• Sometimes you have to be the first dollar in. When the foundation made its $750 million contribution to the GAVI Fund in 1999, we were its only funder, and we did not require that the grantee match our commitment (as we often do today). It was a risky bet—and other contributions to GAVI were initially slow to come in. It is now clear, however, that the foundation’s early contribution worked to jumpstart worldwide progress on immunization, which proved to be an effective way of bringing other donors to the table.
  
  
• “Five Years Only” Does Not Work. One of the key assumptions underlying the business model for GAVI was that the alliance would help a country introduce a new vaccine such as Hib by providing it for free for five years—during which time the price would come down—and then the country and donor governments would be willing and able to step in to sustain the program. GAVI planned to continue to support the immunization needs of each country, but it would refocus its resources to support the introduction of another needed vaccine on the country’s priority list. This approach has not worked as planned. Vaccine prices did not decline as anticipated, and in some cases actually went up.  Although many countries increased their spending on immunization, they were unable to increase their funding to meet the higher cost of the new and improved immunization program. And some traditional donors that we hoped would lend support have redirected vaccine funds to other pressing health needs. As a result, GAVI has recently launched a new model we hope will work better.  Rather than providing free vaccines, GAVI is now requiring countries to “co-pay” from the start—that is, to assume a small percentage of the cost of the program right away, with the percentage increasing every year.
  
   
• GAVI did not pay enough attention to developing-country priorities. Driven in large part by the foundation’s eagerness (even impatience) for results, GAVI often moved at a pace that made it impossible to get full buy-in from developing countries.  For example, in the early years of the alliance, GAVI convinced some countries to introduce a $3.50-a-dose vaccine that was available right away and protected children against five different diseases.  Some countries preferred a vaccine that was not immediately available and protected against only four of these diseases but cost only $1.25 a dose. Now that the five-year introductory period is up, some of these countries are reluctant to pick up the costs of the $3.50 vaccine. They rightly say, “It wasn’t our idea in the first place.” 
  
  
• Funding operating expenses is essential. The foundation was initially reluctant to allow GAVI to use a significant amount of our funding to support grantee operating expenses—as opposed to purchasing vaccines. But we’ve learned that funding operating expenses is vital for getting new entities up and running, for attracting donors that have greater funding restraints, and for giving grantees the ability to innovate. For example, our operating support enabled GAVI to work with legal and financial experts to develop the initial conceptual framework for the IFFIm and make it a reality. IFFIm would not have been born without this support.
  
  
• GAVI was too narrowly focused. From the start, GAVI has recognized that funding vaccine purchases alone would not allow the 75 poorest countries in the world to overcome their infrastructure challenges, so it has provided “immunization system support” to help countries improve their ability to deliver vaccines to a large portion of their children. GAVI has learned that in many cases the ability of the country to reach more children is not related simply to its immunization program. Broader health-system failures are often to blame. So GAVI has now expanded its funding criteria and set aside about half of its resources to help meet broader health-system needs. Under the new criteria, for example, countries can now apply for support for improving the frequency of supervisory visits to remote health clinics or for reforming the system for repairing and maintaining the vehicles used for delivering vaccines and other health supplies to rural areas.
  
  
• Creating an alliance of existing entities was a better strategy than building a new entity. Global partnerships such as GAVI often run the risk of being slow and bureaucratic, and initially we considered funding a new, independent organization rather than an alliance of existing organizations. But as the old African proverb says, “If you want to go fast, go alone. If you want to go far, go together.” Each of GAVI’s partners has a unique role, and brings essential relationships and insights to the table. For example, WHO and UNICEF have been working on the ground for decades in the world’s poorest countries and have trusted relationships with health ministers in places where no one has ever heard of GAVI.

Next: Next Steps