| By Michael Bennon

On August 3, in his first visit to the Asia-Pacific region, new U.S. secretary of defense Mark Esper called out several examples of aggressive conduct by China, including “using predatory economics and debt-for-sovereignty deals.” The Belt and Road Initiative (BRI) has created some conflicts between recipient governments and international institutions in the past. Perhaps the latest and starkest example is in Pakistan, where a wave of BRI projects was followed by this summer’s bailout by the International Monetary Fund (IMF).

When the $62 billion China-Pakistan Economic Corridor (CPEC) was announced in 2015, it should have been easy for followers of the BRI to foresee where the initiative was heading. Even at its initial announced value of $46 billion, the initiative would have amounted to more than 20 percent of Pakistan’s GDP. While details were unclear, as they often are with the BRI, the majority of lending would inevitably come in the form of direct bilateral loans from The Export-Import Bank of China (EXIM) or the China Development Bank (CDB). This was the case for Pakistan’s precedents in Sri Lanka, Kenya, Montenegro, Congo, and other recipients of ambitious bilateral lending initiatives, each of which led to a debt crisis several years later.

Any remaining hope that Pakistan would remain the exception to the group ended this spring as the IMF approved a $6 billion bailout for Pakistan as its finances deteriorated. The United States did not attempt to block the bailout despite threats in 2018 to object unless there were assurances that the funding would not be used to service bad Chinese debt. The fact is that money is fungible, and any bailout would, and this bailout will, service bad Chinese debt. According to the IMF’s own analysis, more than 25 percent of Pakistan’s outstanding public debt, and nearly 40 percent of the debt coming due during the IMF program (through 2023) is bilateral or commercial debt to China. Washington may as well have written China a check.

That is not to say the United States should have followed through on its threats. Pakistan agreed to very ambitious fiscal reforms (arguably too ambitious to achieve) as part of the bailout. In addition, informal reports in Pakistan have noted that approvals/funding for CPEC and other BRI initiatives have dropped precipitously, though Chinese state-owned media reported no change.

Still, Washington and the IMF missed an invaluable opportunity in the Pakistan bailout. The IMF didn’t need to prevent, and indeed could not have prevented, the bailout from indirectly supporting China-financed projects. It could have simply required Pakistan to publicly disclose its contingent liabilities under CPEC by releasing the contracts governing the $62 billion of infrastructure projects planned under the program.

Contingent liabilities can have significant negative impacts on public finances, and they are particularly prominent in the infrastructure sector. The massive contracts governments sign for complex infrastructure projects include provisions that often share risk between the government sponsor and the companies/investors involved. If a natural disaster or an unforeseen site condition causes project costs to increase, for instance, the sponsoring government may need to foot the bill. Keeping tabs on the liabilities governments take on for these projects is an important task for governments and the industry of ratings agencies and bond analysts that evaluate public credits.

Contingent liabilities for a project can create unforeseen costs for a government, regardless of how a project is financed. The sponsoring government virtually always takes on some risk. However, some standards have emerged in capital markets and banks (promoted by multilateral development banks like the World Bank) regarding whether a general sovereign guarantee supports an infrastructure project or not. In other words, a particular infrastructure project could be financed using public finance (i.e., on the sponsoring government’s balance sheet) or using project finance (i.e., not generally recourse to the sponsoring government and supported only by the economics of the project itself).

That is a big distinction in infrastructure finance, and it is a big distinction in sovereign debt markets. A container port, toll road, railway, or airport that doesn’t garner enough shipping, traffic, rail cargo, or air traffic to support its operations can default if it is financed at the project level. If it is financed publicly and underwater economically, it is a liability of the state. Note that this doesn’t necessarily mean the project in question isn’t economically beneficial in aggregate, just that it is not financially viable without the host government’s support.

Either public or project finance is a perfectly viable option for financing an infrastructure project, and the alternative that is optimal usually depends on the details of the project in question. The point here isn’t that one option is better than the other, it is that it must be clear which one of these options is being used—to the market, to the lender, and, most certainly, to the government sponsoring the project.

This brings us back to Pakistan, the broader BRI, and the IMF bailout. As Chinese bilateral lending has come to dominate the infrastructure bilateral and multilateral lending industry, one significant and critically important shift has been in the prevalence of directly negotiated and not publicly disclosed concession contracts for infrastructure. The specific contractual terms and risk allocation between Chinese state-owned enterprises, Chinese bilateral lenders, and the host nation are often unclear, and it often isn’t even clear whether the project in question is actually happening or just an idea until construction begins.

In finance, problems tend to gravitate toward opacity around sovereign guarantees. While some recipients of BRI lending have required the contracts that govern projects to be published, those that have not have clearly been exposed to unexpected and unforeseen contingent liabilities. The delineation of public versus project borrowing is subverted when the contracts used to deliver what is on the surface project finance actually include sovereign guarantees. This has clearly been the case for some BRI projects and led to the allegation of “debt-trap diplomacy” for some BRI recipients. In many cases, the poster children of BRI excess have had difficulty determining just how much they owe as their finances have deteriorated.

At this year’s BRI forum, Chinese officials stated that they would push for more transparency in BRI projects. This is laudable. But transparency in international finance should not be subject only to the bilateral policies of lending nations. It should also be enforced by international institutions, like the IMF. The IMF did miss an opportunity in Pakistan this summer, but it wasn’t to prevent or ring-fence a bailout. The IMF should have simply required transparency in the form of published contracts as a precondition for IMF support.

The IMF should consider doing so in the future when the next potential exception to BRI excess eventually confirms the rule.

Michael Bennon is a non-resident adjunct fellow with the Reconnecting Asia Project and Simon Chair in Political Economy at the Center for Strategic and International Studies in Washington, D.C. He is currently the managing director at the Global Projects Center at Stanford University.