Climate damages in the FUND model: A disaggregated analysis

The authors analyze the FUND model, which provided a significantly lower Social Cost of Carbon (SCC) estimate of $6 compared to other models like DICE ($28) and PAGE ($30). By disaggregating the model, they find that the low SCC is primarily driven by a large estimated net benefit in agriculture, largely due to CO2 fertilization. The paper identifies three technical issues in FUND's agricultural modeling: a risk of division by zero in the optimum temperature equation, the use of physically implausible temperature ranges in Monte Carlo analysis, and the use of truncated normal distributions that inflate the carbon fertilization effect. The authors argue that updating FUND with recent empirical research—which suggests smaller CO2 fertilization benefits and non-quadratic temperature impacts—would significantly increase the SCC estimate.

• The FUND model's central estimate of the Social Cost of Carbon (SCC) is $6 per ton of CO2, which is substantially lower than the $30 estimate from PAGE and $28 from DICE.
• The $6 SCC estimate in FUND is composed of a $6 net benefit in agriculture, a $8 net cost in cooling and heating, and $4 of net costs across 12 other damage categories.
• The net benefit attributed to agriculture in FUND is almost entirely driven by the CO2 fertilization effect, which is estimated to provide a net benefit exceeding $14 per ton of CO2 emissions.
• A mathematical flaw in the optimum temperature equation of FUND version 3.5 creates a risk of division by zero when the Monte Carlo parameter Topt equals 1.6, a value within 0.25 standard deviations of the mean for every region.
• FUND's Monte Carlo analysis for optimum temperatures uses confidence intervals that exceed physically plausible ranges; for example, the 95% confidence interval for South America ranges from 17°C below to almost 18°C above 1990 levels.
• The use of truncated normal distributions for the carbon fertilization parameter in four regions (USA, Canada, Japan/Korea, and Southeast Asia) results in a true mean that is more than 50% higher than the reported mean of the unconstrained curve.
• Recent empirical research contradicts FUND's quadratic temperature-yield relationship; studies on U.S. corn, soybeans, and cotton show slow increases up to an optimum temperature followed by rapid declines, rather than a symmetric quadratic curve.
• Implementing simple fixes for the identified optimum temperature equation problems would increase the FUND SCC estimate by $10 to $12.

APA

Ackerman, F., & Munitz, C. (2011). Climate damages in the FUND model: A disaggregated analysis. Stockholm Environment Institute. https://www.sei.org/mediamanager/documents/Publications/SEI-WorkingPaper-Ackerman-ClimateDamagesInTheFUNDmodel-2011.pdf

Chicago

Ackerman, Frank, and Charles Munitz. Climate damages in the FUND model: A disaggregated analysis. Stockholm Environment Institute, 2011. https://www.sei.org/mediamanager/documents/Publications/SEI-WorkingPaper-Ackerman-ClimateDamagesInTheFUNDmodel-2011.pdf.

Wikipedia

{{cite report |last1=Ackerman |first1=Frank |last2=Munitz |first2=Charles |title=Climate damages in the FUND model: A disaggregated analysis |publisher=Stockholm Environment Institute |date=March 2011 |url=https://www.sei.org/mediamanager/documents/Publications/SEI-WorkingPaper-Ackerman-ClimateDamagesInTheFUNDmodel-2011.pdf |access-date=17 August 2026 |via=Climate Insights Directory}}

BibTeX

@techreport{ackerman2011climate, author = {Ackerman, Frank and Munitz, Charles}, title = {{Climate damages in the FUND model: A disaggregated analysis}}, institution = {Stockholm Environment Institute}, year = {2011}, month = mar, url = {https://www.sei.org/mediamanager/documents/Publications/SEI-WorkingPaper-Ackerman-ClimateDamagesInTheFUNDmodel-2011.pdf}, urldate = {2026-08-17}, note = {Indexed by Climate Insights Directory} }