Innovative Financing Mechanisms
for Preparedness and Response
Highlights and Main Points Made by Individual Speakers and Participantsa
- The International Finance Facility for Immunisation (IFFIm) secures long-term, legally binding commitments from donor governments whose pledges provide the capital to issue bonds for 20 years or longer. (Sison)
- IFFIm provides Gavi with cheap capital, allowing the organization fiscal space to invest in health systems in the 73 poorest countries in the world and negotiating power with manufacturers to reduce the cost of vaccines. (Egerton-Warburton, Sison)
- Pandemic financing poses two problems: short-term liquidity and risk management. There is probably room for both insurance and banking products in solving these problems. (Sison)
- UNITAID is a global health agency that funds work in HIV/AIDS, tuberculosis, and malaria. Sixty percent of UNITAID funding comes from a levy on air tickets in certain countries. It is a stable and predictable revenue source. Legislators must approve the levy, but it never goes to the national budget, so it is not vulnerable to shifts in political will. (Marmora)
- Funding in international development is a means to do work, not an end in itself. It takes administrative discipline to use funding for its designated purpose, rather than redirecting it to other worthy causes. (Marmora)
- The Global Fund uses revolving funds to draw donor contributions over a period of 5 or 6 years rather than all at once. Donors to a revolving fund are assured that, if their contribution is not used, it will be returned. (Bornstein)
- The Global Fund also has a $30 million fund that can be drawn down (for expenses related to HIV/AIDS, tuberculosis, and malaria) during health crises and humanitarian emergencies. (Bornstein)
- Social impact bonds might be a good way to invest in pandemic preparedness. As long as there are course correction measures to ensure efficient spending and investors are compensated for their lending, the rate of return need not be particularly high. (Bornstein)
- Pandemic financing for the poorest countries should draw on the credit risk of the rich countries to buy capital as cheaply as possible. (Egerton-Warburton)
- The problem of having money for pandemic response should not be confused with the problem of spending it. Developing countries should be expected to invest in their own people and to improve public financial management. (Marmora)
a This list is the rapporteurs’ summary of the main points made by individual speakers and participants and does not reflect any consensus among workshop participants.
After lunch, participants reconvened with a panel on innovative finance for pandemic preparedness and response. Juan Costain of the World Bank moderated this discussion, which started with Paolo Sison of Gavi1 discussing IFFIm. He described IFFIm as a simple concept—a way to secure long-term, legally binding pledges from nine donor governments (Australia, France, Italy, the Netherlands, Norway, South Africa, Spain, Sweden, and the United Kingdom). Their pledges form the capital base that is used to issue bonds on the capital markets, accelerating 20-plus-year funding commitments from donors. IFFIm bonds are investment grade—the organization has an AA/AA1 credit rating, allowing it to provide cheap funding. Historically, IFFIm has been able to borrow at a funding cost lower than the aggregate of its donor governments. Since IFFIm’s inception in 2006, it has issued $5 billion in bonds, half of which has been disbursed to Gavi, accounting for one-third of the organization’s funding.
IFFIm, Sison continued, has allowed Gavi to be more innovative and to act earlier, before donor pledges would be due. The financial space has
1Officially, Gavi, the Vaccine Alliance.
allowed the organization to invest more broadly in strengthening health systems in the countries it serves—the 73 poorest countries in the world—and to have more leverage with vaccine manufacturers. The more capital Gavi can deploy against vaccine purchases, the lower they can bring vaccine prices, and the more sustainable the market becomes.
He explained that IFFIm is a charity with no staff registered in the United Kingdom. The IFFIm board members serve voluntarily and lend their expertise to the facility’s management and strategy. The World Bank manages the IFFIm treasury, and Gavi is its only beneficiary. He praised the risk and financial mechanisms the World Bank placed in the IFFIm structure, citing them as the reason why the facility is able to borrow from the market at such attractive rates. The gearing ratio limit, for example, restricts the amount of leverage that IFFIm can have at any point in time. Leverage, in turn, is based on the present value of donor pledges, making IFFIm always overcollateralized. There is also a liquidity requirement that IFFIm have cash on hand to pay for interest and principle on any of its bonds 1 year in advance. All this makes the facility’s inherent risk equivalent to AAA and extremely attractive to investors. It is also difficult for any investor to decline to buy a fairly priced bond for immunization, so IFFIm is in the fortunate position of investor interest exceeding Gavi’s funding needs, Sison observed.
IFFIm has a unique ability to bridge the time when donors can fund a vehicle and the time when the funds are needed, providing funding quickly and at low cost. These are all features that would be valuable in pandemic response, Sison continued. An IFFIm-like facility for pandemics could do well to replicate the long-term donor commitments that are part of IFFIm. He encouraged the audience to consider what might happen to the donor commitments if they were not needed during the 10- to 20-year period they were made for. Perhaps there could be conditions put on the commitments, he continued, so that if there was no outbreak in the time frame, the commitment could fall away and then be renewed. It might also be possible to have the fund recipient countries pay into the financing vehicle in an amount proportionate to the country’s economy.
Sison called back to the previous session on insurance products and described two basic problems to solve. First is the short-term liquidity problem, the need to have cash on hand quickly when in a crisis. The other is the market problem, the need for a market where pandemic risk is better understood, better quantified, and better managed. There is probably room for both insurance and banking products in managing the two distinct problems, he concluded.
Christopher Egerton-Warburton of Lion’s Head Global Partners continued the discussion of IFFIm, as he was involved in its development in the mid-2000s. When asked if the current regulatory environment would allow
for developing a similar tool for pandemics, he said that he found political will to be the main driver of government support. If political will is absent, governments tend to give regulatory obstacles as their reason for not participating; but if the will is there, any regulatory challenge can be overcome.
He echoed Sison’s point about the value of frontloading (the allocation of costs at the beginning of a program) and emphasized IFFIm’s role in providing predictability of funding. The IFFIm balance sheet allows Gavi to procure ahead of donor funding cycles. Similar predictability would be valuable in health systems work, and in most areas of global health and development. Still, Egerton-Warburton cautioned that pandemic financing might not do well to introduce a totally new bond issuer into the market, as that would only complicate the buying and issuing of bonds.
The discussion then switched to analysis of a different innovative financing tool, the UNITAID air ticket levy. Lelio Marmora of UNITAID described the program, saying that, along with mobile communication, the Internet, extractive industries, banking, and finance, air transport is one of the industries that has most benefited from the globalization of the past 30 years. Air transport has seen growth rates of 5 percent a year over the past 30 years and is forecasted to double its passenger load from 3.3 billion to 7 billion per year over the next decade.
The air ticket levy was put in place in 2006 in an agreement between 11 countries and UNITAID; it yields about €200 million per year—a cumulative 2.5 billion for funding the response to HIV/AIDS, tuberculosis, and malaria, working to complement the Global Fund, the President’s Emergency Plan for AIDS Relief (PEPFAR), the President’s Malaria Initiative, and other programs in the same field. Were the same levy applied to all European countries, he continued, it would make another $1.7 billion available for international development. The levy is in place, Marmora explained, because of political will in participating countries. But after being made law, the contribution is less vulnerable to shifts in political will. The tax on air tickets cannot be redirected to the national budget, so it is relatively stable. The stability allows recipient countries to know how much money they can expect from the levy, easing their planning and ability to spend.
UNITAID gets about 60 percent of its funding from the air ticket levy, and Marmora mentioned the administrative discipline it takes to keep the funding tied to its purpose, avoiding the temptation to use it for outbreak response, climate change, or any number of other worthy development projects. There are also administrative challenges to spending the money properly within its designated boundaries. He explained that proper spending means minimizing administrative costs, having openness in the management of the organization, and also managing with flexibility and tolerance for risk. Risk taking is an important part of the process, he continued, and
can pay off in greatly increased funding streams. Still, he cautioned, in international development, money is a means, not an end in itself. Increasing funding is not a victory; the victory is what the funding enables.
When asked how to evaluate the value for money of different funding plans, Marmora hesitated. Value for money is a popular concept now, but it is not clear what it means, though he credited Partners in Health, Results for Development, and the UK Department for International Development (DFID) with bringing intellectual rigor to the question. Sometimes “value for money” is used to refer to return on investment, but it can also mean the public health effects of the investment; these are different and sometimes conflicting concepts. For UNITAID in particular, its goal is to support more innovative ways to save lives, so the organization is somewhat removed from the achievement. UNITAID worked with the U.S. Agency for International Development and the Gates Foundation to develop a rapid diagnostic for tuberculosis, but the value for money from that device, while clearly substantial, is hard to calculate. He suggested that part of the value comes from time saved, and therefore lives saved and new infections avoided, and part of it from the more efficient allocation of doctors’ time, to say nothing of the avoided opportunity costs to the patient and his or her family from time spent at the hospital. By its best estimates, the UNITAID portfolio has a return on investment between $1:$5 and $1:$20. He also mentioned the value for money to the airline industry, a business that stands to be seriously hurt by an epidemic. The industry wants to see the maximum possible investment in a ticket levy.
Adam Bornstein of the Global Fund was the final panelist in the session. He described three main ways his organization raises and deploys capital. There are contributions from donor governments, either through direct donation over 3 years or a promissory note. Recently, some donors have asked that their contributions not be drawn down all at once but over 5 or 6 years. In response, the Global Fund adapted a World Bank system of pulling donor capital from a country or from a promissory note, which allows for a donation to be made on a revolving basis over a fixed number of years. In practice, he continued, if the organization gets a $500 million donation from one country in a lump sum, and the same amount from another country in a 5-year revolving fund, the donation is available in five $100 million chunks. If the money is not drawn down in the specified period, it is returned to the country. The money does not need to be repurposed because it is only used on as-needed basis, which Bornstein described as more efficient. He explained that a revolving fund might be useful for pandemic financing. Revolving funds avoid some of the administrative expenses associated with permanent funds. Donors to a revolving fund can also be assured that, if their money is not used, it will be returned.
The Global Fund has a new investment vehicle for emergencies: a $30
million fund that can be drawn against for health crises and also during complex humanitarian emergencies, but only for expenses related to HIV/AIDS, tuberculosis, and malaria. Bornstein admitted that Gavi had an advantage on the Global Fund in that Gavi is able to leverage its balance sheet for health systems strengthening and similar activities. He speculated that, if funds like the Global Fund had a blanket mandate to use their money during crises, things would change, but otherwise they must spend according to their mandate.
When asked about involving the private sector in pandemic financing, Bornstein emphasized that these people would be interested in the risk-adjusted return on their investment. When asking a private lender for capital during an emergency, there is essentially no risk-related return because the risk is already 100 percent. If there were a structure that allowed money to come in ahead of the risk, that might be more attractive to many lenders.
He suggested that social impact bonds might be a good way to invest in pandemics. Social impact bonds take investors’ money and put it into an intervention, with measures in place to course correct and to repay governments and investors upon success. He saw the necessary pieces of pandemic management—surveillance, rapid response, a credit line, a cash distribution system, and capacity building—as an interesting structure that would lend itself to a social impact bond, perhaps with returns of 3 to 10 percent.
Private-sector investors, he continued, are happy when their money goes out and is put to work. As long as they are compensated for the lending, he felt the rate of return need not be particularly high. Nowadays, he continued, all investors—public or private—want to see that their capital is being recycled. Provided there are course-correction measures in place to ensure the money is being spent properly and efficiently, finding investors should not be difficult.
In the open question session, the panelists were asked about the value of investing in epidemic prevention. Marmora said that, even from a financial perspective, prevention and building health systems are much cheaper than treatment or response. The problem, he continued, is that politicians do not always see it that way; they are under pressure to spend on the most visible projects with a political impact. Tore Godal of the Ministry of Foreign Affairs, Norway, disagreed. He said that governments are willing to pay for prevention but that the distinction between preparedness and response spending was artificial, as the financing instrument will probably be used for both. Like any investor, his country’s government wants to spend efficiently and leverage its balance sheet, perhaps by guaranteeing financing from 2025 to 2035 for any outbreak between 2015 and 2025. This kind of financing could take some risk out of the equation, he observed.
Da Victoria Lobo then asked the panelists how to manage credit in countries with high debt-to-GDP2 ratios. Sison replied that funding should be tied to the recipient countries’ ability to pay, and Bornstein agreed that the International Monetary Fund guidelines on concessionary loans have already set out the terms by which loans do not set back poor or indebted nations. Egerton-Warburton harkened back to the story the audience had heard that morning of how Mexico managed the 2009 H1N1 outbreak. The government had to raise money for response, then ask the World Bank for a loan to pay it back—a loan that was ultimately canceled. Such financing is an option in Mexico because it is a middle-income country and a sophisticated user of capital markets, and it has a good relationship with the World Bank. The same is not true of Liberia, for example. Egerton-Warburton suggested that for the poorest countries and fragile states it would be necessary to invert the process: to give the credit first, borrowing against the credit risk of the donor countries and using the strongest balance sheets in the world to buy fast capital at the lowest possible cost. The domestic contribution would be the last money accessed, not the first. A pandemic financing structure, he concluded, should rely not on Liberia’s ability to borrow, but on Norway’s. Bornstein agreed, as it is not reasonable to ask any poor country to set aside money for an extreme and unlikely situation when they cannot meet routine expenses. At the same time, a participant observed that the foreign direct investment and private capital flows evaporate with an infectious outbreak, a problem that catches the attention of heads of state and finance ministers more than routine health problems ever do.
Marmora then pointed out that there are really two problems to solve in pandemic financing: the problem of getting money and the problem of spending it. The two, he said, should not be conflated. Funding is just the means to do the work of development, not an end in itself. He cautioned against relying on the budget of G8 countries to support the whole development agenda and asked that more pressure be put on developing countries to invest in their own people. Marmora spent 8 years at the Global Fund before going to UNITAID, so he was aware that the Global Fund has $1 billion unspent in West and Central Africa because it is extremely difficult to disburse more than $17 to $20 million in many countries in the region. “Of course,” he continued, “we can throw the money from a plane and say you disbursed it, which is what happened for many years, but if you want to be respectful of the taxpayers’ money . . . you have to do the right thing.” In some countries this will mean a zero-cash policy, in other places pressure to change the procurement policy, and always considerable attention to fiduciary arrangements for avoiding graft.
2Gross domestic product.
A pandemic, Egerton-Warburton observed, is by definition a situation where things have gotten out of control. Grant capital is needed in such situations, and that money is not to be recouped. At the same time, he welcomed the emerging expansion of private-sector investment in development, and the opportunity for donor governments to make long-term commitments through instruments like IFFIm. He commented on the inefficiency of donors planning for disasters by holding large amounts on their balance sheets until the end of the year, not knowing if that cash would be used.
Bornstein reflected that nobody wants to suggest people should profiteer from pandemics, but people do make money during them. Some people lose money, and others gain. During a crisis, logistics companies and drug manufacturers will be asked to contribute, and there is considerable good will to their contribution. But they also need to make a living. If the private sector is to be involved with pandemics, then it needs to earn a return on its investment if for no other reason than that there is a cost to capital and an opportunity for loss. Therefore, he continued, social impact bonds are a promising financial tool for pandemics.