Hillman InfluenceandInfrastructure WEB v3
Hillman InfluenceandInfrastructure WEB v3
Hillman InfluenceandInfrastructure WEB v3
Influence and
Infrastructure
The Strategic Stakes of Foreign Projects
AUTHOR
Jonathan E. Hillman
Influence and
Infrastructure
The Strategic Stakes of Foreign Projects
AUTHOR
Jonathan E. Hillman
Established in Washington, D.C., over 50 years ago, the Center for Strategic and
International Studies (CSIS) is a bipartisan, nonprofit policy research organization
dedicated to providing strategic insights and policy solutions to help decisionmakers chart
a course toward a better world.
In late 2015, Thomas J. Pritzker was named chairman of the CSIS Board of Trustees. Mr.
Pritzker succeeded former U.S. senator Sam Nunn (D-GA), who chaired the CSIS Board of
Trustees from 1999 to 2015. CSIS is led by John J. Hamre, who has served as president and
chief executive officer since 2000.
Founded in 1962 by David M. Abshire and Admiral Arleigh Burke, CSIS is one of the
world’s preeminent international policy institutions focused on defense and security;
regional study; and transnational challenges ranging from energy and trade to global
development and economic integration. For the past seven years consecutively, CSIS
has been named the world’s number one think tank for international security by the
University of Pennsylvania’s “Go To Think Tank Index.”
The Center’s over 220 full-time staff and large network of affiliated scholars conduct
research and analysis and develop policy initiatives that look to the future and anticipate
change. CSIS is regularly called upon by Congress, the executive branch, the media, and
others to explain the day’s events and offer recommendations to improve U.S. strategy.
CSIS does not take specific policy positions; accordingly, all views expressed herein should
be understood to be solely those of the author(s).
© 2019 by the Center for Strategic and International Studies. All rights reserved.
Acknowledgments
This report is made possible by general support to CSIS. No direct sponsorship contributed
to this report.
Introduction 1
Finance 4
Guarantees and Conditions 5
Controlling Funds 6
Debt 8
Design and Construction 11
Standards 11
Technology Transfer 13
Intelligence 14
Ownership and Operation 17
Intelligence 17
Access Denial and Ownership Risks 18
Network Advantages 19
Conclusion 23
About the Author 25
Endnotes 26
British prime minister Margaret Thatcher explained to a London audience in 1985, “You might
have heard a lot lately about “infrastructure”—the new ‘in’ word. Some of you might even ask
exactly what it is. You and I come by road or rail. But economists travel on infrastructure.”1
That word is making a comeback. Around the world, and especially in Asia, countries are
racing to build new railways, ports, pipelines, fiber-optic cables, and other infrastructure
and to reap the benefits that come with greater connectivity. The Association of Southeast
Asian Nations (ASEAN), China, Japan, South Korea, Turkey, Russia, and others have
unveiled ambitious initiatives for regional connectivity.2 Countries with more connections
to global flows of trade, finance, people, and data grow by up to 40 percent more than less-
connected countries.3 Linking communities and powering businesses, infrastructure is
often described as the backbone of economies.
But the stakes extend beyond economics, as recent developments show. China’s port
investments in the Indian Ocean have set off alarms,4 and its attempt to extend a road
near a disputed border with Bhutan led to a military standoff with India in 2017.5 Citing
security concerns, Australia is bankrolling its own fiber-optic cable to Papua New Guinea
and the Solomon Islands. South Korean president Moon Jae-in has proposed connecting
to North Korea’s railway network as part of his strategy for regional peace and economic
integration.6 Countless other examples, from proposals for new pipelines to changes in port
management, underscore the strategic importance that infrastructure projects can carry.7
Increasingly, foreign policy experts travel on infrastructure, but they lack a reliable
map. The literature on the economic implications of infrastructure is robust, but it is
relatively thin when considering strategic implications. There is research on related issues,
especially foreign aid and finance as elements of economic statecraft.8 No one appreciates
the strategic importance of infrastructure more than military logisticians, for whom
supply lines are a matter of life and death.9 But with a few important exceptions, little
has been written about infrastructure as a foreign policy tool.10 Without a framework for
thinking about the strategic aspects of foreign infrastructure,11 defined broadly to include
projects pursued by actors outside their home markets, it is difficult to formulate policy
for competing in today’s global infrastructure contest.
This report takes a small step toward filling that gap by illustrating how states use
foreign infrastructure to advance strategic objectives. Specifically, it takes an initial look
1
at connectivity infrastructure in three sectors: transportation, energy, and information
and communications technology (ICT).12 The report’s primary perspective is the foreign
state pursuing infrastructure projects—as a financier, builder, owner, or operator—during
peacetime. Host state conditions are noted primarily as they enable or constrain the
foreign state’s objectives. With an eye toward illuminating current issues, this report draws
from examples throughout history. While technology has changed, opening new avenues
for influence and restricting others, there is plenty from the past that rhymes with today’s
projects. Most strikingly, China is updating and exercising tactics used by Western powers
during the nineteenth and twentieth centuries.
Influence is explored through three stages of the infrastructure project cycle, from the
perspective of the foreign state, as Figure 1 summarizes below. Financing is the first
and broadest avenue, providing the opportunity to extract diplomatic concessions,
reward supporters, shape project plans, access resources, and gain operational control.
The second stage, design and construction, provides an avenue for setting standards,
transferring technology, and collecting intelligence. During the final stage, ownership and
operation, which are considered together since the owner selects the operator, can be
leveraged for deeper intelligence collection and to restrict or deny a competitor’s access.
States can accrue more influence as their firms own and operate a broader network of
infrastructure assets, becoming more resilient to disruptions and monopolizing critical
skills and technologies. Network advantages can become even more valuable during
conflicts, disaster response, and other contingencies. In practice, avenues for influence
often overlap, with financing tied to construction and operation. This report notes those
connections but considers each phase separately to highlight different mechanisms for
exercising influence.
A few caveats are needed. First, most projects are not strategically important. For every
Hambantota Port, a case that is examined in this report, there are scores of rural roads that
serve their intended economic purposes. Understandably, they generate fewer headlines.
Second, there are examples of infrastructure projects pursued primarily for their economic
benefits that also confer strategic benefits. The political benefits of successful ownership
and operation are examined briefly in this report but deserve further attention, especially
the conditions that help translate commercial success into political success. Third,
By focusing on exceptional cases, this report aims to provoke a deeper discussion of the
strategic implications of infrastructure, particularly among U.S. national security officials
and scholars of international relations. In that spirit, a final section concludes with
recommendations for further research. Among other related issues, there is a pressing
need for work that helps conceptualize how outside influence could be minimized or
countered in developing economies, where infrastructure demands greatly exceed the
availability of financing options. With developing Asia alone requiring $26 trillion in
additional infrastructure investment by 2030 to maintain growth momentum and adapt
to climate change,13 these issues, and the strategic implications they carry, are likely to
intensify in the coming years.
Jonathan E. Hillman | 3
Finance
Infrastructure financing is the broadest avenue for exercising influence, which this section
explores as it relates to three stages of a loan.14 The first and most important phase is the
negotiation of terms, through which a lender can extract political concessions, shape
project specifics, and set repayment terms. The second phase is the disbursement of
funds, which allows a lender to reward supporters or withhold funding. The third phase is
repayment, which can be leveraged if the recipient is overly indebted and unable to fulfill
the initially agreed upon terms.
Financiers exercise influence well before any funds are transferred. To attract foreign
investment, recipient countries often make concessions before formal project negotiations
are started. In recent years, for example, the prospect of Chinese infrastructure loans has
helped persuade the Philippines and Cambodia to reevaluate military or diplomatic ties
with the United States.15 Following actual investments, and the promise of additional funds,
Greece, Hungary, and Poland have softened European Union’s statements on China’s human
rights record and territorial claims.16 In practice, it is difficult to disentangle concessions
made to attract investment in general versus those made to attract investment for specific
projects. Prospective borrowers are always competing for the attention of foreign investors.
Several conditions impact a financer’s influence during project negotiations. The greater
difficulty a recipient country faces in attracting investment, the greater influence a
prospective financier wields.17 The size of the proposed investment matters as well. State
lenders usually accumulate influence through a portfolio of projects rather than one-
off investments. There are exceptions, however, particularly among smaller economies
pursuing high-speed railways, large pipelines, and other big-ticket projects. Laos, for
example, is building a high-speed railway with Chinese financing that is roughly half of
its GDP.18 The speed of the investment and the recipient’s existing debt levels are two
additional conditions that are considered later in this section.
Bilateral financing models favor lenders. To be sure, multilateral lenders still wield
influence over recipient countries, and they are not immune from their leading members
During the nineteenth and early twentieth centuries, European powers frequently pressed
for unrealistic lending guarantees. In India, Britain guaranteed 5 percent profits for its
railway companies, with the Indian Treasury paying any deficit.23 In Turkey before World
War I, Germany and France secured hefty guarantees from the Ottoman government for
each kilometer of railway, extensive mining and construction concessions, and the right
to import their own construction supplies and technical experts.24 Lenders argued these
concessions were necessary, and when the parties sat down at the negotiating table, it was
the lenders that had highly-trained accountants, lawyers, and other specialists on their
side to help justify the guarantees.
The net effect on the host country could be disastrous, as Turkey’s experience highlights.
In the early 1920s, a general manager of the National Bank of Turkey explained all the
groups that profited from the loan and concession system at Turkey’s expense, noting: “the
concessionaries . . . received in payment shares which they promptly converted into cash
. . . the financial groups associated with the concessionaries made money . . . the railway
contractors made money . . . the [debt] holders sporadically received interest.”25 As for
Turkey itself, the manager noted, “the debt charges of an already more or less bankrupt
country were increased.” This was the natural result of a system in which powerful foreign
groups faced little effective oversight.
Modern international financing guidelines are intended to limit these risks for recipients,
but not everyone subscribes to the same rules. The Organisation for Economic Co-
operation and Development (OECD) prohibits its members from engaging in these
types of activities. Specifically, OECD guidelines limit the use of tied aid (which requires
that money being lent is spent in the lending country), regulate credit practices,
impose maximum repayment terms, set minimum interest rates based on country
Jonathan E. Hillman | 5
risk classification, and include social, environmental, and governance standards.26
Unintentionally, these guidelines provide an opening for countries not bound by them.
Limiting repayment terms, for example, can discourage lenders from supporting riskier
projects, which naturally require a higher rate of return. Lenders not bound by those
guidelines have fewer competitors in riskier markets. The preferred solution is not to
lower or remove these guidelines but to extend them to all major lenders, but that has
proved difficult in practice.
Lenders can also shape project specifications in ways that advance their own interests.
For example, if a lender pushes the recipient to expand the scope of a proposed project, it
can increase the overall size of the required loan. Unsolicited proposals, in which a lender
proposes a project rather than the recipient, are especially vulnerable to specification
inflation. A lender might also have a strategic interest in ensuring that a certain standard
is used for the project, as the section on design and construction explains in greater detail.
Standard setting can happen from the top down, as required by a lender, or from the
bottom up, as a consequence of contracting decisions.
Financing negotiations can impact contracting decisions as well. This can occur indirectly,
as project specifications and rules for evaluating bids among contractors are decided.
Lowest-cost estimates, for example, will favor proposals that promise to deliver a project
cheaply, without regard to costs for operating and maintaining the project. Life-cycle
estimates take a longer view and include those costs, favoring proposals that might cost
more initially but require fewer additional expenditures after construction. Decisions to
include or exclude environmental and social impact evaluations, and how those evaluations
are carried out, can also tilt the playing field toward the financier’s preferred contractor.
There are more direct approaches to influence contracting decisions. Through tied aid, a
lender can make using its firms a precondition for the deal. China has been effective in
using both direct and indirect means to favor Chinese construction firms along the Belt and
Road. A CSIS analysis found that Chinese-funded infrastructure projects disproportionately
favored Chinese contractors at the expense of local contractors.27 This can shift the project’s
economic benefits toward the lender, benefitting foreign contractors and leaving the
recipient country dependent upon the foreign lender’s capabilities and expertise. There are
plenty of historical precedents for these tactics. According to one estimate, 40 percent of the
capital raised for India’s colonial-era railways was spent in Britain.28 The strategic stakes of
design and construction decisions are explored later in this report.
Controlling Funds
The second contract phase is the disbursement of funds, which a lender can expedite or
delay. Research suggests that electorates reward incumbent governments for completing
Contracts are intended to limit uncertainty about disbursement, but in practice, many
projects encounter difficulties that can lead to disputes. About 32 percent of joint
construction ventures experience a dispute, according to the consulting firm Arcadis.31
The average dispute takes 14 months to resolve and costs nearly $43 million. Globally,
the most common reason for construction disputes is failure to properly administer
a contract. Next on the list are poorly drafted or incomplete claims, failure to comply
with obligations, and errors or omissions in the contract, respectively. Legitimate or not,
disputes can provide legal cover for a lender to halt the disbursement of funds for non-
economic purposes.
In this murky environment, lenders have three basic options for dealing with
corruption: counter, tolerate, or encourage. A lender might halt funding when evidence
of corruption emerges to promote better governance in the host country, avoid legal
and reputational risk, and minimize commercial losses, among other reasons. When
evidence surfaced of corruption in Bangladesh’s largest development project, the Padma
Bridge, the World Bank, Asian Development Bank (ADB), and other organizations placed
conditions on continued funding, including putting officials on leave, appointing a
special inquiry, and providing access to information from the investigation. When those
conditions were not met, they withdrew funding.34
Alternatively, a lender could allow funding to continue. If the project carries a high
strategic value, for example, or if revealing corruption would jeopardize other objectives,
the lender might express its concerns privately and refrain from taking public action. The
Jonathan E. Hillman | 7
U.S. special inspector general for Afghanistan reconstruction has faulted U.S. officials for
making these tradeoffs early on during the war, writing, “Policymakers tended to believe
that confronting the corruption problem—for instance, by taking a hard stand against
corrupt acts by high-level officials—would impose unaffordable costs on the U.S. ability
to achieve security and political goals.”35 Others have argued that these tradeoffs were
necessary given the primacy of U.S. security goals and the reality of existing practices in
Afghanistan, including patronage networks.36
A lender could actively encourage corruption. For example, a lender looking to exercise
additional political influence through the disbursement of funds could inflate project costs
with the understanding that officials or other participants in the project would benefit.
In extreme cases, the primary motivation for a project is funneling resources to favored
individuals.37 In one recent example, as part of a broader corruption probe, Malaysia’s
government is investigating missing payments for pipeline projects.38 Malaysia’s 2018
election, in which Mahathir Mohamad, the opposition candidate and now prime minister,
made Chinese loans a major issue during the campaign, also highlights the risks of this
approach. Even if these tactics yield short-term political concessions, they can also create
long-term blowback, destabilizing institutions, eroding public trust, and damaging a
lender’s reputation.39
Debt
The third and final contract phase is repayment. Repayment might simply entail a
recipient country paying back its loans on schedule. Lenders can also structure loans
around the delivery of commodities or other natural resources. Sometimes called the
“Angola model” or “resource-financed infrastructure,” this approach can carry a higher
debt burden for host countries, since interest often continues to accrue after project
construction has been completed, at which point resource development activities
begin.40 Only when the pledged resource revenue stream starts does the borrower start to
effectively repay the loan. China has used this model extensively in Africa,41 but it is far
from the first to do so. In ancient Greece, a foreign contractor was hired to drain a marsh
and granted exclusive rights to develop and harvest the land for 10 years.42 Public-private
partnerships are now used for a wide range of projects that generate revenue, from toll
roads to airports, power infrastructure, and telecommunications assets. When projected
revenues fall short, government funds are often required.
Debt is another source of leverage. When borrowers are unable to repay, lenders have
an opportunity to ask for financial concessions, bargain for equity in a project, or
extract non-financial concessions as part of the renegotiations. In 2017, for example,
when Sri Lanka was unable to repay China, the countries agreed to a debt-for-equity
deal that reduced Sir Lanka’s debt by $1.12 billion and gave China a controlling stake
in Hambantota Port and a 99-year lease for operating it.43 When these transactions
are done amidst financial distress, and without full transparency, they understandably
Sri Lanka’s experience illustrates how a lender can exercise influence during the
repayment stage even when it is not a country’s leading lender. When it agreed to a
debt-for-equity deal, China only held roughly 10 percent of Sri Lanka’s foreign debt stock.
In comparison, the World Bank held 11 percent, Japan held 12 percent, and the Asian
Development Bank held 14 percent.45 The largest share of Sri Lanka’s foreign debt is bond
issuances and term loans.46 Despite not being the leading creditor, China’s influence
stemmed from two factors. First, Sri Lanka’s overall debt levels were unsustainable, and
secondly, it was looking for relief. China contributed significantly to that problem because
of the speed at which it lent. In 2008, China only held 2 percent of Sri Lanka’s debt. But
between then and 2017, the beginning of its renegotiations over Hambantota port, it lent
$8 billion,47 raising its overall stake as other lenders were calling for caution.48 The lesson
is that the scale of lending matters but so does speed.
Critically, China was also willing to accept a form of payment that other lenders were
not. For example, the ADB has three major financing guidelines: it seldom takes an
equity stake larger than 25 percent of total share capital, it will seldom be the largest
single investor in an enterprise, and it will not assume responsibility for managing an
enterprise. In the case of Hambantota port, a Chinese state-owned firm took an 80
percent share, becoming the largest investor, and assumed responsibility for managing
the enterprise.49 Equity and management responsibilities are not without risks, of
course, but they provide flexibility for negotiations. A lender’s influence increases when
it can accept a wider range of terms.
The risks that infrastructure lending poses for borrowers are not new, as Britain’s
lending to Egypt highlights. Between 1862 and 1875, Egypt’s external debt increased
23-fold, as it borrowed heavily for infrastructure projects, including the Suez Canal.50
During this period, European banks were issuing loans to the Egyptian government
at effective rates ranging from 7 to 10 percent.51 By 1868, Egypt’s debt had become
unsustainable, but its government continued to borrow, and in 1873, it took out the
largest loan in its history. Two years later, it was forced to sell its 45 percent stake in
the Suez Canal to Britain.
Ultimately, Egypt’s infrastructure binge came at the expense of its sovereignty. When
Egypt went bankrupt in 1876, its creditors moved to take control of its governing
functions. A commission was established to service Egypt’s debt, and its members
were nominated by Austria, Britain, France, and Italy. Representatives from Britain and
France were put in charge of the government’s revenue and expenditures. International
administrators oversaw Egypt’s railway and port commission. As one scholar summarizes,
“The political consequence of all these new regulations was . . . [a] government by an
executive council whose leading members were foreigners.”52
The economist Thomas Schelling once observed that burglars steal what we have, while
loan sharks sell what we want.53 But for global projects, those lines are easily blurred.
When foreign powers offer infrastructure loans, they sell what developing countries want.
Jonathan E. Hillman | 9
When those loans cannot be repaid, foreign powers can “steal” what developing countries
have: taking control of assets and either directly benefitting from their operations or using
unrepayable loans to influence the host government’s actions.
States designing and building infrastructure beyond their borders can advance their
strategic interests in at least three ways: standard setting, technology transfer, and
intelligence collection. Although private firms are typically the most important actors in
this phase, they can independently help set standards and develop expertise that bring
broader benefits to the countries where they are headquartered. State-owned enterprises
can do the same, and their closer relationships with government agencies provide an even
wider window for intelligence activities.
Standards
Standard setting can be tied to financing,
as noted earlier, but it can also occur
during project design and construction.54
Even decisions about who manages a
project can impact what standards are
used and where supplies are sourced. For
example, a major section of the Baghdad
Railway, financed by Germany and built
through Turkey, relied almost entirely
on German suppliers. By 1914, German
suppliers had provided the rolling stock,
approximately 200 steam locomotives
and 3,500 freight or passenger cars.
German dominance of the material used
Source: Jonathan E. Hillman/CSIS
was not only due to its financing of the
project, but also because the construction
was overseen by a manager who graduated from an engineering school in Saxony.55 The
manager was using the supplies he was trained to use.
Standards determine whether the host country’s infrastructure is compatible with the
foreign state’s infrastructure, enabling or restricting access. For example, China and
Russia use different railway gauges, creating a dilemma for Mongolia, which sits between
them and has a mining industry that relies on railway transport. As Mongolia seeks to
strike a balance between its two larger neighbors, debates over which gauge to use have
Jonathan E. Hillman | 11
touched broader security concerns. Mongolian president Khaltmaagiin Battulga, who
previously served as transportation minister, made this a signature issue during the 2016
election, saying, “Tanks can easily penetrate Mongolia in no time if we build a railway
with a [narrower] gauge track, the same used in China.”56 While the threat was naturally
overstated for political effect, and Battulga has softened his criticism of China since taking
office, it highlights how decisions about standards can be political as much as technical.57
For economic and strategic reasons, the European Union is funding a new railway,
Rail Baltica, to connect Poland and three Baltic states. The Baltic states—Estonia,
Latvia, and Lithuania—were once part of the Soviet Union, and despite joining the
European Union in 2004, their railways are compatible with Russia’s network and not
the European Union’s. This requires rail cargo to be swapped at the Lithuanian-Polish
border. A seamless connection is expected to reduce commercial shipment times and
has also been explicitly tied to military mobility.58 The Baltic states are also members
of NATO, which supports construction of the railway to increase its ability to move
troops and personnel.59 Russian state media have tried to cast down on the project’s
economic merits, suggesting that NATO will be the only beneficiary.60 The challenge
that Rail Baltica is addressing underscores the long shadow that infrastructure decisions
cast. Once adopted, standards are often difficult to change, even if a more practical or
advanced technology becomes available.61
Decisions about standards can impact security interests even if they are not
strategically-motivated. Popular legend holds that Tsarist Russia chose a different gauge
to defend against invasion. In truth, it is more likely that the decision was initially
made for efficiency and safety reasons.62 As more track was constructed, the cost of
switching to standard gauge became more prohibitive. That path-dependency became
an asset during World War II, when German engineers occasionally struggled to convert
the Russian railways to the standard gauge that was compatible with German trains.
Similar operational complications arose during the U.S. Civil War, when there were
23 different railway gauges in use across the United States.63 Many of today’s choices
about standards, such as which 5G technology to adopt, have implications well beyond
the immediate technology at hand, impacting a wider array of related technologies and
shaping what might be thought of as a broader “ecosystem.” As a result, switching costs
becoming higher and higher.
Of course, the importance of standards for enabling or restricting access are not limited to
railways. Standards for electricity grids can enable or complicate regional power trading
agreements.64 Oil and natural gas vary widely in quality and composition, creating the need
for common standards to promote interchangeability. The United States, China, European
Union, and Russia all have different global navigation systems and interests in expanding
these systems to other countries. China has made expanding its BeiDou navigation system a
Technology Transfer
States can exercise influence through transferring technology and related expertise. As one
scholar observes, technology is “not only artifacts but also the body of skills, knowledge,
and practice that make them work.”66 For example, an agreement to build a nuclear power
plant typically comes with a package of long-term contracts to operate, maintain, and even
refuel the plant. These agreements can also include regulatory consulting, which allows
a foreign state to help shape the laws governing strategic sectors, as Russia has done for
Turkey’s nuclear power sector.67 After the plant is operational, suppliers can threaten to
raise the price of fuel or close the reactor.68 This leverage may decline over time, as the
host country trains more technical specialists.
The impact of new technology depends on the supplier’s willingness to encourage local
development. As the historian Daniel Headrick has argued, there is a difference between
technology transfer and technology diffusion.69 The latter requires education and
training, which foreign states may attempt to monopolize. During 1850-1940, European
powers restricted technical education opportunities and discouraged non-European
entrepreneurs, contributing to underdevelopment in colonial territories.70 For example,
the French-backed Suez Company, which Egypt granted a concession for developing
the Suez Canal in 1854, made little effort to develop a local workforce. As one scholar
writes, “Investments were made in physical, but not human, capital,” such that, “the
concession did not serve the national economy, but, on the contrary, favored European
capital, widening the gap between the economic structures of rich and poor countries.”71
Restricting access protects the foreign state’s technological edge and prolongs the host’s
dependence on foreign support.
Jonathan E. Hillman | 13
Intelligence
Finally, states can use the design and construction phase to collect intelligence. Designers
and contractors have access to detailed information, which can become valuable during
contingencies. For example, Soviet intelligence agents produced highly-detailed maps that
included not only a bridge’s basic location and dimensions, but also its height above water,
the construction material used, its weight limit, and other details.75 Advances in satellite
technology have increased the amount of information that can be gathered remotely, and
other technology has allowed plans to be copied more easily, but specialized information
available to contractors is still valuable. Foreign contractors can also collect intelligence
unrelated to the project they are using for cover. Although not unique to connectivity
infrastructure, this challenge is growing as more sensors and other technology is
embedded into physical infrastructure.
Detailed information about a project might provide insights into an adversary’s intentions
and capabilities. During the Cold War, the United States and the Soviet Union invested
heavily in intelligence activities to determine each other’s nuclear strike capabilities. One
avenue was recruiting technical experts and construction workers involved in nuclear
infrastructure development. To limit the risk of foreign intelligence, the Soviet Union
relied heavily on prison labor, preferring workers with at least five-year sentences and
sending them to remote regions when they finished.76 Given that most infrastructure is
dual-use, it can be difficult to determine a project’s primary purpose and whether non-
commercial enhancements are being made. In some cases, technical information might
reveal design choices that favor military or commercial activities.
Another method for intelligence collection during the design and construction phase is
planting surveillance technology. After spending $23 million to build an embassy in the
Soviet Union during the 1970s and 1980s, the United States spent more than twice that
trying to neutralize listening devices, including some that were so deeply planted that
they could not be removed from the building’s structure.79 A U.S. Senate committee in
1987 called it “the most massive and skillfully executed bugging operation in history.”
Ultimately, the building was scrapped and a new one was built that finally opened in
2000. It is now more common for states to insist on importing their own construction
workers and materials when building embassies. Although embassies are not connectivity
infrastructure, these basic challenges apply to other types of large projects near valuable
sources of information.
use, meaning they can perform the same function for commercial or military purposes. They
can be multifunction, providing a platform for primary and ulterior functions.
In the coming years, the Internet of Things and related developments that incorporate
more sophisticated sensing technology into highways, electric grids, and other
Jonathan E. Hillman | 15
connectivity infrastructure could boost productivity but also create new security
challenges. These technologies can be leveraged to increase the security of infrastructure
projects during their design and construction, but new systems also create new risks. As
one computer security expert said in 2000, “The future of digital systems is complexity,
and complexity is the worst enemy of security.”84 Of course, many projects do not offer
surveillance rewards commensurate with the investment and risk required for intelligence
collection. But functionally, “smart” infrastructure is surveillance infrastructure. Whether
it is an asset or threat hinges on what is being collected and who is watching.
States can leverage ownership and operation of individual projects as well as larger sets of
related projects. Individual projects can aid intelligence activities and limit a competitor’s
access to strategically important territory. States accrue more influence as they own and
control a broader network of infrastructure assets. These “network advantages” can include
reputational benefits, resilience to disruptions, and dominance of strategically-relevant skills
and knowledge. Owning or operating a network also carry risks, including vulnerabilities to
personnel and assets in dangerous areas as well as reputational damage if operations are not
perceived to benefit local communities. This section considers each of these in turn.
Intelligence
Intelligence operations can benefit from the access that ownership and operational control
provides. There is a long history of countries using commercial ports for intelligence
operations, whether to gain information or conceal the movement of goods or people. In
the early 1850s, roughly a decade after Canton (modern Guangzhou) was established as
one of five Chinese treaty ports, British officials were operating an intelligence network
from there using local agents.85 As a declassified U.S. government intelligence review
summarizes, “It is common knowledge that intelligence services use seamen of their own
or friendly countries’ merchant fleets to make clandestine port observations in denied
areas.”86 Ownership and operational control is not essential for these methods, but it does
make avoiding detection easier.
A foreign state’s ability to use these tactics hinges on the counter-intelligence and
governance capabilities of the host state. Many states have policies that prohibit foreign
companies from owning or operating critical infrastructure. In less critical cases, they
rely on domestic security personnel and procedures to limit security risks. Of course, port
security challenges are not new. Aware that merchants could gather useful intelligence,
the Carthaginians limited Roman traders’ movements and mandated that sales occur with
a government official present.87 Countries hosting foreign-owned and operated ports still
use their own border security and customs agents.
Signals intelligence is another concern in recent debates about foreign ownership. When
a Chinese firm acquired Australia’s Port Darwin in 2015, for example, some observers
worried it would become a foreign listening post. As one analyst explained, “Even if
communications could not be decoded, this information could be logged so that in a
Jonathan E. Hillman | 17
future conflict a foreign power could register that emissions were from a particular type
of communications system aboard a particular ship. By referring to a computer library, an
enemy would know that ship was in the area and to some extent what it was capable of.”88
The proximity of a port to major shipping lanes, military installations, or other high-value
targets could increase its utility for signals intelligence.
Ownership also brings security challenges. In recent years, for example, Chinese
workers in Pakistan have come under attack while working on projects related to the
China-Pakistan Economic Corridor (CPEC). The foreign state has three basic options,
all of which depend on the host state’s preferences. It can help bolster the security
capacity of the host state by providing equipment, training, and funding, it can fund
private security forces, or it can send its own security forces. Another challenge is
deciding whether to protect the asset during a conflict. If the asset cannot be protected
effectively, or is of relatively little value, a foreign owner looking to cut its losses will
have to overcome political pressure to defend the asset. The host country’s insurance
can become the foreign owner’s liability.
Network Advantages
As a state assumes ownership and operational control of additional projects, it can
gain two types of broad network advantages, one political and one logistical, both of
which come with risks. A foreign state can secure political advantages if it successfully
demonstrates the appeal of its way of conducting business. Theoretically, the benefits
might extend beyond local goodwill and have a broader positive impact on the foreign
state’s reputation. In exceptional cases, it could even convey that an economic model is
preferable to alternative approaches being offered by competing states. In Africa, China’s
special economic zones have been motivated in part by the desire to demonstrate the
advantages of China’s approach to development.95 While attractive in theory, these
positive demonstration effects are difficult to achieve in practice.
The reputational effects of infrastructure projects are not necessarily determined by their
commercial performance. There is no guarantee that commercially successful projects
translate into goodwill. Sometimes commercial success is at odds with a host country’s
political environment. Since China assumed control of Greece’s Piraeus Port, for example,
throughput has increased significantly.96 But reports also suggest the Chinese firm has
cut workers’ wages and fired others, creating local resentment.97 If China were primarily
interested in generating political goodwill from the project, it might be moved to strike
more generous deals with local labor groups. Over the longer-term, unless the foreign
state is willing to provide endless subsidies, commercial success becomes essential for the
project to continue operating.
Projects of limited or uncertain commercial value can still produce political benefits, at least
in the short term. For example, China has been willing to heavily subsidize new China-Europe
railway services, which have captured headlines despite their uncertain commercial future.98
When a train from Yiwu, China, arrived in London in January 2017, The Telegraph called it
“a new chapter in the history of the centuries-old trading route,”99 and The Guardian said it
Jonathan E. Hillman | 19
“heralds the dawn of a new commercial era.”100 In the past two years, historic firsts have been
celebrated for trains arriving in France, Latvia, and Finland, among other countries. Even
countries that have been reluctant to endorse China’s BRI have embraced these new routes.
The second type of network advantage is logistical: contingency readiness. States owning
and operating global infrastructure networks can enjoy greater mobility (in the case of
transportation infrastructure), resilience to disruptions, and dominance of dual-use skills
and knowledge. These advantages are accumulated during peacetime, often through
private enterprises, and become even more valuable during conflict. It is beyond the
scope of this report, which focuses on peacetime activities, to fully examine each of these
dimensions. Instead, they are introduced briefly to illustrate the broader implications of
foreign ownership and operation.
Before World War I, the British government invested heavily in a system of telegraph
cables, establishing itself as the dominant owner of cables globally. It also developed a
smaller system of cables, known as the “All Red” routes, which touched only Britain and its
possessions. As the network grew, the British treasury opposed some of these projects on
economic grounds. But it was largely outmaneuvered by the British army, navy, and other
defense organizations, which “developed a virtual fetish” for the routes, as the historian
Paul Kennedy writes.108 Some of these investments had little or no commercial value,
but during WWI they allowed Britain to maintain continuous communications with its
territories while monitoring and disrupting enemy communications.
Extensive ownership of physical assets can also enable dominance of the skills and
knowledge required to operate them. Indeed, Britain’s advantages in the global
telegraph system stemmed not only from owning and operating physical infrastructure
but also the abilities of its companies and the services they provided. Britain’s largest
telegraph company manufactured two-thirds of the cables used during the nineteenth
century and almost half thereafter. In 1896, there were 30 cable-laying ships in the
world, and 24 were British-owned.109 Monopolizing the expertise to lay cables meant
that Britain’s rivals struggled to repair damaged cables. On August 5, 1914, a day after
declaring war on Germany, Britain cut Germany’s direct telegraph cables to the world
outside Europe. These cables remained disabled for the duration of the war, leaving
Germany with one indirect cable route that ran through London and allowing British
Jonathan E. Hillman | 21
intelligence to intercept its messages.110 As its state-owned firms build a growing share
of the world’s underseas fiber-optic cables, China could enjoy similar advantages in
the future.
This report has highlighted how states can use foreign infrastructure projects to advance
non-economic objectives. Infrastructure projects directly impact traditional foreign policy
issues, including diplomacy, intelligence, and military operations. Some of these impacts
are intuitive, while others require a more detailed understanding of how infrastructure
projects are conceived, financed, built, and operated. As part of their knowledge of the
broader U.S. foreign economic policy toolkit, U.S. foreign policy experts should become
familiar with the opportunities and challenges that infrastructure projects present.
Many of the avenues for influence explored are not new. Without access to highly-
privileged information, including private discussions among governments, judgements
about the motives behind recent foreign infrastructure projects are largely speculative.
But history is filled with examples of states using foreign infrastructure to access territory,
harvest resources, shape government policy, dominate technology, and undercut their
competitors. To be sure, not every project is strategically valuable, even those that
appear to lack commercial viability. But with the world now embarking on the greatest
infrastructure push in history, it would be a mistake to ignore the strategic implications of
today’s projects and their broader networks.
Above all, money matters. Finance is the broadest avenue for influence and sets the terms
not only for repayment but often for what follows in the construction and operation
phases. Without a willingness to finance foreign projects, a state forfeits its seat at the
table for the most important phase of negotiations. In theory, it can participate in later
phases, but in practice, its companies are less likely to participate in those projects, as
contractors or operators. The global need for infrastructure far outstrips the financing
ability of any single country. Countries with few resources, or limited political will, for
foreign infrastructure spending will need to find financial force-multipliers, through
partners and allies, and private-sector investors.
Jonathan E. Hillman | 23
This report also identifies several areas for further research. The war-time contingencies
that were introduced in the final section deserve further attention and would benefit
from collaboration between infrastructure experts and military strategists. Emerging
technologies will open new avenues for influence and restrict others, warranting
further attention from scholars and policymakers. Finally, having focused primarily on
the perspective of the foreign state, an equally (if not more) important perspective to
consider is the host state. Examining each of the same phases—financing, design and
construction, and ownership and operation—from the host’s perspective could shed light
on decisionmaking processes and help inform options for preventing and countering
malign foreign influence.
Jonathan E. Hillman is a senior fellow with the CSIS Simon Chair in Political Economy and
director of the Reconnecting Asia Project. At CSIS, he leads an effort to map and analyze
new roads, railways, ports, and other infrastructure emerging across the supercontinent
of Eurasia. His research focuses on the intersection of economics and foreign policy,
including trade, globalization, economic statecraft, and China’s Belt and Road initiative. He
has written on U.S. foreign policy and national security issues for the Washington Post, the
Wall Street Journal, the Los Angeles Times, the National Interest, and other outlets.
Prior to joining CSIS, he served as a policy adviser at the Office of the U.S. Trade
Representative, where he directed the research and writing process for essays, speeches,
and other materials explaining U.S. trade and investment policy. At USTR, he contributed
to the 2015 U.S. National Security Strategy, the President’s Trade Agenda, and numerous
Congressional testimonies. He has also worked as a researcher at the Belfer Center for
Science and International Affairs, the Council on Foreign Relations, and in Kyrgyzstan
as a Fulbright scholar. He is a graduate of the Harvard Kennedy School, where he was a
Presidential Scholar, and Brown University, where he was elected to Phi Beta Kappa and
received the Garrison Prize for best thesis in international relations.
Jonathan E. Hillman | 25
Endnotes
1. Margaret Thatcher, “Speech to Conservative Woman’s Conference,” (speech, the Barbican, central London, May
22, 1985), Margaret Thatcher Foundation, https://www.margaretthatcher.org/document/106056.
2. Center for Strategic and International Studies, “Competing Visions,” Reconnecting Asia, https://reconnectin-
gasia.csis.org/analysis/competing-visions/.
3. James Manyika et al., Global Flows in a Digital Age (McKinsey & Company, 2014), https://www.mckinsey.com/
business-functions/strategy-and-corporate-finance/our-insights/global-flows-in-a-digital-age.
4. Michael J. Green et al., China’s Maritime Silk Road: Strategic and Economic Implications for the Indo-Pacific Region
(Washington, DC: CSIS, 2018), https://www.csis.org/analysis/chinas-maritime-silk-road.
5. Steven Lee Myers, Ellen Barry, and Max Fisher, “How India and China Have Come to the Brink Over a Remote
Mountain Pass,” New York Times, July 26, 2017, https://www.nytimes.com/2017/07/26/world/asia/dolam-pla-
teau-china-india-bhutan.html.
6. Simon Denyer, “South Korean Leader Wants Rail, Road Links with North as First Step toward Economic Inte-
gration,” Washington Post, August 15, 2018, https://www.washingtonpost.com/world/asia_pacific/south-korea-
leader-promises-audacious-step-to-achieve-peace-treaty-with-north/2018/08/15/79397096-a075-11e8-b562-
1db4209bd992_story.html?utm_term=.5abcdad317b7.
7. Niharika Mandhana, “Trouble in Paradise: China-India Rivalry Plays Out in Maldives,” Wall Street Journal,
March 6, 2018, https://www.wsj.com/articles/trouble-in-paradise-china-india-rivalry-plays-out-in-mal-
dives-1520332203; Griff Witte and Luisa Beck, “How a proposed Russian pipeline to Europe is dividing the West,”
Washington Post, April 25, 2018, https://www.washingtonpost.com/world/europe/a-giant-gas-pipeline-raises-the-
spectre-of-russian-influence-in-europe/2018/04/24/415a63f0-4199-11e8-b2dc-b0a403e4720a_story.html?utm_
term=.01590e555958.
8. See, for example: David Baldwin, Economic Statecraft (Princeton University Press, 1985); Daniel W. Drezner,
“Bad Debts: Assessing China’s Financial Influence in Great Power Politics,” International Security 34, no. 2 (2009):
7-45, https://www.belfercenter.org/sites/default/files/legacy/files/IS3402_pp007-045_Drezner.pdf; Robert D.
Blackwill and Jennifer M. Harris, War by Other Means (Harvard University Press, 2016).
9. See, for example: M. Erbel and C. Kinsey, “Think Again – Supplying War: Reappraising Military Logistics and
its Centrality to Strategy and War,” Journal of Strategic Studies 41, no. 4 (2018): 519-544; Martin van Creveld, Sup-
plying War: Logistics from Wallenstein to Patton (Cambridge University Press, 2004); Thomas Kane, Military Logistics
and Strategic Performance (Routledge, 2012).
10. Historians have done some of the best work. See, especially: Daniel R. Headrick, The Tools of Empire: Technolo-
gy and European Imperialism in the Nineteenth Century (Oxford University Press, 1981); Daniel R. Headrick, (Oxford
University Press on Demand, 1991).
11. “Foreign infrastructure” is a general term intended to cover projects with involvement by foreign govern-
ments or non-government actors headquartered outside the country where the project resides.
12. Subsequent research could extend this inquiry to other sectors. For example, the U.S. Department of Home-
land Security lists 16 critical sectors for U.S. domestic infrastructure: Department of Homeland Security, U.S.
Government, “Critical Infrastructure Sectors,” https://www.dhs.gov/critical-infrastructure-sectors.
13. “Asia Infrastructure Needs Exceed $1.7 Trillion Per Year, Double Previous Estimates,” Asian Development
Bank, February 28, 2018, https://www.adb.org/news/asia-infrastructure-needs-exceed-17-trillion-year-dou-
ble-previous-estimates.
14. This is a simplification of the loan cycle, which can be broken into more stages. For example, the World Bank
describes nine stages, from preparation to final maturity, in its official handbook: http://siteresources.worldbank.
org/LOANS/Resources/Disbursement09.pdf.
15. Prak Chan Thul, “U.S. Navy Aid Unit Told to Leave Cambodia,” Reuters, April 4, 2017, https://www.reuters.
com/article/us-cambodia-usa-navy-idUSKBN1760TA.
16. Thornsten Benner et al., “Authoritarian Advance: Responding to China’s Growing Political Influence in
Europe,” Global Public Policy Institute, February 5, 2018, https://www.gppi.net/2018/02/05/authoritarian-ad-
Jonathan E. Hillman | 27
tion=com_content&view=article&id=86&Itemid=307&lang=en#debt-stock.
47. Yogita Limaye, “Sri Lanka: A Country Trapped in Debt,” BBC, May 26, 2017, https://www.bbc.com/news/busi-
ness-40044113.
48. In 2014, for example, the IMF noted: “low tax revenue mobilization remains a concern—particularly given a
relatively high debt level and the ongoing shift from concessional to more expensive loans on commercial terms.”
IMF Press Release No. 14/270, https://www.imf.org/en/News/Articles/2015/09/14/01/49/pr14270.
49. Kiran Stacey, “China Signs 99-Year Lease on Sri Lanka’s Hambantota Port,” Financial Times, December 11,
2017, https://www.ft.com/content/e150ef0c-de37-11e7-a8a4-0a1e63a52f9c.
50. Eric Toussiant, “Debt as an instrument of the colonial conquest of Egypt,” June 2016, http://www.cadtm.org/
spip.php?page=imprimer&id_article=13562.
51. Ali Coşkun Tunçer, “Political Control and Military Conquest: Egypt, 1862-1914,” in Sovereign Debt and Inter-
national Financial Control: The Middle East and the Balkans, 1870-1914 (Basingstoke: Palgrave Macmillan, 2015),
29-52, https://warwick.ac.uk/fac/arts/history/students/modules/hi173/classesandreading/colonialegypt/tunc-
er_-_middle_east_and_the_balkans_1870-1914_chapter_3_-_egypt.pdf.
52. Ibid.
53. Thomas C. Schelling, “What Is the Business of Organized Crime?” The American Scholar 40, no. 4 (1971): 643-
52, http://www.jstor.org/stable/41209902.
54. In addition to offering attractive financing, states can tip the scales in favor of their own standards by subsi-
dizing their companies at home.
55. Marian Kent ed., The Great Powers and the End of the Ottoman Empire (London: Routledge, 2005), 111-112.
56. Jacopo Dettoni, “Mongolia’s Railway to China Turns Into National Security Threat,” bne IntelliNews,
July 30, 2014, http://www.intellinews.com/mongolia-s-railway-to-china-turns-into-national-securi-
ty-threat-500416552/?source=mongolia&archive=bne.
57. Maria Krol Sinclair, “How Sinophobia Blocked a $485 Million Rail Project in Mongolia,” cogitASIA, CSIS, Sep-
tember 6, 2018, https://www.cogitasia.com/how-sinophobia-blocked-a-485-million-rail-project-in-mongolia/.
58. “Lithuania Hopes for ‘military mobility’ Investment Into Rail Baltica, Roads,” The Baltic Times, July 9, 2018,
https://www.baltictimes.com/lithuania_hopes_for___military_mobility___investment_into_rail_baltica__roads/.
59. Olevs Nikers, “New Railroad Agreement a National Security Milestone for Baltic Allies, Poland, EU, and
NATO,” Atlantic Council, October 24, 2016, http://www.atlanticcouncil.org/blogs/natosource/new-railroad-agree-
ment-a-national-security-milestone-for-baltic-allies-poland-eu-and-nato.
60. “Rail Baltica: Who Benefits From $4 Bln Railway Project in the Baltics?” Sputnik, June 23, 2016, https://sput-
niknews.com/europe/201606231041824067-baltic-region-railway-nato/.
61. National Research Council, Realizing the Potential of C4I: Fundamental Challenges (Washington, DC: National
Academy Press, 1999), 78.
62. Jeff Hays, “Railroads in Russia,” Facts and Details, http://factsanddetails.com/russia/Education_Health_Trans-
portation_Energy/sub9_6d/entry-5155.html.
63. William R. Siddall, “Railroad Gauges and Spatial Interaction,” Geographical Review 59, no. 1 (1969): 29-57,
https://www.jstor.org/stable/213081?seq=1#page_scan_tab_contents.
64. United Nations, “Technical Aspects of Grid Interconnection,” in Multi Dimensional Issues in International Elec-
tric Power Grid Interconnections (New York: United Nations, 2006), http://www.un.org/esa/sustdev/publications/
energy/chapter2.pdf.
65. Keshav Kelkar, “China is building a new Silk Road, and this one is digital,” World Economic Forum, August 18,
2018, https://www.weforum.org/agenda/2018/08/china-is-building-a-new-silk-road-and-this-one-s-digital.
66. Daqing Yang, Technology of Empire
Telecommunications and Japanese Expansion in Asia, 1883-1945 (Cambridge: Harvard University Press, 2011), 9.
67. Aaron Stein, Turkey’s Nuclear Program: Challenges and Opportunities (Washington, DC: Atlantic Council, 2016),
http://www.atlanticcouncil.org/images/publications/Turkey_Nuclear_Program_web_1221.pdf.
68. “The world relies on Russia to build its nuclear power plants,” The Economist, August 2, 2018, https://www.
economist.com/europe/2018/08/02/the-world-relies-on-russia-to-build-its-nuclear-power-plants.
69. Daniel R. Headrick, The Tentacles of Progress: Technology Transfer in the Age of Imperialism, 1850-1940 (New
York: Oxford University Press, 1988).
70. Ibid.
71. Caroline Piquet, “The Suez Company’s Concession in Egypt, 1854–1956: Modern Infrastructure and Local
Economic Development,” Enterprise & Society 5, no. 1 (2004): 107-127.
72. Philip H. Gordan, “Don’t Reject the U.S.-India Nuclear Deal,” Brookings Institution, September 24, 2008,
https://www.brookings.edu/opinions/dont-reject-the-u-s-india-nuclear-deal/.
73. Jonathan Hillman and Matthew Goodman, “Asia’s Competing Visions,” CSIS, September 26, 2017, https://
www.csis.org/analysis/asia%E2%80%99s-competing-visions.
74. Japan International Cooperation Agency, “JICA’s Cooperation to Develop State-of-the-Art Training Institute
for Mumbai-Ahmedabad High Speed Rail Project,” September 15, 2017, https://www.jica.go.jp/india/english/
office/topics/press170915_01.html.
75. John Davies, Alexander Kent, and James Risen, The Red Atlas: How the Soviet Union Secretly Mapped the World
(Chicago: University of Chicago Press, 2017).
76. Oleg A. Bukharin, “The Cold War Atomic Intelligence Game, 1945-70: From the Russian Perspective,” Central
Intelligence Agency, April 14, 2017, last modified on June 27, 2018, https://www.cia.gov/library/center-for-the-
Jonathan E. Hillman | 29
nytimes.com/2009/01/07/world/europe/07gazprom.html.
106. “China ‘Blocks’ Mongolia Border after Dalai Lama Visit,” Al Jazeera, December 10, 2016, https://www.alja-
zeera.com/news/2016/12/china-blocks-mongolia-border-dalai-lama-visit-161210060313417.html.
107. Ishaan Tharoor, “The Persian Gulf Crisis over Qatar, Explained,” Washington Post, June 6, 2017, https://www.
washingtonpost.com/news/worldviews/wp/2017/06/06/the-persian-gulf-crisis-over-qatar-explained/?utm_ter-
m=.372c5b3fb776.
108. Paul M. Kennedy, “Imperial Cable Communications and Strategy, 1870-1914,” The English Historical Re-
view 86, no. 341 (1971): 728-752.
109. Daniel R. Headrick and Pascal Griset, “Submarine Telegraph Cables: Business and Politics, 1838–1939,” Busi-
ness History Review 75, no. 3 (2001): 553.
110. Walter S. Rogers, “International Electrical Communications,” Foreign Affairs 1, no. 2 (1922): 156, https://
archive.org/stream/jstor-20028220/20028220#page/n13.