Climate change will alter the savings-investment balances of many countries as they undertake investments in adaptation, mitigation and resilience, and savings patterns react to greater uncertainty and changing income levels. The UN estimates that globally adaptation costs alone are likely to range from USD 140 billion to 300 billion per annum by 2030 and could rise to between USD 280 billion and 500 billion per annum by 2050. By definition that will alter current accounts and external positions in a potentially significant manner.

The IMF is mandated by the Articles of Agreement to conduct "firm surveillance" over the exchange rate policies of its members in order to ensure the effective operation of the international monetary system. Understanding member's external positions and their drivers is at the core of this mandate, and indeed, the Fund conducts external sector assessments (ESAs) on each member, each year. Despite its growing centrality to economic trends, climate change risks are not always well linked with the IMF's core macro frameworks, including current practice for ESAs.1 It is imperative that the IMF develop a sufficient ESA toolkit to incorporate the effects of climate change.

This is not a straightforward task for three main reasons.

1. Consistent estimates of the investments needed and likely to occur in response to climate challenges are not available across a wide basket of countries.
2. How investment and savings will respond is a complex question involving expectations of future income, future investment needs and returns, budget constraints and aid flows.
3. The entire exercise must be done on a multilateral basis. Climate change affects every country, but not every country can borrow externally at once to increase investment. Each country's capacity to respond will depend on what other countries are doing at the same time.

This project aims to address those three challenges to develop the data and analytical tools to incorporate the impact of climate change into the IMF's ESAs, and specifically into the EBA-lite suite of models, which covers the preponderance of the Fund's membership. We will compile a database of the necessary data to provide a useful resource and produce of models to formally and consistently incorporate climate change into ESAs. We will leverage machine learning to predict investment needs. Econometric techniques will explore how savings will likely respond. We will embed the entire exercise in a multilaterally consistent framework to ensure the results add up consistently across the global economy.

1. Heike Mainhardt, 2020. "Build Back Better? IMF's policy advice hampers green COVID19 recovery." Recourse, Greenpeace, Earthlife Africa, Centre for Financial Accountability.