21 OCTOBER 2019
China has not been engaged in debt trap diplomacy — at least not yet
China has not been the primary driver behind rising debt risks in the Pacific, although a continuation of business as usual would risk future debt problems in several countries.There is scope for a new Australian infrastructure financing facility to provide loans to the Pacific without causing debt problems, particularly as it has adopted key sustainable lending rules.Pacific nations have an opportunity to obtain more favourable financing from official development partners but care must be taken to avoid overly geopolitical aid.
China’s Belt and Road Initiative has raised important questions about the risk of debt problems in less-developed countries. The risks are especially acute for the small and fragile economies of the Pacific. Our analysis, however, finds a nuanced picture. The evidence to date suggests China has not been engaged in deliberate ‘debt trap’ diplomacy in the Pacific. Nonetheless, the sheer scale of China’s lending and its lack of strong institutional mechanisms to protect the debt sustainability of borrowing countries poses clear risks. Chinese lending is more intense as a share of GDP in smaller economies. If China wants to remain a major development financier in the Pacific without fulfilling the debt trap accusations of its critics, it will need to substantially restructure its approach, including by adopting formal lending rules similar to those of the multilateral development banks.
By contrast, there is scope for Australia’s more modest infrastructure lending plans to be sustainable. If Australia wants to do more in the Pacific though, it should reverse the current stagnation in its overall aid budget. Pacific nations, meanwhile, have an opportunity to push for more favourable financing from external development partners. Care must be taken, however, to avoid overly geopolitical aid that prioritises short-term wins over the need for domestic reform and good governance.
The Pacific has seen a surge in interest from major powers looking to lend more money to the region. With this, the Pacific has once again become an arena for geostrategic competition among much larger players. China has emerged as a major new financier, extending attention-grabbing loans to Pacific governments now officially brought under its sprawling Belt and Road Initiative (BRI). This has raised concerns about debt sustainability and accusations that China is pursuing ‘debt trap’ diplomacy in the region. It has also prompted the Australian Government to respond with its own new debt-financing initiatives as part of its broader Pacific ‘step-up’.
Debt can play a useful role in financing development if there is due attention to ensuring debt sustainability. Nonetheless, the scale, nature, and opacity of China’s lending activities under the BRI raise important questions about potential debt sustainability problems in many less- developed countries. Pacific countries are prima facie among those most at risk, given their small size and structural vulnerabilities. Several Pacific states are also some of the most heavily indebted countries to China anywhere in the world. The Pacific is therefore a crucial part of the global story surrounding the debt sustainability implications of the BRI.
In the vortex of geopolitics and the rush from larger players to win influence in the region, objective economic analysis has been missing from much of the policy discourse about China’s lending activities in the Pacific. Some analysts are dismissive of the China debt trap narrative, concluding that such concerns are “without foundation”. Others continue to warn of predatory lending practices. This Analysis therefore seeks to provide a more systematic investigation of the available evidence. We find the picture is more nuanced than either camp presents. The evidence suggests China has not been engaged in such problematic debt practices in the Pacific as to justify accusations of debt trap diplomacy, at least not to date. Still, the sheer scale of Chinese lending and the lack of strong institutional mechanisms to protect the debt sustainability of borrowing countries mean a continuation of business as usual would pose clear risks. China will need to substantially restructure its approach if it wants to remain a major player in the Pacific without fulfilling the debt trap accusations of its critics.
There have been some recent signs of greater caution on the part of both China and Pacific Island governments. At the Second Belt and Road Forum held in Beijing in April 2019, China’s President Xi Jinping emphasised the need to ensure debt sustainability in future BRI projects. Pacific leaders have also become more cautious about taking on additional Chinese debt. Six Pacific governments are currently debtors to China — Cook Islands, Fiji, Papua New Guinea, Samoa, Tonga, and Vanuatu — although only Papua New Guinea and Vanuatu have taken on new Chinese loans since 2016.
Other signs, however, suggest that the issue of China’s impact on debt sustainability in the Pacific will only grow in importance. First, several very large loan-financed projects are officially in the pipeline in Papua New Guinea and Vanuatu. Second, all six Pacific governments currently indebted to China officially signed up to the BRI in late 2018, joining around 130 other countries China lists as part of the initiative. This suggests these governments remain interested in further financing from China. Chinese lending may also expand to more countries in the region as Pacific governments look to maximise the amount of external financing available to them. Most recently, Solomon Islands and Kiribati have both announced a switch in diplomatic relations from Taiwan to China.
Australia is also looking to become an important lender in the Pacific. While Australia has long been the dominant aid provider to the region, its development financing had been provided only in the form of grants rather than loans. In November 2018, the Australian Government launched its Pacific ‘step-up’. This included a new $2 billion Australian Infrastructure Financing Facility for the Pacific (AIFFP) — comprising $1.5 billion in loans and $0.5 billion in grants — as well as another $1 billion in callable capital for Export Finance Australia (EFA) and an expanded remit for EFA to finance overseas infrastructure projects. These initiatives are in the early stages of operation. Still, there are concerns that in seeking to compete directly with loans from China, Australia might simply exacerbate existing debt sustainability problems in the Pacific.
This Analysis reviews the evidence surrounding China’s debt practices in the Pacific and the extent to which they have contributed to rising debt sustainability risks in the region. Our analysis draws on the macroeconomic surveillance work of the International Monetary Fund (IMF) and combines this with newly updated data from the Lowy Institute Pacific Aid Map — a unique dataset tracking official financing flows to the region, including from China. We focus on benchmarking China’s practices against other official financiers operating in the Pacific, notably the World Bank and Asian Development Bank (ADB) as the standard-bearers for international good practice. We then conduct a simple quantitative exercise to examine the risk of future debt sustainability problems related to Chinese lending, as well as the Australian Government’s plans to become a major infrastructure lender to the region. Finally, we draw out important policy implications that flow from this analysis.
DEBT SUSTAINABILITY CONCERNS SURROUNDING THE BRI
Globally, the vast majority of Chinese official development finance comes in the form of loans rather than grants, with only a minority of those loans being concessional. The core source of Chinese development financing has been its state-owned policy banks — namely, the Export-Import Bank of China (EXIM Bank) and the China Development Bank. Increasingly, China’s state-owned commercial banks are also becoming major overseas lenders under the BRI banner. Combined with the large volume of planned Chinese lending under the BRI, this has given rise to concerns that the initiative will generate debt sustainability problems in developing countries around the world. It has also led to accusations of so-called ‘debt trap’ diplomacy, which contend that China actively seeks to push countries into debt problems in order to extract geopolitical concessions. Many geostrategic analysts are alarmed by the example of the Hambantota Port in Sri Lanka, in which a state-owned Chinese firm gained a majority equity stake in the strategically located port after the country ran into debt-related difficulties. More recently, there is a concern that something similar could happen in the Pacific.
A lack of transparency and official information about the BRI severely limits objective economic analysis of the situation and raises questions about the planned scale, and often commercial terms, of Chinese lending. For example, in 2018 the IMF identified eight low-income countries in Africa experiencing debt difficulties in which total external debt-to-GDP had risen by over 20 percentage points, with more than half of the increase reflecting bilateral loans from China. One recent academic study suggests that half of China’s overseas lending is “hidden” and not captured by official global debt statistics or private credit rating agencies. Looking to the future, an important 2018 study by the Center for Global Development (CGD) found eight countries at particular risk of debt problems based on the pipeline of debt-financed projects planned under the BRI. The World Bank conducted a similar exercise in 2019 and found 12 countries likely to experience increased debt vulnerability as a result of the BRI over the medium term. Notably, no Pacific countries were included in either the CGD or World Bank study. For that reason, we later present a similar forward-looking exercise for the Pacific.
DEBT SUSTAINABILITY RISKS IN THE PACIFIC
China has cultivated its attractiveness as a development partner for Pacific governments by operating in sharp contrast to traditional development partners. Chinese assistance is perceived to be faster, more responsive to the needs of local political elites, and have fewer conditions attached. As one senior Pacific bureaucrat put it: “We like China because they bring the red flags, not the red tape.” Chinese aid activity in the Pacific region has been predominantly in infrastructure development, fuelled largely by loans rather than grants.
Pacific countries are among the most vulnerable countries in the world to potential debt sustainability problems and by extension possible ‘debt trap’ diplomacy. The countries most at risk of debt problems from the BRI are smaller economies with weaker institutions, even if the majority of Chinese financing goes to larger countries with a better ability to absorb such debt. On the basis of available evidence, China’s overseas lending as a share of GDP appears to be much larger in smaller countries (Figure 1, Panel A). Three small Pacific economies — Tonga, Samoa, and Vanuatu — also appear to be among those most heavily indebted to China anywhere in the world (Figure 1, Panel Bl). The Pacific is therefore a crucial part of the global debt sustainability questions surrounding the BRI.
Source: Government documents and IMF for Pacific countries; for all other countries, Eugenio Cerutti and Haonan Zhou, “The Chinese Banking System: Much More than a Domestic Giant”, VoxEU.org, 9 February 2018, https://voxeu.org/article/chinese-banking-system
Debt can play a useful role in financing development if there is due regard for debt sustainability. However, Pacific countries face a host of structural challenges that substantially heighten their vulnerability to potential debt problems. Much of this derives from the region’s unfavourable economic geography including: remoteness from major international economic centres; internal dispersion within countries (both across rural areas and between a country’s different islands); incredibly small size by most key measures (such as population, land, and GDP); dependence on a narrow set of uncertain income sources (notably commodities, aid, remittances, and tourism); and high vulnerability to major natural disasters and the effects of climate change.
Nine Pacific countries are among the smallest 25 countries in the world. Tonga, Samoa, and Vanuatu — mid-sized Pacific countries — have populations of only 100 000–300 000 people. Papua New Guinea is the exception, with a population of almost nine million people. In terms of remoteness, the average Pacific country weighted by economic size lies 11 500 kilometres away from the rest of the world, a distance 40 per cent greater than that for the Caribbean islands. Regarding natural disasters, Vanuatu tops the list as the country facing the highest risks in the world. Tonga is considered second. Three other Pacific countries are in the top ten, including Papua New Guinea.
Difficult economic geography in turn drives enormous development financing needs, creating a predictable pressure towards potentially unsustainable fiscal policies and debt accumulation as governments seek to satisfy the needs and demands of local populations. The Pacific is, by some margin, the most aid-dependent region in the world. In 2017, the Pacific received aid equal to 5.2 per cent of its gross national income. By comparison, in Sub-Saharan Africa it was 3 per cent. Even with this aid, the Pacific still faces one of the largest estimated financing gaps in the world. For example, the ADB estimates an infrastructure financing gap of 6.2 per cent of GDP every year, the highest such gap of any sub-region within the Asia-Pacific.
Structurally weak and volatile growth further reduces the ability of Pacific countries to sustainably absorb large amounts of debt. Reflecting their difficult economic geography as well as weak institutions, Pacific countries have struggled to sustain even a modest pace of economic growth over the long term. As Figure 2 shows, economic growth in the Pacific is more volatile than it is fast. Growth accelerations tend to be short-lived and often driven by the transitory stimulus effects of large-scale infrastructure construction. This means many investments struggle to generate an adequate return to justify the cost — even for investments in economic infrastructure that are usually considered to be growth-enhancing, such as roads, ports, and power generation. It also means Pacific countries have limited scope to grow their way out of any debt overhang. Meanwhile, frequent negative shocks risk short-term debt-servicing problems and, in the case of larger shocks, can fundamentally alter debt sustainability prospects.
Debt sustainability risks have been rising in the Pacific over the past decade. Figure 3, Panel A shows the distribution of IMF debt distress risk ratings across Pacific countries and how this has evolved over time. While no Pacific country is currently considered by the IMF to be in debt distress, the risks have become notably worse over time. By 2017, no Pacific country was considered at low risk of debt distress. Among the six countries with IMF ratings available since 2011, four saw their IMF ratings worsen: Papua New Guinea, Vanuatu, Samoa, and Tonga. China is an important creditor to all four. For these countries, average debt-to-GDP has risen by 17 percentage points while it has declined on average for other Pacific economies (Figure 3, Panel B).
Notes: 1. For Fiji, Palau, and Nauru, the IMF uses a ratings system for higher-income countries where it simply rates their debt outlook as “sustainable” or “unsustainable”; 2. Debt-to-GDP ratios are calculated as the simple average of debt ratios for individual Pacific countries in each grouping
Source: Authors’ calculations; IMF; World Bank
A review of IMF debt sustainability analyses for these four Pacific countries reveals multiple drivers behind rising debt risks. Primary among these is the impact of large economic shocks (especially major natural disasters) combined with the policy response of Pacific governments that have prioritised economic recovery over fiscal prudence. Recent tropical cyclones have, for example, caused estimated damages of over 60 per cent of GDP in Vanuatu and almost 40 per cent and 30 per cent of GDP in Tonga and Samoa, respectively. Economic mismanagement has also been important, particularly in Papua New Guinea where large budget deficits saw a sharp increase in public debt even before a collapse in key commodity prices derailed its economy. Methodological changes by the IMF to better incorporate the impact of frequent natural disasters, as well as other technical factors, have also been important reasons behind recent changes in IMF debt ratings in the Pacific. Overall, it is clear that rising debt risks in the region, as assessed by the IMF, have been driven by a multitude of factors, rather than by Chinese lending alone.
THE SUPPLY OF OFFICIAL FINANCING IN THE PACIFIC
Pacific governments clearly hold the primary decision-making power over how much debt they wish to incur, including in the aftermath of natural disasters and other negative external shocks. Nonetheless, the availability of loans from different partners is a key conditioning factor. Pacific governments have agency, but a willing financier is also a necessary requirement. This section therefore considers the role of different actors in providing official financing to the Pacific.
We first trace the broad picture of official financing flows to the region and its composition, before focusing on the role of China compared to other financiers. According to Pacific Aid Map data, in 2011–2017 the region received US$2.4 billion in official financing flows on average each year, of which about 81 per cent was in the form of grant assistance, 15 per cent in loans, and the remainder in other flows. In total, official financing flows were equal to 7.5 per cent of the region’s GDP in 2017. However, this figure understates the importance of such flows to many individual Pacific economies since it is skewed by Papua New Guinea and Fiji, which have much larger economies and are far less aid dependent than the rest of the region. If each country is equally weighted, the average Pacific country received official financing equal to about 24 per cent of its GDP in 2017.
Official financing flows to the Pacific have, however, been declining in importance and becoming less concessional. Figure 4, Panel A shows the total amount of official financing disbursed to the region from 2011 to 2017. Although essentially flat in nominal terms, official financing flows fell by 1.7 percentage points relative to regional GDP (Figure 4, Panel B). Importantly, the decline was entirely driven by a sharp reduction in grant financing from traditional bilateral donors — notably from Australia, reflecting a combination of stagnant grant assistance in Australian dollars and unfavourable exchange rate movements. Overall, there has been a substantial fall in grants relative to regional GDP — dropping from 8.2 per cent in 2011 to 5.9 per cent in 2017 (Figure 4, Panel B). Meanwhile, debt financing has risen from 7 per cent of total official financing flows in 2011 to 16 per cent in 2017. Official financing flows to the Pacific have therefore not only been declining in importance but shifting in a less concessional direction.
It is within this environment that China has emerged as a major lender in the Pacific. In 2011–2018, China made official loan commitments totalling US$6 billion (about 21 per cent of regional GDP). By dollar value, the majority of this has flowed to Papua New Guinea as the region’s largest economy and includes a US$4.1 billion roads project announced in 2017, which is yet to be drawn on but remains officially in the pipeline. Excluding this particular loan and focusing on the period 2011–2017 for which we have more comparable data across donors, China committed to loan projects with a total value of about US$1.7 billion. However, we estimate that only around US$942 million had been disbursed by the end of 2017.
Yet, while China has become a major new lender in the Pacific, it has not been the dominant source of new loans in the region. Figure 5 shows the key creditors in the region and the countries to which they have extended loans. China was responsible for 37 per cent of all official sector loans disbursed to the region in 2011–2017. The ADB provided a slightly higher share, at 41 per cent. However, combined with the smaller lending activities of the World Bank, the two multilateral development banks (MDBs) together accounted for 53 per cent of all loan disbursements to the Pacific and have therefore continued to serve as the leading source of new official loans in the Pacific. Japan also contributed 7 per cent of all lending. Taken together with minor contributions from the IMF and Canada, this means traditional creditors provided over 60 per cent of all official lending to the Pacific from 2011 to 2017.
Looking at China’s role as a creditor to individual Pacific countries presents a similar picture (Figure 6). China is the single largest creditor in Tonga, Samoa, and Vanuatu. However, it is only in Tonga that China accounts for more than half of total outstanding debt. Elsewhere, either traditional official lenders or domestic creditors dominate. Meanwhile, China is not an active lender to the rest of the region, which remains dominated by traditional creditors, notably the MDBs. With the important exception of Tonga, China is currently not a dominant creditor in the Pacific.
Note: Most recent available figures
Source: Authors’ calculations; government budget documents; IMF
FURTHER SCRUTINY OF CHINESE LENDING IN THE PACIFIC
The analysis so far suggests that China has not been the primary factor behind rising debt risks in the Pacific. Deteriorating IMF debt ratings reflect a confluence of factors, and official sector lending to the Pacific continues to be led by the two MDBs and traditional creditors more generally, both in terms of the existing stock of debt as well as new loan flows. This provides a strong initial argument against accusations of debt trap diplomacy. However, it is not necessarily sufficient. One might still consider China’s lending practices to be ‘predatory’ or otherwise problematic if its projects are of especially poor quality, if its loan terms are particularly expensive, or if it has been engaged in lending to countries already at a heightened risk of debt problems. We now consider each of these issues.
Quality of projects
The Pacific is littered with infrastructure projects that have failed for a variety of reasons, not least due to the inherently limited economic viability of many investments given the region’s lack of economic dynamism. Chinese infrastructure projects in the Pacific have been sharply criticised, with some going as far as to accuse many of these projects of being “white elephants” and “roads to nowhere”. Tied financing, little due diligence, outsized projects, weak project oversight, and fraudulent and corrupt practices are among the many criticisms that have been directed at Chinese projects.
While there has been a surfeit of commentary about Chinese projects, there has been much less analysis of the actual impact and quality of the projects. A 2014 study provides the best analysis of the effectiveness of Chinese aid, and the determinants of that effectiveness, drawn from case studies in four countries — Cook Islands, Samoa, Tonga, and Vanuatu. The study presented a mixed report card on the quality of Chinese projects, with some performing much better than others. The key determinants of effectiveness were the approaches taken by Pacific Island governments to negotiations, and overall oversight of the projects. The evidence suggests that, if left alone, Chinese state firms will cut corners and inflate prices. If managed properly, they can deliver good quality infrastructure.
The mixed quality of Chinese projects notwithstanding, there is a more fundamental question about whether infrastructure investment in many Pacific countries is likely to generate faster economic growth, given the region’s difficult economic geography. The World Bank has noted that in the Pacific “the value of infrastructure is often the direct improvement of their populations’ livelihoods without necessarily being able to generate growth dividends that could be used to service debt”. Viewed in this light, the mixed quality of Chinese-financed projects is arguably of less concern from a pure debt sustainability perspective, even if it remains crucial from a broader welfare perspective (for example, due to financial waste or lower-quality projects crowding out better ones). If little growth dividend is to be expected, then the impact of debt-financed infrastructure projects becomes more analogous to general deficit financing or government consumption expenditure, for which the principal concerns are the scale of borrowing and whether it is sufficiently concessional.
According to the Pacific Aid Map data, the vast majority (97 per cent) of China’s official loans in the Pacific has been in the form of concessional loans from its EXIM Bank. This is a markedly different situation to the global picture, where lending on market terms by China Development Bank and commercial Chinese banks also play a sizeable role. Standard EXIM concessional loans are denominated in renminbi, with an interest rate of 2 per cent, a 5–7 year grace period, and a 15–20-year maturity. All of the largest loans to Samoa, Tonga, and Vanuatu have carried these standard terms, including a 20-year maturity.
One immediate implication is that the macro-debt dynamics implied by China’s lending terms are quite favourable for the Pacific. With nominal GDP growth across the region generally well above the 2 per cent interest being charged, even slow-growing Pacific economies can potentially grow their way out of Chinese debt. China’s bilateral lending terms are also vastly more favourable than those available from the market. Among Pacific countries, only Fiji and Papua New Guinea have meaningful access to market-based financing. Long-term government domestic borrowing costs are currently around 6 per cent in Fiji and 11 per cent in Papua New Guinea. PNG’s maiden issue of a US$500 million bond is also instructive in terms of access to external foreign currency debt, carrying a ten-year maturity and interest rate of 8.4 per cent.
Nonetheless, exposure to natural disasters and other vulnerabilities lower the ability of Pacific economies to carry large amounts of debt, particularly external debt in foreign currencies. This raises an important question as to whether China’s loans are concessional enough by international standards. To create a consistent basis of comparison across different loan term components, the OECD measures overall concessionality by calculating the effective ‘grant element’ of a loan. According to the OECD, a loan is concessional if it contains a grant element equal to 25 per cent or more of the loan’s face value using a discount rate of 10 per cent. Following the OECD’s methodology, we calculate that a 15–20-year concessional EXIM loan would carry a grant element of 48–56 per cent — concessional enough by international standards to qualify as aid.
How does China compare with other official creditors? Concessional loan terms on offer from the MDBs tend to carry an interest rate in the range of 1–2 per cent, a grace period of five–ten years, and loan maturity of 25–40 years — resulting in a grant element in the range of 60–70 per cent, significantly larger than that calculated for EXIM loans. Japan’s terms are even more concessional. The Japanese-financed Port Vila port redevelopment project, for example, carried a 0.5 per cent interest rate, ten-year grace period, and maturity of 40 years — implying a grant element of around 80 per cent. China’s loans are therefore much less concessional than those available from other official lenders. China’s loan terms, however, compare more favourably with the non-concessional MDB lending windows available to more creditworthy countries — which in the Pacific include Fiji and Papua New Guinea — and allow borrowing at a much greater scale. MDB non-concessional financing terms vary with prevailing global interest rates and specific projects. With global interest rates at historically low levels, a 20-year MDB non-concessional loan with a five-year grace period would carry a grant element using the OECD method of 50 per cent, compared with 56 per cent for a 20-year concessional Chinese EXIM loan.
While most Chinese bilateral lending in the Pacific appears to be concessional, it is not necessarily appropriate for all circumstances. For countries where debt sustainability is already a concern, there is a recognised need for greater concessionality. In these situations, the World Bank and IMF apply a more stringent concessionality standard and will at times impose limits on any non-concessional borrowing by countries receiving their financial assistance — both to protect a country’s debt sustainability and to prevent other creditors, private and official, from free-riding on highly concessional World Bank and IMF financing. Eight Pacific countries are currently subject to strict World Bank limits on non-concessional borrowing, including Samoa, Tonga, and Vanuatu. Using these stricter standards, China’s standard EXIM loans contain a grant element of 23–28 per cent — below the minimum required threshold of 35 per cent.
Lending to countries already at risk
The most important question about China’s debt practices in the Pacific is the extent to which it lends to countries that are already at a high risk of debt distress. To analyse this, we matched all loan-financed projects in the Pacific Aid Map with the risk of debt distress according to the IMF at the time each loan agreement was signed. This allowed us to examine the broad patterns of lending across the Pacific’s major official creditors.
Overall, the majority of Chinese loans appear to have been broadly sustainable for the borrowing country at the time they were made (Figure 7). Like other major creditors, the largest share of Chinese lending, at nearly half the total, has been directed to countries facing a low risk of debt distress. Relative to others, China has lent a lower share to those facing a moderate risk of debt distress while it has lent relatively more to higher-income Pacific countries judged by the IMF to be facing a sustainable debt situation (specifically Fiji). In total, 90 per cent of China’s bilateral loans have gone to countries that the IMF debt ratings indicate could sustainably absorb such debt.
Note: 1. For Fiji, Palau, and Nauru, the IMF uses a ratings system for higher-income countries where it simply rates their debt outlook as “sustainable” or “unsustainable”
Source: Authors’ calculations; Lowy Institute Pacific Aid Map; IMF; World Bank
The remaining 10 per cent of bilateral loans made to countries at high risk of debt distress signals a potential problem with China’s overseas lending practices. China has been the dominant source of loans to countries already judged by the IMF to be at high risk of debt distress, accounting for almost three-quarters of such loans. The two MDBs by contrast are governed by formal rules requiring them to provide more concessional financing to countries at higher risk of debt distress, including by switching entirely to grant financing in the case of countries considered to be at high risk.
However, China is not the biggest outlier in this regard. For instance, 70 per cent of Taiwan’s bilateral loans were provided to countries already judged to be at a high risk of debt distress. While Taiwanese lending is much smaller in scale than China’s, it has nonetheless been large relative to the size of recipient economies, with loans to Kiribati equal to about 10 per cent of GDP. The ADB experience also highlights the difficulty of avoiding potentially problematic loans. For example, earlier this decade the ADB provided sizeable loans to Kiribati and Marshall Islands (each amounting to 7–8 per cent of GDP) prior to the establishment of an IMF rating for these countries. Shortly after, both were judged by the IMF to be at high risk of debt distress — underscoring the difficulty of judging sustainability in the absence of dedicated forward-looking exercises such as that provided by the IMF.
Overall, China’s relatively substantial lending to countries already considered to be at high risk of debt distress can therefore be viewed as the product of the considerable scale of Chinese lending (relative to Taiwan’s), combined with the lack of strong institutional mechanisms to prevent potentially unsustainable lending (compared to the MDBs).
DEBT DIPLOMACY AND FUTURE DEBT RISKS
Unlike elsewhere in the world, most Chinese financing in the Pacific appears to be concessional enough to qualify as aid. The majority of these Chinese loans have also flowed to countries with greater scope to absorb such debt. Nevertheless, the combination of the sheer scale of China’s lending, inadequate controls to avoid loans to countries already at high risk of debt distress, and financing terms that are insufficiently concessional for lending to such countries (including without free-riding on the MDBs) presents an important risk to future debt sustainability in the Pacific.
To assess this risk, we have conducted a similar analytical exercise to that of the BRI debt studies by the CGD and World Bank noted earlier. As well as examining the potential implications of future Chinese lending, we also incorporate Australia’s new bilateral infrastructure lending plans to assess its potential macro-debt implications.
Methodology and assumptions
Following the methodology of the CGD and World Bank studies, we project the future debt-to-GDP ratio for 13 Pacific countries and compare it to a threshold of 50 per cent of GDP as a simple warning indicator of the potential for future debt problems. For each country, we construct a scenario for future Chinese lending projected out to 2024. We then combine these with estimates of future debt-to-GDP as projected by the IMF to reflect existing government policies and other debt-financed projects already in the pipeline. We make no adjustments to economic growth given the doubts raised previously about the ability of infrastructure investment to sustainably boost economic growth in the Pacific.
There is no single point beyond which debt systematically becomes a problem. However, using a 50 per cent of GDP warning threshold provides a simple and transparent assessment that allows our results to be broadly comparable with the CGD and World Bank studies. At the same time, a 50 per cent of GDP warning threshold is generally consistent with the results of the more complex assessments done by the IMF. Finally, the IMF ratings themselves are ill-suited as an alternative single-warning indicator for several technical reasons.
For the six Pacific countries that already borrow from China, we construct a business-as-usual scenario to project out future debt to 2024. This is done on the assumption that China is principally interested in maintaining its current level of lending, and therefore influence, among the six countries that it currently lends to. However, we take a slightly different approach for Papua New Guinea, where our projection is based on the implementation of projects already officially in the pipeline which in aggregate dwarf the size of China’s previous loans.
We also examine whether loans from China could create debt problems in other Pacific countries that have not yet borrowed bilaterally from China but may do so in the future, particularly if they were to switch diplomatic recognition from Taiwan to China. To construct a scenario, we assume that China is interested in gaining a similar amount of influence to that which it has in the first group of six Pacific countries to which it already lends. For our scenario we therefore assume China provides total bilateral loans equal to 11.5 per cent of GDP in each of these countries (the five-year pro rata average for the group of existing borrowers from China).
Finally, we incorporate potential new official loan financing from Australia. This includes $1.5 billion in loans under the AIFFP as well as a $1 billion increase in callable capital for EFA. For our scenario, we focus on the AIFFP, since the amount of potential EFA lending to the Pacific is unclear and is not subject to any formal quantitative target or allocation. Moreover, given EFA operates on a commercial basis, it will likely be attracted to directing the use of its increased capital base to more promising infrastructure projects in much larger and faster-growing emerging markets in Asia, rather than small and slow-growing Pacific markets. We therefore assume that EFA will only play a niche financing role in the Pacific.
As the AIFFP is in its early stages of operation, our analysis proceeds on the following assumptions. First, that AIFFP financing would be provided over a decade, meaning that half of it would be disbursed over our period of analysis to 2024. Second, that Australia would not lend to any country already judged to be at high risk of debt distress or where the result would be to breach our 50 per cent debt-to-GDP warning threshold, and that the loans would be distributed equally relative to the size of each economy. Finally, although the AIFFP will also lend to the private sector, we assume that the entire loan amount under the AIFFP would be used for sovereign lending (that is, to governments, government-controlled entities, or where otherwise guaranteed by the government). This incorporates the possibility of some limited sovereign Pacific lending by EFA.
The overall results of the analytical exercise are shown in Figure 8, and confirm that there are significant risks of future debt sustainability problems under a business-as-usual scenario for bilateral Chinese lending. Four of the six countries that currently borrow from China — Vanuatu, Samoa, Tonga, and Fiji — are already effectively at our 50 per cent warning threshold and, with the exception of Fiji, would be pushed well beyond this under our business-as-usual scenario. Vanuatu stands out as a particular concern, given in late 2018 it signed up for another large loan-financed Chinese roads project. Papua New Guinea would also see a dramatic increase in debt, beyond our warning threshold. This primarily reflects the 2017 agreement for a US$4.1 billion roads project that has yet to be drawn on, and highlights the significant debt implications should this specific project go ahead at its present scale.
Notes: 1. Based on IMF World Economic Outlook and individual country Article IV Staff Reports; 2. Based on BAU bilateral Chinese lending/borrowing for existing debtors and assumed similar average rates for potential new borrowers; 3. Based on A$1.5 billion lending target over ten years to countries with sufficient debt space
Source: Authors’ calculations; Lowy Institute Pacific Aid Map; IMF; ADB
For potential new borrowers from China, there appears to be more scope for Chinese lending. In our scenario analysis, only Nauru and Kiribati would exceed the debt-warning threshold. In Nauru, this reflects its already high debt level. In Kiribati, which recently switched diplomatic ties to China, debt is at a moderate level. The IMF, however, projects this will rise dramatically in order to finance substantial climate change adaptation requirements. The key question therefore is whether Chinese loans would be used for this purpose or additional to this projected debt. Other countries, including Solomon Islands, would appear to have more scope to sustainably absorb sizeable amounts of additional borrowing given relatively low existing and projected debt levels.
The results for new bilateral Australian loans show that there is scope to sustainably deliver the targeted funds under the AIFFP. This is because the AIFFP loans would be relatively small, with the full $1.5 billion only equal to 3.6 per cent of regional GDP today (and potentially spread over a decade or so). Our conclusion, however, hinges on two factors. First, we have assumed that EFA lending would be relatively small. If this proves not to be the case, then the impact on Pacific debt sustainability would be more significant. Second, the question of whether Australian loans to Papua New Guinea can be sustainable is uncertain. Much will depend on whether bilateral Australian loans would displace at least part of the sizeable pipeline of loans from China. There is also considerable uncertainty about the true debt situation in Papua New Guinea, with the IMF figures excluding uncertain amounts of debt at state-owned firms and the government exposed to sizeable unfunded pension liabilities. PNG’s debt situation may therefore be considerably worse than it appears. If sustainable Australian lending to Papua New Guinea proves heavily constrained, it will be virtually impossible to deliver the total amount of planned new loans without fuelling increased debt sustainability risks.
Our findings show that a continuation of business as usual for bilateral Chinese lending in the Pacific would quickly give rise to potential debt sustainability problems. China will therefore need to reconfigure its approach significantly if it wants to disprove the debt trap accusations made by its critics.
China has begun exercising greater caution over the potential debt sustainability implications of the BRI and taken a number of initial steps to address this. China has supported an IMF training centre to help improve the debt management capacity of countries involved in the BRI, and China’s Ministry of Finance has agreed with major multilateral financing institutions to establish a new cooperation platform. In 2017, China also committed itself to the G20 Operational Guidelines for Sustainable Financing and in 2019 to the G20 Principles for Quality Infrastructure Investment, both of which contain debt-related provisions including complying with World Bank and IMF policies for countries where debt sustainability is already a concern.
Concrete action from China to operationalise these commitments is now required. China’s Ministry of Finance has issued a new BRI debt sustainability framework (BRI-DSF), modelled on that of the IMF and World Bank, to guide BRI-financing activities in less-developed countries. Yet, the BRI-DSF remains a “non-mandatory policy tool” and provides little guidance as to how Chinese financial institutions are expected to alter their behaviour in response to identified debt sustainability risks.
To remedy this, China should adopt formal lending rules similar to those of the MDBs. These could mandate the use of the BRI-DSF by China’s policy banks, notably EXIM Bank, when undertaking sovereign lending to less-developed countries, and require the provision of more concessional financing to countries at greater risk of debt distress. For example, where countries are deemed to be at low risk, standard concessional EXIM loan terms might be appropriate. For those at moderate risk, more concessional loans could be provided, involving a larger interest subsidy from the Ministry of Finance. China could also employ alternative approaches, such as blending EXIM loans with grants from the Ministry of Commerce (MOFCOM) or replacing EXIM loans with interest-free MOFCOM loans (an existing instrument). For countries at high risk, China would ideally only provide grants via MOFCOM.
Implementing such formal lending rules would offer certain advantages for China. By applying only to China’s policy banks and MOFCOM, it would be relatively straightforward to implement and require only modest additional coordination efforts. This approach would cover the majority of what might be considered Chinese development finance, therefore bringing China into a substantially more analogous position to traditional development financiers. It would also encourage much greater cooperation and coordination between China, the IMF, and other official creditors — with greater information sharing also potentially helping to reduce some of the geopolitical tensions surrounding the BRI. Ideally, the results of the BRI-DSF would be made publicly available to enhance overall transparency.
Adopting formal lending rules would also help to support more sustainable debt management by borrowing countries. As elsewhere in the developing world, China’s state firms often engage directly with political elites when seeking agreement for new projects. In weaker governance environments, this can mean effectively bypassing or otherwise sidelining local finance officials normally responsible for advising on such matters. In a recent example, Vanuatu agreed to another large Chinese EXIM loan in late 2018 on terms that appear to violate the finance ministry’s desire to borrow only on more concessional terms. While final decisions will always rest at the political level, such practices undermine effective debt management by Pacific nations, despite many governments having formal rules designed to protect fiscal sustainability. Clear rules anchoring Chinese lending to a formal assessment of debt sustainability could help address these issues by effectively mandating early engagement with the finance officials of recipient countries on any new lending plans.
Ad hoc debt restructurings (as has been China’s approach globally) are also not a panacea should future debt problems emerge. Uncertainty about future debt restructuring can itself undermine prudent management. The case of Tonga is an example, in which China has twice agreed to defer debt repayments, but in a way that risks creating short-term debt-servicing problems only a few years down the track. To reduce these problems, China should set out a clear policy framework for its approach to debt restructurings, and cooperate closely with other official creditors in its approach. Ideally, China should join the longstanding Paris Clubgroup of official creditors (currently it is an observer) or establish some new arrangement.
The most important conclusion from our scenario projections is that China cannot remain a major lender in the Pacific at the same scale as in the past without fuelling significant debt sustainability risks in most of the countries in which it is currently active. Even the provision of more concessional loans would likely prove problematic if at the same scale as the past. While China has begun to provide more grant financing to the Pacific, it is starting from a low base (Figure 9). If China wants to remain a major financier in the region without fulfilling the debt trap accusations of its critics, it will need to shift dramatically to provide far more grant funding than loans.
Australia, like China, needs to avoid lending to countries that are already facing a high risk of debt distress, and provide more concessional financing instead. The AIFFP appears to have adopted rules along these lines, with the AIFFP Design Document indicating the facility will not lend to countries already assessed by the IMF to be at high risk of debt distress. It will also follow the debt limit policies of the World Bank and IMF in other cases where potentially unsustainable borrowing is a concern. Meanwhile, EFA already adheres to such rules through its commitments to OECD sustainable lending rules for export credit agencies.
Protecting debt sustainability in Pacific countries will also require Australian loans to be as concessional as possible, given elevated debt risks and the often limited economic viability of many infrastructure projects in the Pacific. This reinforces that the EFA’s role in the region should remain relatively niche, given its commercial nature, and that it should direct its focus to Asia.
That would leave the AIFFP to play the primary role in the Pacific. According to the AIFFP Design Document, AIFFP financing terms will include pure loans, grants only, or a mixture of both (effectively blending the two together to create a more concessional loan). For sovereign lending, the terms of the pure AIFFP loans will be benchmarked against the non-concessional loan terms of the MDBs. As discussed above, given low global interest rates at present, this would mean a grant element of around 50 per cent — technically well above the OECD minimum for concessional loans. This, however, means AIFFP loans could be marginally more expensive than equivalent Chinese concessional EXIM loans, depending on the specific loan terms used for individual projects. The exception is where loans are blended with grants, although the AIFFP design indicates this will not be in all cases.
The AIFFP has indicated that its key differentiating quality will be its emphasis on high standards. This means ensuring prudent project selection and design, competitive open procurement, transparency and good governance, and strong environmental and social safeguards. Such commitments are appropriate and necessary for an OECD donor, and are the same high standards to which the MDBs adhere. However, while the MDBs are good at ensuring quality, this typically comes at the expense of speed and responsiveness — creating a key gap that China has been able to fill.
The AIFFP’s intention may be to operate in a more nimble manner than the MDBs while still adhering to high standards; however, its prospects of doing so are uncertain at best, at least in the short to medium term. Its institutional capacity will have to be built up. In the meantime, using loan terms similar to the non-concessional loans already available from the MDBs could hinder the ability of the AIFFP to compete effectively with Chinese financing. This merits consideration of more concessional financing terms, including more blending with grants. More concessional financing terms would also clearly assist in preserving debt sustainability. Even if the AIFFP restricts itself to ‘high-quality’ projects, the limited ability of infrastructure to sustainably catalyse faster growth in the Pacific means highly concessional financing remains critical.
All of this implies that Australia should also rethink the size of its overall aid budget. Following a series of deep cuts earlier this decade, Australia’s total aid budget has stagnated in nominal terms. Today, Australia’s strategic goal of doing more in the Pacific is boxed in by a limited aid budget, the desire to avoid cutting back on other important development priorities (such as health and education, or aid to countries outside the Pacific), and the need to avoid causing debt sustainability problems by relying too heavily on non-concessional lending. If Australia wants to do more, one of these constraints needs to be relaxed. Increasing the overall aid budget would be the most desirable option. Finally, China itself might begin providing substantially more grant financing to the Pacific. In that case, a stagnant aid budget would increasingly place Australia at a geostrategic disadvantage.
For Pacific nations
Pacific governments are clearly in the driver’s seat as to whether their own borrowing policies are sustainable and in using their limited debt space wisely for the best projects on the best terms. The major priority lies in strengthening their own fiscal and infrastructure management institutions, for which development partners can also usefully provide technical support, capacity building, and policy-linked budget support to reinforce reform. Pacific nations also have an opportunity to use the current climate of competition among major powers to their advantage, for example by pushing for more concessionality (including more grants), better project management, and more responsiveness to local priorities.
However, care will also be needed. For example, at the time of writing a newly formed government in Papua New Guinea appears to be looking to both Australia and China to secure sizeable general budget financing support. On the one hand, such support would provide immediate fiscal relief at a domestically critical moment and help to restructure the government’s debt profile away from its current reliance on expensive market-based financing. However, whether this ultimately proves beneficial for PNG will depend on whether it uses this opportunity to push forward with reforms to put its fiscal and economic trajectory on a more sustainable path, as opposed to avoiding difficult reforms and facilitating ongoing fiscal profligacy. Similarly, external players have a responsibility to avoid overly geopolitically-driven financial assistance that risks prioritising short-term wins at the expense of undermining the incentives for reform and better governance that are critical to sustainable development. Amid rising competition for influence, the risk of inadvertently creating long-term ‘governance traps’ instead is an equally if not more concerning risk than the potential for debt traps.
The evidence presented in this Analysis suggests that China has not been engaged in such problematic debt practices in the Pacific as to justify accusations of debt trap diplomacy, at least not so far. Nonetheless, the sheer scale of Chinese lending and the lack of strong institutional mechanisms to protect the debt sustainability of borrowing countries poses clear risks. If China wants to remain a major player in the Pacific, without fulfilling the ‘debt trap’ accusations made by its critics, it will need to substantially restructure its approach. By contrast, there is more scope for Australia’s relatively modest infrastructure lending plans to remain sustainable, particularly as it has adopted lending rules to protect the sustainability of borrowing countries. If Australia wants to do more though, it will need to reverse the current stagnation in its overall aid budget. Pacific nations, meanwhile, have an opportunity to push for more favourable financing from external partners. Care from all parties will, however, be needed to avoid overly geopolitical aid that risks prioritising short-term wins over the need for domestic reform and good governance.
* The data for this Analysis was finalised on 19 September 2019.
 Brahma Chellaney, “China’s Debt-trap Diplomacy”, Project Syndicate, 23 January 2017, https://www.project-syndicate.org/commentary/china-one-belt-one-road-loans-debt-by-brahma-chellaney-2017-01?barrier=accesspaylog.
 Academics at the Australian National University have refuted the China debt trap narrative based largely on the limited share of debt owed to China compared to other creditors across the Pacific and the circumstances, particularly in Tonga, in which large Chinese loans came about: see Rohan Fox and Matthew Dornan, “China in the Pacific: Is China Engaged in ‘Debt-trap Diplomacy’?”, DevPolicy Blog, 8 November 2018, https://devpolicy.org/is-china-engaged-in-debt-trap-diplomacy-20181108/?utm_source=Devpolicy&utm_campaign=22334c1efb-Devpolicy+News+Dec+15+2017_COPY_01&utm_medium=email&utm_term=0_082b498f84-22334c1efb-227683262. However, this does not take into account other elements of China’s lending practices that might be considered particularly problematic, as explained later in this Analysis.
 For example, the US ambassador to Australia recently warned of Chinese “payday loan” diplomacy in the Pacific. Stephen Dziedzic, “China’s ‘Payday Loans’ Attacked by Arthur B Culvahouse Jr, New US Diplomat to Australia”, ABC News, 13 March 2019, https://www.abc.net.au/news/2019-03-13/china-accused-payday-loans-pacific-us-ambassador-australia/10896280.
 For example, in addition to citing a new BRI debt sustainability framework, Xi Jinping said: “We also need to ensure the commercial and fiscal sustainability of all projects so that they will achieve the intended goals as planned.” Xi Jinping, “Working Together to Deliver a Brighter Future for Belt and Road Cooperation”, Keynote speech at the opening ceremony of the second Belt and Road Forum for International Cooperation, Beijing, 26 April 2019, https://eng.yidaiyilu.gov.cn/qwyw/rdxw/88232.htm.
 China State Information Center, Belt and Road Portal, https://eng.yidaiyilu.gov.cn/info/iList.jsp?cat_id=10076&tm_id=80; “Tonga Gets Five Years’ Grace on Chinese Loan as Pacific Nation Joins Belt and Road Initiative”, ABC News, 19 November 2018, https://www.abc.net.au/news/2018-11-19/china-defers-tongas-loan-payments-as-nation-signs-up-to-bri/10509140.
 Natalie Whiting, Christina Zhou and Kai Feng, “What Does It Take for China to Take Taiwan’s Pacific Allies? Apparently, $730 Million”, ABC News, 19 September 2019, https://www.abc.net.au/news/2019-09-20/kiribati-to-switch-diplomatic-ties-from-taiwan-to-china/11532192.
 On 1 July 2019, the Export Finance and Insurance Corporation changed its trading name to Export Finance Australia.
 Department of Foreign Affairs and Trade, “Stepping-up Australia’s Pacific Engagement”, November 2018, https://dfat.gov.au/geo/pacific/engagement/Pages/stepping-up-australias-pacific-engagement.aspx; Marise Payne and Simon Birmingham, “Enhancing Infrastructure in the Pacific”, Joint Media Release, 1 July 2019, https://foreignminister.gov.au/releases/Pages/2019/mp_mr_190701.aspx?w=E6pq%2FUhzOs%2BE7V9FFYi1xQ%3D%3D.
 A lack of consistent data tracking official financing in the Pacific is a major challenge for analysts and policymakers alike. The Lowy Institute Pacific Aid Map tracks official financing flows to the region including more than 20 000 projects in 14 Pacific countries provided by 62 donors to the region, including China, from 2011 up to and including 2017 (data for 2018 and 2019 is incomplete): see Lowy Institute Pacific Aid Map, https://pacificaidmap.lowyinstitute.org/. Unless otherwise stated, all data presented in this Analysis is derived from the Pacific Aid Map.
 Gregory Chin and Kevin Gallagher, “Coordinated Credit Spaces: The Globalization of Chinese Development Finance”, Development and Change 50, Issue 1 (2019), 245–274, https://onlinelibrary.wiley.com/doi/10.1111/dech.12470.
 Chellaney, “China’s Debt-trap Diplomacy”.
 Jonathan Hillman, “Game of Loans: How China Bought Hambantota”, CSIS Briefs, March 2018, https://www.csis.org/analysis/game-loans-how-china-bought-hambantota.
 David Wroe, “China Eyes Vanuatu Military Base in Plan with Global Ramifications”, The Sydney Morning Herald, 9 April 2018, https://www.smh.com.au/politics/federal/china-eyes-vanuatu-military-base-in-plan-with-global-ramifications-20180409-p4z8j9.html.
 IMF, “Macroeconomic Developments and Prospects in Low-Income Developing Countries”, IMF Policy Paper (Washington, DC: IMF, 2018), https://www.imf.org/~/media/Files/Publications/PP/2018/pp021518-macroeconomic-developments-and-prospects-in-low-income-developing-countries.ashx.
 John Hurley, Scott Morris and Gailyn Portelance, “Examining the Debt Implications of the Belt and Road Initiative from a Policy Perspective”, CGD Policy Paper (Washington, DC: Center for Global Development, 2018), https://www.cgdev.org/sites/default/files/examining-debt-implications-belt-and-road-initiative-policy-perspective.pdf.
 Luca Bandiera and Vasileios Tsiropoulos, “A Framework to Assess Debt Sustainability and Fiscal Risks under the Belt and Road Initiative”, Policy Research Working Paper No WPS 8891 (Washington, DC: World Bank Group, 2019), http://documents.worldbank.org/curated/en/723671560782662349/pdf/A-Framework-to-Assess-Debt-Sustainability-and-Fiscal-Risks-under-the-Belt-and-Road-Initiative.pdf.
 Confidential discussion with the authors, Sydney, April 2019.
 Another study presents different figures but also concludes that the same Pacific countries are among the most heavily indebted to China in the world: see Horn, Reinhart and Trebesch, “China’s Overseas Lending”.
 World Bank, Pacific Futures: World Bank Discussion Note, 2011, https://devpolicy.anu.edu.au/pdf/2012/Discussion_Note_Pacific_Futures.pdf.
 ADB, Key Indicators for Asia and the Pacific 2018, 49th edition (Manila: Asian Development Bank, 2018), 76, https://www.adb.org/publications/key-indicators-asia-and-pacific-2018.
 World Population Review, “Papua New Guinea”, http://worldpopulationreview.com/countries/papua-new-guinea-population/.
 John Gibson and Karen Nero, “Are Pacific Island Economies Growth Failures? Geo-Political Assessments and Perspectives”, 2007, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.465.3391&rep=rep1&type=pdf.
 WorldRiskReport 2018 (Berlin: Bündnis Entwicklung Hilft and Ruhr University Bochum — Institute for International Law of Peace and Armed Conflict, 2018), 40, https://reliefweb.int/sites/reliefweb.int/files/resources/WorldRiskReport-2018.pdf.
 ADB, Meeting Asia’s Infrastructure Needs (Manila: Asian Development Bank, 2017), 50, https://www.adb.org/sites/default/files/publication/227496/special-report-infrastructure.pdf. For a broader discussion of various other estimates, see also ADB, Pacific Economic Monitor, December 2018, https://www.adb.org/sites/default/files/publication/471196/pem-december-2018.pdf.
 Ron Duncan, “Sources of Growth Spurts in Pacific Island Economies”, Asia and Pacific Policy Studies 3, Issue 2 (2016), 351–365, https://onlinelibrary.wiley.com/doi/full/10.1002/app5.125.
 Construction costs are also structurally higher in the Pacific as a result of its difficult economic geography: ADB, Meeting Asia’s Infrastructure Needs, 99.
 Debt sustainability analyses are prepared jointly by the IMF and the World Bank; however, in this Analysis we refer to these simply as IMF debt ratings both as the IMF tends to lead the analysis and for brevity.
 Government of Vanuatu, Vanuatu Post-Disaster Needs Assessment: Tropical Cyclone Pam (Port Vila: Government of Vanuatu, 2015), ix, https://www.gfdrr.org/sites/default/files/publication/pda-2015-vanuatu.pdf; Government of Tonga, Post Disaster Rapid Assessment: Tropical Cyclone Gita (Nuku’alofa: Government of Tonga, 2018), 11, https://www.gfdrr.org/sites/default/files/publication/WB_Tonga_Report_FA07.pdf; Government of Samoa, Samoa Post-disaster Needs Assessment: Cyclone Evan 2012 (Apia: Ministry of Finance, 2013), 17, https://openknowledge.worldbank.org/bitstream/handle/10986/15977/ACS44320ESW0wh00Box0379812B00OUO090.pdf?sequence=1&isAllowed=y.
 The IMF now explicitly incorporates the impact of frequent natural disasters into its debt sustainability analyses in the Pacific, causing it to recognise greater risks in its assessments than it had previously. For the purposes of its assessments, the IMF also assumes all lending from the multilateral development banks will be provided as loans when in reality the banks then use this rating to adjust their financing terms towards providing more grants instead of loans. With the banks scaling up their activities in the Pacific in recent years, this has then contributed to worsening debt risk ratings.
 This is much larger than the OECD figure reported earlier primarily as we include non-aid flows but also as the Pacific Aid Map includes aid flows not otherwise captured by the OECD (mostly from China but other countries also): see Lowy Institute Pacific Aid Map, https://pacificaidmap.lowyinstitute.org/.
 Sarah O’Dowd, “Power Plays in the Pacific”, East Asia Forum, 21 August 2018, https://www.eastasiaforum.org/2018/08/21/power-plays-in-the-pacific/.
 Primrose Riordan, “Coalition Attack on China over Pacific Aid”, The Australian, 10 January 2018, https://www.theaustralian.com.au/nation/foreign-affairs/coalition-attack-on-china-over-pacific-aid/news-story/29eb518cae5b114272a664aa56a67166; Catherine Graue and Stephen Dziedzic, “Federal Minister Concetta Fierravanti-Wells Accuses China of Funding ‘Roads that Go Nowhere’ in Pacific”, ABC News, 10 January 2018, https://www.abc.net.au/news/2018-01-10/australia-hits-out-at-chinese-aid-to-pacific/9316732.
 Matthew Dornan and Philippa Brant, “Chinese Assistance in the Pacific: Agency, Effectiveness and the Role of Pacific Island Governments”, Asia Pacific and Policy Studies 1, Issue 2 (2014), 349–363, https://onlinelibrary.wiley.com/doi/full/10.1002/app5.35.
 See the discussion by World Bank Lead Economist Robert Utz, in ADB, Pacific Economic Monitor, December 2018, 40, https://www.adb.org/sites/default/files/publication/471196/pem-december-2018.pdf.
 IMF, “Vanuatu: 2019 Article IV Consultation”, Country Report No 19/162, 13 June 2019, https://www.imf.org/en/Publications/CR/Issues/2019/06/12/Vanuatu-2019-Article-IV-Consultation-Press-Release-Staff-Report-and-Statement-by-the-46991; Ministry of Finance, Government of Samoa, Quarterly Public Debt Bulletin, Issue No 26, December 2018, https://www.mof.gov.ws/Portals/195/AID/AID%20Quarterly%20Debt%20Bulletin/QDB%20Dec%202018.pdf; Ministry of Finance, Government of Tonga, Budget Statement 2018/19: Recovery with Greater Resilience, http://www.finance.gov.to/node/349.
 Over 2000–2017, average nominal GDP growth for the median Pacific country was 5.6 per cent and the lowest figure was 2.6 per cent. Over 2011–2017, the same figures were 5.5 per cent and 2.6 per cent, respectively.
 Export Finance Australia, “PNG—Sovereign Bond Issue Bodes Well; but FX Shortages Remain”, World Risk Developments, October 2018, https://www.exportfinance.gov.au/resources-news/news-events/world-risk-developments/2018/world-risk-developments-october-2018/.
 This equates to comparing the face value of the loan to its present value given an assumed discount rate.
 The OECD has recently revised its technical definition for concessional loans to reflect a more complex approach tailored to the characteristics of individual debtor countries and different creditor types. Here we use the earlier simpler methodology which provides a uniform approach across all debtors/creditors. Qualitatively, the conclusions remain the same using either method. See OECD, “Official Development Assistance — Definition and Coverage”, https://www.oecd.org/dac/stats/officialdevelopmentassistancedefinitionandcoverage.htm.
 Japan International Cooperation Agency, “Signing of Japanese ODA Loan Agreement with Vanuatu”, Press Release, 13 June 2012, https://www.jica.go.jp/english/news/press/2012/120613.html.
 World Bank, “IBRD Financial Products: Lending Rates and Fees”, https://treasury.worldbank.org/en/about/unit/treasury/ibrd-financial-products/lending-rates-and-fees#1; ADB, “Overview of LIBOR-based Loans: Sovereign and Sovereign-Guaranteed Borrowers”, November 2018, https://www.adb.org/sites/default/files/lbl-overview-20181127.pdf.
 This refers to the World Bank Non-Concessional Borrowing Policy (http://ida.worldbank.org/debt/non-concessional-borrowing) and IMF Debt Limits Policy (https://www.imf.org/external/np/spr/2015/conc/index.htm).
 The World Bank and IMF use a lower discount rate of 5 per cent compared to the OECD, which reduces the calculated grant element contained in any given loan.
 World Bank, “IDA Lending Terms”, http://ida.worldbank.org/financing/ida-lending-terms; and ADB, “Financial Products: Public Sector Financing”, https://www.adb.org/site/public-sector-financing/financial-products.
 In the absence of an IMF risk rating, the ADB used a simplified approach to judge the risk of debt distress based on a static analysis of existing debt sustainability indicators, rather than the forward-looking analysis used by the IMF. Around 1 per cent of ADB lending has also still flowed to countries judged at the time by the IMF to be at a high risk of debt distress. For the projects of concern, changes in IMF debt ratings took place several months prior to the signing of loan agreements between the ADB and borrowing governments. It therefore appears to reflect operational lags between changes in IMF debt ratings and changes in ADB lending terms, particularly given the project design cycle.
 The Cook Islands, Fiji, Kiribati, Marshall Islands, Nauru, Palau, Papua New Guinea, Samoa, Solomon Islands, Tonga, Tuvalu, Vanuatu, and the Federated States of Micronesia.
 The CGD study also made no growth adjustment.
 The IMF considers several countries (including Kiribati, Papua New Guinea, Marshall Islands, Samoa, and Tonga) to be at high/heightened risk of overall debt distress based on debt ratios that are either close to 50 per cent of GDP or projected to reach this level under its DSA projections.
 There are three main issues. First, the IMF ratings assume all financing from the multilateral development banks is provided in the form of loans. In reality, the banks adjust their policies to provide more grants to countries at greater risk of debt distress (including entirely grant financing for those at high risk). Second, for higher-income countries (Fiji, Nauru, and Palau) the IMF only produces a binary rating of “sustainable” or “unsustainable”, which is of limited use for our purposes. Third, using the IMF ratings would not allow us to compare the effects of different magnitudes of additional lending from China and Australia, again reducing the utility for our purposes.
 Theoretically, EFA financing based on this increased capital could be considerably larger than AIFFP financing. However, the use of this capital is not limited to the Pacific and no specific allocation has been set for the Pacific.
 We view this as relatively ambitious given the likely practical challenges in building a pipeline of high-quality projects.
 Stephen Dziedzic, “What’s in the Secret Documents Signed by China and Vanuatu at APEC?”, ABC News, 27 November 2018, https://www.abc.net.au/news/2018-11-27/the-secret-documents-china-and-vanuatu-signed-at-apec/10557438?pfmredir=sm.
 O’Dowd, “Power Plays in the Pacific”.
 IMF, “Kiribati: 2018 Article IV Consultation”, Country Report No 19/26, 24 January 2019, https://www.imf.org/~/media/Files/Publications/CR/2019/cr1926-kiribati.ashx.
 If Papua New Guinea is excluded, Australian loans would equal about 18 per cent of GDP for the remaining three countries eligible for lending under our scenario.
 IMF Institute Training at China–IMF Capacity Development Center (CICDC), Dalian, China, https://www.imf.org/en/Capacity-Development/Training/ICDTC/Schedule/CT; and “Memorandum of Understanding on Collaboration Matters to Establish the Multilateral Cooperation Center for Development Finance”, 25 March 2019, https://www.aiib.org/en/about-aiib/who-we-are/partnership/_download/collaboration-on-matters.pdf.
 “G20 Operational Guidelines for Sustainable Financing”, G20 2017, Germany, March 2017, https://www.bundesfinanzministerium.de/Content/EN/Standardartikel/Topics/world/G7-G20/G20-Documents/g20-operational-guidelines-for-sustainable-financing.pdf?__blob=publicationFile&v=1; and “G20 Principles for Quality Infrastructure Investment”, G20 2019, Japan, https://www.mof.go.jp/english/international_policy/convention/g20/annex6_1.pdf.
 Ministry of Finance of People’s Republic of China, “Debt Sustainability Framework for Participating Countries of the Belt and Road Initiative”, 25 April 2019, 1, http://www.mof.gov.cn/zhengwuxinxi/caizhengxinwen/201904/P020190425513990982189.pdf.
 In a practical sense, this would be necessary to inform their respective debt sustainability assessments and any consequent adjustment in lending plans.
 For a discussion of the experience in Tonga and Vanuatu, see Dornan and Brant, “Chinese Assistance in the Pacific: Agency, Effectiveness and the Role of Pacific Island Governments”.
 Vanuatu’s 2015–2017 Debt Management Strategy targets to avoid any non-concessional borrowing (using the IMF and World Bank standard) and, according to the more recent IMF Article IV report, the finance ministry intends to retain this target under an updated strategy.
 Agatha Kratz, Allen Feng and Logan Wright, “New Data on the ‘Debt Trap’ Question”, Rhodium Group, 29 April 2019, https://rhg.com/research/new-data-on-the-debt-trap-question/.
 China recently agreed to extend the grace period on debt owed by Tonga for the second time: “Tonga Gets Five Years’ Grace on Chinese Loan as Pacific Nation Joins Belt and Road Initiative”, ABC News, 19 November 2018, https://www.abc.net.au/news/2018-11-19/china-defers-tongas-loan-payments-as-nation-signs-up-to-bri/10509140. On the one hand, this implies a total grace period of 15 years on the original 20-year loans, making the overall terms even more concessional. On the other hand, Tonga will now have to repay the entire debt within a five- to seven-year window. Given the size of the debt (about a quarter of Tonga’s GDP), this will make the short-term fiscal pain of servicing the debt much more difficult, particularly in the absence of prudent planning if the Tongan Government believes it will to be able to negotiate for further debt restructuring.
 The Paris Club is an informal group of major creditor nations that convene to renegotiate the official agreements of debtor countries unable to meet their external debt obligations.
 This was suggested in the CGD study.
 “AIFFP Design Document”, https://www.aiffp.gov.au/sites/default/files/2019-07/AIFFP%20Design%20Document_0.pdf.
 OECD, “Sustainable Lending”, http://www.oecd.org/trade/topics/export-credits/sustainable-lending/.
 See “AIFFP Design Document”.
 Roland Rajah, “Stepping Up on Pacific Infrastructure”, The Interpreter, 20 December 2018, https://www.lowyinstitute.org/the-interpreter/stepping-pacific-infrastructure.
 “Papua New Guinea Asks China for Help with $11.8b National Debt”, ABC News, 7 August 2019, https://www.abc.net.au/news/2019-08-07/png-to-ask-china-for-help-refinancing-national-debt/11391186.
 There is an extensive literature on the effect of aid on governance, which generally suggests that aid can undermine governance — particularly if not carefully delivered to avoid such negative impacts — and highlights how geopolitically-driven aid can be especially damaging (for an overview, see Geske Dijkstra, “Aid and Good Governance: Examining Aggregate Unintended Effects of Aid”, Evaluation and Program Planning 68 (2018), 225–232, https://www.sciencedirect.com/science/article/pii/S0149718917302975). There are also studies that look at Chinese aid in particular which suggest a negative effect overall: see, for example, Samuel Brazys and Krishna Chaitanya Vadlamannati, “Aid Curse with Chinese Characteristics? Chinese Development Flows and Economic Reforms”, AidData Working Paper 52, May 2018, https://www.aiddata.org/publications/aid-curse-with-chinese-characteristics-chinese-development-flows-and-economic-reforms.
Photo: Hussan Amir on Unsplash