China began to make loans to African countries in the 1960s to gain access to raw materials.
Featured - Belt and Road
Initiative - External
Reports 3-18-21
Risky
Business: New Data on Chinese Loans and Africa’s Debt Problem.
By Deborah Brautigam, Yufan Huang, and Kevin Acker
FROM MODEST BEGINNINGS IN 1960, CHINA HAS RECENTLY become a highly visible actor in Africa’s lending landscape. African borrowers have built roads, installed electrical grids, and modernized their airports with Chinese finance. Yet when commodity prices and growth rates began to tumble in 2015, the specter of a new debt crisis arose. These fears expanded sharply with the impact of the COVID‑19 pandemic.
Are the African countries most vulnerable to debt distress those with high Chinese debt? Who are the Chinese lenders in Africa and how do they manage lending in risky environments? Is China a bigger lender than the World Bank? What kind of terms do we see on Chinese loans in Africa? Why have Chinese banks lent so much in risky environments? How often are loans collateralized with natural resource exports? Do Chinese banks require property as collateral for loans to African governments or their state‑owned enterprises (SOEs)?
In
this paper we attempt to answer these questions, using data on Chinese loan
commitments from the SAIS China Africa Research Initiative and the World Bank,
and data on African borrowing and debt levels from the World Bank and
International Monetary Fund’s International Debt Statistics. Our analysis finds
that Chinese loans play a more modest role in Africa’s struggle with debt
sustainability than conventional wisdom would suggest. The picture varies
sharply across the continent’s 54 countries, however. New data released by the
World Bank in June 2020 suggest that Chinese lending is over 25 percent of the
debt stock in seven countries in Africa deemed to be at risk of, or already in,
debt distress: Djibouti (57 percent), Angola (49 percent), Republic of Congo
(45 percent), Cameroon (32 percent), Ethiopia (32 percent), Kenya (27 percent),
and Zambia (26 percent). However, in another 12 countries, more than half of
those at highest risk of debt distress, Chinese lending is relatively modest,
making up less than fifteen percent of all debt. This data is helpful, but many
questions remain as to the details of debt and its evolution.
RISING AFRICA AND RISKY
AFRICA: ENTER CHINA
IN
THE 1980s AND 1990s, MANY AFRICAN COUNTRIES went through a long struggle with
economic crisis, reluctant reform, and mounting debt. Penalty charges imposed
by Paris Club creditors and other bilateral lenders on debtors’ arrears added
billions to debt woes. After the millennium, thirty‑one low income African
countries received substantial debt cancellation through the highly indebted
poor countries (HIPC) initiative spearheaded by the Paris Club, the World Bank
and the International Monetary Fund (IMF). Yet as their debts were eased, many
countries began to borrow again to fill an annual gap in infrastructure finance
estimated to be between US$ 68 and US$ 108 billion. As their traditional
sources of credit, the Paris Club, had shifted their areas of concern away from
infrastructure, countries turned to non‑traditional financiers, including
China.
Chinese
lending took off in a period of rapid growth in Africa. In 2014, a year after
the peak in nonAngolan Chinese lending commitments, the IMF predicted that sub‑Saharan
Africa would continue to grow at a rate of 5.5 percent. Yet already in a May
2014 interview, IMF Managing Director Christine Lagarde presciently warned that
the “Africa rising” story could be at risk. “Governments should be attentive
and they should be cautious about not overloading the countries with too much
debt.” Oil prices fell dramatically from US$100/bbl in 2014 to only US$ 44/bbl
in 2016. Copper prices showed a similar decline.
In
2016, the African region’s growth rate slumped to 1.4 percent, the lowest rate
since 1995. Although some countries continued to post strong economic growth,
such as Senegal, Ethiopia, and Kenya, others struggled with political
instability including civil war, or saw their economies contract with a fall in
commodity prices. Still, as of 2019, before the COVID‑19 crisis hit, analysts
at Brookings argued that fears were overblown. Although some countries were
facing difficulties, “an African sovereign debt crisis is not imminent.” Even
with the economic recession caused by COVID‑19, the picture remains nuanced.
Countries dependent on tourism, or commodities with depressed demand, are more
at risk than those without these income sources. Many countries are facing a
liquidity crisis but not all are facing insolvency.
By
2017, the Paris Club accounted for only five percent of public and publicly
guaranteed debt in sub‑Saharan Africa. Private lenders from wealthy countries
filled a large part of this gap, responding to opportunities in a continent
with high risks and high rewards. Over the past decade, more than fifteen
African countries issued foreign currency sovereign bonds, many for the very
first time. Eight African countries issued 30‑year bonds in 2018. As Table 1
points out, in 2018, African governments owed US$ 117 billion to the bond
market, 32 percent of public and publicly guaranteed debt.
Much
of the rest of the gap was filled by Chinese lenders. According to the World
Bank’s International Debt Statistics, in 2017, official bilateral credits from
China accounted for 62 percent of bilateral official credits, or about 23
percent of all public and publicly guaranteed debt in sub‑Saharan Africa. In
2018, for the 40 low income African countries, Chinese debt came to 60 percent
of bilateral lending, and 17 percent of public and publicly guaranteed (PPG)
debt in this subset of countries. Chinese commercial banks are also part of the
picture, and would be accounted for in this data as non‑official private
creditors, even if they are owned by the state.
OUR DATA
Overview of our data
CHINESE
FINANCIERS ARE NOT VERY transparent, and do not systematically provide data on
the loans they offer to individual overseas borrowers. International media
sources do sometimes report on loans from China to Africa, but these reports
are often inaccurate. Many reported loans are based on the wish lists of
African governments or vague memorandums of understanding (MoUs) signed between
African and Chinese officials, and never end up being officially signed and
disbursed.
To
sift through the noise and collect accurate data on signed Chinese loan
commitments to African governments or their SOEs, CARI deploys a highly
trained, multilingual team of research assistants (RAs). CARI RAs are trained
to follow a rigorous set of steps to triangulate and confirm the existence of
officially signed loans using African, Chinese, and international sources. Most
loans are not considered confirmed until our researchers find them included in
African ministry of finance or central bank documents or reported by Chinese
embassies in Africa. Furthermore, loans we have identified as signed are
periodically re‑checked, and loans for projects that do not enter disbursement
are removed. Deskwork is supplemented with field research conducted by CARI
fellows, and interviews with our contacts in Africa.
Our
data on loan commitments should not be viewed as “debt”. Each loan takes an
average of five years to disburse, and for large projects the disbursement time
is longer. Some loans have already been repaid. It is not uncommon for
outstanding debt to only reach 49 percent of total loan commitments between
2000 and 2018. Nigeria provides a good illustration. Between 2000 and 2018,
Nigeria signed 16 loan contracts totaling US$ 6 billion with Chinese
financiers. However, their outstanding debt to China as of 2018 was only US$
2.5 billion. This is because Nigeria has repaid US$ 660 million, and another
US$ 2.8 billion remains undisbursed. Most of the undisbursed debt is accounted
for by the US$ 2.6 billion worth of loans signed in 2017 and 2018. As each loan
takes around five years to be fully disbursed, Nigeria has yet to receive most
of the funds from these loans.
With
US$ 43 billion, Angola makes up 29 percent of all Chinese loan commitments in
Africa. Both the volume and the modalities of Chinese lending there are quite
different from most of China’s other development partners on the continent.
The Shifting Landscape of
Chinese Lenders
ALTHOUGH
THE CHINESE GOVERNMENT made its first official loan to an African country
(Guinea) in 1960, Chinese banks are relatively new to the continent. Between
1994 and the present, Chinese lenders proliferated in Africa.
China
Eximbank today counts clients in 45 African countries, and Africa accounts for
a third of the bank’s overseas business. The average annual growth rate of
China Eximbank’s Africa lending surpassed 40 percent between 2006 and 2018 (see
Figure 3: Loans by Lender). China Development Bank has provided finance to over
a dozen African governments. Loans from Chinese commercial banks are growing.
The largest syndicated loan in our database is US$ 4.1 billion in support of
the massive 2,170 mw Caculo Cabaca hydropower project in Angola. Here,
Industrial and Commercial Bank of China (ICBC) brought together a number of
China‑based banks, including Bank of China’s Beijing branch, China Construction
Bank’s Beijing Branch, China Minsheng Bank, Ping An Bank, and Bank of China’s
Shanghai Pilot Trade Zone Branch.
Terms of Chinese Loans
RECENTLY
A PAPER PUBLISHED BY THE Center for Global Development analyzed the terms of
lending by China and compared these terms with those from the World Bank. This
research showed that the World Bank’s loans to poor countries come with 1.54
percent fixed interest rates, 10‑year grace periods, and 40‑year maturities.
The Bank also offers market‑based loans to countries that no longer qualify for
the most concessional lending. These use the six‑month LIBOR rate, plus 205
basis points (bp), and have maturities of 18 to 20 years.
Our
data suggests that Chinese terms and interest rates vary considerably by
lender, and type of project. China offers interest‑free loans as one instrument
in its foreign aid program, but these make up less than five percent of all
Chinese loan commitments. Concessional loans are subsidized by the Chinese
government from the foreign aid budget, while preferential export buyer’s credits
are subsidized from other budget lines. In the first years of the millennium,
China Eximbank’s concessional loans had interest rates around four percent, but
over the past decade, interest rates for China Eximbank’s concessional and
preferential export buyer’s credit have stabilized at a fairly uniform two
percent, with 20‑year maturity. Grace periods for these loans vary by the
expected construction time of the project, but average five years.
Like
the World Bank’s IBRD (International Bank for Reconstruction and Development)
loans, Chinese commercial loans are provided at a variable interest rate linked
to the sixmonth LIBOR plus a margin. The lowest margin in our data is 30 bps
(0.3 percent) charged by the Industrial and Commercial Bank of China for a
rural electrification project in Ghana. The highest margin was 450 bp for
a loan from China Development Bank to Zambia to make up the 15 percent of
project cost it was required to contribute for the Mansa‑Luwingu Road Project.
Most commercial loans had margins in the 300 pb range. The maturity for these
loans ranges from three to 17 years with an average of eight years.
Collateralized Lending
ONE
OF THE misunderstood features of Chinese overseas lending is the extent to
which loans are secured by collateral. For example, a New York Times article in
May 2020 contended that borrowers of Chinese loans “put up ports, mines and
other crown jewels as collateral”. We have seen only one case, a Chinese‑funded
petroleum refinery in Chad, where a sovereign borrower put up property as
collateral. We believe this statement is a misreading of the practice of
collateralized lending or using future revenue earnings to secure
infrastructure project loan finance.
Collateralized
lending, or “resource‑secured infrastructure finance,” is a form of project
finance in which repayment is secured not through existing assets as in a
mortgage (i.e. property) but through future receivables (i.e. future cocoa,
tobacco, oil, copper, and other export revenues). While Chinese banks did not
invent this project finance model, China Eximbank in particular has been an
enthusiastic user of it in Africa. This is because reducing the risk of a loan
lowers its cost, which enables borrowing governments to use more finance (and
employ more Chinese construction firms, since, like all export credit agencies,
China Eximbank’s lending is largely tied to the use of Chinese goods and
services).
For
example, in Ghana, a Chinese loan to support construction of the Bui Dam was
secured through an arrangement whereby the Ghanaian marketing board for cocoa,
Cocobod, arranged to export cocoa beans to a Chinese buyer. The dam is
also secured by an off‑take arrangement based on future electricity revenues
paid by consumers. The cocoa security was arranged through a sales agreement
between Genertec Corporation of China and Cocobod for up to 40,000 metric tons
of cocoa beans annually for the first five years of the loan. The cocoa
beans will be sold for foreign exchange at the prevailing market price, and the
proceeds placed in an escrow account with China Eximbank. The cocoa security
arrangement was scheduled to last for five years, and after this, loan
repayment is done from the electricity sales. This off‑take arrangement
requires Bui Hydropower to have a power purchase agreement with the Electricity
Company of Ghana: 85 percent of energy sales from Bui will be deposited into an
escrow account to help repay the loan. In the first instance, Bui Dam revenues
(in cedi) will reimburse Cocobod for the cocoa exports. In both cases of cocoa
and electricity, excess funds in the account can be withdrawn by Ghana, or they
can stay in the account and earn interest.
Whereas
many banks and commodity traders have used commodity‑secured loans in Africa,
it is far less common to see them used to lower risks for project finance in
infrastructure construction. Yet even though we have many examples in our data,
this lending mode is less widespread than many believe. Commodity‑secured loans
account for 25 percent of the loan commitments in our data. However, lending to
Angola, much of which is secured by oil exports, makes up 75 percent of the
export commodity‑secured loans. This model has also been used in nine other
countries, and this amounts to only six to 12 percent of the loan commitments
in our data. In some cases (such as Nigeria or Ghana) only a handful of
projects were financed this way. Chinese banks have used this model more
extensively in the Republic of Congo, Democratic Republic of the Congo, Sudan,
and Equatorial Guinea, where at least 57 projects were constructed with
resource‑secured financing.
We
expect that, as with Ghana’s Bui Dam electricity off‑take, other kinds of
revenue securities that are not export‑related exist. Kenya’s Standard Gauge
Railway, for example, uses escrow accounts filled with revenues from the
railway itself, and, if necessary, funds from the Railway Development levy, a
tax of 1.5 percent imposed on all imports into Kenya, and collected by the
Kenya Port Authority. In 2018, the Railway Development Levy would have provided
US$ 261 million to supplement revenues earned by the railway. By
mechanisms like these, low income countries have found revenue sources that add
to their ability to finance expensive infrastructure projects.
As
noted above, we found only one case in our data‑‑a US$ 330 million preferential
export buyer’s credit from China Eximbank that financed China National
Petroleum Corporation’s joint venture oil refinery in Chad‑‑where the loan was
secured with the actual asset. Chad had provided a government guarantee but
Sinosure, China’s export credit insurer, refused to insure the loan given the
risk. CNPC arranged for Chad’s shares to be collateral for the loan. CNPC
also used its own shares as collateral for its portion of the investment loan,
a common practice in China.
DEVELOPMENT SUSTAINABILITY
VERSUS DEBT SUSTAINABILITY
EAST
ASIA’S DEVELOPMENT SUCCESS was based on debt financing. China’s own growth
model has involved extensive public sector borrowing. Local governments in
China have created numerous special purpose vehicles (LSPVs): companies
established to access loan funding for infrastructure projects and real estate
development. Yet across East Asia, including in China, external borrowing was
fairly conservative during periods of rapid growth. In China, external
borrowing never rose above 20 percent of gross domestic product (GDP).
Li
Ruogu, the former head of China’s export credit agency, China Export Import
Bank, argued that post‑HIPC concerns about debt sustainability meant that low
income borrowers were restrained from investing in projects that could provide
the necessary infrastructure for their economic development. The World Bank and
IMF, having lived through several decades of debt crisis, and implemented debt
write‑offs under the HIPC initiative, have been more wary about optimism on the
sustainability of borrowing. The International Financial Institution’s (IFIs)
Debt Sustainability Framework provided limits on borrowing for post‑HIPC
countries, and these limits have been the subject of intense negotiations with
some countries that wanted to borrow at commercial rates from Chinese lenders.
In
April 2019 China’s Ministry of Finance published China’s own Debt
Sustainability Framework. In many ways, the framework is quite similar to the
one used by the Washingtonbased IFIs. However, as Johanna Malm has pointed out,
there are some important differences. For example, Malm notes, “the framework
makes it clear that China does not see debt distress as an obstacle to
continued borrowing.”
“[I]t
should be noted that an assessment for a country as “high risk” of debt
distress, or even “in debt distress”, does not automatically mean that debt is
unsustainable in a forward‑looking sense. In general, when a country is likely
to meet its current and future repayment obligations, its [public and publicly
guaranteed] external debt and overall public debt are sustainable.”
The
crux of the Chinese argument is the belief that “Productive investment, while
increasing debt ratios in the short run, can generate higher economic growth
[…] leading to lower debt ratios over time” [emphasis added]. The concern
of course is whether investment is truly productive.
CHINESE LENDING AND DEBT
DISTRESS: WHAT DO THE NUMBERS SAY?
IN
THIS SECTION WE USE EXAMPLES FROM our data, and information on Chinese lending
contained in the World Bank’s monitoring of the COVID‑19 Debt Service
Suspension Initiative (DSSI), to present a picture of the Chinese contribution
to debt in Africa. All figures refer to PPG. The DSSI data suggest that for the
40 low income African countries, debt to China adds up to US$ 64 billion and accounts
for 22 percent of the PPG debt stock in 2018, and an estimated 29 percent of
debt service due in 2020. The outstanding debt to the World Bank is very close
to this figure. Low income African countries owe the World Bank US$ 62 billion,
but due to generous subsidies the World Bank is able to offer its clients, debt
service is lower.
Countries
in debt distress or at high risk of debt distress as of June 2020 are marked.
Along the left axis, we show where Chinese debt lies as a percent of GNI. Debt
sustainability is based on the country’s entire portfolio of debt and its
terms, but this provides a quick snapshot of one important variable.
The
ratio of debt to national income that can be sustained depends on a number of
factors. In the European Union, for example, member countries have committed to
keep debt below 60 percent of national income. The IMF’s debt sustainability
analysis uses a ratio of 30 percent for countries with weakmacroeconomic
performance and management capacity but allows countries with stronger
management to go up to 50 percent of GDP. We can see that Chinese debt as a
percent of income is below 10 percent for most African borrowers. The
exceptions are Angola, Djibouti, Republic of Congo, Mozambique, Ethiopia, and
Zambia.
In
the next section, we look at the level of Chinese debt in the individual
countries that, before the COVID‑19 pandemic, were judged to be at high risk
of, or already in, debt distress. While the list of countries facing debt
distress will undoubtedly grow as a result of COVID‑19, we start with the 20
African countries rated by the World Bank and IMF as at “high” risk or “in debt
distress” in the June 2020 DSSI list, and in other publications. Figure 4 shows
those countries. In the case studies below, we add Angola, which concluded a
program with the IMF in 2018, an act that signals a high level of debt risk,
although there is no formal rating. We group these 22 countries into three
categories, those with a small share of debt owed to China (less than 15 percent),
those with a medium share (15 to 25 percent), and those with a high share (over
25 percent).
Chinese Loans a Small
Share (under 15%) of Debt Stock: 12 debt-distressed countries
IN
OVER HALF OF THE COUNTRIES IN, or at high risk of, debtdistress, China accounts
for less than 15 percent of PPG debt and debt service is approximately in the
same range. Some of these countries have borrowed very little from China.
Burundi: Saudi Arabia and India
have both lent more to Burundi than China, whose debt comes to just two percent
of the total. Almost all of Chinese lending went to telecoms projects.
Cape Verde: China held less than
two percent of Cape Verde’s debt, having financed airport scanners for US$ 13
million and two e‑government projects (US$ 30 million).
Chad: In Chad, a US$ 1.45
billion oil‑secured loan from the Anglo‑Swiss company Glencore meant that by
the end of 2016, 85 percent of Chad’s oil exports (its primary source of
revenue) were going to repay Glencore. Only seven percent of Chad’s debt
is owed to China, financing a cement factory, Chad’s share of an oil refinery
and power station built by China National Petroleum Corporation, and
electricity transmission lines from the refinery to the capital. After the oil
price crash in 2014‑2015, Chad was having trouble paying three loans with
upcoming maturities to China Eximbank, and in 2017 the outstanding debts on
these loans were rescheduled.
Eritrea: China makes up 4
percent of the country’s debt stock. The loans were to support a telecom
project, a thermal power plant, food storage, purchase of Chinese machinery,
and a road.
The Gambia: As of 2018, China
had made only one loan to The Gambia, for a telecoms project.
Ghana: The government of
Ghana has a total external debt of US$ 19.4 billion. They owed US$ 1.86 billion
to China (under 10 percent), while nearly half of the external debt is owed to
bondholders and commercial banks, and a third to multilateral banks. Our data
show US$ 3.7 billion in loan commitments between 2000 and 2018. Ghana’s largest
Chinese‑financed projects include several loans for the 2007 Bui Dam (US$ 673
million), and the 2013 Western Corridor gas processing plant (US$ 850 million).
The latter loan had a 10‑year term, so should be close to being repaid.
Mauritania: Saudi Arabia is the
largest single creditor in Mauritania, where China holds less than 10 percent
of debt. China built Mauritania’s Friendship Port in the 1980s as a foreign aid
project. The largest project in Mauritania since 2000 has been an expansion of
this port, for US$ 293 million.
São Tomé and Príncipe: The DSSI lists
outstanding official bilateral debt to China as US$ 10 million. This is likely
misreported by São Tomé and Príncipe. Our data records no lending to São Tomé
and Príncipe from official Chinese bilateral lenders. However, we do record a
US$ 30 million loan from the China International Fund, a private company from
Hong Kong, from which only US$ 10 million was disbursed.
Sierra Leone: In Sierra Leone,
Chinese lenders have financed several fiberoptic and telecoms projects, for a
total outstanding debt of US$ 48 million, about three percent of the country’s
total debt according to the DSSI. However, the World Bank figure does not include
a contingent liability for a public private partnership (PPP) toll road
financed by China Railway 7th Group (the US$ 165 million, 67 km Wellington‑Masiaka
Toll Road).
Somalia: According to the
DSSI, Chinese debt, at US$ 83.9 million, is less than four percent of the total
in Somalia, where the largest creditors are Italy, the United Arab Emirates,
and the United States, all creditors to whom Somalia‑‑a highly indebted poor
country that embarked on the HIPC debt relief process only in March 2020‑‑is in
arrears. We have recorded no loans from China to Somalia between 2000 and 2018.
Since there is no debt service scheduled for these Chinese loans, we believe
the World Bank records refer to Chinese debt from before 2000 that has gone
into arrears.
Sudan: In Sudan, which is
also not eligible for the DSSI, China is a less significant lender than many
might assume. Paris Club debt, including arrears, at US$ 20.6 billion is 37.7
percent of the total, and non‑Paris Club debt at US$ 20.2 billion, is 36.9
percent. According to the DSSI, Saudi Arabia is currently Sudan’s major
creditor, with US$ 1.6 billion outstanding. The data released by the World Bank
in June 2020 note that China accounts for only eight percent of Sudan’s
outstanding debt. While according to our data, China has signed off on at least
US$ 6.8 billion to Sudan over the years, a portion of this debt has already
been repaid through oil shipments. After the loss of the oil‑producing South in
2012, China granted a delay of five years to Sudan’s debt repayment.
South Sudan: The World Bank did not
include data for South Sudan on its DSSI website (it is possible that the data
for South Sudan was combined with that for Sudan). However, the IMF reported
that as of the end of March 2019, South Sudan owed China US$ 150 million for
the Juba International Airport, out of a total debt stock of US$ 1.196 billion,
or 12.5 percent. Our data shows two loans signed as of 2018, for the
airport, and for an air traffic system, for a total of US$ 407 million.
Chinese Loans a Medium
Share (15 to 25 percent) of Debt Stock: 3 Debt Distressed Countries
IN
A SECOND GROUP, INCLUDING Central African Republic, Mozambique, and Zimbabwe,
Chinese debt stock is from 15 to 25 percent of the total. It is important to
note that in all three countries, debt service due in 2020 rises considerably
above 25 percent. These cases show the impact of higher interest rates for
Chinese loans compared with other funders.
Central African Republic: In the Central
African Republic (CAR), China accounted for 18 percent of the debt stock, but
31 percent of debt service for 2020. The DSSI lists outstanding debt to China
as US$ 130 million. According to the IMF, CAR is in arrears to Taiwan (we also
see this in Liberia and Burkina Faso, which are at only moderate risk of debt
distress). These loans appear to have been folded into the China figures,
since our figures show only US$ 71 million in Chinese loan commitments. Our
data shows several small loans to help make up budgetary shortfalls, and two
other projects (US$ 21 million for the Boali hydropower project and a US$ 36
million for a telecoms project) for a total of US$ 73 million in lending
commitments.
Mozambique: The DSSI records
show that Mozambique owed China US$ 2 billion out of a total debt of US$ 11.4
billion (18 percent), but debt service to Chinese creditors was expected to be
27 percent of all debt service for 2020. It is possible that this debt service
figure reflects debt reprofiling that was done in 2017 and pushed some debt
payments off to later dates. In 2017, Mozambique was in default on its external
public debt, and reached an agreement to reschedule its debts to China.
According to media reports, Mozambique was granted a grace period on payments
due on US$ 2.02 billion, although the original maturities were
maintained. Our data show signed loan commitments of US$ 2.5 billion.
Mozambique’s largest loans from China have financed the Maputo‑Catembe Bridge
(US$ 686 million), Beira to Machipanda EN6 road repair (287 km) for US$ 416
million, and the 74 km Maputo ring road (US$ 300 million). Aside from one zero‑interest
foreign aid loan signed in 2018, there have been no new Chinese loans since the
country reprofiled its Chinese debt.
Zimbabwe: Although Zimbabwe is
not eligible for the DSSI because of its arrears to the World Bank and IMF, it
is in debt distress. Because of these arrears, Zimbabwe’s debt sustainability
analysis published in March 2020 noted that 77 percent of Zimbabwe’s debt was
owed to Paris Club and multilateral creditors like the World Bank, with “non‑Paris
Club lenders”, including China, making up 20 percent. However, the DSSI
figure for Zimbabwe states that China accounts for 24.8 percent of Zimbabwe’s
outstanding debt, around US$ 1.1 billion. It is not clear why this discrepancy
exists.
Our
database includes US$ 3 billion in Chinese loan commitments in Zimbabwe since
2000. The largest projects include US$ 998 million for the Hwange coal‑fired
power plant expansion, signed in 2017 and currently underway, and US$ 360
million for the Kariba South Bank hydropower project, and US$ 219 million for
an upgrade to state‑owned telecoms company NetOne. As in other countries, a
significant portion of these loan commitments remain to be disbursed.
According
to the DSSI, Zimbabwe was scheduled to repay Chinese lenders US$ 72 million in
2020, which would have been 54 percent of all debt service. Zimbabwe has
defaulted on multiple loans to China Eximbank, and each time was granted
extensions of the repayment period. A US$ 35 million loan for ZISCOSteel was
granted maturity extensions in 2003, 2007, and 2010. Two loans totaling US$ 18
million for the SinoZimbabwe Cement Plant were granted maturity extensions and
interest rate reductions in 2004, and a US$ 200 million buyer’s credit for
agricultural machinery was granted a maturity extension in 2012.
Chinese Loans a Large
Share (over 25 percent) of Debt Stock
IN
SEVEN LOW INCOME COUNTRIES with significant debt problems, China now holds over
25 percent of all external debt, according to the DSSI figures: Angola,
Cameroon, Republic of Congo, Djibouti, Ethiopia, Kenya, and Zambia.
Angola: With over US$ 19
billion in debt to China according to the DSSI, Angola is China’s largest
borrower in Africa, and indeed, owes China the most among all the low‑income
countries in the DSSI. China makes up 49 percent of outstanding government
debt. For 2020, debt service on Chinese loans is scheduled to account for 58
percent of all debt service due. Our database shows over US$ 40 billion in loan
commitments over the past 20 years. Many of these have been repaid with oil
exports, and some have not yet been disbursed. Chinese lines of credit have
financed over 100 projects in Angola, including the US$ 2.5 billion Kilamba
Kiaxi new city with over 20,000 apartments, US$ 509 million for the Nzeto‑Soyo
road, and US$ 835 million for Soyo I, Africa’s largest gas turbine power
station. Angola also refinanced a number of loans that its state‑owned oil
company Sonangol owed to Chinese banks. Between 2010 and 2014, Sonangol signed
US$ 10 billion worth of loans with CDB and ICBC. When the oil price crashed in
2015, Sonangol was having trouble paying them back. In late 2015, CDB extended
a US$ 15 billion line of credit to Angola, US$ 10 billion of which Angola used
to recapitalize Sonangol. Over the next two years Sonangol used some of this
money to pre‑pay debts to Chinese banks.
Cameroon: Cameroon has
borrowed extensively from Chinese banks. Some of the largest loan commitments
include Kribi port (US$ 948 million over two phases), Yaounde water supply
project from the Sanaga river (US$ 678 million), and US$ 541 million for the
211 Mwh Memve’ele hydropower dam. We have data on the terms for 27 loans
out of a total of 44 loans from China, all but two are at fixed interest rates,
two percent or less. According to the World Bank’s DSSI data, China accounted
for 32 percent of Cameroon’s public debt as of 2018 (US$ 3.2 billion), but 45
percent of all debt service estimated to be due in 2020. This is a function of
the loan restructuring Cameroon agreed to in 2019. Cameroon’s debt to China was
reprofiled so that only one third of the debt service due between 2019 and 2022
would have to be paid over those three years, with the other two thirds (US$
253 million) reprofiled to be paid in following years within the existing
maturity. Cameroon’s debt difficulties meant that disbursement of funds committed
to existing projects slowed. As of late September 2019, China accounted for
28.5 percent of undisbursed external loans. Chinese banks could slow or
pause disbursement if Cameroon was unable to finance its share (usually 15
percent) of a project, or unable to complete tasks like compensating
landholders in the project area, a responsibility usually left with host
governments.
Republic of the
Congo: Chinese lending to the Republic of Congo accounts for 45 percent
of the country’s external debt, and in 2020, 43 percent of debt service. The
largest loans were for highways. Chinese loans have financed a new highway,
National Route 1, linking Brazzaville with Pointe Noire on the coast (US$ 1.8
billion) and National Route 2 (US$ 537 million). In 2019, China Eximbank agreed
to restructure US$ 1.6 billion of outstanding debt from loans signed by the ROC
between 2010 and 2014, extending the maturities by 15 years and reducing the
interest rates.
Djibouti: In Djibouti, where
Chinese banks financed the Djibouti portion of the Djibouti‑Addis railway, a
large water project, and three port upgrade projects, China holds 57 percent of
PPG debt at US$ 1.2 billion, according to the DSSI. Multilateral creditors,
with US$ 600 million, hold 29 percent. Djibouti’s debt service due to China
accounts for 58 percent of the total due in 2020. Negotiations between Djibouti
and China Eximbank are ongoing regarding the restructuring of the US$ 490
million loan for Djibouti’s section of the Addis‑Djibouti railway. In 2019, an
MOU was signed to extend the maturity by 10 years and reduce the interest from
LIBOR + 300 bps to LIBOR + 210 bps. The agreement has yet to be finalized.
Ethiopia: Our data suggest
that Ethiopia has signed loan agreements with Chinese lenders of almost US$ 14
billion between 2000 and 2018. These loans have funded over 50 projects, the
most significant being telecoms expansion (US$ 3 billion), wind farms (US$ 600
million), hydropower plants and associated transmission lines (US$ 2.3
billion), the Addis Ababa light rail system (US$ 475 million), the Addis‑Djibouti
railway (US$ 2.5 billion), and a number of sugar complexes including mills (US$
1.7 billion). The DSSI data lists China as the single most significant creditor
in Ethiopia, with outstanding debt of US$ 8.7 billion, 32 percent of all public
debt. The World Bank is very close, with 31 percent of outstanding debt. Yet
higher interest rates for Chinese loans mean that China makes up 42 percent of
all debt service due in 2020. In 2018, China Eximbank granted a restructuring
for the loan for the Ethiopian section of the AddisDjibouti railway, extending
the maturity by 20 years.
Kenya: The DSSI records
that Chinese lending to Kenya makes up 27 percent (US$ 7.5 billion) of Kenya’s
outstanding debt of US$ 27.8 billion. Large projects financed by Chinese
lenders in Kenya include US$ 867 million for a number of geothermal wells at Olkaria,
US$ 229 million for the Karimenu Dam Water Supply Project, US$ 156 million for
the Nairobi southern bypass highway, and US$ 5.1 billion for two phases of the
controversial Standard Gauge Railway between Mombasa and Malaba. In Kenya,
commercial interest rates for some large loans, including part of the Standard
Gauge Railway, mean that in 2020, debt service on Chinese loans was scheduled
to take up 38 percent of all debt service.
Zambia: Zambia has been
hovering on the precipice of debt distress for several years, even though a
joint IMF/World Bank debt sustainability analysis (DSA) published in August
2019 noted that at that point, Zambia was servicing its debt and had “remained
current on all its debt obligations.” According to the DSSI, as of 2018,
PPG debt to China was less (US$ 2.8 billion) than to bondholders (US$ 3
billion), and Chinese lenders held about 26 percent of PPG debt in Zambia. The
DSA recorded that Zambia’s public electricity utility ZESCO held an additional
US$ 700 million in non‑guaranteed debt, likely to be entirely from Chinese
lenders.
These
figures are quite a bit smaller than our CARI data on signed Chinese loan
commitments (US$ 9.7 billion). Aside from repayments, the most obvious reason
for this difference is the distinction between loan commitments and
disbursements, as we noted above with the case of Nigeria. The DSA noted that
Zambia had contracted US$ 9.7 billion of PPG loans that were expected to be
disbursed between 2019 and 2024. A large portion of this is likely to be
Chinese. For example, ZESCO signed off on US$ 5.6 billion in Chinese loan
commitments for power projects just between 2016 and 2018. Because of Zambia’s
debt problems and inability to contribute its side of some project costs
(compensation for land acquisition, for example), disbursement on existing
projects has been halting.
DISCUSSION
IN
2020, HOW DIFFERENT IS AFRICA’S debt crisis compared with the crisis that began
in the late 1970s? The earlier crisis had its origins in global and domestic
policy factors, and the balance between these differed country by country.
There is no dispute about the two most important global shocks: the dramatic
rise in oil prices caused by the Yom Kippur War (1974) and the IranIraq War
(1980) which hurt all oil importers, and the dramatic rise in global interest
rates caused by the July 1981 decision of the US Federal Reserve to raise US
interest rates to 22.36 percent. Developing countries that had borrowed at
variable rates found their debts far more difficult to service, and capital
fled from the south to the north.
When
countries went into arrears, they faced steep penalties from Paris Club
creditors. Domestic policy factors such as overvalued exchange rates, loss‑making
public companies, and expensive subsidies also played a role. In the late
1980s, the members of the Paris Club recognized the fact that dozens of low‑income
countries were essentially bankrupt. In low income sub‑Saharan Africa, the
external debt to GNI ratio rose from 49 percent to 104 percent between 1980 and
1987, considerably above today’s pre‑COVID‑19 levels. This recognition of
insolvency slowly led to programs of bilateral and multilateral debt writeoffs,
culminating in the HIPC initiative in 1996.
In
2020, after decades of reform, most African countries are in far better
economic shape than they were in the 1980s, although there are exceptions such
as Zimbabwe. Global interest rates are at record low levels. In some countries
such as South Sudan, the Central African Republic, Burundi, and Somalia, armed
conflict and civil war hamper economic output and debt sustainability. Pre‑COVID
19, the collapse of commodity prices was the chief factor driving debt distress
in Angola, Chad, the Republic of Congo, Mauritania, and Sudan. This also played
a role in Ghana and Zambia, but expansive spending around elections was an
additional factor. Still, for most of the 21 countries in this analysis, debt
problems are likely to be liquidity problems rather than reflections of
insolvency.
Chinese
lenders are now a significant part of the debt picture in Africa, but their
role should not be overestimated. In over half of the low‑income countries most
at risk of, or already in, debt distress, Chinese lending is relatively small,
with less than 15 percent of debt stock. That is to say, their debt problems
are largely caused by lenders other than China.
In
another 3 countries, borrowing from China appears to be between 15 and 25
percent of debt stock: Central African Republic (CAR), Mozambique, and
Zimbabwe. Yet in CAR, some of the DSSI debt data is from Taiwan, which
complicates analysis.
In
just seven African countries, as of 2018, China contributed between 26 and 58
percent of PPG debt stock. These are the countries where Chinese lending is
central to the African debt distress picture.
Debt
service is another important variable. In two countries, Angola and Djibouti,
more than 50 percent of debt service was owed to Chinese lenders in 2020. Yet
even here, looking at averages creates a somewhat misleading picture. With
Angola included, 29 percent of all debt service in the DSSI countries in Africa
is due to China. With Angola excluded, only 18 percent of debt service is
Chinese.
The
World Bank’s DSSI data is a gold mine of information for Chinese lending in low
income countries and Angola. Our CARI database gives the details on each loan
commitment made by Chinese lenders across Africa, including countries that are
not part of the DSSI. These two resources should allow analysts to make a great
leap forward in understanding the myths and realities of Chinese lending in
risky markets in Africa.
Please
click to read full report.
FROM
MODEST BEGINNINGS IN 1960, CHINA HAS RECENTLY become a highly visible actor in
Africa’s lending landscape. African borrowers have built roads, installed
electrical grids, and modernized their airports with Chinese finance. Yet when
commodity prices and growth rates began to tumble in 2015, the specter of a new
debt crisis arose. These fears expanded sharply with the impact of the COVID‑19
pandemic.
Are
the African countries most vulnerable to debt distress those with high Chinese
debt? Who are the Chinese lenders in Africa and how do they manage lending in
risky environments? Is China a bigger lender than the World Bank? What kind of
terms do we see on Chinese loans in Africa? Why have Chinese banks lent so much
in risky environments? How often are loans collateralized with natural resource
exports? Do Chinese banks require property as collateral for loans to African
governments or their state‑owned enterprises (SOEs)?
In
this paper we attempt to answer these questions, using data on Chinese loan
commitments from the SAIS China Africa Research Initiative and the World Bank,
and data on African borrowing and debt levels from the World Bank and
International Monetary Fund’s International Debt Statistics. Our analysis finds
that Chinese loans play a more modest role in Africa’s struggle with debt
sustainability than conventional wisdom would suggest. The picture varies
sharply across the continent’s 54 countries, however. New data released by the
World Bank in June 2020 suggest that Chinese lending is over 25 percent of the
debt stock in seven countries in Africa deemed to be at risk of, or already in,
debt distress: Djibouti (57 percent), Angola (49 percent), Republic of Congo
(45 percent), Cameroon (32 percent), Ethiopia (32 percent), Kenya (27 percent),
and Zambia (26 percent). However, in another 12 countries, more than half of
those at highest risk of debt distress, Chinese lending is relatively modest,
making up less than fifteen percent of all debt. This data is helpful, but many
questions remain as to the details of debt and its evolution.
RISING AFRICA AND RISKY
AFRICA: ENTER CHINA
IN
THE 1980s AND 1990s, MANY AFRICAN COUNTRIES went through a long struggle with
economic crisis, reluctant reform, and mounting debt. Penalty charges imposed
by Paris Club creditors and other bilateral lenders on debtors’ arrears added
billions to debt woes. After the millennium, thirty‑one low income African
countries received substantial debt cancellation through the highly indebted
poor countries (HIPC) initiative spearheaded by the Paris Club, the World Bank
and the International Monetary Fund (IMF). Yet as their debts were eased, many
countries began to borrow again to fill an annual gap in infrastructure finance
estimated to be between US$ 68 and US$ 108 billion. As their traditional
sources of credit, the Paris Club, had shifted their areas of concern away from
infrastructure, countries turned to non‑traditional financiers, including
China.
Chinese
lending took off in a period of rapid growth in Africa. In 2014, a year after
the peak in nonAngolan Chinese lending commitments, the IMF predicted that sub‑Saharan
Africa would continue to grow at a rate of 5.5 percent. Yet already in a May
2014 interview, IMF Managing Director Christine Lagarde presciently warned that
the “Africa rising” story could be at risk. “Governments should be attentive
and they should be cautious about not overloading the countries with too much
debt.” Oil prices fell dramatically from US$100/bbl in 2014 to only US$ 44/bbl
in 2016. Copper prices showed a similar decline.
In
2016, the African region’s growth rate slumped to 1.4 percent, the lowest rate
since 1995. Although some countries continued to post strong economic growth,
such as Senegal, Ethiopia, and Kenya, others struggled with political
instability including civil war, or saw their economies contract with a fall in
commodity prices. Still, as of 2019, before the COVID‑19 crisis hit, analysts
at Brookings argued that fears were overblown. Although some countries were
facing difficulties, “an African sovereign debt crisis is not imminent.” Even
with the economic recession caused by COVID‑19, the picture remains nuanced.
Countries dependent on tourism, or commodities with depressed demand, are more
at risk than those without these income sources. Many countries are facing a
liquidity crisis but not all are facing insolvency.
By
2017, the Paris Club accounted for only five percent of public and publicly
guaranteed debt in sub‑Saharan Africa. Private lenders from wealthy countries
filled a large part of this gap, responding to opportunities in a continent
with high risks and high rewards. Over the past decade, more than fifteen
African countries issued foreign currency sovereign bonds, many for the very
first time. Eight African countries issued 30‑year bonds in 2018. As Table 1
points out, in 2018, African governments owed US$ 117 billion to the bond
market, 32 percent of public and publicly guaranteed debt.
Much
of the rest of the gap was filled by Chinese lenders. According to the World
Bank’s International Debt Statistics, in 2017, official bilateral credits from
China accounted for 62 percent of bilateral official credits, or about 23
percent of all public and publicly guaranteed debt in sub‑Saharan Africa. In
2018, for the 40 low income African countries, Chinese debt came to 60 percent
of bilateral lending, and 17 percent of public and publicly guaranteed (PPG)
debt in this subset of countries. Chinese commercial banks are also part of the
picture, and would be accounted for in this data as non‑official private
creditors, even if they are owned by the state.
OUR DATA
Overview of our data
CHINESE
FINANCIERS ARE NOT VERY transparent, and do not systematically provide data on
the loans they offer to individual overseas borrowers. International media
sources do sometimes report on loans from China to Africa, but these reports
are often inaccurate. Many reported loans are based on the wish lists of
African governments or vague memorandums of understanding (MoUs) signed between
African and Chinese officials, and never end up being officially signed and
disbursed.
To
sift through the noise and collect accurate data on signed Chinese loan
commitments to African governments or their SOEs, CARI deploys a highly
trained, multilingual team of research assistants (RAs). CARI RAs are trained
to follow a rigorous set of steps to triangulate and confirm the existence of
officially signed loans using African, Chinese, and international sources. Most
loans are not considered confirmed until our researchers find them included in
African ministry of finance or central bank documents or reported by Chinese
embassies in Africa. Furthermore, loans we have identified as signed are
periodically re‑checked, and loans for projects that do not enter disbursement
are removed. Deskwork is supplemented with field research conducted by CARI
fellows, and interviews with our contacts in Africa.
Our
data on loan commitments should not be viewed as “debt”. Each loan takes an
average of five years to disburse, and for large projects the disbursement time
is longer. Some loans have already been repaid. It is not uncommon for
outstanding debt to only reach 49 percent of total loan commitments between
2000 and 2018. Nigeria provides a good illustration. Between 2000 and 2018,
Nigeria signed 16 loan contracts totaling US$ 6 billion with Chinese
financiers. However, their outstanding debt to China as of 2018 was only US$
2.5 billion. This is because Nigeria has repaid US$ 660 million, and another
US$ 2.8 billion remains undisbursed. Most of the undisbursed debt is accounted
for by the US$ 2.6 billion worth of loans signed in 2017 and 2018. As each loan
takes around five years to be fully disbursed, Nigeria has yet to receive most
of the funds from these loans.
With
US$ 43 billion, Angola makes up 29 percent of all Chinese loan commitments in
Africa. Both the volume and the modalities of Chinese lending there are quite
different from most of China’s other development partners on the continent.
The Shifting Landscape of
Chinese Lenders
ALTHOUGH
THE CHINESE GOVERNMENT made its first official loan to an African country
(Guinea) in 1960, Chinese banks are relatively new to the continent. Between
1994 and the present, Chinese lenders proliferated in Africa.
China
Eximbank today counts clients in 45 African countries, and Africa accounts for
a third of the bank’s overseas business. The average annual growth rate of
China Eximbank’s Africa lending surpassed 40 percent between 2006 and 2018 (see
Figure 3: Loans by Lender). China Development Bank has provided finance to over
a dozen African governments. Loans from Chinese commercial banks are growing.
The largest syndicated loan in our database is US$ 4.1 billion in support of
the massive 2,170 mw Caculo Cabaca hydropower project in Angola. Here,
Industrial and Commercial Bank of China (ICBC) brought together a number of
China‑based banks, including Bank of China’s Beijing branch, China Construction
Bank’s Beijing Branch, China Minsheng Bank, Ping An Bank, and Bank of China’s
Shanghai Pilot Trade Zone Branch.
Terms of Chinese Loans
RECENTLY
A PAPER PUBLISHED BY THE Center for Global Development analyzed the terms of
lending by China and compared these terms with those from the World Bank. This
research showed that the World Bank’s loans to poor countries come with 1.54
percent fixed interest rates, 10‑year grace periods, and 40‑year maturities.
The Bank also offers market‑based loans to countries that no longer qualify for
the most concessional lending. These use the six‑month LIBOR rate, plus 205
basis points (bp), and have maturities of 18 to 20 years.
Our
data suggests that Chinese terms and interest rates vary considerably by
lender, and type of project. China offers interest‑free loans as one instrument
in its foreign aid program, but these make up less than five percent of all
Chinese loan commitments. Concessional loans are subsidized by the Chinese
government from the foreign aid budget, while preferential export buyer’s credits
are subsidized from other budget lines. In the first years of the millennium,
China Eximbank’s concessional loans had interest rates around four percent, but
over the past decade, interest rates for China Eximbank’s concessional and
preferential export buyer’s credit have stabilized at a fairly uniform two
percent, with 20‑year maturity. Grace periods for these loans vary by the
expected construction time of the project, but average five years.
Like
the World Bank’s IBRD (International Bank for Reconstruction and Development)
loans, Chinese commercial loans are provided at a variable interest rate linked
to the sixmonth LIBOR plus a margin. The lowest margin in our data is 30 bps
(0.3 percent) charged by the Industrial and Commercial Bank of China for a
rural electrification project in Ghana. The highest margin was 450 bp for
a loan from China Development Bank to Zambia to make up the 15 percent of
project cost it was required to contribute for the Mansa‑Luwingu Road Project.
Most commercial loans had margins in the 300 pb range. The maturity for these
loans ranges from three to 17 years with an average of eight years.
Collateralized Lending
ONE
OF THE misunderstood features of Chinese overseas lending is the extent to
which loans are secured by collateral. For example, a New York Times article in
May 2020 contended that borrowers of Chinese loans “put up ports, mines and
other crown jewels as collateral”. We have seen only one case, a Chinese‑funded
petroleum refinery in Chad, where a sovereign borrower put up property as
collateral. We believe this statement is a misreading of the practice of
collateralized lending or using future revenue earnings to secure
infrastructure project loan finance.
Collateralized
lending, or “resource‑secured infrastructure finance,” is a form of project
finance in which repayment is secured not through existing assets as in a
mortgage (i.e. property) but through future receivables (i.e. future cocoa,
tobacco, oil, copper, and other export revenues). While Chinese banks did not
invent this project finance model, China Eximbank in particular has been an
enthusiastic user of it in Africa. This is because reducing the risk of a loan
lowers its cost, which enables borrowing governments to use more finance (and
employ more Chinese construction firms, since, like all export credit agencies,
China Eximbank’s lending is largely tied to the use of Chinese goods and
services).
For
example, in Ghana, a Chinese loan to support construction of the Bui Dam was
secured through an arrangement whereby the Ghanaian marketing board for cocoa,
Cocobod, arranged to export cocoa beans to a Chinese buyer. The dam is
also secured by an off‑take arrangement based on future electricity revenues
paid by consumers. The cocoa security was arranged through a sales agreement
between Genertec Corporation of China and Cocobod for up to 40,000 metric tons
of cocoa beans annually for the first five years of the loan. The cocoa
beans will be sold for foreign exchange at the prevailing market price, and the
proceeds placed in an escrow account with China Eximbank. The cocoa security
arrangement was scheduled to last for five years, and after this, loan
repayment is done from the electricity sales. This off‑take arrangement
requires Bui Hydropower to have a power purchase agreement with the Electricity
Company of Ghana: 85 percent of energy sales from Bui will be deposited into an
escrow account to help repay the loan. In the first instance, Bui Dam revenues
(in cedi) will reimburse Cocobod for the cocoa exports. In both cases of cocoa
and electricity, excess funds in the account can be withdrawn by Ghana, or they
can stay in the account and earn interest.
Whereas
many banks and commodity traders have used commodity‑secured loans in Africa,
it is far less common to see them used to lower risks for project finance in
infrastructure construction. Yet even though we have many examples in our data,
this lending mode is less widespread than many believe. Commodity‑secured loans
account for 25 percent of the loan commitments in our data. However, lending to
Angola, much of which is secured by oil exports, makes up 75 percent of the
export commodity‑secured loans. This model has also been used in nine other
countries, and this amounts to only six to 12 percent of the loan commitments
in our data. In some cases (such as Nigeria or Ghana) only a handful of
projects were financed this way. Chinese banks have used this model more
extensively in the Republic of Congo, Democratic Republic of the Congo, Sudan,
and Equatorial Guinea, where at least 57 projects were constructed with
resource‑secured financing.
We
expect that, as with Ghana’s Bui Dam electricity off‑take, other kinds of
revenue securities that are not export‑related exist. Kenya’s Standard Gauge
Railway, for example, uses escrow accounts filled with revenues from the
railway itself, and, if necessary, funds from the Railway Development levy, a
tax of 1.5 percent imposed on all imports into Kenya, and collected by the
Kenya Port Authority. In 2018, the Railway Development Levy would have provided
US$ 261 million to supplement revenues earned by the railway. By
mechanisms like these, low income countries have found revenue sources that add
to their ability to finance expensive infrastructure projects.
As
noted above, we found only one case in our data‑‑a US$ 330 million preferential
export buyer’s credit from China Eximbank that financed China National
Petroleum Corporation’s joint venture oil refinery in Chad‑‑where the loan was
secured with the actual asset. Chad had provided a government guarantee but
Sinosure, China’s export credit insurer, refused to insure the loan given the
risk. CNPC arranged for Chad’s shares to be collateral for the loan. CNPC
also used its own shares as collateral for its portion of the investment loan,
a common practice in China.
DEVELOPMENT SUSTAINABILITY
VERSUS DEBT SUSTAINABILITY
EAST
ASIA’S DEVELOPMENT SUCCESS was based on debt financing. China’s own growth
model has involved extensive public sector borrowing. Local governments in
China have created numerous special purpose vehicles (LSPVs): companies
established to access loan funding for infrastructure projects and real estate
development. Yet across East Asia, including in China, external borrowing was
fairly conservative during periods of rapid growth. In China, external
borrowing never rose above 20 percent of gross domestic product (GDP).
Li
Ruogu, the former head of China’s export credit agency, China Export Import
Bank, argued that post‑HIPC concerns about debt sustainability meant that low
income borrowers were restrained from investing in projects that could provide
the necessary infrastructure for their economic development. The World Bank and
IMF, having lived through several decades of debt crisis, and implemented debt
write‑offs under the HIPC initiative, have been more wary about optimism on the
sustainability of borrowing. The International Financial Institution’s (IFIs)
Debt Sustainability Framework provided limits on borrowing for post‑HIPC
countries, and these limits have been the subject of intense negotiations with
some countries that wanted to borrow at commercial rates from Chinese lenders.
In
April 2019 China’s Ministry of Finance published China’s own Debt
Sustainability Framework. In many ways, the framework is quite similar to the
one used by the Washingtonbased IFIs. However, as Johanna Malm has pointed out,
there are some important differences. For example, Malm notes, “the framework
makes it clear that China does not see debt distress as an obstacle to
continued borrowing.”
“[I]t
should be noted that an assessment for a country as “high risk” of debt
distress, or even “in debt distress”, does not automatically mean that debt is
unsustainable in a forward‑looking sense. In general, when a country is likely
to meet its current and future repayment obligations, its [public and publicly
guaranteed] external debt and overall public debt are sustainable.”
The
crux of the Chinese argument is the belief that “Productive investment, while
increasing debt ratios in the short run, can generate higher economic growth
[…] leading to lower debt ratios over time” [emphasis added]. The concern
of course is whether investment is truly productive.
CHINESE LENDING AND DEBT
DISTRESS: WHAT DO THE NUMBERS SAY?
IN
THIS SECTION WE USE EXAMPLES FROM our data, and information on Chinese lending
contained in the World Bank’s monitoring of the COVID‑19 Debt Service
Suspension Initiative (DSSI), to present a picture of the Chinese contribution
to debt in Africa. All figures refer to PPG. The DSSI data suggest that for the
40 low income African countries, debt to China adds up to US$ 64 billion and accounts
for 22 percent of the PPG debt stock in 2018, and an estimated 29 percent of
debt service due in 2020. The outstanding debt to the World Bank is very close
to this figure. Low income African countries owe the World Bank US$ 62 billion,
but due to generous subsidies the World Bank is able to offer its clients, debt
service is lower.
Countries
in debt distress or at high risk of debt distress as of June 2020 are marked.
Along the left axis, we show where Chinese debt lies as a percent of GNI. Debt
sustainability is based on the country’s entire portfolio of debt and its
terms, but this provides a quick snapshot of one important variable.
The
ratio of debt to national income that can be sustained depends on a number of
factors. In the European Union, for example, member countries have committed to
keep debt below 60 percent of national income. The IMF’s debt sustainability
analysis uses a ratio of 30 percent for countries with weakmacroeconomic
performance and management capacity but allows countries with stronger
management to go up to 50 percent of GDP. We can see that Chinese debt as a
percent of income is below 10 percent for most African borrowers. The
exceptions are Angola, Djibouti, Republic of Congo, Mozambique, Ethiopia, and
Zambia.
In
the next section, we look at the level of Chinese debt in the individual
countries that, before the COVID‑19 pandemic, were judged to be at high risk
of, or already in, debt distress. While the list of countries facing debt
distress will undoubtedly grow as a result of COVID‑19, we start with the 20
African countries rated by the World Bank and IMF as at “high” risk or “in debt
distress” in the June 2020 DSSI list, and in other publications. Figure 4 shows
those countries. In the case studies below, we add Angola, which concluded a
program with the IMF in 2018, an act that signals a high level of debt risk,
although there is no formal rating. We group these 22 countries into three
categories, those with a small share of debt owed to China (less than 15 percent),
those with a medium share (15 to 25 percent), and those with a high share (over
25 percent).
Chinese Loans a Small
Share (under 15%) of Debt Stock: 12 debt-distressed countries
IN
OVER HALF OF THE COUNTRIES IN, or at high risk of, debtdistress, China accounts
for less than 15 percent of PPG debt and debt service is approximately in the
same range. Some of these countries have borrowed very little from China.
Burundi: Saudi Arabia and India
have both lent more to Burundi than China, whose debt comes to just two percent
of the total. Almost all of Chinese lending went to telecoms projects.
Cape Verde: China held less than
two percent of Cape Verde’s debt, having financed airport scanners for US$ 13
million and two e‑government projects (US$ 30 million).
Chad: In Chad, a US$ 1.45
billion oil‑secured loan from the Anglo‑Swiss company Glencore meant that by
the end of 2016, 85 percent of Chad’s oil exports (its primary source of
revenue) were going to repay Glencore. Only seven percent of Chad’s debt
is owed to China, financing a cement factory, Chad’s share of an oil refinery
and power station built by China National Petroleum Corporation, and
electricity transmission lines from the refinery to the capital. After the oil
price crash in 2014‑2015, Chad was having trouble paying three loans with
upcoming maturities to China Eximbank, and in 2017 the outstanding debts on
these loans were rescheduled.
Eritrea: China makes up 4
percent of the country’s debt stock. The loans were to support a telecom
project, a thermal power plant, food storage, purchase of Chinese machinery,
and a road.
The Gambia: As of 2018, China
had made only one loan to The Gambia, for a telecoms project.
Ghana: The government of
Ghana has a total external debt of US$ 19.4 billion. They owed US$ 1.86 billion
to China (under 10 percent), while nearly half of the external debt is owed to
bondholders and commercial banks, and a third to multilateral banks. Our data
show US$ 3.7 billion in loan commitments between 2000 and 2018. Ghana’s largest
Chinese‑financed projects include several loans for the 2007 Bui Dam (US$ 673
million), and the 2013 Western Corridor gas processing plant (US$ 850 million).
The latter loan had a 10‑year term, so should be close to being repaid.
Mauritania: Saudi Arabia is the
largest single creditor in Mauritania, where China holds less than 10 percent
of debt. China built Mauritania’s Friendship Port in the 1980s as a foreign aid
project. The largest project in Mauritania since 2000 has been an expansion of
this port, for US$ 293 million.
São Tomé and Príncipe: The DSSI lists
outstanding official bilateral debt to China as US$ 10 million. This is likely
misreported by São Tomé and Príncipe. Our data records no lending to São Tomé
and Príncipe from official Chinese bilateral lenders. However, we do record a
US$ 30 million loan from the China International Fund, a private company from
Hong Kong, from which only US$ 10 million was disbursed.
Sierra Leone: In Sierra Leone,
Chinese lenders have financed several fiberoptic and telecoms projects, for a
total outstanding debt of US$ 48 million, about three percent of the country’s
total debt according to the DSSI. However, the World Bank figure does not include
a contingent liability for a public private partnership (PPP) toll road
financed by China Railway 7th Group (the US$ 165 million, 67 km Wellington‑Masiaka
Toll Road).
Somalia: According to the
DSSI, Chinese debt, at US$ 83.9 million, is less than four percent of the total
in Somalia, where the largest creditors are Italy, the United Arab Emirates,
and the United States, all creditors to whom Somalia‑‑a highly indebted poor
country that embarked on the HIPC debt relief process only in March 2020‑‑is in
arrears. We have recorded no loans from China to Somalia between 2000 and 2018.
Since there is no debt service scheduled for these Chinese loans, we believe
the World Bank records refer to Chinese debt from before 2000 that has gone
into arrears.
Sudan: In Sudan, which is
also not eligible for the DSSI, China is a less significant lender than many
might assume. Paris Club debt, including arrears, at US$ 20.6 billion is 37.7
percent of the total, and non‑Paris Club debt at US$ 20.2 billion, is 36.9
percent. According to the DSSI, Saudi Arabia is currently Sudan’s major
creditor, with US$ 1.6 billion outstanding. The data released by the World Bank
in June 2020 note that China accounts for only eight percent of Sudan’s
outstanding debt. While according to our data, China has signed off on at least
US$ 6.8 billion to Sudan over the years, a portion of this debt has already
been repaid through oil shipments. After the loss of the oil‑producing South in
2012, China granted a delay of five years to Sudan’s debt repayment.
South Sudan: The World Bank did not
include data for South Sudan on its DSSI website (it is possible that the data
for South Sudan was combined with that for Sudan). However, the IMF reported
that as of the end of March 2019, South Sudan owed China US$ 150 million for
the Juba International Airport, out of a total debt stock of US$ 1.196 billion,
or 12.5 percent. Our data shows two loans signed as of 2018, for the
airport, and for an air traffic system, for a total of US$ 407 million.
Chinese Loans a Medium
Share (15 to 25 percent) of Debt Stock: 3 Debt Distressed Countries
IN
A SECOND GROUP, INCLUDING Central African Republic, Mozambique, and Zimbabwe,
Chinese debt stock is from 15 to 25 percent of the total. It is important to
note that in all three countries, debt service due in 2020 rises considerably
above 25 percent. These cases show the impact of higher interest rates for
Chinese loans compared with other funders.
Central African Republic: In the Central
African Republic (CAR), China accounted for 18 percent of the debt stock, but
31 percent of debt service for 2020. The DSSI lists outstanding debt to China
as US$ 130 million. According to the IMF, CAR is in arrears to Taiwan (we also
see this in Liberia and Burkina Faso, which are at only moderate risk of debt
distress). These loans appear to have been folded into the China figures,
since our figures show only US$ 71 million in Chinese loan commitments. Our
data shows several small loans to help make up budgetary shortfalls, and two
other projects (US$ 21 million for the Boali hydropower project and a US$ 36
million for a telecoms project) for a total of US$ 73 million in lending
commitments.
Mozambique: The DSSI records
show that Mozambique owed China US$ 2 billion out of a total debt of US$ 11.4
billion (18 percent), but debt service to Chinese creditors was expected to be
27 percent of all debt service for 2020. It is possible that this debt service
figure reflects debt reprofiling that was done in 2017 and pushed some debt
payments off to later dates. In 2017, Mozambique was in default on its external
public debt, and reached an agreement to reschedule its debts to China.
According to media reports, Mozambique was granted a grace period on payments
due on US$ 2.02 billion, although the original maturities were
maintained. Our data show signed loan commitments of US$ 2.5 billion.
Mozambique’s largest loans from China have financed the Maputo‑Catembe Bridge
(US$ 686 million), Beira to Machipanda EN6 road repair (287 km) for US$ 416
million, and the 74 km Maputo ring road (US$ 300 million). Aside from one zero‑interest
foreign aid loan signed in 2018, there have been no new Chinese loans since the
country reprofiled its Chinese debt.
Zimbabwe: Although Zimbabwe is
not eligible for the DSSI because of its arrears to the World Bank and IMF, it
is in debt distress. Because of these arrears, Zimbabwe’s debt sustainability
analysis published in March 2020 noted that 77 percent of Zimbabwe’s debt was
owed to Paris Club and multilateral creditors like the World Bank, with “non‑Paris
Club lenders”, including China, making up 20 percent. However, the DSSI
figure for Zimbabwe states that China accounts for 24.8 percent of Zimbabwe’s
outstanding debt, around US$ 1.1 billion. It is not clear why this discrepancy
exists.
Our
database includes US$ 3 billion in Chinese loan commitments in Zimbabwe since
2000. The largest projects include US$ 998 million for the Hwange coal‑fired
power plant expansion, signed in 2017 and currently underway, and US$ 360
million for the Kariba South Bank hydropower project, and US$ 219 million for
an upgrade to state‑owned telecoms company NetOne. As in other countries, a
significant portion of these loan commitments remain to be disbursed.
According
to the DSSI, Zimbabwe was scheduled to repay Chinese lenders US$ 72 million in
2020, which would have been 54 percent of all debt service. Zimbabwe has
defaulted on multiple loans to China Eximbank, and each time was granted
extensions of the repayment period. A US$ 35 million loan for ZISCOSteel was
granted maturity extensions in 2003, 2007, and 2010. Two loans totaling US$ 18
million for the SinoZimbabwe Cement Plant were granted maturity extensions and
interest rate reductions in 2004, and a US$ 200 million buyer’s credit for
agricultural machinery was granted a maturity extension in 2012.
Chinese Loans a Large
Share (over 25 percent) of Debt Stock
IN
SEVEN LOW INCOME COUNTRIES with significant debt problems, China now holds over
25 percent of all external debt, according to the DSSI figures: Angola,
Cameroon, Republic of Congo, Djibouti, Ethiopia, Kenya, and Zambia.
Angola: With over US$ 19
billion in debt to China according to the DSSI, Angola is China’s largest
borrower in Africa, and indeed, owes China the most among all the low‑income
countries in the DSSI. China makes up 49 percent of outstanding government
debt. For 2020, debt service on Chinese loans is scheduled to account for 58
percent of all debt service due. Our database shows over US$ 40 billion in loan
commitments over the past 20 years. Many of these have been repaid with oil
exports, and some have not yet been disbursed. Chinese lines of credit have
financed over 100 projects in Angola, including the US$ 2.5 billion Kilamba
Kiaxi new city with over 20,000 apartments, US$ 509 million for the Nzeto‑Soyo
road, and US$ 835 million for Soyo I, Africa’s largest gas turbine power
station. Angola also refinanced a number of loans that its state‑owned oil
company Sonangol owed to Chinese banks. Between 2010 and 2014, Sonangol signed
US$ 10 billion worth of loans with CDB and ICBC. When the oil price crashed in
2015, Sonangol was having trouble paying them back. In late 2015, CDB extended
a US$ 15 billion line of credit to Angola, US$ 10 billion of which Angola used
to recapitalize Sonangol. Over the next two years Sonangol used some of this
money to pre‑pay debts to Chinese banks.
Cameroon: Cameroon has
borrowed extensively from Chinese banks. Some of the largest loan commitments
include Kribi port (US$ 948 million over two phases), Yaounde water supply
project from the Sanaga river (US$ 678 million), and US$ 541 million for the
211 Mwh Memve’ele hydropower dam. We have data on the terms for 27 loans
out of a total of 44 loans from China, all but two are at fixed interest rates,
two percent or less. According to the World Bank’s DSSI data, China accounted
for 32 percent of Cameroon’s public debt as of 2018 (US$ 3.2 billion), but 45
percent of all debt service estimated to be due in 2020. This is a function of
the loan restructuring Cameroon agreed to in 2019. Cameroon’s debt to China was
reprofiled so that only one third of the debt service due between 2019 and 2022
would have to be paid over those three years, with the other two thirds (US$
253 million) reprofiled to be paid in following years within the existing
maturity. Cameroon’s debt difficulties meant that disbursement of funds committed
to existing projects slowed. As of late September 2019, China accounted for
28.5 percent of undisbursed external loans. Chinese banks could slow or
pause disbursement if Cameroon was unable to finance its share (usually 15
percent) of a project, or unable to complete tasks like compensating
landholders in the project area, a responsibility usually left with host
governments.
Republic of the
Congo: Chinese lending to the Republic of Congo accounts for 45 percent
of the country’s external debt, and in 2020, 43 percent of debt service. The
largest loans were for highways. Chinese loans have financed a new highway,
National Route 1, linking Brazzaville with Pointe Noire on the coast (US$ 1.8
billion) and National Route 2 (US$ 537 million). In 2019, China Eximbank agreed
to restructure US$ 1.6 billion of outstanding debt from loans signed by the ROC
between 2010 and 2014, extending the maturities by 15 years and reducing the
interest rates.
Djibouti: In Djibouti, where
Chinese banks financed the Djibouti portion of the Djibouti‑Addis railway, a
large water project, and three port upgrade projects, China holds 57 percent of
PPG debt at US$ 1.2 billion, according to the DSSI. Multilateral creditors,
with US$ 600 million, hold 29 percent. Djibouti’s debt service due to China
accounts for 58 percent of the total due in 2020. Negotiations between Djibouti
and China Eximbank are ongoing regarding the restructuring of the US$ 490
million loan for Djibouti’s section of the Addis‑Djibouti railway. In 2019, an
MOU was signed to extend the maturity by 10 years and reduce the interest from
LIBOR + 300 bps to LIBOR + 210 bps. The agreement has yet to be finalized.
Ethiopia: Our data suggest
that Ethiopia has signed loan agreements with Chinese lenders of almost US$ 14
billion between 2000 and 2018. These loans have funded over 50 projects, the
most significant being telecoms expansion (US$ 3 billion), wind farms (US$ 600
million), hydropower plants and associated transmission lines (US$ 2.3
billion), the Addis Ababa light rail system (US$ 475 million), the Addis‑Djibouti
railway (US$ 2.5 billion), and a number of sugar complexes including mills (US$
1.7 billion). The DSSI data lists China as the single most significant creditor
in Ethiopia, with outstanding debt of US$ 8.7 billion, 32 percent of all public
debt. The World Bank is very close, with 31 percent of outstanding debt. Yet
higher interest rates for Chinese loans mean that China makes up 42 percent of
all debt service due in 2020. In 2018, China Eximbank granted a restructuring
for the loan for the Ethiopian section of the AddisDjibouti railway, extending
the maturity by 20 years.
Kenya: The DSSI records
that Chinese lending to Kenya makes up 27 percent (US$ 7.5 billion) of Kenya’s
outstanding debt of US$ 27.8 billion. Large projects financed by Chinese
lenders in Kenya include US$ 867 million for a number of geothermal wells at Olkaria,
US$ 229 million for the Karimenu Dam Water Supply Project, US$ 156 million for
the Nairobi southern bypass highway, and US$ 5.1 billion for two phases of the
controversial Standard Gauge Railway between Mombasa and Malaba. In Kenya,
commercial interest rates for some large loans, including part of the Standard
Gauge Railway, mean that in 2020, debt service on Chinese loans was scheduled
to take up 38 percent of all debt service.
Zambia: Zambia has been
hovering on the precipice of debt distress for several years, even though a
joint IMF/World Bank debt sustainability analysis (DSA) published in August
2019 noted that at that point, Zambia was servicing its debt and had “remained
current on all its debt obligations.” According to the DSSI, as of 2018,
PPG debt to China was less (US$ 2.8 billion) than to bondholders (US$ 3
billion), and Chinese lenders held about 26 percent of PPG debt in Zambia. The
DSA recorded that Zambia’s public electricity utility ZESCO held an additional
US$ 700 million in non‑guaranteed debt, likely to be entirely from Chinese
lenders.
These
figures are quite a bit smaller than our CARI data on signed Chinese loan
commitments (US$ 9.7 billion). Aside from repayments, the most obvious reason
for this difference is the distinction between loan commitments and
disbursements, as we noted above with the case of Nigeria. The DSA noted that
Zambia had contracted US$ 9.7 billion of PPG loans that were expected to be
disbursed between 2019 and 2024. A large portion of this is likely to be
Chinese. For example, ZESCO signed off on US$ 5.6 billion in Chinese loan
commitments for power projects just between 2016 and 2018. Because of Zambia’s
debt problems and inability to contribute its side of some project costs
(compensation for land acquisition, for example), disbursement on existing
projects has been halting.
DISCUSSION
IN
2020, HOW DIFFERENT IS AFRICA’S debt crisis compared with the crisis that began
in the late 1970s? The earlier crisis had its origins in global and domestic
policy factors, and the balance between these differed country by country.
There is no dispute about the two most important global shocks: the dramatic
rise in oil prices caused by the Yom Kippur War (1974) and the IranIraq War
(1980) which hurt all oil importers, and the dramatic rise in global interest
rates caused by the July 1981 decision of the US Federal Reserve to raise US
interest rates to 22.36 percent. Developing countries that had borrowed at
variable rates found their debts far more difficult to service, and capital
fled from the south to the north.
When
countries went into arrears, they faced steep penalties from Paris Club
creditors. Domestic policy factors such as overvalued exchange rates, loss‑making
public companies, and expensive subsidies also played a role. In the late
1980s, the members of the Paris Club recognized the fact that dozens of low‑income
countries were essentially bankrupt. In low income sub‑Saharan Africa, the
external debt to GNI ratio rose from 49 percent to 104 percent between 1980 and
1987, considerably above today’s pre‑COVID‑19 levels. This recognition of
insolvency slowly led to programs of bilateral and multilateral debt writeoffs,
culminating in the HIPC initiative in 1996.
In
2020, after decades of reform, most African countries are in far better
economic shape than they were in the 1980s, although there are exceptions such
as Zimbabwe. Global interest rates are at record low levels. In some countries
such as South Sudan, the Central African Republic, Burundi, and Somalia, armed
conflict and civil war hamper economic output and debt sustainability. Pre‑COVID
19, the collapse of commodity prices was the chief factor driving debt distress
in Angola, Chad, the Republic of Congo, Mauritania, and Sudan. This also played
a role in Ghana and Zambia, but expansive spending around elections was an
additional factor. Still, for most of the 21 countries in this analysis, debt
problems are likely to be liquidity problems rather than reflections of
insolvency.
Chinese
lenders are now a significant part of the debt picture in Africa, but their
role should not be overestimated. In over half of the low‑income countries most
at risk of, or already in, debt distress, Chinese lending is relatively small,
with less than 15 percent of debt stock. That is to say, their debt problems
are largely caused by lenders other than China.
In
another 3 countries, borrowing from China appears to be between 15 and 25
percent of debt stock: Central African Republic (CAR), Mozambique, and
Zimbabwe. Yet in CAR, some of the DSSI debt data is from Taiwan, which
complicates analysis.
In
just seven African countries, as of 2018, China contributed between 26 and 58
percent of PPG debt stock. These are the countries where Chinese lending is
central to the African debt distress picture.
Debt
service is another important variable. In two countries, Angola and Djibouti,
more than 50 percent of debt service was owed to Chinese lenders in 2020. Yet
even here, looking at averages creates a somewhat misleading picture. With
Angola included, 29 percent of all debt service in the DSSI countries in Africa
is due to China. With Angola excluded, only 18 percent of debt service is
Chinese.
The
World Bank’s DSSI data is a gold mine of information for Chinese lending in low
income countries and Angola. Our CARI database gives the details on each loan
commitment made by Chinese lenders across Africa, including countries that are
not part of the DSSI. These two resources should allow analysts to make a great
leap forward in understanding the myths and realities of Chinese lending in
risky markets in Africa.
Norb Leahy, Dunwoody GA Tea Party Leader
No comments:
Post a Comment