Moving People:

The efficiency of transportation infrastructure growth in large-scale economies

by
Roy M. Antoun

Abstract:

Transportation infrastructure - the roads, bridges, and tunnels that we often take for granted - is
an imperative investment that states use to promote internal growth and development. Most of
the literature exploring infrastructure economics has explained the industry’s effects on GDP or
wages; the focus remained on internal, macro-level growth. Transportation infrastructure,
however, has evolved into a globalized industry, heavily dependent on international support for
growth and innovation. For wealthier states, building a road isn’t just building a road anymore.
This study explores how foreign economic involvement affects transportation by looking at
international trade, foreign direct investments, and four types of transportation: road, rail, air, and
shipping. The results show that increases in multinational corporations yield generally negative
effects on transportation infrastructure whereas increases in construction multinationals actually
make infrastructure more efficient.

Keywords: infrastructure, transportation, FDI, construction, trade

Antoun, 1
INTRODUCTION

Transportation infrastructure is a development essential. The roads, bridges, and tunnels that we

often take for granted are key to a state’s internal growth. Until now, most of the literature

exploring infrastructure economics has focused on internal, macro-level growth, such as the

industry’s effects on GDP or wages. Transportation infrastructure, however, has evolved into a

globalized industry; one that is heavily dependent on international support for growth and

innovation. According to Engineering News Record (ENR), for example, four of the top ten

general contractors in the world are Chinese firms, two from France, one from Germany, one

from Spain, and only one American firm (Bechtel, Inc. placing in tenth place in 2014). As an

increasing amount of foreign firms enter domestic markets looking for opportunities in

transportation, infrastructure economics research must take into consideration how foreign

involvement affects transportation infrastructure.

Large construction firms have moved beyond business at home. Skanska (a Swedish

General Contractor) has expanded from its origins in European markets and has taken on

milestone projects in the United States, such as the $1.29 billion Bayonne Bridge project and

construction work for the new Freedom Tower in downtown Manhattan.1 China State

Construction and Engineering Corporation’s (CSCEC) subsidiary CCA Civil branched out to the

United States from Beijing and recently undertook a $112 million project reconstructing the

Staten Island Expressway in New York City and later won more bids on New Jersey’s Pulaski

Skyway.2 Research presented in this paper explores how foreign influences, like these

international contractors, impact the domestic transportation this precise phenomenon.

1
http://www.panynj.gov/bayonnebridge/#faqsBayonneBridgeClearQu02 “The Port Authority
remains committed to the completion of the Bayonne Bridge project. In April 2013, the Board of
Commissioners awarded $1.29 Billion towards completion of the Project.”
2
http://www.silive.com/news/index.ssf/2014/04/14_m_shelled_out_by_state_to_c.html
Antoun, 2
Specifically, this study examines how international trade and foreign direct investments

influence four types of transportation: road, rail, air, and shipping.

The findings presented in this study are significant because they help identify some

aspects of foreign influence that affect transportation utility and quality, something that the

existing literature has failed to extensively cover. Moreover, government agencies can use this

study to gauge how transportation is affected by multinational corporations both inside and

outside of the construction and transportation industries. These results also show the so-called

“winners” and “losers” in the transportation industry when more foreign influences are

introduced to domestic markets. Exploring new trends in transportation, as an industry and

public good, is vital to the growth and sustainability of large economies.

To understand how foreign involvement impacts transportation infrastructure, this paper

proceeds in the following way: First, I discuss the relevant available literature and what this

paper lends to the discussion regarding infrastructure economics. Second, I suggest some

hypothesizes related to the influence of foreign elements on transportation efficiency and

develop a model to test these hypothesizes. I then detail the results of these tests, offer an

analysis, and suggest some policy implications of my findings. I conclude with some suggestions

for further research.

LITERATURE REVIEW

Much of the existing literature surrounding infrastructure economics has focused on the effects

infrastructure has on GDP and wages in local economies; this methodology is largely expected

considering that infrastructure is inherently a domestic financial investment. Calderon and

Serven (2004) provide an interesting perspective regarding the effects infrastructure development

Antoun, 3
has on long-term economic growth and income inequality. Spanning 40 years (1960 - 2000) and

encompassing over 100 countries in their dataset, Calderon and Serven (2004) propose that

infrastructure has positive effects on the economy, trade openness, and trade shocks. These types

of investments help poorer individuals and underdeveloped areas get access to “core economic

activities” which, in turn, gives them access to more productive opportunities (page number).

Calderon and Serven (2004) specifically focus on four major types of infrastructure: (1)

telecommunications (which they find to be the most important of all), (2) power capacity, (3)

road and railway networks, and (4) water and sanitation. They test these categories on two levels,

quantity and quality, and have four main findings. First, the volume of infrastructure in any given

country has a positive effect on long-term economic growth, trade openness, and trade shocks.

There is a positive correlation between infrastructure stocks and trade openness for all categories

of infrastructure across 5-year spans (totaling 40 years). Second, their results suggest that

infrastructure quantity and quality both decrease income inequality. Third, their statistical tests

fail to show any misspecification in their econometrics. Lastly, their results are also statistically

significant, suggesting that their results are valid and their substitutions for any variable lags

have worked. Overall, Calderon and Serven (2004) find a positive correlation (0.21 coefficient)

between growth and infrastructure quality.

Egert, Kozluk, and Sutherland (2009) provide valuable research regarding infrastructure

economics and regulations that help or hinder a firm’s access to markets. They argue that “more

is not always better” and some evidence suggests that increased investments in infrastructure can

be misallocated. Dividing infrastructure stock into four categories (energy, water, transport, and

communications), they look to quantity, quality, and existing policies to determine which policy

is best to provide the most efficient means of developing infrastructure. They acknowledge three

Antoun, 4
limitations to their study. First, the research is narrow and focuses on network industries. Second,

there are data limitations as distinguishing public and private sector funds is difficult. Lastly,

there is also limited information regarding the quality and congestion of infrastructure.

In spite of these limitations, Egert, Kozluk, and Sutherland (2009) identify important

trends in infrastructure across several countries. At the time of their study, telecommunications

— over all other forms of infrastructure — was experiencing an increased amount of investment.

Moreover, they find that the “physical level of infrastructure provision has generally increased,”

and that quality in energy infrastructure has largely facilitated the improvements in

telecommunications infrastructure (p. 7). Their study contributes insights into infrastructure

investments made by both the public and private sectors. They use a regression model to analyze

how firms operate under particular regulatory measures, such as barriers to entry, vertical

integration, regulatory independence, and public ownership. Regulatory independence can

increase infrastructure investment by 11%, whereas changes in barriers to entry increase

infrastructure investments by only 6%. However, they find that an increase in infrastructure

investment does not necessarily translate to increases in the maintenance and operation of

infrastructure as well. Their data shows that policies in which the private sector has increased

authority over the maintenance and operation of infrastructure do better than purely public sector

ownership.

Easterly and Rebelo (1993) explore the relationship between multiple public expenditures

and growth. Despite their more broad focus, their study helps explain infrastructure investment’s

effect on long-term economic growth and trade. Including 100 countries between the years 1970

and 1988 in their sample, they used cross-section regressions to detail the relationships between

several public expenditure variables — agriculture, education, health, housing, transport and

Antoun, 5
communication, industry and mining, and general government. Their research also looks at

infrastructure investments’ effects on the private sector. Easterly and Rebelo (1993) suggest that

these types of investments increase “the social return to private investment,” but do not actually

raise private investment itself. Assuming that investments in infrastructure increase incomes,

their study suggests an inverse relationship between income and international trade taxes.

Meaning, as income rises, international trade taxes fall (making international trade more open).

The most important aspect to take away from Easterly and Rebelo (1993) is their analysis of

transport and communication’s effect on growth. Their study shows that these infrastructure

stocks are robustly significant and the coefficient for transport and communication are more than

double that of general government investment (0.7 to 2.0, respectively). Easterly and Rebelo

explain that public investments in transport and communication yield high returns, which in turn

explain the large discrepancies in coefficients.

Similar to Calderon and Serven (2004), this study looks at road and railway; however,

this paper is focused on purely transportation infrastructure. It adds on to their model by

introducing air and shipping transportation but excludes their non-transportation infrastructure

variables. Moreover, instead of exploring the effects infrastructure has on economic factors, this

paper investigates the opposite effect. I take Calderon and Serven’s (2004) model but reverse it

and specialize the dataset for transportation variables in wealthy countries. The results found in

this paper should add information about this inverse relationship between transportation and

foreign economic influences and provide greater insight on this relationship’s effects in wealthy

states.

Antoun, 6
HYPOTHESES

Hypothesis #1

Imports have some positive impact on transportation efficiency; however, total FDI inflows

should have negative effects on transportation efficiency.

Imports should have positive effects on transportation efficiency because they increase

the utility of roads and ports as incoming goods are increased. FDI inflows are expected to have

negative effects on transportation efficiency because an increase in multinational corporations

has the potential to oversaturate infrastructure, making it inefficient. All too often countries

improve their transportation while playing “catch-up”; funds are allocated to fix or modernize

transportation systems only when it is clear that the systems are almost at the point of utter

dysfunction. Increases in total FDI flows are expected to move transportation systems closer to

that point of deterioration.

Null Hypothesis #1

Imports and FDI inflows have no effect on transportation efficiency.

Hypothesis #2

FDI inflows from specifically construction and transportation industries make transportation

infrastructure more efficient.

Unlike total FDI Inflows, Construction and Transportation FDI inflows should improve

infrastructure efficiency due to the nature of how government contracts are awarded. Introducing

foreign firms should increase domestic competition and lower construction costs as firms attempt

to win low-bid contracts.3

3
All countries in this paper claim to engage in low-bid government contracting.
Antoun, 7
Null Hypothesis #2

FDI inflows from construction and transportation industries have no effect on transportation

efficiency.

VARIABLES

To test the above hypothesizes and effectively operationalize all elements involved with

transportation infrastructure, transportation infrastructure efficiency (TIE) and foreign influence

(FI) have been categorized and measured. TIE was divided into four major modes of

transportation: road, railway, air, and shipping. FI variables were identified as any public or

private economic inflows. Data was collected for 15 countries, all of which were states with

GDP greater than $1 trillion as of 2013. The time series for this study began in 1993 and ended

in 2013 and included the most recent data available from the World Bank Databank and OECD

databases on transportation and infrastructure. Data prior to 1993 was not included for several

reasons. Trade and FDI variables were largely missing or unrecorded prior to the 1990’s.

Moreover, capturing data from earlier years would have skewed the results; expanding the time

series would have measured development and not growth. Several of the countries included in

the dataset are wealthy today but were significantly less wealthy (some still developing) through

the 1960’s, 70’s, and 80’s.

Data was collected for the dependent variable, transportation infrastructure efficiency

(TIE), without consideration for any variables that measured physical or geographical indicators

— such as total length of road or rail networks or road density. Instead, transportation utility data

was collected (how often transportation was used). These data include goods transported on

roads, measured by millions of tons of goods per kilometer. Passengers carried on roads were

accounted for by every million passengers per kilometer. The research also took into

Antoun, 8
consideration the percentage of roads paved as a percentage of total roads in the network. The

same measurements were taken for railway transportation. Total kilometers of rail lines, millions

of tons of goods transported on railways per kilometer, and millions of railway passengers

carried kilometer. Air transportation was measured by passengers carried annually as well as

freight carried in millions of tons per kilometer. Liner shipping was also taken into account by a

connectivity index standardized by the World Bank in which larger numbers meant more port

connectivity. Transportation budgets for each country were included in TIE to measure how

budgets would react to international demand; these budgets were measured in euros as per the

International Road Federation’s standards.

Several independent variables have been identified to represent foreign involvement (FI).

Foreign influences were operationalized by identifying any economic inflows that would affect a

domestic market. Total FDI inflows as a percent of GDP was measured to include the effects all

industries (like banking, insurance, and manufacturing) had on transportation. To be thorough

and to compare these values against total FDI, FDI inflows from specifically construction and

transportation industries were included. International trade influence was measured by capturing

imports only because influence, at its core, requires an external element to impose on domestic

boundaries and not the other way around.

Several other variables have been identified to control for FI. Unemployment was

included to see its effects on transportation utility. Population was also a control variable.

Exports were added as a control to test whether the reverse of import had any significant role. All

variables were captured for the following countries: United States, China, Japan, Germany,

France, United Kingdom, Brazil, Italy, Russia, India, Canada, Australia, Spain, South Korea, and

Mexico all for years 1993 to 2013.

Antoun, 9
DATA AND METHODOLOGY

Data was collected for each country and each year, mostly from the International Road

Federation (a source that the World Bank utilizes for most of its transportation data). Data on

transportation spending for China was absent from the IRF database; instead, this data was

sourced from the National Bureau of Statistics of the People's Republic of China (NSC) online

database.

Regression models were created for each transportation efficiency indicator separately.

The first hypothesis tested for the effects total FDI Inflows (FI), Imports (I), Population (P),

Unemployment (U), and Exports (E) had on each dependent variable indicated by TIE

(Transportation Infrastructure Efficiency) in the time series (t).

Model 1:
△TIEt = FI(x)t + TIEt + I(x)t + P(x)t + U(x)t + E(x)t + e

The second hypothesis tested for the effects FDI Construction and Transportation Inflows

(FCT), Imports (I), Population (P), Unemployment (U), and Exports (E) had on each dependent

variable indicated by TIE (Transportation Infrastructure Efficiency) in the time series (t).

Model 2:
△TIEt = FCT(x)t + TIEt + I(x)t + P(x)t + U(x)t + E(x)t + e

For both hypotheses, the regressions were executed for each dependent variable individually to

maintain the integrity of the linear slope equation. Thus, the results will show the effects of all

independent variables on each transportation efficiency indicator separately to gauge what types

of relationships exist with each type of infrastructure variable.

Antoun, 10
RESULTS

The outputs of each regression included only statistically significant results. Thus, any omission

of independent variables denotes significance levels higher that the 5% tolerance set in the

regressions. All results yielded ANOVA significance levels of 0.000, suggesting that all

independent variables collectively explained the respective variances; however, I also chose to

analyze the independent variables individually.

The null hypothesis for Hypothesis 1 was rejected on the grounds that most of the

independent variables were statistically significant and yielded negative effects on TIE. For

goods transported on roads, exports showed a negative slope (-0.472), suggesting that goods

decrease as exports increase (See Table 2). Unemployment also had negative effects on goods

transported on roads (-0.230). FDI inflows and exports had negative effects on passengers

carried on roads, yielding (-0.122) and (-0.448), respectively; however, imports, population, and

unemployment had positive effects on the amount of passengers traveling on roads (See Table

3). Both FDI inflows (-0.327) and unemployment (-0.230) had negative effects on the percentage

of roads paved (See Table 4). It is important to note, however, that this particular transportation

indicator (percentage of roads paved) eventually plateaus in production, especially for

developed, rich nations. Eventually, all the roads that need to be paved are paved; thus, the

percentage stalls or stagnates at some point for all countries. This indicator was primarily

included because previous research, such as …, had acknowledged it in their data.

Imports and unemployment had negative effects (-0.358 and -0.291, respectively) on the

amount of good transported by rail as well as the amount of passengers carried by rail (-0.401

and -0.102, respectively) (See Table 5). However, exports and population yielded positive (0.

311 and 0.682, respectively) effects on passengers carried by rail. Exports had substantially

Antoun, 11
negative effects on air transportation, yielding negative correlations for air freight and passengers

carried by air (-0.743 and -0.756, respectively) (See Table 7 and Table 8). Imports had largely

positive effects on air and shipping but negative effects on transportation budgets (-0.780) (See

Tables 7 through 10). When the average slope was taken from all transportation indicators, the

correlation was generally negative. R-Square was (0.293).

The null hypothesis for Hypothesis 2 was rejected as well. Overall, results showed

largely positive correlations between FDI inflows from the construction industry and the

dependent variables. Most importantly, FDI construction inflows had positive effects on goods

transported on roads (0.194) and goods transported by rail (0.096) (See Table 11 and Table 14).

Imports had positive effects on the percentage of roads paved (0.956) (See Table 13), air freight

(0.175) (See Table 17), and shipping connectivity (0.688) (See Table 18); however, imports had

negative effects on passengers carried by road and rail and transportation budgets. Exports

showed a negative correlation with goods transported on roads but a positive correlation with

passengers carried by rail. The average R-Squared was taken and showed that 63.6% of the

variance was explained by these results.

ANALYSIS

Results for Hypothesis 1 showed an overall negative relationship between total FDI Inflows (and

other foreign influences) and transportation infrastructure efficiency. These statistically

significant results can be explained by an oversaturation of transit users coupled with increases in

population. Thus, there are inefficiencies in domestic markets when too many foreign elements

(trade included) are introduced. Domestic firms may not be able to compete and increases in

transportation utility can cause transportation methods to break down, causing infrastructure

utility to decrease. However, the null hypotheses explaining relationships specifically between

Antoun, 12
total FDI Inflows and goods transported on roads, goods transported on rail, and transportation

budgets could not be rejected. It is possible that there is no relationship between these variables

or that FDI Inflows can actually improve these particular transportation indicators; an increase in

the number of foreign firms could cause increases in the amount of money coming into domestic

markets, especially if local labor is hired.

The same cannot be said about trade. Imports and exports yielded mixed results that were

statistically significant because of the way transportation is planned by government agencies. For

instance, imports had positive effects on the amount of passengers traveled by road, but exports

had negative effects on this particular dependent variable. The same concept was also found with

air freight (exports had negative impacts on air freight, while imports were positive). An inverse

relationship was found with passengers carried by rail in which imports had negative impacts on

this dependent variable, whereas exports had positive impacts. The results were sporadic because

infrastructure generally plays “catch-up” with demand. For example, roads are improved only

after traffic builds up or asphalt and concrete deteriorates. Rarely are transportation channels

improved or expanded in anticipation of an increase in demand.

Hypothesis 2 results were more consistent because independent variables FDI

Construction Inflows and FDI Transportation Inflows were specifically focused on industries

related to infrastructure. Moreover, the R-Squared results were much higher for Hypothesis 2,

suggesting more consistent data in the variance. Correlations between FDI Construction Inflows

and goods transported on roads and rail were positive, indicating that increases in foreign

construction firms make transportation infrastructure more efficient.

Why is this the case? Government infrastructure contracts are awarded to low-bid

contractors. When, for instance, an agency requires bridge rehabilitation or a new road to be

Antoun, 13
built, contractors engage in a sealed-bid auction in which all bidders simultaneously submit their

price to the agency so that no bidder knows how the other contractors have priced the bid. The

contractor with the lowest bid is usually awarded the job, pending safety audits, bonding

capacity, and proof of insurances. This process naturally increases competition between firms.

However, when foreign construction firms enter the market and further increase competition in

these bids, they inadvertently lower the cost of completing a project. In order to win a project

against a larger pool of bidders, firms must find new ways of cutting costs to have the lowest

price. The data reflects this phenomenon in that road and railway utility increases as more

foreign construction firms are introduced into domestic markets. Although corporations and

managers may lose in profits to bring their prices down, labor wins because labor is mobile,

especially if these firms are forced to hire locals. While the percentage of roads paved may

gradually decrease over time due to spatial limitations, the results still showed that

unemployment had adverse effects on the number of roads paved.

Imports and exports continued to show mixed results, just as they have in Hypothesis 1,

suggesting that infrastructure still lags behind trade demands. Nonetheless, all independent

variables collectively yielded positive correlations with transportation indicators, indicating that

FDI construction inflows play a significant role in positively impacting transportation

infrastructure quality and efficiency. The null hypothesis could not be rejected for FDI

Transportation Inflows. Unlike the construction industry, the transportation industry does not

include firms that actually build the products for government agencies. Rather, their industry

focuses primarily on services such as traffic analysis or consultation. Therefore, FDI

transportation inflows are more likely to affect traffic congestion in local municipalities over the

Antoun, 14
actual build quality of shipping, air, road, or rail (especially since shipping, air, and rail are less

influenced by traffic congestion).

POLICY IMPLICATIONS

Several policy implications can be suggested given the results shown in the analysis.

Governments looking to increase infrastructure utility and economic growth are encouraged to

do three things. First, they should increase construction competition by introducing more foreign

firms, Second, states are discouraged from relying on import substitution to increase commerce.

Third, the findings suggest that public agencies should continue investing in transportation

infrastructure to promote economic growth.

Increases in Competition
The nature of how construction projects are awarded by public sector agencies is a cost-

effective procedure for maximizing infrastructure quality and efficiency. Introducing foreign

construction firms to participate in sealed-bid and low-bid auctions increases competition with

domestic firms. There is a policy danger, however, with attracting increased competition.

Foreign firms have the potential to outperform their domestic competition and drive them out of

business. Nonetheless, if foreign firms are forced to hire locals, labor can still benefit since labor

is mobile. The only economic “losers” would be domestic managers.

Import Substitution
Imports had mixed effects on transportation efficiency. As such, governments are

encouraged to continue avoiding import substitution due to its poor performance on domestic

economic mobility. Although exports showed a negative relationship to goods transported on

roads, it is important to note that exporting goods tends to yield a natural, physical decrease of

Antoun, 15
those goods domestically. Ultimately, domestic firms gain when selling their products abroad,

particularly if they have the infrastructure to do so through air, shipping, and occasionally rail

and road.

Continued Investments
Overall, governments and state agencies are still encouraged to prioritize infrastructure

development. Doing so will increase the mobility and connectivity of goods and services.

Although their metrics differed slightly from the study presented here, Calderon and Servin

(2004) have demonstrated that increases in infrastructure quality have positive effects on the

economy, trade openness, and trade shocks while also giving poorer communities better access

to new markets. The results in this study show that the reverse of this is also somewhat true —

that trade can improve certain types of infrastructure and that foreign firms make infrastructure

more efficient.

CONCLUSION

Research presented in this paper shows us that increasing the amount of foreign construction

firms in domestic markets has positive effects on transportation growth and efficiency.

Moreover, an increase in all types of multinational corporations in domestic markets causes

transportation efficiency to decrease because of infrastructure overuse. It’s suggested that an

increase in these firms coupled with fluctuations in trade causes increased uses of roads, rail, air,

and shipping, which quickly results in the breakdown of these transit mediums. Consider, for

example, a newly paved highway designed to withstand a traffic flow of 100,000 vehicles per

day. Now consider that same highway a few months later with a traffic flow of 150,000 vehicles

per day and ten new foreign firms needing to use this road. These firms require heavy trucks to

travel on this road on a daily basis. Naturally, these trucks will deteriorate the road. The number

Antoun, 16
of potholes increase and people choose to drive on the road less because of its condition. When

these mediums breakdown, their use decreases as the data in this study has shown.

Although this study presents some compelling evidence about foreign elements and their

effects on transportation infrastructure, future research can expand on the methodology. Data

collection was difficult for several reasons. Detailed transportation information is rarely sought

out by international scholars and information that is currently and readily available is limited.

Increasing the dependent variable size could give researchers a better understanding of the exact

effects foreign elements have on, for example, the concrete industry, if that data was available

for every country. The same could be said for the steel manufacturing industry, asphalt, and

lumber. Moreover, future research could include more detailed independent variables regarding

labor mobility, specifically in the construction and transportation industries. Lastly, including

data on low-bid contract corruption can control for construction costs and explain if the

transportation budget is operating efficiently or not. I look forward to seeing if this data can be

found and organized into future research and welcome any improvements that can be made to

this paper.

Antoun, 17
Table 1: Summary Information
Variable Mean Data Source
Goods Transported by
World Bank (International Road Federation)
Roads 306,911.83
Passengers Carried by
World Bank (International Road Federation)
Roads 898,320.02
Percentage of Roads Paved World Bank (International Road Federation)
39.03
Goods Transported by Rail World Bank
362,638.29
Passengers Carried by Rail World Bank
118,524.51
World Bank (International Civil Aviation
Passengers Carried by Air
63,083,438.91 Organization)
World Bank (International Civil Aviation
Air Freight
5,561.62 Organization)
Shipping Connectivity World Bank
61.00
Transportation Spending OECD
10,833,628,694.26
Total FDI Inflows World Bank
2.18
FDI Inflows (Construction) OECD
618.54
FDI Inflows
OECD
(Transportation) 1,590.19
Imports World Bank
23.67
Population World Bank
135,908,217.09
Exports World Bank
24.40
Unemployment World Bank
7.59
Countries: United States, China, Germany, France, United Kingdom, Brazil, Italy, Russia,
Canada, Australia, Spain, South Korea, Mexico
Years: Annual from 1993 to 2013

Antoun, 18
For Tables 2 through 19
Notes: (i) Beta values shown. (ii) In parentheses are significance values.

Table 2
Goods Transported by Roads
1 2 3 4
Total FDI Inflows 0.091
(0.236)
Imports -0.257 -0.196
(0.167) (0.273)
Exports -0.247 -0.281 -0.452 -0.472
(0.158) (0.104) (0.000) (0.000)
Population 0.099 0.097 0.088
(0.181) (0.193) (0.232)
Unemployment -0.222 -0.227 -0.219 -0.23
(0.002) (0.002) (0.002) (-0.001)
R-Squared 0.262 0.255 0.249 0.242
Overall Model Significance 0.000 0.000 0.000 0.000

Table 3
Passengers Carried by Roads
1
Total FDI Inflows -0.122
(0.016)
Imports 0.311
(0.034)
Exports -0.448
(0.002)
Population 0.79
(0.000)
Unemployment 0.154
(0.006)
R-Squared 0.746
Overall Model Significance 0.000

Antoun, 19
Table 4
Percentage of Roads Paved
1 2 3 4
Total FDI Inflows -0.295 -0.285 -0.343 -0.327
(0.001) (0.001) (0.000) (0.000)
Imports 0.476 0.187 0.104
(0.093) (0.052) (0.23)
Exports -0.306
(0.276)
Population 0.108 0.136
(0.225) (0.112)
Unemployment -0.194 -0.184 -0.191 -0.23
(0.026) (0.034) (0.027) (0.004)
R-Squared 0.208 0.2 0.177 0.168
Overall Model Significance 0.000 0.000 0.000 0.000

Table 5
Goods Transported by Rail
1 2 3 4
Total FDI Inflows 0.044 0.048
(0.462) (0.416)
Imports -0.254 -0.338 -0.332 -0.358
(0.17) (0.000) (0.000) (0.000)
Exports -0.089
(0.628)
Population 0.08 0.088 0.084
(0.202) (0.149) (0.168)
Unemployment -0.286 -0.284 -0.285 -0.291
(0.000) (0.000) (0.000) (0.000)
R-Squared 0.171 0.17 0.168 0.162
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 20
Table 6
Passengers Transported by Rail
1 2
Total FDI Inflows 0.074
(0.164)
Imports -0.442 -0.401
(0.003) (0.005)
Exports 0.341 0.311
(0.016) (0.026)
Population 0.689 0.682
(0.000) (0.000)
Unemployment -0.094 -0.102
(0.077) (0.053)
R-Squared 0.501 0.496
Overall Model Significance 0.000 0.000

Table 7
Passengers Carried by Air
1 2
Total FDI Inflows -0.003
(0.958)
Imports 0.39 0.388
(0.035) (0.034)
Exports -0.744 -0.743
(0.000) (0.000)
Population -0.105 -0.104
(0.09) (0.089)
Unemployment -0.2 -0.199
(0.001) (0.001)
R-Squared 0.168 0.168
Overall Model Significance 0.000 0.000

Antoun, 21
Table 8
Air Freight
1
Total FDI Inflows -0.12
(0.042)
Imports 0.442
(0.018)
Exports -0.756
(0.000)
Population -0.153
(0.014)
Unemployment -0.204
(0.001)
R-Squared 0.163
Overall Model Significance 0.000

Table 9
Shipping Connectivity
1 2 3 4
Total FDI Inflows -0.049 -0.046
(0.576) (0.593)
Imports 0.351 0.281 0.276 0.321
(0.174) (0.003) (0.003) (0.000)
Exports -0.075
(0.772)
Population -0.142 -0.134 -0.131
(0.122) (0.124) (0.128)
Unemployment -0.212 -0.213 -0.219 -0.169
(0.025) (0.024) (0.019) (0.058)
R-Squared 0.194 0.193 0.191 0.162
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 22
Table 10
Transportation Spending
1 2 3
Total FDI Inflows 0.021 0.024
(0.735) (0.676)
Imports -0.743 -0.786 -0.78
(0.000) (0.000) (0.000)
Exports -0.045
(0.793)
Population -0.311 -0.308 -0.309
(0.000) (0.000) (0.000)
Unemployment -0.191 -0.19 -0.189
(0.001) (0.001)
R-Squared 0.46 0.46 0.459
Overall Model Significance 0.000 0.000 0.000

Table 11
Goods Transported by Roads
1 2 3 4
FDI Construction Inflows 0.211 0.19 0.182 0.203
(0.047) (0.04) (0.044) (0.013)
FDI Transportation Inflows -0.04
(0.671)
Imports 0.171 0.162 0.115
(0.479) (0.496) (0.599)
Exports -0.474 -0.472 -0.43 -0.319
(0.059) (0.058) (0.065) (0.001)
Population 0.567 0.565 0.561 0.566
(0.000) (0.000) (0.000) (0.000)
Unemployment 0.062 0.048
(0.523) (0.596)
R-Squared 0.702 0.701 0.699 0.698
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 23
Table 12
Passengers Carried by Roads
1 2 3 4 5
FDI Construction Inflows 0.043 0.041 0.038 0.026
(0.529) (0.487) (0.513) (0.615)
FDI Transportation Inflows -0.003
(0.958)
Imports -0.154 -0.155 -0.19 -0.109 -0.109
(0.461) (0.452) (0.298) (0.054) (0.051)
Exports 0.064 0.064 0.09
(0.759) (0.755) (0.64)
Population 0.878 0.878 0.876 0.867 0.87
(0.000) (0.000) (0.000) (0.000) (0.000)
Unemployment 0.024 0.023
(0.719) (0.714)
R-Squared 0.848 0.848 0.847 0.847 0.846
Overall Model Significance 0.000 0.000 0.000

Table 13
Percentage of Roads Paved
1 2 3 4
FDI Construction Inflows -0.108 -0.136 -0.059
(0.429) (0.288) (0.504)
FDI Transportation Inflows 0.093 0.106
(0.471) (0.4)
Imports 0.664 0.929 0.948 0.956
(0.156) (0.000) (0.000) (0.000)
Exports 0.316
(0.541)
Population 0.882 0.832 0.816 0.8
(0.000) (0.000) (0.000) (0.000)
Unemployment -0.35 -0.386 -0.354 -0.357
(0.008) (0.001) (0.001) (0.001)
R-Squared 0.823 0.821 0.816 0.813
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 24
Table 14
Goods Transported by Rail
1 2 3 4 5
FDI Construction Inflows 0.065 0.07 0.067 0.096 0.096
(0.24) (0.164) (0.18) (0.034) (0.035)
FDI Transportation Inflows 0.012
(0.81)
Imports 0.299 0.303 0.243 0.031
(0.112) (0.102) (0.151) (0.54)
Exports -0.281 -0.284 -0.233
(0.14) (0.133) (0.189)
Population 0.903 0.903 0.902 0.924 0.912
(0.000) (0.000) (0.000) (0.000) (0.000)
Unemployment 0.037 0.041
(0.49) (0.416)
R-Squared 0.867 0.867 0.865 0.862 0.861
Overall Model Significance 0.000 0.000 0.000 0.000 0.000

Table 15
Passengers Transported by Rail
1 2 3 4
FDI Construction Inflows -0.119 -0.116 -0.112
(0.281) (0.276) (0.247)
FDI Transportation Inflows 0.013 0.009
(0.897) (0.922)
Imports -1.458 -1.442 -1.444 -1.585
(0.000) (0.000) (0.000) (0.000)
Exports 0.865 0.852 0.855 1.01
(0.015) (0.011) (0.01) (0.001)
Population -0.185 -0.185 -0.185 -0.186
(0.066) (0.064) (0.062) (0.06)
Unemployment -0.013
(0.903)
R-Squared 0.494 0.494 0.494 0.494
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 25
Table 16
Passengers Carried by Air
1 2 3 4 5 6
FDI Construction Inflows -0.02 -0.018
(0.853) (0.852)
FDI Transportation Inflows 0.004
(0.965)
Imports 0.653 0.654 0.634 0.44
(0.066) (0.063) (0.056) (0.137)
Exports -0.709 -0.71 -0.687 -0.522 -0.093
(0.049) (0.047) (0.039) (0.088) (0.359)
Population 0.598 0.598 0.599 0.593 0.616 0.66
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Unemployment 0.123 0.125 0.127
(0.237) (0.207) (0.194)
R-Squared 0.495 0.495 0.495 0.482 0.466 0.46
Overall Model Significance 0.000 0.000 0.000 0.000 0.000 0.000

Table 17
Air Freight
1 2 3 4
FDI Construction Inflows -0.045 -0.054 -0.031
(0.551) (0.45) (0.621)
FDI Transportation Inflows 0.048 0.048
(0.489) (0.494)
Imports 0.098 0.19 0.197 0.201
(0.689) (0.011) (0.007) 0.006
Exports 0.098
(0.694)
Population 0.938 0.931 0.931 0.93
(0.000) (0.000) (0.000) (0.000)
Unemployment 0.105 0.114 0.128 0.136
(0.152) (0.1) (0.052) (0.032)
R-Squared 0.753 0.752 0.75 0.749
Overall Model Significance 0.000 0.000 0.000 0.000

Antoun, 26
Table 18
Shipping Connectivity
1 2 3 4
FDI Construction Inflows 0.041
(0.738)
FDI Transportation Inflows 0.041 0.135 0.129
(0.305) (0.187) (0.195)
Imports 0.472 0.506 0.622 0.64
(0.266) (0.216) (0.000) (0.000)
Exports 0.164 0.121
(0.701) (0.764)
Population 0.44 0.437 0.428 0.436
(0.000) (0.000) (0.000) (0.000)
Unemployment 0.354 0.344 0.361 0.388
(0.006) (0.005) (0.001) (0.000)
R-Squared 0.474 0.473 0.472 0.457
Overall Model Significance 0.000 0.000 0.000 0.000

Table 19
Transportation Spending
1 2 3 4 5
FDI Construction Inflows -0.116 -0.112 -0.137
(0.294) (0.293) (0.161)
FDI Transportation Inflows 0.121 0.122 0.146 0.085
(0.257) (0.247) (0.137) (0.336)
Imports -0.577 -0.629 -0.656 -0.649 -0.646
(0.1) (0.000) (0.000) (0.000) (0.000)
Exports -0.051
(0.876)
Population 0.311 0.31 0.306 0.304 0.299
(0.001) (0.001) (0.001) (0.001) (0.001)
Unemployment 0.069 0.063
(0.541) (0.548)
R-Squared 0.572 0.526 0.532 0.523 0.524
Overall Model Significance 0.000 0.000 0.000 0.000 0.000

Antoun, 27
Bibliography

Boarnet, Marlon G. & Haughwout, Andrew F., “Do Highways Matter? Evidence and Policy
Implications of Highways’ Influence on Metropolitan Development”, University of California
Transportation Center (2000)

Cesar Calderon & Luis Serven, “The Effects of Infrastructure Development on Growth and
Income Distribution”, Central Bank of Chile, Economic Research Division, JEL Classification:
H54, 054 (2004)

César Calderón & Luis Servén, “Trends in Infrastructure in Latin America, 1980-2001”, Central
Bank of Chile, Economic Research Division, JEL Classification: H54, O54 (2004)

Cesar Calderon & Luis Serven, “Infrastructure and Economic Development in Sub-Saharan
Africa”, The World Bank Development Research Group Macroeconomics and Growth Team,
Policy Research Working Paper 4712 (September 2008)

Don Pickrell, “Transportation and Land Use” in Essays in Transportation Economics and Policy.
The Brookings Institution: Washington, DC (1999)

Dulani Seneviratne & Yan Sun, “Infrastructure and Income Distribution in ASEAN-5: What are
the Links?”, International Monetary Fund Working Paper (2013)

Égert, B., T. Kozluk and D. Sutherland, “Infrastructure Investment: Links to Growth and the
Role of Public Policies”, OECD Economics Department Working Papers, No. 686, OECD
publishing, © OECD. doi:10.1787/225678178357 (2009)

Marta Bengoa & Blanca Sanchez-Robles, “Foreign direct investment, economic freedom and
growth: new evidence from Latin America”, European Journal of Political Economy Vol. 19
529–545 (2003)

Ronald R. Braeutigam, “Learning About Transportation Costs” in Essays in Transportation

Economics and Policy. The Brookings Institution: Washington, DC (1999)

Sylvie D´emurger, “Infrastructure Development and Economic Growth: An Explanation for

Regional Disparities in China?”, Journal of Comparative Economics 29, 95–117
doi:10.1006/jcec.2000.1693 (2001)

William Easterly & Sergio Rebelo, “Fiscal policy and economic growth: An empirical
investigation”, Journal of Monetary Economics 32 417-458. North-Holland (1993)

Young Hoon Kwak, Ting Yi Chih, & C. William Ibbs. “Towards a Comprehensive
Understanding of Public Private Partnerships for Infrastructure Development”. California
Management Review: Vol. 51, No. 2 (Winter 2009)

Antoun, 28