GCD's Mission is to help banks understand and model credit risks. The comprehensive data pools are collected over a decade and distributed back to members for their own research and modelling.


Learn More

GCD is a unique data consortium that owns banks internal data for both PD and LGD. GCD’s data pools support the key parameters of banks’ credit risk modelling: Probability of Default (PD), Loss Given Default (LGD), Exposure at Default (EAD).

Learn More

GCD’s library gives access to wide variety of publications on risk related topics. Global Credit Data members work together to analyse the data and discuss methodology issues. GCD has published numerous papers and is actively promoting academic research on the data collected.

Access the Library 

Members not only benefit from exclusive rights and access to credit databases and analytics, but also from knowledge and research facilitation possible via the unique industry association.

Through a variety of forums such as workshops, webinars and surveys, GCD is an active industry participant facilitating the discussion in key strategic areas.

Learn More

Global Credit Data collects raw data from its members and distributes it back to them for use in their own analysis and modelling. GCD supports its members by providing a flexible high-end tool on the data pool: the GCD Visual Analyzer. Member banks can create dynamic Reference Data Sets and generate instant views on the data.

Learn More

-> menu code <-

NGFS Climate Scenarios Q&A

by | Apr 22, 2024 | 2024, ESG Climate Risk, Public

  1. Home
  2. /
  3. News
  4. /
  5. Working Groups
  6. /
  7. ESG Climate Risk
  8. /
  9. NGFS Climate Scenarios Q&A

Presentation can be accessed here.

Are compound risks also included?

No compound risks are not yet covered by the scenarios at this point in time. This limitation has also been noted by us, and we have drafted a note that covers the topic, and explains how this could be potentially incorporated in the future of climate scenarios here
Short-term NGFS climate scenarios are currently also in development, and those should cover the compound effect of physical risks already to some extent.

When is the no return point of an disorderly transition achieved?

There is no specific point where this happens. Rather, every year that effective transition is delayed, the transition becomes more disorderly (if the same goal is to be achieved).

Does low demand refer to only lower "energy" demand or also in general lower general consumption?

Low Demand is not a 'de-growth' scenario. Lower energy demand is achieved by changes in preferences that do not affect total demand (by assumption).

Is there an additional information on Frequency of Physical risk events and losses by region?

Physical risk damages are available on country-level (or sometimes region-level, depending on the exact variable you are looking for). The frequency of such hazards occurring is not covered.

Are increased fuel prices for example also a part of shadow carbon price?

It depends on the source of these price increases. If they are the result of fuel taxes, yes. If they are the results of geo-political turmoil, no.

What is the starting year for the new scenarios?

The starting point in the data is a result of the underlying models. For simplicity, consider the year of release as starting year. Scenario modelling/results considered infromation until march 2023 for the november release last year.

Do policy actions in Net Zero 2050 begin in 2024?


How are the impacts on GDP broken down by different sources (acute, chronic and transition)? Are there different variables that drive these impacts? what are the input variables behind these?

Yes, the GDP impacts are provided by NiGEM, with different inputs per climate risk type. Transition considers fossil fuel consumption, carbon price and revenue, as well as useful energy from the IAMs as input to the NiGEM output. Chronic takes into account temperature pathways from the IAMs that flow into the damage function and then feed into NiGEM. Acute is based on seperate modelling per hazard, but underlying temperature pathways define the severity levels for each of them. Cyclones and floods cause capital asset damages, heatwaves cause decreases in labour productivity, and drought cause crop yield declines.
A detailed description can be found in the Technical Documentation.

Which physical risks are being accounted for? How are the economic effects transmitted and estimated?

GDP damages are a result of different scenario narratives implemented in the IAMs feeding into NiGEM. As such, GDP damages presented in slide 21 are the sum of transition impacts (from the green transition) and physical impacts (from climate change). Physical risk considers chronic risk (based on a damage function) and acute risk (which considers damages from four natural hazards being cyclones, floods, heatwaves, droughts.

Are IAM on a global scale, appropriate for regional modelling?

IAM data is also provided in IAM-native regional clusters as well as in seperately downscaled country-level data. Those might be more applicable to your needs.

Is there a dictionnary for all the variables available in those scenarios (to get a precise definition of underlying variables) ?

We provide an overview of variables in the documentation in the IIASA Scenario Explorer
and in the EnTry Parameter Guide
where you can find an overvieew of all variables per category with corresponding units.

Carbon Prices are higher in Net Zero scenario for all scenario time horizon compared to Delayed Transition scenario so why GDP Impact in Delayed Transition is higher than in Net Zero scenario?

You may observe the higher damages in Delayed Transition because you are looking at overall GDP damages, which are driven by both transition and physical damages. The Net Zero 2050 scenario keeps global waming to 1.5°C, whereas the Delayed Transition models a global temperature increase up to 2°C, which results in higher physical risk damages. In general, Net Zero 2050 shows higher transition GDP damages than Delayed transition due to the higher carbon pric, but the total damage in Delayed Transition may be higher due to elevated physical risks.

NiGEM is based on historic data, Are the relationships based on historic data appropriate to be used for future climate scenarios?

Macro-models are indeed callibrated on historic data. However, the baseline used in our NiGEM runs is callibrated on the SSP2 pathway, which projects a number of macro variables decades into the future.

What are the confidence intervals of the results on GDP impact?

The macro-model does not provide confidence intervals on its output. There are only a few variables in the NGFS scenarios that come with confidence intervals. The temperature pathway of each scenario is one of the few variables for which multiple percentiles are available (ranging from the 5th to 95th). Output of the chronic damage function is also available for these various temperature percentiles, as well as a low, median and high damage estimate. As such, a confidence interval can be constructed for the chronic damages. Transition risk and acute physical risk impacts do not come with this feature

Are you producing separate scenarios for chronic and acute, or a combines one?

They are based on the same scenarios (specifically the respective temperature pathways). But are modelled differently, i.e. damage function versus nature catastrophe modelling.

What are Financial Institutions need to remember when using the NGFS scenarios?

NGFS Scenarios are not forecasts, not a final product rather a common reference framework.
Users often lacking knowledge on underlying assumptions and differences across IAMs or regional granularity. They should be used as a starting point for institutions' own analyses. The scenario set should not be considered an exhaustive analysis on itself. More details can be found in the recently published guidance note: https://www.ngfs.net/en/ngfs-guidance-note-scenarios

Is there a plan to get a more interactive sessions where small groups of participants can ask more technical questions

After the latest release we held a few data tutorial sessions, which we will likely repeat after this year's release.

Are you considering to include sectoral GVA figures and climate impacted fiscal variables (debt/GDP) in the upcoming releases?

A sectoral disagregation method is being developped, which will allow users to increase the sectoral granularity of the long-term scenarios. In addition, many more granular variables are foreseen to be part of the short-term scenarios, which will likely be published early 2025.

Can you please expand on what is being done towards improving sectoral granularity?

The proposed methodology will show a way to disaggregate the scenarios final energy series using main CPRS sectors to create sectoral GDP impacts by incorporating energy demand, prices as well as elasticities in an estimation with a CES production functions at sectoral level.

It seems counterintuitive that the Current Policies scenario has the most impact on GDP in the short term. Could you provide a quick explanation for that?

This is due to a modelling assumption on chronic physical risk. While transition impacts are quite well captured by the NGFS suite-of-models, we recognize that it is difficult to capture the full potential impact of climate physical risks. At higher warming levels, such as observed in the Current Policies scenario, damage estimates may vastly underestimate total losses due to great uncertainty about what could happen at such high warming levels. To account for this higher uncertainty, and in attempt not to underestimate losses, the choice was made to use a higher damage estimate for Current Policies (i.e., a higher percentile of the damage distribution) than for Orderly scenarios. This difference in modelling assumtions between the scenarios causes the small difference at the start of the time period.

Why do the differences in GDP plots appear quite large in the presented plot, however when comparing at a country/regional level we observe much lower differences?

Regional impacts always differ from the global levels. On regional/country level there are regions above and below the global estimation. We do not observe the described tendency.

Are the short term scenarios gonna be available for the upcoming EBA Stress Test 2025?

They are intended to be published early 2025 but adjusted depending on development and review process

The NGFS scenario portal provides results for the scenarios. can they also provide inputs for these?

The exogenous inputs per model used to produce are in detail described in the technical documentation. In short, IAMs are mainly based on the IPCC SSP2 scenario, which then flow into the other models of th emodelling suite, primarily GDP and temperature pathways.

While the technical documentation has good description, it would be great to see a matrix table where scenarios and actual variables shock are listed with information on shock magnitudes regarding how they differ across scenarios. Is such information easily available?

At the moment, the technical documentation is the most granular type of information publicly available. We have noted this, and will take it into consideration for future updates for the technical documentation.