Getting to Yes on a World Bank Recapitalization

November 08, 2017

The United States is apparently driving a hard bargain with World Bank President Jim Kim, who is hoping the bank’s 189 shareholders will agree to increase the current capital of the bank’s “hard” window (for middle-income countries with over $1200 in per capita income) sometime in 2018. A deeper capital base would allow the bank to lend more—more each year than the almost $30 billion of loans approved in 2016. But the US wants to link any support for a recapitalization to World Bank “graduating” China—and perhaps other member countries with good access to private capital markets who don’t seem to “need” the World Bank.

But cutting off China is not something Jim Kim and senior management at the World Bank want to do. There are sensible arguments on both sides of this divide. And there is a simple way to begin to thread the needle and get to yes, but first some background (readers familiar with the US position and the World Bank counter-arguments may want to skip to the “getting to yes” section below).

The US position has some merit

On the face of it, the US position is not unreasonable. China doesn’t “need” to borrow; it can easily self-finance its own development investments. Indeed, China is a donor and a lender to other countries, with massive lending by its ”policy banks” (China Development Bank and China Exim) both inside and outside China (and long-run plans to lend heavily to other countries as part of its Belt and Road Initiative; China is also the single largest shareholder in what the US might consider a “rival” multilateral bank to the World Bank, the Asian Infrastructure Investment Bank founded under its leadership in 2015. And from the point of view of the US, China’s status as one of the biggest borrowers at the World Bank (with almost $2 billion in approvals in 2016, behind only Peru, India, and Kazakhstan) means that ending its borrowing would free up existing World Bank capital for other smaller and poorer countries that face tougher terms if they borrow on private markets.

Then why does China want to borrow?

China wants to borrow, the World Bank often argues, because along with the money for any particular project or program comes valuable technical and policy advice based on World Bank staff’s expertise and experience acquired in other countries.  China could in principle “buy” advice from the World Bank, but may well prefer to bundle money with advice; that way World Bank management has skin in the game on the success of a major investment. Moreover, for China the process of borrowing and implementing World Bank projects brings lessons on how to manage procurement, avoid corruption, develop high standards of project analysis, and avoid costly environmental and social risks. In addition, the central government as gatekeeper to the World Bank (and the Asian Development Bank) manages a healthy competition among state and local governments; only high-value proposals consistent with central government priorities get approval to open negotiations with the outside lender.  

Why does the World Bank want to lend?

And the World Bank wants to lend—especially to China. The hard window of the World Bank, as in any bank, has to pay for itself if its development mission is to be sustained and furthered. Having “strong” as well as “weak” borrowers is at the heart of its business model as a collective cooperative or global club for lending.  China’s risk of default is low, and it’s a big economy; the transaction costs of doing big loans to big economies are in general lower than for most other borrowers. Net income from lending to China thus helps subsidize the higher costs and greater risks of lending to smaller and less financially hefty borrowers. In addition there is the argument that the bank benefits in non-financial ways from its work in China--quoting President Kim—that “the lessons we learn in China…are very helpful to other middle income countries”. Finally, the Bank argues that 73 percent of the world’s poor (using the $1.90 a day line of absolute poverty) live in middle-income countries; and though fewer than 2 percent of China’s population is estimated to live below the poverty line now, median consumption is still surprisingly low, at $7.66 per person per day compared for example to Poland at $14.20, until recently another large, middle-income borrower.

Is it just China? What else is motivating the US position? 

First, a recapitalization would require the Trump Administration to ask Congress for money. To add $40 billion of paid-in capital to the bank’s current ledger (which would more or less double its current equity capital would require Congress come up with about $6 billion (given its 15.9 percent share in total ownership of the bank)—not a huge amount but politically a heavy lift for an administration arguing the US bears too much of the burden of global peace and security. Even during the Obama Administration, US Treasury sought to keep capital increases at the MDBs to a minimum, even as they acknowledged the need for more capital following aggressive MDB counter-cyclical lending in the face of the global financial crisis.

Second, there is ideology to consider. In 2000, in the waning days of the Clinton Administration, the Republican majority of the members of the “Meltzer” Commission recommended the World Bank end lending to “middle-income” countries on the grounds that those countries had adequate access to foreign capital markets—similar to today’s discussion, and that therefore the World Bank and the other multilateral banks were competing with private capital. (With the subsequent election of Republican George Bush in 2000, the resulting discussion and debate was worrying for supporters of the World Bank’s business model and its development mission in the US and Europe, and for middle-income countries who wanted continued access to loans on the better terms the multilaterals provided, and motivated this report of the Carnegie Endowment for International Peace criticizing the Meltzer commission analysis—and with recommendations I invoke below.)

And third, China is special. Why not formally “graduate” Poland and other borrowers with income per capita higher than China?  China’s huge economy makes it a geopolitical power that now rivals the US—and then there is the Trump Administration unhappiness with China’s persistent trade surplus with the US.    

Moreover, the US has the stronger hand in any negotiation—with President Kim and with other country shareholders

Why? The US’s 15.9 percent of the bank’s capital makes it the largest single shareholder. In principle the US could step aside and allow other countries to contribute to a recapitalization, but that would dilute its current ownership and in current circumstances, with Trump Administration support for multilateral institutions at best unclear, that might make it costlier for the World Bank to borrow. The World Bank’s AAA++ rating and resulting ultra-low costs as a borrower, depends in some unclear measure on continuing and active US support; in the end US steadfast political support is central to what the rating agencies rely on. So the World Bank needs the US in on any recapitalization.

In a sense it’s a good sign that the Trump Administration hasn’t just said no, and prefers to get something back by negotiating. Perhaps even the Trump administration doesn’t want to give up entirely the big leverage the World Bank  provides in support of US foreign policy, market-driven growth, enhanced global security, reduced poverty, and in general a better world—all at bargain rates for US taxpayers.

Getting to yes: Differential pricing

The World Bank’s arguments are sensible and technocratic; the United States argument is rooted in its fiscal straits, its longstanding market (laissez-fare) orientation, more marked during Republican administrations, and the Trump Administration’s unfriendliness to its new global rival.

But both sides want to get to yes. The Trump Treasury Department is probably looking to limit the size of any recapitalization, and to see China “graduated.”  The World Bank and its middle-income borrowers don’t want to accept the idea that any borrower that is otherwise in good standing can be excluded from the borrowing club, and particularly in a case where it is one or more non-borrowers that want a club member pushed out.

One way of getting to yes is for the US to take the position that it is time for the World Bank to change its pricing. At the moment, all borrowers face the same rate for any particular type of loan—the poorest IBRD countries (think Guatemala or Zimbabwe) borrow at the same rate as China. All get a subsidy in the amount of the spread between the IBRD loan and the higher cost (and terms in general, including maturity) of loans on the private market. The table below includes countries that have been or are currently among the bank’s biggest borrowers. As the table below shows, using as an example a 10-year fixed loan from IBRD of 2 percent or 200 basis points, the more costly a country’s alternative on the private market the bigger would be its subsidy were it borrowing today. 

Country Name GDP per capita (current US$) 10-year fixed loan rates Subsidy assuming World Bank lending rate of 2%
Brazil $ 8,650 10.05 8.05
Mexico $ 8,201 7.19 5.19
India $ 1,709 6.93 4.93
Indonesia $ 3,570 6.66 4.66
China $ 8,123 3.90 1.90
Turkey $ 10,788 11.83 9.83
Colombia $ 5,806 6.77 4.77
Poland $ 12,372 3.45 1.45
Peru $ 6,046 5.05* 3.05

In most banks, terms of loans are related to the borrower’s creditworthiness—more risky borrowers face higher costs. But the World Bank is a development bank; it would make sense to tie the effective subsidy, whatever it is to some measure of countries’ wealth—so that the richer the country the smaller its subsidy. That form of differential pricing was a recommendation of the 2001 Carnegie Endowment report referred to above (The Role of the Multilateral Development Banks in Emerging Market Economies), co-chaired by Paul Volcker (a firm believer in encouraging countries to “graduate”) and Angel Gurria (then a former Finance and Foreign Affairs minister of Mexico and now the head of the OECD). The first recommendation of that report was that graduation should be “voluntary, but coupled with incentives to avoid prolonged dependence.”

The logic of voluntary graduation

The logic of voluntary graduation is that countries do in fact ‘self-graduate’ as the spread borrowing on the private market and borrowing from the World Bank shrinks. By the year 2000, 26 countries had done so—and demonstrating one logic of voluntary graduation, four (Korea, Malaysia, Chile and Costa Rica) had returned by 2000 to borrowing, showing voluntary graduation gives the MDBs the flexibility to respond when they are needed with counter-cyclical lending during global systemic crises. Since then many more have without fan-fare self-graduated, including apparently Poland.

The Volcker-Gurria commission wanted at the same time to avoid “prolonged dependence” and encourage countries to rely on private capital; so the World Bank and the other MDBs “should increase further the incentive to graduate by differentiating their pricing according to borrowers’ per capita income.”

That idea still makes sense, and could help the parties get to yes on a recapitalization and on graduation soon if not immediately of China. 

A simple and predictable change in pricing according to per capita income would not be new at the World Bank, where already countries in the category “low-income” borrow from IDA, the soft window, at much lower rates than countries in the category “lower middle-income.” Thus, for example, Malawi and Honduras borrow at lower rates than Ghana and Paraguay. However, lower middle-income countries such as Ghana and Paraguay currently borrow at the same rates as countries in the category “upper-middle income” such as South Africa, China, and Brazil. Of the countries listed in the table above, all are “upper middle income” except India, Indonesia, and Ukraine.

As a start on differential pricing, the bank’s shareholders could agree on a different rate for the lower middle-income borrowers than for higher middle-income borrowers. For China, in the latter category, that would still mean a subsidy but a smaller one than now (see table).

Note that an increase in the cost of borrowing for all upper-middle income countries, including China and high-income Poland, would make sense in other ways. For example, it would increase the incentives inside the World Bank to minimize the delays and relatively high “hassle” costs that “strong” borrowers like China face, extracting some efficiency gains from the new pricing policy. And it would not eliminate access of countries like China to policy advice; even now some countries that have voluntarily graduated “purchase” that advice from the bank—including Chile and countries like Saudi Arabia which has never been a borrower.

Of course, a policy tying borrowing costs to per capita income would not be welcomed by upper middle-income borrowers, and the specifics would have to be negotiated not only between the US and the bank, but by all the shareholders.  

But negotiating is all about in the end getting to yes—and in this case getting closer to a healthy recapitalization and a more optimal “development” use of the bank’s lending resources.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.

Image credit for social media/web: Social media image by Wu Zhiyi / World Bank