Social Discounting of Large Dams with Climate Change Uncertainty

Abstract

It has long been known that the economic assessment of large projects is sensitive to assumptions about discounting future costs and benefits. Projects that require high upfront investments and take years to begin producing economic benefits can be difficult to justify with the discount rates typically used for project appraisal. While most economists argue that social discount rates should be below 4%, many international development banks and government planning agencies responsible for project appraisal can be found using rates of 7-12% or more. These agencies justify choosing higher discount rates to account for the opportunity cost of capital. Meanwhile, a new and robust debate has begun in economics over whether social discount rates of even 3-4% are too high in the context of climate change. This paper reviews the recent discounting controversy and examines its implications for the appraisal of an illustrative hydropower project in Ethiopia. The analysis uses an integrated hydro-economic model that accounts for how the dam's transboundary impacts vary with climate change. The real value of the dam is found to be highly sensitive to assumptions about future economic growth. The argument for investment is weakest under conditions of robust global economic growth, particularly if these coincide with unfavourable hydrological or development factors related to the project. If however long-term growth is reduced, the value of the dam tends to increase. There may also be distributional or local arguments favouring investment, if growth in the investment region lags behind that of the rest of the globe. In such circumstances, a large dam can be seen as a form of insurance that protects future vulnerable generations against the possibility of macroeconomic instability or climate shocks.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.