Dollar 2018
Dollar 2018
Dollar 2018
China is a major funder of developing country infrastructure, lending $40 billion annually
through policy banks. Lending does not favor the belt and road above other regions. China’s
lending is indifferent to risk, that is, it is uncorrelated with indices of political stability and rule
of law. Some major borrowers with poor governance are beginning to have debt sustainability
problems, while other borrowers are in good fiscal shape. Chinese banks have been reluctant to
follow global environmental norms but seem to be evolving in that direction. Chinese actions
seem more a revision of the global system than a challenge to it.
Key words: Asian Infrastructure Investment Bank, belt and road initiative, Chinese economy,
debt sustainability, foreign aid
JEL codes: E22, E61, F35
1. Introduction
China in recent years has become a major funder of infrastructure in the developing
world. The big Chinese financing push started shortly after the global financial crisis
(GFC). Starting in 2013 China “branded” the program under the rubric of the Belt and
Road Initiative (BRI). In speeches in Kazakhstan and Indonesia, President Xi Jinping
proposed the Silk Road Economic Belt and the 21st Century Maritime Silk Road. The
BRI involves issues such as policy coordination and harmonization of standards, but at
its heart is infrastructure financing.
In his opening remarks at the Belt and Road Forum in Beijing in May 2017 Presi-
dent Xi noted that:
†Correspondence: David Dollar, John L. Thornton China Center, Brookings Institution, 1775
Massachusetts Ave NW, Washington DC 20036, USA. Email: ddollar@brookings.edu
The initiative has been met with enthusiasm by many developing countries, who
see new opportunities for financing needed infrastructure. Some nearby countries such
as India, on the other hand, have reacted with caution as they question China’s strate-
gic motivations. Developed countries of the West have also been reticent to endorse
the program until they learn more about the details. A concern of Western countries is
that China’s efforts will undermine global norms, especially in the areas of debt sus-
tainability and environmental and social safeguards.
This paper examines China’s role in development finance and in particular
addresses the issue of whether China is challenging global norms. China’s effort is
recent and data problems are legion. Most of the lending is coming from two big state-
owned policy banks, China Development Bank and China EXIM Bank. It would be
straight-forward for them to publish up-to-date data on lending to different countries,
including the terms. In the absence of that kind of transparency, researchers have to
make heroic efforts to compile their best estimates. The next section of the paper looks
at the available estimates and paints a picture of the scale of Chinese financing and
where it goes in terms of countries and sectors.
The third section of the paper takes up the issue of debt sustainability. The coun-
tries receiving significant finance from China vary significantly in the quality of eco-
nomic governance. Some have very good governance, and some, quite poor. Not
surprisingly, some of the countries with poor governance are beginning to have prob-
lems servicing their external debts. This raises some important questions of global gov-
ernance: will the troubled countries go to the International Monetary Fund (IMF), as
has been the practice in the past? Will China change its behavior as recipient country
debt reaches unsustainable levels?
The fourth section of the paper examines another important global issue, environ-
mental and social safeguards in infrastructure projects. Most large projects have
environmental risks and involve resettlement of communities. The multilateral devel-
opment banks (MDBs) have developed through experience a set of rules and proce-
dures for assessing and mitigating risks. Many client countries, however, find these
procedures cumbersome. China’s approach is to follow the laws and regulations of the
country in which it is operating, which is a reasonable position. However, in some
countries with poor governance regulations are mostly honored in the breech. How is
the Chinese approach playing out on the ground? And is there evidence of any shift in
the Chinese approach?
An interesting recent development covered in Section 5 is the establishment of the
Asian Infrastructure Investment Bank (AIIB), a Chinese-led MDB that has quickly
attracted 56 member countries. In 2 years of operation the AIIB has lent about $3 bil-
lion, a very small share of China’s development finance; and three-quarters of the pro-
jects have been co-financed with the other MDBs. AIIB’s lending will grow over time,
however, and has the potential to directly and indirectly address some of the concerns
about Chinese practices.
Section 6 concludes. It is too early to make a definitive judgment on whether
China’s finance is a challenge to the global economic order. There are certainly things
to worry about such as growing indebtedness of some of China’s big clients and envi-
ronmental and social safeguards on the ground. But there are also signs of evolution.
China has reduced its exposure to Venezuela, which is the most problematic of its
major clients. Several of the major borrowers from China have turned to IMF pro-
grams, which is the traditional medicine for unsustainable debt. Chinese companies
and lenders have been given voluntary environmental and social standards by the Min-
istry of Commerce, which is a step in the right direction. Most encouraging, AIIB is
off to a good start and may bring about improvements in the operation of the whole
MDB system.
It is difficult to pinpoint exactly how much finance China is providing to the BRI
because of data problems. There are two main types of financing: outward direct
investment (ODI) and Chinese development finance (CDF), that is, loans to develop-
ing countries primarily for infrastructure and largely coming from the two policy
banks, China Development Bank (CDB) and China EXIM Bank (EXIM). ODI is com-
mercial, and the state and private firms making these investments expect to make a
profit. CDB and EXIM are referred to as policy banks in China. They borrow on
domestic and international capital markets and lend with a spread, so they expect to
be financially self-sufficient. But as the name “policy bank” suggests, they carry out
various policy directives of the government. In recent years they have been tasked with
lending to other developing countries for infrastructure and other development pro-
jects. These activities are analogous to the nonconcessional lending of the World Bank,
which is known as development finance. The motivation for China is partly economic:
the economy has excess savings and under-employed construction companies and
heavy industry. Also, if infrastructure is improved in neighboring countries then China
benefits indirectly as trade expands. There is also strategic motivation as China gains
friends and influence through these projects.
China’s Ministry of Commerce reports data by recipient country on China’s ODI.
However, more than half of the outflow is recorded as going to Hong Kong. It is
unlikely that that is the final destination for all of this investment so it is impossible to
know where this finance is going. For CDF, the policy banks do not report detailed
lending to individual countries. They do report that their overall portfolio of overseas
lending was $675 billion at end-2016, more than twice the size of the World Bank. At
the time of the Belt and Road Forum, in May 2017, they announced that as of end-
2016 about one-third of their lending had gone to BRI countries.
A data set on China’s development finance has been compiled by Dreher et al.
(2017) under the title AidData. This data set contains project-level information on
Chinese official development finance to Africa, Asia, Europe, and Latin America from
2000 to 2014. The data set was collected in two stages. In stage one, AidData identified
projects undertaken in a particular country and supported by a specific supplier of
development finance using Factiva, a Dow Jones-owned media database that draws on
Figure 1 Chinese development finance: Overall total and amount to belt and road countries.
Source: AidData.
any other source at such attractive rates, so in that sense it is a benefit to those coun-
tries. The attraction for borrowing countries is that they get access to a large amount
of financing in order to meet their serious infrastructure gaps. The projects are gener-
ally carried out by Chinese construction companies, who often bring many of their
workers with them.
In earlier work, I showed that China’s ODI is similar to Western direct investment
in that it is attracted to larger markets and to natural resource wealth, as measured by
natural resource rents as a share of gross domestic product (GDP) (Dollar, 2017). It is
unlike Western investment, however, in that it is uncorrelated with a measure of prop-
erty rights and rule of law. It is interesting to introduce BRI into that analysis. After
controlling for those other variables, an indicator for the 64 BRI countries has a nega-
tive, insignificant coefficient (Table 1, column 1).1 The dependent variable is the stock
of ODI at end-2015, the most recent year for which there is comprehensive, cross-
country data. It probably should be no surprise that BRI has had no impact on China’s
direct investment, as that should be largely commercial with much of it going to the
USA and other advanced economies. In column 2, the advanced economies are
dropped to show the allocation of ODI among developing economies. Now political
stability is significant, but it is still the case that the belt and road does not affect the
allocation.
Columns 3–6 present analogous regressions for CDF. The dependent variable is the
average of CDF for 131 developing countries over the most recent 3 years of data:
2012–2014. BRI is a relatively recent initiative, so that earlier financing would not neces-
sarily be tied to it. The explanatory variables relate to different ideas about the objective
of China’s financing, ideas that are not mutually exclusive. GDP and population mea-
sure market size; natural resource rents capture resource wealth (World Bank, 2017a);
political stability and rule of law are risk indicators; the indicator for belt and road tests
the influence of this initiative; and strategic considerations are measured by the correla-
tion of each country’s voting in the UN with China and with the US. China and the US
voting in 2015 had a negative correlation of −0.80 so they clearly have different strategic
interests.
A curious thing about the cross-country allocation of CDF is that it is hard to explain
it at all. Population is the main variable that has consistent explanatory power. Neither
the size of GDP nor natural resource wealth matters. Unlike China’s direct investment,
its development finance is not aimed at natural resource rich countries. Political stability
and rule of law are uncorrelated with the allocation. Rule of law has a negative, but insig-
nificant coefficient. And an indicator variable for belt and road countries has an insignif-
icant negative coefficient. The R2 is only 0.11, reflecting the fact that it is hard to explain
the cross-country pattern of China’s lending. Column 4 replaces BRI with a finer
regional breakdown. There is a modest tendency for Africa to get more financing, but it
is not statistically significant. Neither maritime Asia nor landlocked Asia gets more
financing than other regions. Just looking at the raw data, 37% of China’s financing in
the 2012–2014 period went to Africa; 25% to maritime Asia; 14% to Latin America; and
only 14% to landlocked Asia. In Africa in recent years, China has been providing about
Specification 1 2 3 4 5 6
Number of countries 172 128 131 131 126 126
Coefficient t-Statistics Coefficient t-Statistics Coefficient t-Statistics Coefficient t-Statistics Coefficient t-Statistics Coefficient t-Statistics
China’s Development Finance
Ln GDP 2015 0.43* 2.19 −0.04 −0.79 0.01 0.07 −0.02 −0.14 0.05 0.35 0.03 0.19
Ln population 2015 0.43* 1.95 0.85*** 7.76 0.85* 2.57 0.98** 2.97 0.80* 2.35 0.92** 2.65
Natural rents 0.08*** 3.61 0.04* 1.94 0.02 0.37 −0.03 −0.42 −0.00 −0.07 −0.04 −0.54
(% of GDP)
Political Stability 2015 0.52 1.59 0.75** 2.64 0.96 1.09 1.26 1.42 1.32 1.45 1.49 1.59
Rule of Law 2015 0.18 0.53 −0.39 −1.13 −0.46 −0.44 −0.56 −0.53 −0.61 −0.57 −0.67 −0.61
Belt and Road Initiative −0.34 −0.92 −0.18 −0.48 −0.62 −0.59 - - −0.01 −0.00 - -
(BRI)
Neighboring Country 2.54*** 3.67 1.68** 3.02 2.28 1.33 - - 2.40 1.36 - -
Africa - - - - - - 1.29 0.76 - - −0.35 −0.15
Maritime Asia - - - - - - −0.82 −0.48 - −1.53 −0.70
Land-locked Asia - - - - - - 1.42 0.62 - - 0.17 0.06
Latin America and - - - - - - −0.55 −0.31 - - −1.53 −0.67
Caribbean
UN Voting with China - - - - - - - - −12.63 −1.55 −9.19 −1.09
UN Voting with the - - - - - - - - −19.46* −1.96 −18.05 −1.60
U.S.
R2 0.46 0.54 0.11 0.12 0.13 0.12
*, **, and ** denote statistical significance at the 0.1, 0.01 and 0.001 levels, respectively.
David Dollar
one-third of the external financing for infrastructure, which is very welcome given the
infrastructure deficit on the continent (Dollar, 2016).
There is modest evidence that strategic considerations are important. First, no
country that does not recognize the People’s Republic of China received development
finance. Second, the UN voting patterns do have influence. The coefficient on voting
with China is negative but not significant; the coefficient on voting with the USA is
negative, larger in absolute value, and statistically significant. China and the USA tend
to vote in opposite directions so a relevant counterfactual would be: imagine a country
voting one standard deviation more often with China and one standard deviation less
often with the USA. The coefficients in column 5 imply that the country would receive
50% more CDF. In column 6 all of the regional dummies are added; the coefficients
on the UN variables remain about the same, but the statistical significance of the USA
UN vote correlation is now marginal.
Some additional insight can be gained by focusing on the top 20 recipients of CDF,
2012–2014 (Table 2). The list does include some Asian economies that are along the
Belt and Road, such as Iran, Pakistan, Kazakhstan, and Indonesia. But it also includes
eight African countries: Angola, Cote d’Ivoire, Ethiopia, Kenya, Nigeria, South Africa,
Sudan, and Tanzania; and three Latin ones: Venezuela, Ecuador, and Argentina. In the
regressions, the rule of law index has an insignificant negative coefficient. Looking at
the top 20 recipients, several have rule of law that is above the mean for developing
countries, such as Indonesia, Sri Lanka, Kazakhstan, Ethiopia, Kenya, South Africa,
and Tanzania; but others are rated very poorly on rule of law: Venezuela, Ecuador,
Angola, Nigeria, Sudan, Iran, and Pakistan. This means that significant amounts of
Chinese finance are going to risky environments, an issue to which I will return in the
next section. The most important thing that we have established so far is that there is
no geographic pattern to China’s development finance: it seems more demand-driven,
by which countries are willing to borrow, than supply-driven by a Chinese master plan.
But the regressions also suggest that strategic considerations play some role.
Among the top 20 borrowers there are some that are estranged from the USA, which
can be seen in their UN voting. The average developing country has a correlation with
US voting of 0.21, and with China of 0.73. The reason that the UN voting does not
have more explanatory power is that most countries vote with China, and few with the
USA. But in Table 2 it can be seen that Venezuela, Angola, Sudan, and Iran have espe-
cially low voting correlation with the USA. This could mean that China is particularly
willing to finance countries that are not allies of the USA. It could also mean that these
countries have poor access to global capital markets and that therefore Chinese financ-
ing is especially welcome.
China’s growing development finance raises several issues of global governance, the
first of which is debt sustainability. Developing countries have suffered severe external
debt crises from time to time: Latin America in the 1980s, East Asia in the 1990s, and
Russia in 1998 are just some of the examples. As a result of these bitter experiences
developing countries have become more aware of the issue of debt sustainability.
External debt is different from domestic debt in that it has to be serviced ultimately
through exports. Capital flows to developing countries go through cycles: at times, in
the search for yield, global investors are willing to lend a lot at relatively low interest
rates. It is attractive then to borrow externally in order to fund infrastructure. There is
always a risk, however, of capital flow reversal and increases in interest rates. Chinese
banks are secretive about their lending terms, but most of these loans are in dollars at
flexible, commercial rates. Only about one-quarter of China’s development finance,
2012–2014, is concessional enough to meet the standard of “official development assis-
tance.” Kitano (2014) analyzes the implementation of this highly concessional part of
China’s financing.
For the nonconcessional lending, as interest rates rise in New York and London,
the cost of servicing loans from China will rise. The ability to service external debt also
depends on the value of one’s exports. Looking at the list of major borrowers from
China, many are exporters of energy or minerals: Venezuela, Ecuador, South Africa,
Nigeria, Angola, Iran, and Sudan, for example. Servicing debt may be reasonable at
one price for exports, but become burdensome if the price falls significantly. The fall in
the prices of energy and minerals in recent years is raising the specter of a new round
of debt crises. The current trend of low commodity prices and rising dollar interest
rates is putting the squeeze on the finances of developing countries.
Some, but not all, of the countries that have borrowed heavily from China in recent
years are at risk of debt distress. The World Development Indicators include recent
data on external debt relative to gross national income for 106 of the countries
included in the database on CDF, including all of the top 20 borrowers. For these
20 countries, debt to GNI increased from 35% in 2008 to 50% in 2015. For the other
77 developing countries there was a modest increase in external debt, from an average
of 45% of GNI in 2008 to an average of 48% in 2015. The average level of debt for the
major borrowers from China is not alarming. But the rapid increase is something of a
concern. More important, the average disguises large variation at the country level. In
the last couple of years large increases in debt, taking countries to risky levels, were
experienced by Angola, Belarus, Cote d’Ivoire, Ethiopia, Kenya, South Africa, Ukraine,
Venezuela, and Tanzania. A number of these countries have very poor governance,
and it is not surprising that debt has not been used productively. The rise of Chinese
development financing is too recent a phenomenon to have careful studies of its
impact on growth. The rise in external debt to GDP is an indicator to watch because a
strong growth impact would increase GDP and tend to keep the ratio stable; whereas a
weak growth effect would show up in debt to GDP rising to unsustainable levels.
China’s willingness to lend to countries with very poor governance may well pro-
long the poor governance. Ricardo Hausmann (2015), an economist and a former
planning minister for Venezuela, made this point in a 2015 op-ed for Project
Syndicate:
Venezuela has tried to finance itself with the help of the China Development Bank, which
does not impose the kind of conditionality the IMF bashers dislike. Instead, the CDB
lends on secret terms, for uses that are undisclosed and corrupt, and with built-in privi-
leges for Chinese companies in areas like telecommunications (Huawei), appliances
(Haier), cars (Chery), and oil drilling (ICTV). The Chinese have not required that Vene-
zuela do anything to increase the likelihood that it regains creditworthiness. They merely
demand more oil as collateral. Whatever the IMF’s faults, the CDB is a disgrace.
Is China violating norms of global finance? At this point it would be hard to argue
that. In the case of Venezuela, China has had to renegotiate loan terms in favor of
Venezuela because the country was unable to service the original loans once the price
of oil fell. As Venezuela’s economic crisis has deepened in the face of poor economic
management, China has stopped making new loans to the country, indicating that it is
unwilling to underwrite the regime with a blank check.
Of the countries that have borrowed heavily from China, several currently have
IMF programs to help with unsustainable fiscal and balance of payments problems:
Cote d’Ivoire, Kenya, and Ukraine (IMF, 2017a). Other countries that have borrowed
heavily from China, on the other hand, are in good fiscal and financial shape: Kazakh-
stan and Indonesia would be examples.
An interesting recent development is that China is providing $50 million to fund a
China-IMF Capacity Development Center (IMF, 2017b). This virtual center will be
under IMF administration, will be anchored in Beijing, and will offer courses both
inside and outside China on core Fund topics. Roughly half the participants will be
Chinese officials, and half, officials from other developing countries, including coun-
tries along the BRI. One of the important topics that will be emphasized initially is
debt sustainability analysis. The People’s Bank of China (PBC) is the driving force
behind this initiative, and curiously PBC represents China in MDBs such as the Afri-
can Development Bank and the Inter-American Development Bank. PBC naturally has
more awareness of this issue than other Chinese agencies and wants the knowledge to
be spread within China, and also wants to strengthen the capacity of other developing
countries. In the end, it is the governments of borrowing countries that need to dem-
onstrate discipline and far-sightedness.
On the issue of debt sustainability, a balanced assessment is that most of the devel-
oping countries taking advantage of Chinese finance for infrastructure are in sound fis-
cal condition. A few have taken on excessive amounts of debt, and they have turned to
the IMF for the traditional medicine of adjustment policies and emergency finance.
Venezuela is the one case in which China’s financing may have enabled poor economic
policies to persist. But China has reduced its exposure and it seems likely that Venezu-
ela will go the IMF in the end.
One aspect of the business practices of the World Bank … that has a particularly strong
impact on infrastructure investment is environmental and social safeguard policies.
These procedures developed by the World Bank, the gold standard of environmental
and social safeguards, have had a number of unintended consequences. It has become
time-consuming and expensive to do infrastructure projects with the World Bank and,
as a result, developing countries have turned to other sources of funding. Infrastructure
was the original core business of the World Bank, accounting for 70% of lending in
the 1950s and 1960s. That has steadily declined to about 30% in the 2000s.
Given this situation, the emergence of China as a major funder of infrastructure pro-
jects has been welcomed by most developing countries. China is seen as more flexible
and less bureaucratic. It completes infrastructure projects relatively quickly so that the
benefits are realized sooner. However, China’s approach of relying on a recipient coun-
try’s own laws and regulations has its own risks. Some of the infrastructure projects that
China has proposed in Latin America, such as the Nicaragua Canal or the Brazil-to-Peru
rail across the Amazon and the Andes, carry serious environmental and social risks.
Such projects call for carefully balancing development needs with environmental risks.
The Working Group on Development and Environment in the Americas, a multiu-
niversity effort, carried out case studies for eight countries on the question of whether
Chinese trade and investment had led to environmental degradation. On the one hand,
they conclude that “Chinese trade and investment in Latin America since the turn of
the 21st century was a major driver of environmental degradation in the region, and
was also a source of social conflict.” (Ray et al., 2015, p. 2) On the other hand, they
find evidence of positive evolution: “Chinese investors show an ability to exceed local
standards, but their performance varies widely across different regulatory regimes and
between more experienced and newer firms. There is an important role for Latin
American governments and civil society to raise the performance level across the
board, through holding firms accountable and facilitating learning between firms.”
(Ray et al., 2015, p. 3).
A case study of Chinese construction of hydro plants in Cameroon found a similar
result (Chen & Landry, 2016). One Chinese contractor executed a World Bank project
for a hydro plant. The implementation carefully addressed environmental and resettle-
ment issues and provided a mechanism for redress for affected people. However, it
took 7 years from initiation of the project until the start of construction. At the same
time another hydro project, financed by China EXIM and implemented by a Chinese
contractor, moved much faster – 4 years from initiation to start of construction. The
Chinese financing was also much more expensive: a flexible rate of Euribor +310 basis
points; 15-year term with 5 years grace. The World Bank financing was at a fixed 0.5%
interest rate with 40-year term and 10 years grace. The case study concludes
(p. 17) that:
The Chinese-financed project complied with all legal requirements and with Eximbank’s
own policies. However, in managing both foreseen and unforeseen social and environ-
mental impacts, the type of response, and degree of responsibility taken by the financiers
differed. Both projects struggled with problems surrounding the Chinese contractors and
labor issues. However, the autonomous managers of the World Bank project enjoyed far
better capacity and enforcement mechanisms, and a greater willingness to use them to
force compliance. Both China Eximbank and the World Bank have upped their game in
prioritizing environmental norms and standards: however, the severity of these standards,
and the means by which they are applied, differ. … China Eximbank is evolving in its
stance towards norms of environment and social impact mitigation. In the case of Mem-
ve’ele, it appears to have played a silent role in the overall project implementation: while
environmental impact mitigation and assessments were a condition of the loan disburse-
ment, enforcement and monitoring was largely the responsibility of the GOC, for better
or worse, showing a gap between theory and practice. There is something of a trade-off,
as some GOC respondents intimated, between Chinese and Western IFI financing: while
financing from China is faster and less onerous, it comprises a risk of more issues arising
at later stages of project implementation.
While the bulk of the current funding for BRI comes from the existing policy banks, in
the future the AIIB will also play a role. AIIB was established under Chinese leadership
AIIB’s leadership hopes that the bank can meet international standards but be
more timely and cost-effective. This is largely a matter of implementation and it will
take time and experience on the ground to see if the effort is a success. In its first
2 years of operation AIIB lent about $3 billion, with three-quarters of its projects co-
financed with the World Bank or regional development banks. It will take time for
AIIB to build up a portfolio of projects that it developed on its own. If AIIB can meet
environmental standards more efficiently, that would be a very positive innovation. If
AIIB’s activities can put pressure on the World Bank and the regional development
banks to streamline their procedures and speed up their infrastructure projects, then
this would be a positive change to the global system that emanated from China.
6. Conclusions
Note
1 The belt and road countries are identified by the Hong Kong Trade Development Coun-
cil (2017).
References
Atkins L., Brautigam D., Chen Y. & Hwang J. (2017). Challenges of and opportunities from the
commodity price slump. CARI Economic Bulletin no. 1. Washington DC: China Africa
Research Initiative (CARI), Johns Hopkins University School of Advanced International