)(0.439)(0.232) Documents required for1.229***0.537*** exporting (number)(0.297)(0.153) Time taken to export (days)0.177***0.071*** (0.051)(0.025) Time taken to import (days)0.202***0.089*** (0.050)(0.026) Population density (people/km2)0.018***0.018***0.019***0.016***0.007**0.007**0.007**0.006** (0.006)(0.006)(0.006)(0.005)(0.003)(0.003)(0.003)(0.002) 2005BaseBase omittedomitted 2006-0.613-0.216-0.257-0.4340.6560.8230.7950.728 (2.329)(2.323)(2.235)(2.150)(1.310)(1.300)(1.283)(1.230) 2007-3.196-2.417-2.776-2.161-0.781-0.440-0.613-0.326 (2.639)(2.550)(2.612)(2.554)(1.173)(1.146)(1.172)(1.148) 2008-3.147-2.292-2.301-1.688-0.789-0.413-0.451-0.144 (1.989)(1.906)(1.944)(1.982)(0.954)(0.926)(0.960)(0.982) 2009-6.809***-5.720***-6.408***-5.657***-2.463**-1.978**-2.299**-1.946* (2.155)(2.058)(2.062)(2.090)(1.012)(0.972)(0.992)(1.006) 20100.7581.4531.7242.5221.1011.4371.5001.911 (3.131)(3.069)(3.100)(3.142)(1.848)(1.851)(1.856)(1.865)
East Asia and the PacificBase omitted Europe and Central Asia-15.125***-15.460***-15.403***-15.774***-3.398***-3.614***-3.537***-3.756*** (2.599)(2.545)(2.457)(2.351)(0.698)(0.688)(0.646)(0.621) Latin America and Caribbean-9.595***-9.031***-9.108***-9.466***-0.546-0.305-0.354-0.495 (2.587)(2.518)(2.445)(2.329)(0.742)(0.746)(0.704)(0.660) Middle East and North Africa-16.211***-16.292***-17.276***-17.989***-3.838***-3.876***-4.263***-4.622*** (2.923)(2.767)(2.949)(3.012)(0.795)(0.751)(0.854)(0.914) South Asia1.7121.5741.7551.877-0.823-1.010-0.858-0.883 (3.905)(3.748)(3.840)(3.653)(1.364)(1.242)(1.239)(1.136) Sub-Saharan Africa27.255***25.980***26.520***24.617***14.306***13.654***13.974***13.047*** (4.848)(4.722)(4.701)(4.738)(2.418)(2.360)(2.336)(2.318) Constant11.888***7.541**13.389***12.215***1.562-0.7331.9981.291 (4.210)(3.695)(3.149)(3.173)(1.955)(1.674)(1.222)(1.294) Observations224224224224224224224224 R-squared0.6860.7030.6970.7060.5640.5830.5740.586 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
An2nex table 4. (continued) ExplanatoryDependent variable is poverty rate (%)Dependent variable is poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4
Annex table 5. OLS regression of poverty indexes at the poverty line of $2 per day (PPP) ExplanatoryDependent variable is poverty rate (%)Dependent variable is poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents required for1.705***0.703** exporting (number)(0.577)(0.315) Documents required for1.783***0.910*** importing (number)(0.422)(0.214) Time taken to export (days)0.314***0.137*** (0.079)(0.038) Time taken to import (days)0.317***0.154*** (0.066)(0.036) Population density (people/km2 )0.017**0.016**0.018***0.014*0.011***0.011***0.011***0.010*** (0.007)(0.007)(0.007)(0.007)(0.004)(0.004)(0.004)(0.004) 2005BaseBase omittedomitted 2006-3.700-3.089-3.062-3.393-0.428-0.133-0.154-0.291 (2.941)(2.970)(2.746)(2.719)(1.710)(1.706)(1.634)(1.572) 2007-6.810*-5.683-6.066*-5.191-2.416-1.839-2.091-1.628 (3.627)(3.511)(3.521)(3.470)(1.855)(1.791)(1.832)(1.788) 2008-6.531**-5.311*-5.036*-4.258-2.346-1.715-1.692-1.235 (2.814)(2.723)(2.704)(2.779)(1.434)(1.373)(1.403)(1.432) 2009-11.045***-9.526***-10.347***-9.281***-4.955***-4.153***-4.644***-4.077*** (3.192)(3.095)(2.968)(3.027)(1.557)(1.488)(1.484)(1.503) 2010-2.697-1.892-1.027-0.0770.3250.8261.0761.666 (3.565)(3.372)(3.414)(3.506)(2.316)(2.270)(2.295)(2.322)
East Asia and the PacificBaseBase omittedomitted Europe and Central Asia-32.786***-32.834***-33.184***-33.489***-11.317***-11.536***-11.539***-11.806*** (4.955)(4.878)(4.703)(4.542)(1.866)(1.829)(1.755)(1.676) Latin America and Caribbean-25.407***-24.557***-24.537***-25.182***-7.055***-6.636***-6.679***-6.957*** (4.925)(4.747)(4.636)(4.461)(1.873)(1.815)(1.759)(1.669) Middle East and North Africa-30.088***-30.189***-31.972***-32.863***-11.444***-11.503***-12.268***-12.798*** (5.835)(5.577)(5.691)(5.772)(2.142)(2.021)(2.139)(2.187) South Asia9.89510.50710.14810.7351.9531.9061.9752.091 (6.532)(6.616)(6.608)(6.515)(2.705)(2.646)(2.667)(2.556) Sub-Saharan Africa23.951***22.706***22.779***20.248***18.860***17.957***18.282***16.858*** (6.430)(6.246)(6.109)(6.142)(3.480)(3.381)(3.348)(3.355) Constant32.571***28.793***35.780***34.892***9.440***6.392**10.564***9.724*** (6.718)(6.051)(5.398)(5.296)(3.054)(2.662)(2.265)(2.260) Observations224224224224224224224224 R-squared0.7210.7340.7370.7420.6750.6930.6890.698 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 5. (continued) ExplanatoryDependent variable is poverty rate (%)Dependent variable is poverty gap index (%) variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4
Annex table 6. OLS regression of GDP per capita, PPP (constant 2005 international $) and Gini index ExplanatoryDependent variable is log of GDP per capitaDependent variable is Gini index variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4 Documents required for-0.106***0.029 exporting (number)(0.023)(0.201) Documents required for-0.084***0.016 importing (number)(0.020)(0.177) Time taken to export (days)-0.019***-0.042* (0.004)(0.023) Time taken to import (days)-0.018***-0.023 (0.003)(0.025) Population density (people/km2)-0.001***-0.001***-0.001***-0.001***-0.002-0.002-0.002-0.002 (0.000)(0.000)(0.000)(0.000)(0.002)(0.002)(0.002)(0.002) 2005Base omitted 20060.024-0.006-0.0150.0050.7340.7410.6760.729 (0.130)(0.134)(0.120)(0.122)(1.215)(1.216)(1.219)(1.223) 20070.0820.0350.037-0.0120.1560.1670.0600.038 (0.145)(0.145)(0.143)(0.142)(1.219)(1.198)(1.194)(1.197) 20080.1150.0610.025-0.012-0.650-0.641-0.864-0.826 (0.120)(0.119)(0.112)(0.116)(1.125)(1.140)(1.135)(1.160) 20090.1800.1180.1370.078-0.088-0.076-0.218-0.240 (0.149)(0.149)(0.133)(0.138)(1.316)(1.311)(1.287)(1.310) 20100.0980.0730.001-0.040-1.372-1.369-1.718-1.651 (0.175)(0.162)(0.149)(0.158)(1.413)(1.423)(1.396)(1.432)
East Asia and the PacificBase omitted Europe and Central Asia0.667***0.644***0.685***0.688***-6.073***-6.070***-5.758***-5.825*** (0.149)(0.149)(0.133)(0.131)(1.266)(1.278)(1.280)(1.302) Latin America and Caribbean0.648***0.611***0.595***0.630***11.173***11.167***11.040***11.187*** (0.138)(0.133)(0.126)(0.122)(1.233)(1.231)(1.202)(1.210) Middle East and North Africa0.1980.2310.3030.325*-3.967**-3.981***-3.741**-3.740** (0.180)(0.170)(0.192)(0.194)(1.549)(1.528)(1.447)(1.500) South Asia-0.075-0.174-0.095-0.133-5.739***-5.717***-5.232***-5.439*** (0.180)(0.200)(0.176)(0.190)(1.804)(1.799)(1.827)(1.828) Sub-Saharan Africa-0.818***-0.794***-0.755***-0.636***3.947*3.945*4.476**4.480** (0.227)(0.223)(0.214)(0.221)(2.033)(2.096)(2.037)(2.152) Constant8.948***8.946***8.732***8.752***40.300***40.379***41.630***41.143*** (0.228)(0.229)(0.193)(0.183)(2.115)(2.082)(1.478)(1.547) Observations222222222222217217217217 R-squared0.5430.5480.5940.5910.6600.6600.6650.662 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 6. (continued) ExplanatoryDependent variable is log of GDP per capitaDependent variable is Gini index variablesModel 1Model 2Model 3Model 4Model 1Model 2Model 3Model 4
Annex table 7. OLS regression of poverty rate at poverty line of $1.25 per day (PPP) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Efficiency of customs clearance process-13.82*** (5.06) Quality of trade and transport-related infrastructure-4.36 (5.40) Ease of arranging competitively priced shipments-2.60 (4.89) Competence and quality of logistics services-8.48** (3.93) Frequency shipments reach consignee within schedule-15.26*** (4.70) Ability to track and trace consignments-12.27*** (3.52) Overall logistic performance index-11.48*** (4.18) Population density (people per km2 of land area)0.026***0.024***0.024***0.023***0.029***0.025***0.026*** (0.004)(0.005)(0.005)(0.004)(0.005)(0.004)(0.005) Year 20103.202.933.053.158.21*5.524.75 (3.77)(3.58)(3.83)(3.70)(4.38)(3.75)(3.87) East Asia and the PacificBase omitted Europe and Central Asia-21.95***-20.83***-19.95***-20.87***-22.18***-21.97***-21.52*** (3.70)(5.63)(6.04)(4.84)(3.87)(4.40)(4.22)
Latin America and Caribbean-16.71***-16.44***-16.10**-15.94***-16.24***-16.82***-16.43*** (3.82)(5.61)(6.04)(4.99)(3.88)(4.53)(4.38) Middle East and North Africa-24.22***-21.36***-20.53***-23.53***-23.78***-29.24***-23.24*** (4.19)(5.71)(6.42)(4.81)(3.93)(4.82)(4.35) South Asia-11.18-8.78-7.94-8.32-14.37**-11.28*-11.01 (6.66)(7.28)(7.45)(6.93)(6.96)(6.47)(6.77) Sub-Saharan Africa33.50***34.35***35.46***33.03***30.42***30.78***32.57*** (10.69)(11.40)(11.90)(10.87)(9.58)(10.50)(10.89) Constant53.34***30.53**26.30*41.29***66.64***52.64***49.79*** (12.05)(13.72)(13.12)(10.16)(14.27)(9.14)(10.61) Observations54545454545454 R-squared0.7980.7680.7660.7820.8270.8050.790 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 7. (continued) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7
Annex table 8. OLS regression of poverty gap at poverty line of $1.25 per day (PPP) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Efficiency of customs clearance process-4.58** (2.19) Quality of trade and transport-related infrastructure0.42 (2.28) Ease of arranging competitively priced shipments1.37 (2.31) Competence and quality of logistics services-2.54* (1.43) Frequency shipments reach consignee within schedule-6.42** (2.43) Ability to track and trace consignments-4.55*** (1.56) Overall logistic performance index-3.30** (1.61) Population density (people per km2 of land area)0.008***0.007***0.006**0.007***0.010***0.008***0.008*** (0.002)(0.002)(0.002)(0.002)(0.002)(0.002)(0.002) Year 20101.471.150.861.433.652.361.88 (2.18)(1.99)(2.09)(2.12)(2.57)(2.20)(2.23) East Asia and the PacificBase omitted Europe and Central Asia-4.91***-4.10**-4.11*-4.52***-5.20***-5.00***-4.69*** (1.04)(1.93)(2.10)(1.29)(1.27)(1.29)(1.16)
Annex table 8. (continued) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Latin America and Caribbean-2.68**-2.45-2.48-2.43*-2.54*-2.75*-2.57* (1.20)(1.88)(2.12)(1.44)(1.28)(1.41)(1.32) Middle East and North Africa-5.91***-4.86**-5.20**-5.60***-6.00***-7.89***-5.49*** (1.54)(1.98)(2.25)(1.50)(1.43)(1.90)(1.45) South Asia-3.67*-2.07-1.71-2.67-5.39**-3.87*-3.43* (1.99)(2.20)(2.50)(1.84)(2.39)(1.92)(1.92) Sub-Saharan Africa19.76***20.65***20.69***19.69***18.26***18.66***19.59*** (6.03)(6.13)(6.19)(5.96)(5.45)(5.82)(6.08) Constant15.26***3.070.6410.56***23.86***16.30***12.76*** (5.23)(6.08)(6.40)(3.88)(7.41)(4.22)(4.27) Observations54545454545454 R-squared0.7090.6910.6940.6990.7450.7180.702 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 9. OLS regression of log of GDP per capita, PPP (constant 2005 international $) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Efficiency of customs clearance process1.20*** (0.22) Quality of trade and transport-related infrastructure1.19*** (0.20) Ease of arranging competitively priced shipments0.82*** (0.21) Competence and quality of logistics services1.08*** (0.19) Frequency shipments reach consignee within schedule1.03*** (0.21) Ability to track and trace consignments0.98*** (0.20) Overall logistic performance index1.26*** (0.19) Population density (people per km2 of land area)-0.001**-0.001***-0.001***-0.001*-0.001***-0.001**-0.001*** (0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000) Year 2010-0.03-0.10-0.16-0.05-0.35*-0.21-0.21 (0.14)(0.14)(0.19)(0.14)(0.19)(0.17)(0.15) East Asia and PacificBase omitted Europe and Central Asia0.91***1.01***0.78***0.86***0.89***0.90***0.92*** (0.20)(0.20)(0.28)(0.24)(0.23)(0.27)(0.21)
Annex table 9. (continued) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Latin America and Caribbean0.93***0.97***0.88***0.86***0.89***0.94***0.92*** (0.15)(0.17)(0.23)(0.20)(0.20)(0.22)(0.17) Middle East and North Africa0.60***0.39**0.140.63***0.51**0.97***0.56*** (0.16)(0.15)(0.25)(0.19)(0.19)(0.23)(0.16) South Asia0.280.420.230.090.410.260.36 (0.36)(0.33)(0.36)(0.36)(0.33)(0.32)(0.33) Sub-Saharan Africa-0.90***-0.71**-1.00***-0.75**-0.74**-0.70**-0.75** (0.29)(0.30)(0.37)(0.29)(0.31)(0.30)(0.29) Constant5.17***5.14***6.02***5.36***4.94***5.48***4.80*** (0.55)(0.50)(0.53)(0.49)(0.64)(0.51)(0.50) Observations54545454545454 R-squared0.7250.7400.6800.7430.7420.7270.754 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 10. OLS regression of log of export volume Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Efficiency of customs clearance process1.95** (0.80) Quality of trade and transport-related infrastructure2.93*** (0.53) Ease of arranging competitively priced shipments1.64** (0.64) Competence and quality of logistics services2.30*** (0.50) Frequency shipments reach consignee within schedule1.70*** (0.51) Ability to track and trace consignments1.64*** (0.56) Overall logistic performance index2.52*** (0.63) Population density (people per km2 of land area)-0.002-0.001-0.002*-0.001-0.002*-0.001-0.001 (0.001)(0.001)(0.001)(0.001)(0.001)(0.001)(0.001) Year 20100.00-0.12-0.220.01-0.58-0.30-0.37 (0.36)(0.32)(0.41)(0.33)(0.42)(0.45)(0.39) East Asia and the PacificBase omitted Europe and Central Asia-1.44*-0.79-1.59*-1.27-1.53*-1.44*-1.16 (0.76)(0.72)(0.82)(0.81)(0.86)(0.84)(0.80)
Annex table 10. (continued) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Latin America and Caribbean-1.16-0.80-1.22*-1.13-1.27-1.14-0.98 (0.73)(0.68)(0.72)(0.74)(0.76)(0.77)(0.74) Middle East and North Africa-0.93-1.01-1.73**-0.57-1.10-0.27-0.71 (0.92)(1.00)(0.83)(1.09)(0.93)(1.10)(0.99) South Asia0.670.730.73-0.080.950.670.76 (1.40)(0.91)(1.33)(1.08)(1.42)(1.56)(1.21) Sub-Saharan Africa-3.56***-2.29**-3.42***-2.68**-3.35***-3.13***-2.86*** (0.94)(1.01)(1.12)(1.05)(1.01)(0.99)(1.01) Constant19.73***16.90***20.27***18.39***19.30***20.07***17.65*** (2.43)(1.70)(2.10)(1.78)(2.00)(2.04)(2.14) Observations48484848484848 R-squared0.4650.6060.4600.5590.4940.4760.540 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Annex table 11. OLS regression of Gini index Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Efficiency of customs clearance process0.97 (2.96) Quality of trade and transport-related infrastructure3.87* (2.06) Ease of arranging competitively priced shipments4.31** (1.60) Competence and quality of logistics services4.17** (1.80) Frequency shipments reach consignee within schedule2.01 (1.80) Ability to track and trace consignments1.88 (2.03) Overall logistic performance index3.66* (2.06) Population density (people per km2 of land area)-0.004-0.005**-0.006**-0.004*-0.005*-0.004-0.005* (0.003)(0.002)(0.003)(0.002)(0.003)(0.003)(0.003) Year 2010-1.82-2.16-2.76*-2.06-2.50-2.25-2.46 (1.50)(1.46)(1.53)(1.48)(1.66)(1.62)(1.52) East Asia and the PacificBase omitted Europe and Central Asia-7.33*-6.78**-7.74***-7.14**-7.17**-7.34**-7.17** (3.94)(3.16)(2.45)(2.69)(3.55)(3.47)(2.95)
Annex table 11. (continued) Explanatory variablesModel 1Model 2Model 3Model 4Model 5Model 6Model 7 Latin America and Caribbean10.98***10.99***10.39***10.65***10.95***10.87***10.79*** (3.82)(2.98)(2.45)(2.63)(3.43)(3.38)(2.84) Middle East and North Africa-1.60-1.87-3.38-0.86-1.48-0.76-1.41 (4.18)(3.35)(2.75)(3.27)(3.76)(4.13)(3.29) South Asia-2.42-0.81-1.16-1.65-1.62-2.13-1.31 (4.45)(3.59)(2.94)(2.93)(4.08)(3.85)(3.42) Sub-Saharan Africa0.000.87-0.160.980.540.420.61 (4.23)(3.50)(2.88)(3.22)(3.99)(3.97)(3.41) Constant38.75***31.74***30.72***30.70***34.93***36.25***31.77*** (8.94)(6.65)(5.16)(5.84)(7.30)(7.07)(6.84) Observations51515151515151 R-squared0.7760.7910.8020.7980.7810.7800.789 Source:Author’s estimation, based on the World Bank database. Heteroskedasticity-robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
References
Angrist J. and J.S. Pischke (2008). Mostly Harmless Econometrics: An Empiricist’s Companion. Princeton University Press.
Arellano, M. and S. Bond (1991). “Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations”, Review of Economic Studies, vol. 58, No. 2; pp. 277-297. Institute for International Economic Studies, Stockholm University
Djankov, S., C. Freund and C.S. Pham (2010). “Trading on time”, Review of Economics and Statistics vol. 92; pp. 166-173.
Dollar, D., M. Hallward-Driemeier and T. Mengistae (2006). “Investment climate and international integration”, World Development, vol. 34, No. 9; pp. 1498-1516.
Engman, M. (2005). “The economic impact of trade facilitation”, OECD Trade Policy Working Paper No. 21.Organisation for Economic Co-operation and Development, Paris.
Foster, J., J. Greer and E. Thorbecke (1984). “A class of decomposable poverty measures”, Econometrica, vol. 52, No. 3; pp. 761-766.
Heckman, J., R. Lalonde and J. Smith (1999). “The economics and econometrics of active labor market programs”, in A. Ashenfelter and D. Card (eds.), Handbook of Labor Economics 1999, vol. 3. Elsevier Science.
Holtz-Eakin, D., W. Newey and H.S. Rosen (1988). “Estimating vector auto-regressions with panel data”, Econometrica, vol. 56; pp. 1371-1395.
Iwanow, T. and C. Kirkpatrick (2007). “Trade facilitation, regulatory quality and export performance”, Journal of International Development, vol. 19; pp. 735-753.
Layton, B. (2007). “Trade facilitation: A study in the context of the ASEAN Economic Community blueprint”, in H. Soesastro (ed.), Deepening Economic Integration in East Asia: The ASEAN Economic Community and Beyond. ERIA Research Project, Economic Research Institute for ASEAN and East Asia, Jakarta.
United Nations (2002). Trade Facilitation Handbook for the Greater Mekong Subregion (ST/
ESCAP/2224), Trade and Investment Division, Economic and Social Commission for Asia and the Pacific, Bangkok.