International Journal of Economics and Finance February, 2009

Import Inflows of Bangladesh: the Gravity Model Approach

Md. Mahmudul Alam (Corresponding author)

CRM, Marketing Division, Grameenphone Ltd.
47 Shantinagar, Dhaka-1217, Bangladesh
Tel: 880-1711-503-782 E-mail: rony000@gmail.com

Md. Gazi Salah Uddin

School of Business, Presidency University
11/A, Road-92, Gulshan-2, Dhaka 1212, Bangladesh
Tel: 880-1715-201-727 E-mail: salahuddin@presidency.edu.bd

Khan Md. Raziuddin Taufique

School of Business, Presidency University
11/A, Road-92, Gulshan-2, Dhaka 1212, Bangladesh
Tel: 880-1711-348-223 E-mail: raziuddint@presidency.edu.bd

Abstract
Bangladesh suffers from a chronic deficit in her trade balance. The paper is an attempt to explore the imports of
Bangladesh which is one of the most significant factors responsible for unfavorable trade balance of the country. The aim
of the study is to intend some initiatives for an attempt to ultimately reshaping the trade balance of Bangladesh with her
foreign trade partners. The paper examines the existence of the gravity theory for the imports of Bangladesh with its eight
major trading partner countries- India, China, Singapore, Japan, Hong Kong, South Korea, USA and Malaysia. The data
set consists of yearly data from 1985 to 2003 in a panel approach. The paper comes across with the findings that the
gravity theory is consistent with the imports of Bangladesh. That is, the geographical distance of Bangladesh with its
partner countries has significant impacts on its imports. But in near future this may change because of different factors
such as profitability, easy trade procedures, product delivery time etc. that influence the imports decision more than does
the geographical distance. This paper finds mixed relationship between the GDP and imports of Bangladesh. It also shows
that the imports of Bangladesh influence the domestic production very little because Bangladesh mostly imports
consumer goods rather than capital goods. Moreover, the population of Bangladesh has significant impacts on imports
which in turn implies that Bangladesh is not capable of producing adequate consumer goods to meet the increased demand
resulted from high population growth. It also shows that partner countries’ GDP has significant positive impacts and
partner countries’ population has mixed impact on imports of Bangladesh. This paper concludes that it will be an alarming
situation for trade balance of Bangladesh if the imports continue to increase in such a pattern that the rate increases five to
eight times more in respect of population increases and at the same time the ratio of capital goods in proportion of total
imports decreases.
Keywords: Gravity Model, Imports, Gross Domestic Product (GDP), Population, Geographical Distance, Panel
1. Introduction
In the past, national economies were relatively self-contained entities, isolated from each other by barriers to cross-border
trade and investment; by distance, time-zones, and language; and by national differences in government regulations,
culture, and business systems. As these barriers are being minimized by taking different initiatives, the contribution of
cross-border trade to the development of national economy is becoming significant. The economic development of
Bangladesh is also, to a significant extent, affected by the trade flows with partner countries. Although Bangladesh suffers
from a chronic deficit in her trade balance (Table-1), scope might be found to explore the most significant factor(s)

131
Vol. 1, No. 1 International Journal of Economics and Finance

responsible for unfavorable trade balance with a view to take some initiatives for an attempt to ultimately reshaping the
trend of Bangladesh’s trade balance with her foreign trade partners.
Import is one of the moist important factors responsible for trade balance. Due to trade deficit where import is higher than
export, the imports of Bangladesh need to look from different dimensions. While analyzing on imports, the import trading
partners also get focus. In addition, recently the imports of Bangladesh also have started to move from border country to
other countries. Though Bangladesh shares three of her borders with India and none with China, China has emerged as the
largest source of import for Bangladesh replacing India for the first time in 2006 (Table-2). The value of imported amount
of Bangladesh was 13.75b taka from China and 12.41b taka from India in 2005-2006 fiscal, where in 2004-2005 it was
12.33b taka from India and 6.94b taka from China.
The gravity model is a popular formulation describing the trade flows between different geographical entities. The theory
says that there is negative relationship between geographical distance and trade volume which indicates that the lower
distance influences for more trade and vice versa. As the imports of Bangladesh are geographically shifting, this paper
aims at analyzing the gravity theory for Bangladesh. Moreover, this paper also finds out the impacts of home GDP,
partner countries’ GDP, home population, and partner countries’ population on the imports of Bangladesh.
2. Literature Review
The empirical studies by Tinbergen (1962) and Linnemann (1966) showed trade flows follow the physical principles of
gravity: two opposite forces determine the volume of bilateral trade between countries - the level of their economic
activity and income, and the extent of impediments to trade. The latter includes, in particular, transportation costs, trade
policies, uncertainty, cultural differences, geographical characteristics, limited overlap in consumer preference schemes,
regulatory bottlenecks, etc. National borders are among these impediments, even for industrialized countries (Anderson
and Van Wincoop, 2003).
Although trade potential primarily results from the matched export capacities and import demand (microeconomic), on a
more aggregated level of analysis, proximity in demand, in per capita income, in space, and in culture, are the potential
macroeconomic determinants of export. Thus various combinations of macroeconomic variables, such as gross domestic
product (GDP) and population with geographic distance, are significant predictors of trade potentials. Hence, gravity
equations have been used extensively in the empirical literature on international trade (Havrylyshin and Pritchett, 1991;
Frankel and Wei, 1993; Bayoumi and Eichengreen, 1997; Evenett and Hutchinson, 2002).
Filippini (2003) used a gravity Equation model to analyze trade flows between East Asian industrializing countries
(including China) and some developed countries in order to show the remarkable trade performance of East Asian
countries. The study showed that all coefficient signs were consistent with model assumptions and found high propensity
of Asian countries (including China but excluding Japan) to exchange high-tech manufactured products with Japan and
USA. Another interesting result was that among the East Asian economies, China plays a very important role as an
exporter and as an importer too in recent years.
Martinez-Zarzoso (2003) applied the gravity model to annual bilateral exports between 19 countries. His results indicated
that incomes of exporter and importer, as expected, had positive influence on bilateral trade flows. Population had a large
and positive impact on import, indicating that bigger countries import more than do small countries. Regarding transport
infrastructure, he found that exporting countries’ infrastructure fosters trade. Achay. (2006) investigated the determinants
of trade flows between various countries of the world. He applied the gravity model on a sample of 146 countries for
five-year sub-periods between 1970 and 2000. According to this study, GDP, distance and regional integration
agreements are the determinants of trade. It is also found that all estimated coefficients were statistically significant and
their signs were in conformity with expectations. The adjustment quality of the model as measured by determination
coefficient (adjusted R2) was quite high, standing at 71 percent. It was found that GDP, GDP per capita, common frontier,
common official language, common currency or common colonial past have a positive impact on the volume of bilateral
trade. On the other hand, the geographical distance had a negative impact on the volume of trade.
Very few studies on Bangladesh have been conducted so far in this area, which is perhaps due to unavailability of
information, particularly regarding trade inflows and outflows with partner countries. Rahman (2004) applied a
generalized gravity model to analyze Bangladesh trade flows with its trading partners using the panel data estimation
techniques. The study estimated the gravity model of trade (sum of imports and exports). The results showed that
Bangladesh’s trade is positively determined by the size of the economies, per capita GNP (gross national product)
differential of the countries involved and openness of the trading countries. The major determinants of Bangladesh’s
exports were found to be exchange rate, partner countries’ total import demand, and openness of the economy. All these
factors affected Bangladesh’s exports positively whereas transportation cost was found to be a significant factor in
influencing trade negatively.
From the extensive literature, gravity equations share common features that can be customized for different purposes:
First, a gravity equation is bilateral; Secondly, gravity equations can be derived from various theoretical trade models

132
International Journal of Economics and Finance February, 2009

(Deardorff, 1995) and finally, a gravity equation may be used in order to estimate either determinants of the volume or
determinants of the nature of trade flows.
3. Data and Model Specification
Classical gravity models generally use cross-section data to estimate trade effects and trade relationships for a particular
time, for example one year. In reality, however, eight cross-section data observed over several periods (panel data
methodology) result in more useful information than does cross-section data alone. There is couple of advantages of this
method. First, panels can capture the relevant relationships among variables over time. Second, panels can monitor
unobservable trading-partner-pairs’ individual effects. If individual effects are correlated with the regressors, OLS
estimates omitting individual effects will be biased. Therefore, the paper used panel data methodology for testing
empirical gravity.
This panel study covers yearly data for Bangladesh and its eight major partner countries from 1985 to 2003; a balance
panel of total 684 observations. The countries - India, China, Singapore, Japan, Hong Kong, South Korea, United States of
America (USA) and Malaysia- are chosen based on the share of total import volumes of Bangladesh. These eight
countries cover most of the imports of Bangladesh (Table-3). Bangladesh imported 64 percent in 1996, 60 percent in 2001
and 2002 of its total import from these eight countries.
<< Table 3 >>
Data for total import flows of Bangladesh (million US$) have been taken from Economic Review of Bangladesh. Then it
has been converted into real value by adjusting with the US consumer price index (constant year 2000). GDP data are also
in real term (constant 2000 million US$) that have been taken from World Development Indicators (WDI-2007) database.
Population data (in million) were also taken from World Development Indicators (WDI-2007) database. Geographical
distance in kilometers has been gathered from a web based database (www.timeanddate.com). All of these data have been
converted into logarithmic (ln) format. For the statistical output, the SAS has been used and for the panel regression, the
TSCSreg procedure has been used.
In 1687, Newton proposed the “Law of Universal Gravitation.” It held that the attractive force between two objects i and
j is given by,
Fij = G (MiMj / D2ij) (1)
Where notation is defined as follows, Fij is the attractive force, Mi and Mj are the masses, Dij is the distance between the
two objects and G is a gravitational constant depending on the units of measurement for mass and force.
This model originates from the Newtonian physics notion. Newton’s Gravity Law in mechanics states that two bodies
attract each other proportionally to the product of each body’s mass (in kilograms) divided by the square of the distance
between their respective centers of gravity (in meters). The gravity model has been applied to a wide variety of goods and
factors of production moving across regional and national boundaries under different circumstances since the early 1940s
(Oguledo and Macphee, 1994).
The gravity model for trade is analogous to this law. The analogy is as follows: “the trade flow between two countries is
proportional to the product of each country’s ‘economic mass’, generally measured by GDP, each to the power of
quantities to be determined, divided by the distance between the countries’ respective ‘economic centers of gravity’,
generally their capitals, raised to the power of another quantity to be determined” (Christie, 2002). This formulation can
be generalized to,
Mij = K Yi ȕ1 Yj ȕ2 Dij Ȗ popi į1popj į2 (2)
The linear form of the model is as follows:
ln (Mij) = K + ȕ1 ln (Yi) + ȕ2 ln (Yj) + Ȗ ln (Dij) + į1 ln (POPi) + į2 ln (POPj) (3)
The basic statistical model is specified in the following logarithmic form in which five variables are assumed to determine
the import flows to Bangladesh from its major trading partners:
ln (Mij) = Į + ȕ1 ln (Yi) + ȕ2 ln (Yj) + Ȗ ln (Dij) + į1 ln (POPi) + į2 ln (POPj)+ ut (4)
where, Mij is the flow of Bangladesh’s (i) imports from partner country (j), Yi is Bangladesh’s GDP, Yj is the partner
countries’ GDP, Dij is the geographical distance between Bangladesh’s (i) capital city to partner countries’ (j) capital city,
POPi is Bangladesh’s population, POPj is the partner countries’ population, u is the random error term and ȕ, Ȗ, į are the
structural parameters.
Here, this paper also observes the result for the basic model excluding the population variable from Equation no.-4,
because there are few very highly populated countries such as China, India, and USA are available in partner countries’
list that may impact the overall results. In this case the model is:
ln (Mij) = Į + ȕ1 ln (Yi) + ȕ2 ln (Yj) + Ȗ ln (Dij) + ut (5)

133
Vol. 1, No. 1 International Journal of Economics and Finance

The general hypothesis of the model is that the economic size of the importing countries, Mij, are usually measured with
GDP. The estimated coefficients are usually close to the predicted value of one. If GDP of a country increases, the country
is expected to import more from foreign countries. The same holds true for the partner countries that are exporters. So,
both ȕ1 and ȕ2 are expected to be positive.
Distance is usually measured using the “great circle” formula. This formula approximates the shape of the earth as a
sphere and calculates the minimum distance along the surface. As the distance between countries becomes larger, imports
will fall. The distance is a factor, which is used as a proxy to consider the impact of transport costs and other transaction
costs. Hummels, et al. (2007) have argued that shipping costs (freight charges and marine insurance) can go a long way
towards explaining why distance matters. Distance indicates the time elapsed during shipment. For perishable goods, the
probability of surviving intact is a decreasing function of time in transit. In gravity model, distance is a resistance factor
and has a negative impact on volume of bilateral trade. As a result, Ȗ is expected to be negative.
The coefficient of population of the exporters may have negative or positive sign depending on whether the country
exports less when it is big (absorption capacity) or whether a big country exports more compared to a small country
(economies of scale). For similar reasons, the coefficient of importer population may have negative or positive sign
(Martinez-Zarzoso and Nowak-Lehmann, 2003). Therefore, the expected signs of į1 and į1 may be negative or positive.
4. Results of Analysis
As the data has been taken from the countries of different regions and different economy sizes, here pooled OLS is not
used. Moreover, geographical distance is fixed and not time sensitive. When working with constant data over time, time
effect provides no output. Due to time constant, variable included in the equation-4 and equation-5, fixed effect model
does not provide a good output. Moreover, m-value of Hausman test for equation-4 is insignificant, even in 90 percent
confidence level which indicates to use fixed effect model. However, m-value of Hausman test for equation-5 is
significant at 99 percent confidence level which indicates to use random effect model.
<< Table 4 >> << Table 5 >> << Table 6 >><< Table 7 >><< Table 8 >>
<< Table 9 >> << Table 10 >><< Table 11 >><< Table 12 >><< Table 13 >>
Domestic GDP has no significant impact on the imports of Bangladesh both in random effect and fixed effect model for
the qeutions-4. However, the GDP of Bangladesh shows significant positive relationships with imports of Bangladesh
both in random effect and one way fixed effect model for the qeutions-5. The coefficients of determination (R2) of random
effect models for both cases indicate that 80 percent of the total variation in the dependent variable is accounted for by the
independent variables that indicate good fit of the model. However, a qualitative focus can give an in depth idea about the
study. Imports criteria show that it influences the domestic production very little because Bangladesh mostly imports
consumer goods rather than capital goods. One third of total imports of Bangladesh were capital goods in 2000, but in
recent years the rate declines to 9 percent in 2004-05, 11 percent in 2005-06, and 12 percent in 2006-07 (Table-14). As a
developing and trade deficit country it is alarming for Bangladesh that the rate of total imports is increasing but the ratio of
capital goods is decreasing.
<< Table 14>>
Due to time irresponsible fixed value, the fixed effect models do not provide any output for geographical distance variable.
The random effect models show significant negative impacts of geographical distance on the imports of Bangladesh,
which is accepted at the 99 percent confidence level for both equation-4 and equation-5. It is noted that here the sign of the
coefficient is negative, means less geographical distance influence for higher import and vice versa. This result supports
the theory of gravity model. However, recently Bangladesh has started to move its trade from lower distance country to
higher distance country. The model of gravity mainly considers the geographical distance for trade, but the reality is more
complex that might cause the inconsistency of this result in near future. There are lots of factors available that go beyond
the assumption of gravity theory, such as, Bangladeshi importers are interested in China because it is offering goods with
a wide price range, easy trade procedures, many duty free goods, shorter timing of shipment etc. China usually sends its
products to Bangladesh within 25 days on an average by completing all the formalities while import from India takes 35
days. Many Chinese products, such as, food stuff, plastic goods, rubber goods, wooden products, raw hides, pulp, papers,
electrical machinery and equipment and parts thereof, sound recorders and reproducers, television accessories of
electronics articles, boilers, machinery and mechanical appliances, parts thereof, Cotton, yarn/thread and cotton fabrics
etc., are cheap compared to the same quality products of India.
Both random effect and one-way fixed effect models show that domestic population has significant positive impacts on
the imports of Bangladesh (equation-4). At 96 percent confidence level, one-way fixed effect model shows that 1 percent
increases in domestic population will lead to 5.65 percent increases in imports of Bangladesh. And at 99 percent
confidence level random effect models show that 1 percent increases in domestic population will lead to 7.19 percent
increases in imports of Bangladesh. That means Bangladesh is not capable to produce enough in proportion of increasing

134
International Journal of Economics and Finance February, 2009

demand due to population growth. As a result, demand for import increases five to eight times of the rate of its population
increases.
Partner countries’ GDP has significant positive impacts on the imports of Bangladesh both for equation-4 and equation-5.
If Bangladesh imports more, the partner countries will produce more to meet the demand of Bangladesh. Fixed effect
model shows no impact of Partner countries’ population on Bangladeshi imports, where random effect models show
significant negative impact.
5. Conclusion
Recently trade of Bangladesh is decreasing with its border countries and trade volume is increasing with far distance
countries, such as, the case of China and India. Therefore, to find out the impacts of gravity model on the trade patterns of
Bangladesh, its import is tested on home GDP, partner countries’ GDP, geographical distance between home capital city
and partner countries’ capital city, home population, and partner countries’ population.
The paper finds mixed results for the impact of Bangladeshi GDP on its imports. If population is not considered, GDP
shows positive relationships with imports of Bangladesh. In a quantities focus, imports criteria show that it influences the
domestic production very little because Bangladesh mostly imports consumer goods rather than capital goods. Further,
population of home country has highly significant impact on imports of Bangladesh. It in turn implies that Bangladesh is
not capable of producing enough in proportion of increasing demand due to population growth. Partner countries’ GDP
has significant positive impacts on the imports of Bangladesh and partner countries’ population has mixed results on
imports of Bangladesh.
Most importantly it is found that geographical distance has significant impact on imports of Bangladesh which means
transport costs and other transaction costs, such as, the probability of surviving intact of perishable goods etc. still have
significant impacts on its import. But in recent phenomenon, the imports are more influenced by profitability, easy trade
procedures, product delivery time etc., rather than the geographical distance. As a consequence, policy makers need to
conduct further serious studies to find out the relationship, if any, between trading pattern, geographical distance and trade
deficits of Bangladesh. Policy makers also need to carefully consider the future alarming situation for trade balance of
Bangladesh if the import increases continuously in such a pattern that the rate increases five to eight times more in respect
of population increases and at the same time the ratio of capital goods in proportion of total imports decreases.
References
Achay, L. (2006). Assessing Regional Integration in North Africa. National Institute of Statistics and Applied economics,
Rabat, Morocco.
Anderson, J. E. and Wincoop, E.v. (2003). Gravity with Gravitas: A Solution to the Border Puzzle. American Economic
Review, 93(1), 170-192.
Bayoumi, T. and Eichengreen, B. J. (1997). Is regionalism Simply a Diversion? Evidence from the Evolution of the EC
and EFTA. Takatoshi Ito and Anne O Krueger (eds.), Regionalism vs. Multilateral Arrangements. Chicago: The
University of Chicago Press.
Bangladesh Bank (2004-2005, 2005-2006). Import Payments (annual). Department of Public Relations & Publication,
Dhaka, Bangladesh.
Bergstrand, J. H. (1989). The Generalized Gravity Equation, Monopolistic Competition, and the Factor-Proportions
Theory of International Trade. Review of Economics and Statistics, 71, 143-153.
Christie, E. (2002). Potential Trade in Southeast Europe: A Gravity Model Approach. Working Paper No. 21. The Vienna
Institute for International Economic Studies-WIIW.
Deardorff, A. V. (1995). Determinants of Bilateral Trade: Does Gravity Work in a Neoclassical World?. Working Paper
No. W5377. National Bureau of Economic Research (NBER).
Evenett, S. J. and Hutchinson,W. K. (2002). The Gravity Equation in International Economics: Theory and Evidence.
Scottish Journal of Political Economy, 49(5), 489-490.
Filippini, C. and Molini, V. (October 2003). The Determinants of East Asian Trade Flows: A Gravity Equation Approach.
Journal of Asian Economics, 14(5), 695-711(17).
Frankel, J. A. and Wei, S. j. (1993). Trade Blocs and Currency Blocs. Working Paper No. W4335. National Bureau of
Economic Research (NBER).
Havrylyshyn, O. and Pritchett, L. (1991). European Trade Patterns after the Transitions. Policy Research Working Paper
Series 748. World Bank.
Hill, L. W. C. (2004). Global Business Today (pp. 5). New York: McGraw Hill Irwin.

135
Vol. 1, No. 1 International Journal of Economics and Finance

Hummels, D., Volodymyr, L., and Skiba, A. (2007). The Trade Reducing Effects of Market Power in International
Shipping. Working paper no. W12914. National Bureau of Economic Research (NBER).
Linnemann, H. (1966). An Economic Study of International Trade Flows. Amsterdam: North-Holland.
Martínez-Zarzoso, I., and Nowak-Lehmann, F. D. (2003). Augmented Gravity Model: An empirical application to
Mercosur-European Union trade flows. Journal of Applied Economics, 6(2), 291-316.
Oguledo, V. I. and Macphee, C. R. (1994). Gravity Models: A Reformulation and an Application to Discriminatory Trade
Arrangements. Applied Economics, 26(2), 107-120.
Rahman, M. (2004). The Determinants of Bangladesh’s Trade: Evidence from the Generalised Gravity Model. The
Economic Society of Australia’s 33 Conference of Economists, University of Sydney, Australia.
Razzaque, A. and Raihan, S. (2007). WTO and Regional Trade Negotiation Outcomes: Quantitative Assessments of
Potential Implications on Bangladesh. Pathak Shamabesh, 150 –152.
Tinbergen, J. (1962). Shaping the World Economy: Suggestions for an International Economic Policy. New York: The
Twentieth Century Fund.
www.timeanddate.com

Table 1. Trade Balance of Bangladesh (Million US $)

Year Goods Imports Goods Exports Trade Balance
1985 2,286 999 -1,287
1986 2,301 880 -1,421
1987 2,446 1,077 -1,369
1988 2,734 1,291 -1,443
1989 3,300 1,305 -1,995
1990 3,259 1,672 -1,587
1991 3,074 1,689 -1,386
1992 3,354 2,098 -1,256
1993 3,657 2,545 -1,113
1994 4,351 2,934 -1,416
1995 6,057 3,733 -2,324
1996 6,285 4,009 -2,275
1997 6,551 4,840 -1,711
1998 6,716 5,141 -1,574
1999 7,536 5,458 -2,077
2000 8,053 6,399 -1,654
2001 8,133 6,085 -2,049
2002 7,780 6,102 -1,678
2003 9,492 7,050 -2,442
2004 11,157 8,151 -3,006
2005 12,292 9,186 -3,105
Source: World Development Indicators, 2007

136
International Journal of Economics and Finance February, 2009

Table 2. Common Commodity Imports by Bangladesh from China and India

China (TK in Millions) India (TK in Millions)
Group-wise Commodity
2005-2006 2004-2005 2005-2006 2004-2005

Cotton,(all types) cotton yarn/thread and cotton fabrics 2,987 2,072 2,284 1,246

Boilers, Machinery and mechanical appliances, parts thereof 2,094 1,595 994 1,147

Electrical machinery and equipment and parts thereof, sound

recorders and reproducers, television image and sound
recorders and reproducers and parts and accessories of such
articles 1,334 636 323 302

Man-made staple fibers 1,016 884 286 267

Knitted or crocheted fabrics 857 665 91 93

Vehicles other than railway or tramway, rolling stock and parts

and accessories thereof 214 200 704 504

Plastics and articles thereof 164 98 474 444

Cereals 21 20 1,425 2,961

Source: Bangladesh Bank (2004-2005, 2005-2006), Import Payments (annual), Department of Public Relations &
Publication, Dhaka, Bangladesh.

Table 3. Goods Import Market Shares of the Eight Major Trading Partners of Bangladesh
Total (a)
S. Sub-Total Imports as %
Year India China Singapore Japan Hong-Kong Korea USA Malaysia (a) (b) of (b)

1985 67 90 291 301 51 56 286 48 1,190 2,286 52.1%

1986 60 74 209 255 46 64 164 27 899 2,301 39.1%

1987 64 82 156 378 68 64 178 24 1,014 2,446 41.5%

1988 87 91 145 456 112 82 216 36 1,225 2,734 44.8%

1989 104 110 186 445 116 103 325 50 1,439 3,300 43.6%

1990 145 132 323 475 157 126 208 41 1,607 3,259 49.3%

1991 181 133 334 336 184 165 181 32 1,546 3,074 50.3%

1992 231 149 275 286 247 181 230 42 1,641 3,354 48.9%

1993 342 248 211 365 299 258 207 53 1,983 3,657 54.2%

1994 414 223 200 498 331 284 202 57 2,209 4,351 50.8%

1995 689 420 275 587 399 340 274 41 3,025 6,057 49.9%

1996 1,100 707 343 695 390 366 330 69 4,000 6,285 63.6%

1997 922 575 297 647 409 360 302 197 3,709 6,551 56.6%

1998 934 593 321 483 443 381 311 172 3,638 6,716 54.2%

1999 1,235 560 553 494 452 287 301 131 4,013 7,536 53.3%

2000 833 568 701 685 455 319 325 108 3,994 8,053 49.6%

2001 1,184 709 824 846 478 411 248 148 4,848 8,133 59.6%

2002 1,019 878 871 655 441 346 261 145 4,616 7,780 59.3%

2003 1,358 938 1,000 605 433 333 223 169 5,059 9,492 53.3%

Source: Bangladesh Economic Review, Ministry of Finance, 2005

137
Vol. 1, No. 1 International Journal of Economics and Finance

Table 4. Model Efficiency Random Effect (Eq No. 4)

Method m-Value m-Prob. dof R2
One-way Random Effect 5.33 0.15 3 0.82
Two-way Random Effect 4.16 0.13 2 0.81

Table 5. Model Efficiency for Random Effect (Eq No. 5)

Method m-Value m-Prob. dof R2
One-way Random Effect 21.01 * 0.000 2 0.79
Two-way Random Effect 8.11 * 0.004 1 0.79
* Significant at 99% confidence level

Table 6. Model Efficiency for Fixed Effect (Eq No. 4)

Numerator Denominator
Method
F-Value F-Prob. dof dof R2
One-way Fixed Effect 40.1 * 0.000 7 139 0.88
Two-way Fixed Effect 31.96 * 0.000 25 121 0.89
* Significant at 99% confidence level

Table 7. Model Efficiency for Fixed Effect (Eq No. 5)

Numerator Denominator
Method
F-Value F-Prob. dof dof R2
One-way Fixed Effect 41.4 * 0.000 7 141 0.87
Two-way Fixed Effect 32.33 * 0.000 25 123 0.89
* Significant at 99% confidence level

Table 8. Both One-way and Two-way Random Effect Regression Output (Eq No. 4) (Dependent Variable: ln (Mij)
Model Coefficient t-Value p-Value
Constant -7.67 ** -1.85 0.07
ln (Yi) -1.59 -1.31 0.19
ln (Yj) 1.07 * 5.52 0.0001
ln (Dij) -2.03 * -4.13 0.0001
ln (POPi) 7.19 * 2.76 0.007
ln (POPj) -0.41 * -3.12 0.002
* Significant at 99% confidence level, ** Significant at 90% confidence level

Table 9. One-way Fixed Effect Regression Output (Eq No. 4)(Dependent Variable: ln (Mij)
Model Coefficient t-Value p-Value
Constant -28.31 * -12.8 0.0001
ln (Yi) -1.29 -1.08 0.28
ln (Yj) 1.43 * 5.28 0.0001
ln (Dij) 0 - -
ln (POPi) 5.65* 2.12 0.04
ln (POPj) -0.39 -0.48 0.63
* Significant at 95% confidence level

138
International Journal of Economics and Finance February, 2009

Table 10. Two-way Fixed Effect Regression Output (Eq No. 4)(Dependent Variable: ln (Mij)
Model Coefficient t-Value p-Value
Constant -13.68 * -2.90 0.004
ln (Yi) 0 - -
ln (Yj) 1.41 * 5.15 0.0001
ln (Dij) 0 - -
ln (POPi) 0 - -
ln (POPj) -0.54 -0.66 0.51
* Significant at 99% confidence level

Table 11. Both One-way and Two-way Random Effect Regression Output (Eq No. 5) (Dependent Variable: ln (Mij))

Model Coefficient t-Value p-Value

Constant -13.67 * -3.39 0.0009
ln (Yi) 2.06 * 9.27 0.0001
ln (Yj) 0.78 * 5.33 0.0001
ln (Dij) -1.57 * -3.35 0.001
* Significant at 99% confidence level

Table 12. One-way Fixed Effect Regression Output (Eq No. 5) (Dependent Variable: ln (Mij))
Model Coefficient t-Value p-Value
Constant -31.06* -21.07 0.0001
ln (Yi) 1.11 * 3.73 0.003
ln (Yj) 1.56 * 7.00 0.0001
ln (Dij) 0 - -
* Significant at 99% confidence level

Table 13. Two-way Fixed Effect Regression Output (Eq No. 5) (Dependent Variable: ln (Mij))
Model Coefficient t-Value p-Value
Constant -15.39* -3.89 0.0002
ln (Yi) 0 - -
ln (Yj) 1.33* 5.49 0.0001
ln (Dij) 0 - -
* Significant at 99% confidence level

139
Vol. 1, No. 1 International Journal of Economics and Finance

Table 14. Category-Wise Yearly Imports of Bangladesh (Million US $)

Major Commodities 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07

FOOD GRAINS 186 409 431 574 418 581

Milk & cream 59 61 61 86 73 83

Oil seeds 72 64 73 86 90 106

Edible oil 251 364 471 441 473 583

Pulses all sorts 88 145 120 159 164 195

Sugar 23 104 110 220 124 294

Clinker 150 144 139 170 210 240

Crude petroleum 242 267 252 350 604 524

POL 481 620 770 1252 1400 1709

Chemical 335 353 406 510 580 668

Pharmaceutical products 39 44 45 41 50 49

Fertilizer 107 109 150 332 342 357

Dyeing, tanning etc. materials 87 86 109 132 148 161

Plastics and rubber articles thereof 250 281 367 477 523 643

Raw cotton 312 393 583 666 742 858

Yarn 283 270 323 393 501 582

Textile and articles thereof 1063 1106 1295 1571 1728 1892

Staple fiber 39 41 57 75 76 97

Iron, steel and other base metals 413 455 479 680 980 985

Capital goods 2617 2735 2875 1115 1539 1929

Others 803 848 870 2823 2887 3401

Subtotal : 7913 8931 10016 12195 13684 16013

Imports of EPZ 627 727 887 952 1062 1144

Grand Total : 8540 9658 10903 13147 14746 17157

Capital Goods as % of Total Import (Except EPZ) 33% 31% 29% 9% 11% 12%

Source: Bangladesh Bank, Import Payments Data, Available at

<http://www.bangladesh-bank.org/econdata/import/categoryimp.html>

140