The focus of this dissertation is on how developing countries mobilize financial resources to support actions on climate change. The existing literature has largely focused on how the preferences of donors shape financial flows and has not paid enough attention to how developing countries exercise their agency in determining how and under what conditions they receive international climate finance. This dissertation analyzes how host country governments negotiate and identifies the factors that constrain the exercise of that agency. This dissertation finds that the credibility of government commitment best explains how developing countries mobilize climate finance. Negotiating capital (charismatic leadership and salience), the policy context, and actor preferences help to explain finance mobilization. Closely tied to the question of the quantity of finance is the design of the delivery vehicle that is used to channel it. Therefore, the institutional design features of the funds have received a lot of attention in this dissertation. These features include: selecting the fund manager (the trustee of the fund), the fund's governing arrangement (the board), scope (what the fund focuses on), and the financing instruments at the fund's disposal. The institutional design features vary across contexts and pose different levels of sovereignty costs to the host country. This dissertation finds that host governments seek to minimize sovereignty costs they incur even if this means increasing the transaction costs associated with the fund. This finding is in contrast with the scholarship on the design of international institutions that expects design features to reflect the maximization of efficiency gains, such as reductions in transaction costs. The cases here suggest that the actors maximize control and reduce sovereignty costs even if that means incurring higher amounts of transaction costs. Four national climate funds form the empirical core of this study. Bangladesh's experience illustrates how a country that tried hard to bargain with donors was hamstrung by the governance risks posed by its administrative and budgetary processes. Even though the government pre-empted negotiations and designed its own fund, donors were too reluctant to use it as the delivery vehicle. Despite having strong negotiating capital, it had to concede sovereignty costs by accepting the World Bank as the trustee of the fund. The lack of existing data also hampered credibility as it created confusion on how the fund was really adding value. Brazil was in a strong negotiating position vis-à-vis Norway. As it already had policies under implementation, with data that could be monitored, it enjoyed low sovereignty costs in the design of the Amazon Fund. As the original policy to control deforestation had the buy-in of the Ministry of Environment and Forests, which was also the lead in the negotiations with Norway, it did not suffer from implementation problems. As new governments followed, the gains achieved and institutionalized during the Lula and Dilma presidencies have been reversed. Former Prime Minister Meles Zenawi's vision for low carbon growth in Ethiopia gained the interest of a few donors such as the UK and Norway. Initially, the emphasis on climate change, however, was not widely shared amongst Ethiopia's donors. Therefore, the CRGE Facility did not attract substantial amounts of finance at the outset. The fund design reflected the concerns of both sides. UNDP was asked to manage one window of the fund while the Ministry of Finance and Economic Cooperation housed the government-managed window. The government had to allow donors to earmark their contributions if they routed their finance through the government-managed window. In effect, this meant setting up parallel governance and reporting frameworks for each earmarked contribution, thereby increasing transaction costs. While the CRGE strategy and vision are officially under implementation, the inability of the government to provide data and indicators has meant that donors remain unconvinced about how much implementation is actually taking place. In Indonesia, former President Yudhoyono's leadership and Indonesia's salience in terms of deforestation-related emissions provided the government with much negotiating leverage. Indonesia did not have the data or the policies in place at the time of negotiations with Norway. Therefore, it was subject to input-based financing instruments, with specified milestones and targets, until it was ready for results-based financing. The lack of policy implementation, at the time of fund design, also meant that policy rivalry between the lead negotiators (President's Office) and the main target of the fund (the Ministry of Forests) impeded implementation. It took Indonesia nearly a decade before it claimed payments for avoided deforestation from Norway.