## Q-CRAFT User Guide (Version 1.0_11-04-2024) — References and Key Elements ## Source details **Canonical URL:** [Q-CRAFT User Guide (Version 1.0_11-04-2024) — References and Key Elements](https://www.imf.org/-/media/files/topics/fiscal/fiscal-risks/tool/qcraft-user-guidev10.pdf) ## Other formats - [Markdown version](/-/media/files/topics/fiscal/fiscal-risks/tool/qcraft-user-guidev10.pdf.md) - [Structured JSON version](/-/media/files/topics/fiscal/fiscal-risks/tool/qcraft-user-guidev10.pdf.json) --- ### Purpose, scope, and coverage - Q-CRAFT supports governments in quantifying macroeconomic fiscal risks of climate change for 171 economies through 2099. - The tool projects a baseline macro-fiscal scenario for GDP and key fiscal indicators using a simple production function and a standard debt dynamic equation, grounded in available budgetary, economic, and demographic data with user-modifiable assumptions. - The analytical approach applies cross-country empirical estimates using the Kahn (2021) methodology to project temperature impacts on GDP per capita under different climate scenarios to 2099. - Q-CRAFT is a partial-equilibrium, stylized tool intended as a starting point for fiscal risk analysis; it is not a forecasting model nor a general equilibrium model. ### Acronyms - DSA Debt Sustainability Analysis - GDP Gross Domestic Product - GHG Greenhouse Gases - IMF International Monetary Fund - Q-CRAFT Quantitative Climate Risk Assessment Fiscal Tool - UN United Nations - WB World Bank - WDI World Development Indicators - WEO World Economic Outlook ### Key empirical findings and scenario benchmarks preserved from the literature used by Q-CRAFT - Kahn et al (2021) estimate global GDP per capita reductions from rising temperatures could be: - as low as one percent by the end of the century if all commitments made in the 2015 Paris Agreement are implemented; and - as high as 7 to 13 percent by the end of the century in pessimistic climate change scenarios relative to a baseline with historical-trend temperature increases. - Q-CRAFT uses updated empirical inputs from the FADCP Climate Dataset (Massetti and Tagklis, 2023), CRU data (Harris et al., 2020), CMIP6 data (Copernicus Climate Change Service, Climate Data Store, (2021): CMIP6 climate projections), and Centorrino, Massetti, and Tagklis (2024). ### Data inputs, horizons, and preloaded datasets - Q-CRAFT comes pre-loaded with: - UN population projections to 2100 (three demographic scenarios: medium, high, low; population data by age groups for 1950-2100); - IMF WEO projections (currently to 2028) for macro-fiscal variables (real GDP; nominal GDP; GDP deflator; revenue; expenditure; overall balance; primary balance; gross debt) for the period 2001-28 in the Macro-fiscal worksheet; - WB WDI historical data for labor productivity (productivity defined as GDP per employed person) through 2021. - The Q-CRAFT projection start year is 2029 (productivity “start” rate is the productivity growth calculated for 2028 and used as the 2029 start). - OECD productivity growth benchmark assumption: OECD productivity continues to grow at 1.1 percent per year (achieved between 1991-2022). ### Baseline construction and key modeling assumptions - Employment: - After the end of the IMF WEO horizon, employment is assumed to grow by the projected growth in working-age (15-64 year) population. - Nominal GDP: - Nominal GDP growth is approximated as the sum of employment growth, productivity growth, and inflation. - Government revenue: - Assumed to remain constant as a share of nominal GDP after the WEO horizon. - Primary expenditure: - Excludes government interest payments and is assumed to grow by productivity, inflation, and total population after the WEO horizon (thus changing as a share of GDP). - Debt projection: - Q-CRAFT projects the debt-to-GDP ratio to 2099 using the debt dynamics equation starting from WEO-based initial debt. - Productivity trajectory: - Users enter a structural long-run productivity growth expectation (end period, 2090 to 2100). Q-CRAFT derives an annual productivity growth path from the 2029 start rate to the end rate using a linear trajectory and a logistic-based convergence of productivity level relative to OECD (parameters: ‘Rate’ set at 0.5; ‘Turning Point’ adjustable by user). - Inflation assumptions: - Users should reflect Central Bank targets; typical ranges cited: - 2-3 percent in advanced economies; - 3-6 percent in emerging and developing economies. - If WEO projection in 2028 is in target range, use it as both ‘start’ and ‘end’; otherwise, use WEO 2028 as ‘start’ and the Central Bank target as ‘end’. - Interest rate options (user choice): - Constant nominal interest rate from 2029; - Constant (nominal) interest growth differential from 2029; - Constant real interest rate from 2029 (user must set long-term real rate, e.g., r-star). ### User operations and spreadsheet management rules - Worksheet color conventions: - Green “Dashboard” for parameter selection; - Blue worksheets contain pre-loaded data (IMF, UN, WB) and user-replaceable inputs; - Red worksheets perform calculations and should not be changed; - Yellow “Output” worksheets show results and should not be changed; - Blank “Read me” worksheet provides operational information. - Cell color coding: - Blue digits = hard-coded data from external sources; - Green digits = links between worksheets; - Black digits = calculations. - Users can hard-paste their own data into blue or shaded cells to override preloaded inputs (note: hard pasting deletes original Q-CRAFT data and cannot be undone). - Q-CRAFT can be aligned with IMF-WB Debt Sustainability Analysis projections by hard-pasting extended projections and adjusting Productivity, Interest Rate, Inflation, and Baseline worksheets accordingly. ### Limitations, caveats, and conservatism of results - Q-CRAFT’s results are stylized and conservative because the tool: - Does not account for climate change–induced natural disasters, sea-level rise risks, tipping points, non-market damages (e.g., morbidity and mortality, conflicts, food insecurity), or spillovers; - Does not model discrete contingent liabilities endogenously (option to manually add such fiscal impacts is provided); - Does not generate the fiscal impact of climate change adaptation spending; - Relies on historical relationships between temperature and economic outcomes; inference to 2099 involves model and deep scenario uncertainty. - The guide emphasizes the need for country expertise to produce a baseline consistent with national macro-fiscal frameworks and other models (e.g., medium-term fiscal frameworks, debt sustainability analysis, general equilibrium models). ### Practical guidance for users - Select country and demographic scenario in the Dashboard to auto-load WEO and UN data. - Form realistic productivity assumptions: Q-CRAFT provides historical averages (1991-2022; 2007-19; 2014-19) and group comparators to inform start and end productivity growth choices. - For productivity data gaps, consider substituting productivity levels of similar economies; the guide lists economies lacking WB WDI productivity data (Andorra; Antigua and Barbuda; Aruba; Dominica; Eritrea; Grenada; Kiribati; Kosovo; Marshall Islands; Micronesia; Nauru; Palau; San Marino; Seychelles; South Sudan; St Kitts and Nevis; Syria; Taiwan Province of China; Tuvalu; Yemen). - Use country-specific data where available (e.g., medium-term fiscal frameworks, DSA projections, national demographic projections) to customize outputs. - Advanced users should account for demographic impacts on interest-growth differentials and consider whether a constant differential assumption is appropriate in the presence of shrinking populations. --- ### Section IV — Further discussion ### Debt dynamics and debt-stabilizing primary balance - Q-CRAFT implements the debt dynamics equation to trace how interest rates, nominal GDP growth, and the primary balance determine the trajectory of public debt. - Key intuitions from the debt-dynamics relationship: - When the rate of economic growth is lower than the interest rate, the debt-to-GDP ratio tends to increase; when economic growth is higher than the interest rate, the debt-to-GDP ratio tends to decrease. - A primary surplus reduces the debt-to-GDP ratio and a primary deficit raises it, ceteris paribus. - Declining economic output from climate change can exacerbate unfavorable debt dynamics, potentially converting otherwise favorable dynamics into unfavorable ones. - Q-CRAFT calculates the debt-stabilizing primary balance — the primary balance necessary to maintain stable debt levels — for every year to 2099. - Use of the debt-stabilizing calculation: - Assess whether current fiscal policy is consistent with a stable debt-to-GDP ratio. - Quantify the budgetary effort required to achieve a stable debt ratio. - Note: Q-CRAFT is partial equilibrium and does not estimate GDP effects of fiscal adjustments required to meet a debt-to-GDP target. - Stable-inflation assumption: - Conceptually implies a fully rational household sector expects authorities to comply with the fiscal rule and the central bank to maintain macroeconomic stability; households smooth consumption, preventing negative consumption effects from consolidation. ### Fiscal rule for debt-to-GDP - To compare the debt-to-GDP evolution with a fiscal-rule target users must: - Initially select “no” and enter 0 for the debt target to run without a fiscal-rule target. - Select “yes” and enter the desired debt-to-GDP level in the dashboard worksheet to apply a fiscal-rule debt target. ### Climate change scenarios simulated - Q-CRAFT simulates fiscal consequences under six climate scenarios (advanced-user details in Section IV B): - Paris: based on SSP1-2.6, meeting 2015 Paris commitments; keeps global temperature increase below 2°C at the end of the century. - Moderate: based on SSP2-4.5; emissions follow present trends and stabilize at the end of the century; temperature increase similar to the 1960-2014 trend. - High: based on SSP3-7.0; continued use of fossil fuels and limited energy-efficiency improvements. - Hot: emissions as in “high” but uses the 90th percentile of temperature increases among models using SSP3-7.0 instead of the median. - Hot adapted: same temperatures as “hot” but countries adapt more quickly. - Hot un-adapted: same temperatures as “hot” but countries adopt very slowly to climate change. ### Macro-fiscal effects of climate change (assumptions and mechanics) - Q-CRAFT uses empirical estimates of slow-moving warming impacts on GDP per capita growth from Kahn et al (2021) as generated by the FAD Climate Policy Division; estimates are provided for 171 economies in the Climate Data-worksheet. - Q-CRAFT assumes climate-driven fiscal effects start in 2030 to isolate long-term impacts from near- and medium-term macro shocks. - Methodological assumptions reflected in scenario outputs: - Revenue-to-GDP ratio: remains the same as the baseline in each climate scenario; because GDP falls under climate scenarios, government revenue is lower in level terms relative to baseline. - Primary expenditure: initially assumed rigid at the baseline level in local currency terms; as GDP declines, the primary expenditure-to-GDP ratio rises. - Consequence: reduced revenue with unchanged primary expenditure worsens the primary balance, raises debt, and can drive the debt-to-GDP ratio much higher than the baseline—even if baseline debt-to-GDP is stable. ### Expenditure Rigidity parameter - The dashboard allows users to set an Expenditure Rigidity parameter between 0 and 1: - Value of 1: government takes a passive stance and does not adjust expenditure in response to decreasing revenues (primary expenditure unchanged in local currency terms). - Value of 0: government reprioritizes expenditure to maintain the same primary expenditure-to-GDP ratio as in the baseline. - Q-CRAFT does not incorporate potential GDP effects of any expenditure cuts chosen by the user. ### Discrete fiscal risks and natural disasters - Users can register materializations of discrete fiscal risks or natural-disaster impacts as rises in primary expenditure or losses of revenue (entered as percent of GDP in the Discrete Risks-worksheet). - Discrete risks are generally idiosyncratic and should reflect country-specific historical experience (frequency and typical fiscal impacts) and expectations that likelihood and impact rise with worse climate scenarios. - Inputs for discrete risks affect revenues and expenditures only; Q-CRAFT does not factor any potential impact of these discrete risks on GDP growth. - Illustrative hypothetical country example pattern: - Historical pattern: risks materialized once a decade, producing a one-off 0.5 percent of GDP rise in primary expenditure in the impact year and a one-off 0.5 percent of GDP revenue loss in the following year. - Paris scenario: same pattern assumed, with the first impact year assumed to be 2030. - Moderate scenario: same likelihood as Paris but fiscal impacts rise to 0.7 percent of GDP (first impact year 2030). - High scenario: risks materialize once every five years with 0.7 percent of GDP fiscal impacts (first impact year 2030). - Hot scenario: risks materialize once every five years with 1 percent of GDP fiscal impacts. - Hot adapted scenario: risks materialize once every five years with 0.8 percent of GDP fiscal impacts. ### Illustrative example — Aurelium - Aurelium: fictional advanced economy used to demonstrate Q-CRAFT mechanics; prone to natural disasters expected to worsen with climate change. - Baseline dashboard assumptions for Aurelium: - A high demography scenario. - Constant productivity growth rate of 1.25 percent a year. - Constant inflation assumption of 2.75 percent. - Expenditure rigidity of 1. - Constant nominal interest rate assumption at the end of the WEO horizon. - No fiscal rule. - Baseline projections and dynamics: - Long-term nominal GDP growth remains relatively stable at around 4.5 percent a year. - Revenue-to-GDP ratio projected to remain constant. - Primary expenditure projected to grow from mid-century by productivity, inflation, and total population, widening primary and overall deficits and raising debt-to-GDP. - Climate scenario impacts for Aurelium: - In the worst-case scenario, Aurelium might lose nearly 6 percent of GDP relative to the baseline. - Nominal GDP level and growth rates decline under climate scenarios, lowering government revenue expressed as percent of nominal GDP while primary expenditure remains rigid. - By the end of the century, climate change could increase Aurelium’s debt by 60 percent of GDP relative to the baseline. - Registering discrete fiscal risks worsens the fiscal outlook: one-off events add to primary deficits on impact, raising debt and interest expenditure in subsequent years and leaving lasting fiscal footprints. --- ### Methodology: baseline macro-fiscal framework and supporting elements ### Foundations and nominal GDP decomposition - Q-CRAFT foundations: - Macro projections rely on a simple production-function framework and empirical temperature–economy relationships for climate impacts. - Fiscal projections use the debt-dynamics equation. - Nominal GDP decomposition and projection: - Nominal GDP N = Y * P, where Y is real GDP and P is the GDP deflator. - Real GDP Y is assumed to follow Y = A * L, where L is employment and A captures factors affecting labor productivity (land productivity; physical and human capital; innovation and technological progress; sectoral composition; resource endowment; history and institutions). - Growth in nominal GDP is approximately the sum of growth in productivity, employment, and the GDP deflator. ### Demography, employment, and productivity - Demography and employment: - Demography is the key driver of long-run employment growth. - Q-CRAFT projects employment to grow in line with the UN’s working-age population projections (age range 15-64 years old) under medium, high, and low scenarios from 2029 onwards. - Q-CRAFT assumes the ratio of employment-to-working-age population stabilizes by the end of the medium-term covered in the IMF WEO (2028 currently). - Box 2 — Demography: - In each demographic scenario, the share of the elderly (over 65) in the total population is expected to rise and the growth in working-age population is expected to slow globally over the century. - Demographic projections vary considerably across countries: some already have stagnating or declining working-age populations; others have growth yet to peak. - Policies (retirement age, pensions, investment in human capital) determine whether a demographic dividend materializes. - Productivity measurement and projection: - Q-CRAFT uses GDP per employed person as productivity. - WB historical labor productivity data cover 1991-2021; Q-CRAFT calculates average productivity growth for 1991-2022; 2007-19; and 2014-19. - Users must assume a realistic end-period structural labor productivity growth rate by the end of the century; Q-CRAFT computes implied productivity level relative to the OECD assuming OECD productivity continues to grow at 1.2 percent. ### Inflation, revenue, primary expenditure, and "no policy change" - Inflation and nominal GDP assumptions: - Inflation is assumed to be stable in line with the Central Bank’s target in the long run. - GDP deflator growth and consumer price inflation can diverge in the near- to medium-term due to terms of trade shocks; Q-CRAFT abstracts from exchange rate adjustments and external-sector feedbacks. - Revenue and primary expenditure under “no policy change”: - Revenue is assumed to remain constant relative to nominal GDP in the long run. - From the end of the WEO-covered period, revenue grows by: the assumed inflation; the projected productivity growth; and the projected employment growth (UN working-age population growth). - Real primary expenditure per capita grows in line with productivity, inflation, and total population. - From the end of the WEO-covered period, primary expenditure grows by: the assumed inflation; the projected productivity growth; and the UN projected growth in total population (not working-age population). - Demographic composition effects: - If working-age population grows faster than total population, primary expenditure will fall relative to GDP. - If working-age population grows slower than total population, the ratio of primary expenditure to GDP will rise. - Fiscal-rule adjustment mechanics: - If a debt ceiling is used and reached at time T, Q-CRAFT calculates the fiscal gap to stabilize debt; primary expenditure at T+1 is reduced (or increased) by the fiscal gap; the process repeats over the horizon. Fiscal adjustment lags imply the debt ceiling target is never precisely achieved. - Q-CRAFT does not include feedback from fiscal consolidation on GDP growth. ### Interest rates and debt dynamics - Interest-rate considerations: - Long-term fiscal projections are highly sensitive to interest rate assumptions; uncertainty is considerable. - Advanced-economy interest rates have been declining for decades; despite recent post-pandemic increases, there are grounds to expect them to remain low into the future. - For developing economies, convergence to OECD productivity implies reduced access to concessional lending—upward pressure on interest rates—while more developed financial markets could increase domestic savings and lower domestic interest rates. - In the long run, real interest rates tend to track productivity growth; convergence to slower productivity growth could imply declining interest rates. - Q-CRAFT interest-rate input options: - Users choose between constant nominal interest rate; (nominal) interest growth differential; or real interest rate. - Debt-dynamics equation used by Q-CRAFT: - D_{t+1} = D_t * [(1+ i) / (1+g)] – pb_t - Variables: nominal GDP growth g; nominal interest rate i; primary balance (percent of nominal GDP) pb; debt-to-GDP ratio D. - Key corollaries: - Larger primary surplus or smaller primary deficit lowers debt-to-GDP; higher initial debt raises debt-to-GDP; higher interest rates raise debt-to-GDP; faster GDP growth lowers debt-to-GDP. - The primary balance is the variable governments can control in the near- to medium-term. - The interest rate–growth differential (i – g) represents “automatic debt dynamics”; for i > g, dynamics are unfavorable; for i < g, favorable. - Q-CRAFT calculates the debt-stabilizing primary balance each year; differences between projected and debt-stabilizing balances quantify the fiscal adjustment task or fiscal space. --- ### Climate scenarios, Kahn et al (2021) summary, and policy implications ### Climate scenarios overview - Q-CRAFT focuses on five IPCC-based climate scenarios plus “Hot adapted” and “Hot Un-Adapted,” projecting global mean temperature increase between 0.7 and 3.5°C by the end of the century with respect to present levels. - Scenario descriptions (synopsis): - Paris: SSP1-2.6, keeps global temperature increase below 2°C at the end of the century. - Moderate: SSP2-4.5; emissions increase in line with current trends and stabilize at the end of the century. - High: SSP3-7.0; high-emissions scenario with scaled-back mitigation. - Hot: same emissions as High but uses the 90th percentile of temperature increase among models using SSP3-7.0. - Hot adapted: same temperatures as Hot, but countries adapt more quickly. - Hot Un-Adapted: same temperatures as Hot, but countries adapt more slowly. ### Box 4 — Summary of Kahn et al (2021) (methodology and key findings) - Methodology: - Links deviations of temperature and precipitation from long-term historical norms to growth in real GDP per capita. - Estimated using data from 174 countries over 1960-2014, allowing for dynamics, nonlinearity, an implicit adaptation model, and distributional changes in weather patterns. - Key empirical findings: - Per-capita real output growth is adversely affected by persistent changes in temperature above or below its historical norm. - A persistent increase in average global temperature by 0.04 degrees Celsius per year, in the absence of mitigation policies, reduces world real GDP per capita by around 7 percent by 2100 compared to a 1960-2014 trend baseline. - Losses increase to 13 percent globally if country-specific variability of climate conditions were to rise commensurate with annual temperature increases of 0.04 degrees Celsius. - Abiding by the Paris Agreement goals, limiting temperature increase to 0.01 degrees Celsius per year, reduces the loss substantially to about 1 percent. - Adaptation can reduce long-run growth effects but is highly unlikely to fully offset them. - Q-CRAFT captures climate effects on real GDP as a slowdown in productivity growth; inflation is assumed unchanged across climate scenarios. ### Fiscal effects and policy-relevant implications - Fiscal transmission of climate-driven GDP declines in Q-CRAFT: - Revenue: declines in level terms as nominal GDP falls; revenue-to-GDP ratio stays at baseline. - Primary expenditure: held at baseline local currency levels unless the Expenditure Rigidity parameter adjusts spending; primary expenditure-to-GDP ratio rises as GDP falls. - Result: worse primary and overall balances, higher debt-to-GDP, and higher interest expenditure under adverse climate scenarios. - Adaptation parameter and scenarios: - Kahn et al (2021) embed an adaptation parameter m (default 30-year adaptation); Q-CRAFT allows adjustments. - ‘Hot Adapted’ sets m = 20; ‘Hot Un-Adapted’ sets m = 50. - Faster adaptation reduces macroeconomic impacts but does not fully offset damages; dataset lacks estimates of adaptation spending. - Policy-relevant implications: - Mitigation reduces long-run GDP losses: limiting temperature increase to 0.01 degrees Celsius per year (Paris goals) substantially reduces global GDP losses compared to 0.04 degrees Celsius per year. - Adaptation matters but is not a full offset. - Maintaining rigid primary expenditure in the face of GDP declines raises primary expenditure-to-GDP ratios and worsens fiscal balances and debt dynamics. - Introducing expenditure flexibility (Expenditure Rigidity parameter toward 0) can restore expenditure-to-GDP ratios but implies lower real primary expenditure per capita and difficult reallocations. - Need to account for discrete fiscal shocks: more frequent and larger natural disasters and contingent liabilities can raise primary expenditure or lower revenue beyond econometric scenario estimates, reducing fiscal space. --- *Source: Q-CRAFT User Guide, Version 1.0_11-04-2024, IMF.* ### References ............................................................................................................. ### Q-CRAFT User Guide (Version 1.0_11-04-2024) — References and Key Elements ### Acronyms - DSA Debt Sustainability Analysis - GDP Gross Domestic Product - GHG Greenhouse Gases - IMF International Monetary Fund - Q-CRAFT Quantitative Climate Risk Assessment Fiscal Tool - UN United Nations - WB World Bank - WDI World Development Indicators - WEO World Economic Outlook ### Purpose, scope, and coverage - Q-CRAFT supports governments in quantifying macroeconomic fiscal risks of climate change for 171 economies through 2099. - The tool projects a baseline macro-fiscal scenario for GDP and key fiscal indicators using a simple production function and a standard debt dynamic equation, grounded in available budgetary, economic, and demographic data with user-modifiable assumptions. - The analytical approach applies cross-country empirical estimates using the Kahn (2021) methodology to project temperature impacts on GDP per capita under different climate scenarios to 2099. - Q-CRAFT is a partial-equilibrium, stylized tool intended as a starting point for fiscal risk analysis; it is not a forecasting model nor a general equilibrium model. ### Key empirical findings and scenario benchmarks preserved from the literature used by Q-CRAFT - Kahn et al (2021) estimate global GDP per capita reductions from rising temperatures could be: - as low as one percent by the end of the century if all commitments made in the 2015 Paris Agreement are implemented; and - as high as 7 to 13 percent by the end of the century in pessimistic climate change scenarios relative to a baseline with historical-trend temperature increases. - Q-CRAFT uses updated empirical inputs from the FADCP Climate Dataset (Massetti and Tagklis, 2023), CRU data (Harris et al., 2020), CMIP6 data (Copernicus Climate Change Service, Climate Data Store, (2021): CMIP6 climate projections), and Centorrino, Massetti, and Tagklis (2024). ### Data inputs, horizons, and preloaded datasets - Q-CRAFT comes pre-loaded with: - UN population projections to 2100 (three demographic scenarios: medium, high, low; population data by age groups for 1950-2100); - IMF WEO projections (currently to 2028) for macro-fiscal variables (real GDP; nominal GDP; GDP deflator; revenue; expenditure; overall balance; primary balance; gross debt) for the period 2001-28 in the Macro-fiscal worksheet; - WB WDI historical data for labor productivity (productivity defined as GDP per employed person) through 2021. - The Q-CRAFT projection start year is 2029 (productivity “start” rate is the productivity growth calculated for 2028 and used as the 2029 start). - OECD productivity growth benchmark assumption: OECD productivity continues to grow at 1.1 percent per year (achieved between 1991-2022). ### Baseline construction and key modeling assumptions - Employment: - After the end of the IMF WEO horizon, employment is assumed to grow by the projected growth in working-age (15-64 year) population. - Nominal GDP: - Nominal GDP growth is approximated as the sum of employment growth, productivity growth, and inflation. - Government revenue: - Assumed to remain constant as a share of nominal GDP after the WEO horizon. - Primary expenditure: - Excludes government interest payments and is assumed to grow by productivity, inflation, and total population after the WEO horizon (thus changing as a share of GDP). - Debt projection: - Q-CRAFT projects the debt-to-GDP ratio to 2099 using the debt dynamics equation starting from WEO-based initial debt. - Productivity trajectory: - Users enter a structural long-run productivity growth expectation (end period, 2090 to 2100). Q-CRAFT derives an annual productivity growth path from the 2029 start rate to the end rate using a linear trajectory and a logistic-based convergence of productivity level relative to OECD (parameters: ‘Rate’ set at 0.5; ‘Turning Point’ adjustable by user). - Inflation assumptions: - Users should reflect Central Bank targets; typical ranges cited: - 2-3 percent in advanced economies; - 3-6 percent in emerging and developing economies. - If WEO projection in 2028 is in target range, use it as both ‘start’ and ‘end’; otherwise, use WEO 2028 as ‘start’ and the Central Bank target as ‘end’. - Interest rate options (user choice): - Constant nominal interest rate from 2029; - Constant (nominal) interest growth differential from 2029; - Constant real interest rate from 2029 (user must set long-term real rate, e.g., r-star). ### User operations and spreadsheet management rules - Worksheet color conventions: - Green “Dashboard” for parameter selection; - Blue worksheets contain pre-loaded data (IMF, UN, WB) and user-replaceable inputs; - Red worksheets perform calculations and should not be changed; - Yellow “Output” worksheets show results and should not be changed; - Blank “Read me” worksheet provides operational information. - Cell color coding: - Blue digits = hard-coded data from external sources; - Green digits = links between worksheets; - Black digits = calculations. - Users can hard-paste their own data into blue or shaded cells to override preloaded inputs (note: hard pasting deletes original Q-CRAFT data and cannot be undone). - Q-CRAFT can be aligned with IMF-WB Debt Sustainability Analysis projections by hard-pasting extended projections and adjusting Productivity, Interest Rate, Inflation, and Baseline worksheets accordingly. ### Limitations, caveats, and conservatism of results - Q-CRAFT’s results are stylized and conservative because the tool: - Does not account for climate change–induced natural disasters, sea-level rise risks, tipping points, non-market damages (e.g., morbidity and mortality, conflicts, food insecurity), or spillovers; - Does not model discrete contingent liabilities endogenously (option to manually add such fiscal impacts is provided); - Does not generate the fiscal impact of climate change adaptation spending; - Relies on historical relationships between temperature and economic outcomes; inference to 2099 involves model and deep scenario uncertainty. - The guide emphasizes the need for country expertise to produce a baseline consistent with national macro-fiscal frameworks and other models (e.g., medium-term fiscal frameworks, debt sustainability analysis, general equilibrium models). ### Practical guidance for users - Select country and demographic scenario in the Dashboard to auto-load WEO and UN data. - Form realistic productivity assumptions: Q-CRAFT provides historical averages (1991-2022; 2007-19; 2014-19) and group comparators to inform start and end productivity growth choices. - For productivity data gaps, consider substituting productivity levels of similar economies; the guide lists economies lacking WB WDI productivity data (Andorra; Antigua and Barbuda; Aruba; Dominica; Eritrea; Grenada; Kiribati; Kosovo; Marshall Islands; Micronesia; Nauru; Palau; San Marino; Seychelles; South Sudan; St Kitts and Nevis; Syria; Taiwan Province of China; Tuvalu; Yemen). - Use country-specific data where available (e.g., medium-term fiscal frameworks, DSA projections, national demographic projections) to customize outputs. - Advanced users should account for demographic impacts on interest-growth differentials and consider whether a constant differential assumption is appropriate in the presence of shrinking populations. *Source: Q-CRAFT User Guide, Version 1.0_11-04-2024, IMF.* ### Section IV for further discussion. ### Section IV — Further discussion ### Debt dynamics and debt-stabilizing primary balance - Q-CRAFT implements the debt dynamics equation to trace how interest rates, nominal GDP growth, and the primary balance determine the trajectory of public debt. - Key intuitions from the debt-dynamics relationship: - When the rate of economic growth is lower than the interest rate, the debt-to-GDP ratio tends to increase; when economic growth is higher than the interest rate, the debt-to-GDP ratio tends to decrease. - A primary surplus reduces the debt-to-GDP ratio and a primary deficit raises it, ceteris paribus. - Declining economic output from climate change can exacerbate unfavorable debt dynamics, potentially converting otherwise favorable dynamics into unfavorable ones. - Q-CRAFT calculates the debt-stabilizing primary balance — the primary balance necessary to maintain stable debt levels — for every year to 2099. - Use of the debt-stabilizing calculation: - Assess whether current fiscal policy is consistent with a stable debt-to-GDP ratio. - Quantify the budgetary effort required to achieve a stable debt ratio. - Note: Further technical details of the debt dynamics equation are provided for advanced users in Section IV. *The tool’s partial-equilibrium framing and stable-inflation assumption* - Q-CRAFT is a partial equilibrium model and does not estimate GDP effects of the fiscal adjustments required to meet a debt-to-GDP target. - The stable inflation assumption conceptually implies a fully rational household sector expects authorities to comply with the fiscal rule and the central bank to maintain macroeconomic stability; households smooth consumption, preventing negative consumption effects from consolidation. ### Fiscal rule for debt-to-GDP - To compare the debt-to-GDP evolution with a fiscal-rule target users must: - Initially select “no” and enter 0 for the debt target to run without a fiscal-rule target. - Select “yes” and enter the desired debt-to-GDP level in the dashboard worksheet to apply a fiscal-rule debt target. ### Climate change scenarios simulated - Q-CRAFT simulates fiscal consequences under six climate scenarios (with advanced-user details in Section IV B): - Paris: based on SSP1-2.6, meeting 2015 Paris commitments; keeps global temperature increase below 2°C at the end of the century. - Moderate: based on SSP2-4.5; emissions follow present trends and stabilize at the end of the century; temperature increase similar to the 1960-2014 trend. - High: based on SSP3-7.0; continued use of fossil fuels and limited energy-efficiency improvements. - Hot: emissions as in “high” but uses the 90th percentile of temperature increases among models using SSP3-7.0 instead of the median. - Hot adapted: same temperatures as “hot” but countries adapt more quickly. - Hot un-adapted: same temperatures as “hot” but countries adopt very slowly to climate change. ### Macro-fiscal effects of climate change - Q-CRAFT uses empirical estimates of slow-moving warming impacts on GDP per capita growth from Kahn et al (2021) as generated by the FAD Climate Policy Division; estimates are provided for 171 economies in the Climate Data-worksheet. - Q-CRAFT assumes climate-driven fiscal effects start in 2030 to isolate long-term impacts from near- and medium-term macro shocks. - Methodological assumptions reflected in scenario outputs: - Revenue-to-GDP ratio: remains the same as the baseline in each climate scenario; because GDP falls under climate scenarios, government revenue is lower in level terms relative to baseline. - Primary expenditure: initially assumed rigid at the baseline level in local currency terms; as GDP declines, the primary expenditure-to-GDP ratio rises. - Consequence: reduced revenue with unchanged primary expenditure worsens the primary balance, raises debt, and can drive the debt-to-GDP ratio much higher than the baseline—even if baseline debt-to-GDP is stable. ### Expenditure Rigidity parameter - The dashboard allows users to set an Expenditure Rigidity parameter between 0 and 1: - Value of 1: government takes a passive stance and does not adjust expenditure in response to decreasing revenues (primary expenditure unchanged in local currency terms). - Value of 0: government reprioritizes expenditure to maintain the same primary expenditure-to-GDP ratio as in the baseline. - Q-CRAFT does not incorporate potential GDP effects of any expenditure cuts chosen by the user. ### Discrete fiscal risks and natural disasters - Users can register materializations of discrete fiscal risks or natural-disaster impacts as rises in primary expenditure or losses of revenue (entered as percent of GDP in the Discrete Risks-worksheet). - Discrete risks are generally idiosyncratic and should reflect country-specific historical experience (frequency and typical fiscal impacts) and expectations that likelihood and impact rise with worse climate scenarios. - Inputs for discrete risks affect revenues and expenditures only; Q-CRAFT does not factor any potential impact of these discrete risks on GDP growth. - Illustrative hypothetical country example (pattern used in Figure 9): - Historical pattern: risks materialized once a decade, producing a one-off 0.5 percent of GDP rise in primary expenditure in the impact year and a one-off 0.5 percent of GDP revenue loss in the following year. - Paris scenario: same pattern assumed, with the first impact year assumed to be 2030. - Moderate scenario: same likelihood as Paris but fiscal impacts rise to 0.7 percent of GDP (first impact year 2030). - High scenario: risks materialize once every five years with 0.7 percent of GDP fiscal impacts (first impact year 2030). - Hot scenario: risks materialize once every five years with 1 percent of GDP fiscal impacts. - Hot adapted scenario: risks materialize once every five years with 0.8 percent of GDP fiscal impacts. ### Illustrative example — Aurelium - Aurelium: fictional advanced economy used to demonstrate Q-CRAFT mechanics; prone to natural disasters expected to worsen with climate change. - Baseline dashboard assumptions for Aurelium: - A high demography scenario. - Constant productivity growth rate of 1.25 percent a year. - Constant inflation assumption of 2.75 percent. - Expenditure rigidity of 1. - Constant nominal interest rate assumption at the end of the WEO horizon. - No fiscal rule. - Baseline projections and dynamics: - Long-term nominal GDP growth remains relatively stable at around 4.5 percent a year. - Revenue-to-GDP ratio projected to remain constant. - Primary expenditure projected to grow from mid-century by productivity, inflation, and total population, widening primary and overall deficits and raising debt-to-GDP. - Climate scenario impacts for Aurelium: - In the worst-case scenario, Aurelium might lose nearly 6 percent of GDP relative to the baseline. - Nominal GDP level and growth rates decline under climate scenarios, lowering government revenue expressed as percent of nominal GDP while primary expenditure remains rigid. - By the end of the century, climate change could increase Aurelium’s debt by 60 percent of GDP relative to the baseline. - Registering discrete fiscal risks worsens the fiscal outlook: one-off events add to primary deficits on impact, raising debt and interest expenditure in subsequent years and leaving lasting fiscal footprints. ### Methodology: baseline macro-fiscal framework - Q-CRAFT foundations: - Macro projections rely on a simple production-function framework and empirical temperature–economy relationships for climate impacts. - Fiscal projections use the debt-dynamics equation. - Nominal GDP decomposition and projection: - Nominal GDP N = Y * P, where Y is real GDP and P is the GDP deflator. - Real GDP Y is assumed to follow Y = A * L, where L is employment and A captures factors affecting labor productivity (land productivity; physical and human capital; innovation and technological progress; sectoral composition; resource endowment; history and institutions). - Growth in nominal GDP is approximately the sum of growth in productivity, employment, and the GDP deflator. - Demography and employment: - Demography is the key driver of long-run employment growth. - Short- and medium-term employment growth can be influenced by business-cycle conditions and labor force participation; informality complicates labor-market analysis in developing countries. - Q-CRAFT projects employment to grow in line with the UN’s working-age population projections (age range 15-64 years old) under medium, high, and low scenarios from 2029 onwards. - Q-CRAFT assumes the ratio of employment-to-working-age population stabilizes by the end of the medium-term covered in the IMF WEO (2028 currently). *Source: IMF staff (Q-CRAFT user guide, Section IV).* ### Box 2. Long-Term Macro-Fiscal Effects of Demography ### Box 2. Long-Term Macro-Fiscal Effects of Demography ### Demographic scenarios and macroeconomic implications - In each of the three demographic scenarios used in Q-CRAFT, the share of the elderly (over 65) in the total population is expected to rise and the growth in working-age population is expected to slow globally over the century. - These demographic projections mechanically weigh down global economic growth by slowing employment growth. - Demographic projections vary considerably across countries, implying important country-specific policy implications: - Some countries are already experiencing stagnation or a decline in working-age population. - Other countries have growth in working-age population that is yet to peak. - Demographic drag and demographic dividend are not automatic: - Changes to retirement age or pensions can induce the elderly to remain in the labor market, delaying or lessening the demographic drag. - Reaping a demographic dividend requires supportive microeconomic policies and investment in human capital. ### Fiscal implications of demographic change - For developing countries, achieving a demographic dividend and strong productivity growth to converge toward advanced economies will likely have fiscal implications. - Ageing societies create fiscal risks: - Public finances are likely to worsen with an ageing population. - Expenditures on health, aged care, and pensions are likely to rise. - Ageing may erode a country’s revenue base, even as revenues tend to rise relative to GDP as a country develops. - Source attribution for the Box: Amaglobeli, D., and Shi, W. (2016), “How to assess fiscal implications of demographic shifts: a granular approach, IMF.” ### Productivity: measurement, projection, and realism checks - Q-CRAFT uses the broadest measure of productivity: GDP (output) per employed person (input). - Historical productivity data and default Q-CRAFT calculations: - The WB publishes historical labor productivity data for 1991-2021. - Q-CRAFT automatically calculates average productivity growth using World Bank data for: 1991-2022; 2007-19; and 2014-19 for reference countries and benchmarks. - Users must assume a realistic end-period structural labor productivity growth rate by the end of the century; Q-CRAFT assists by computing historical productivity growth by World Bank country groupings (1991-2021) for benchmarking. - Realism check relative to OECD: - Q-CRAFT computes implied productivity level relative to the OECD assuming OECD productivity continues to grow at the historical average rate of 1.2 percent. - As a general guide, it may not be realistic for a currently developing or emerging economy to become significantly more productive than the OECD average by the end of the century. ### Inflation and nominal GDP assumptions - Nominal GDP projections are required for fiscal analysis; Q-CRAFT uses the GDP deflator to derive nominal GDP. - Inflation assumptions: - Inflation is assumed to be stable in line with the Central Bank’s target in the long run. - Q-CRAFT assumes countries will have stable inflation consistent with the Central Bank’s target over the long term. - GDP deflator growth and consumer price inflation can diverge in the near- to medium-term due to terms of trade shocks; Q-CRAFT abstracts from exchange rate adjustments and external-sector feedbacks. ### Revenue, primary expenditure, and primary balance under “no policy change” - Revenue: - Revenue is assumed to remain constant relative to nominal GDP in the long run. - From the end of the WEO-covered period, revenue grows by: the assumed inflation; the projected productivity growth; and the projected employment growth (which equals the UN’s projection for growth in working-age population). - Primary expenditure: - In the long run, real primary expenditure per capita grows in line with productivity, inflation, and total population. - From the end of the WEO-covered period, primary expenditure grows by: the assumed inflation; the projected productivity growth; and the UN projected growth in total population (not working-age population). - Intuition: - Nominal primary expenditure G is the basket of goods and services government provides. - Growing G in line with inflation and total population reflects a ‘no policy change’ assumption that the real value of G per citizen is unchanged. - G also benefits from economy-wide productivity gains and is assumed to grow in line with productivity. - Demographic composition effects on primary expenditure ratio to GDP: - If working-age population grows faster than total population (a demographic dividend), primary expenditure will fall relative to GDP over time. - If working-age population grows slower than total population (demographic drag from ageing), the ratio of primary expenditure to GDP will rise. - Using the baseline “no policy change” projection: - Widening long-term primary deficits indicate an unsustainable fiscal trajectory. - Widening long-term primary surpluses imply the country never uses its fiscal space, which is not credible over the long run. - Fiscal financing and fiscal-rule adjustments: - Q-CRAFT abstracts from financing constraints and assumes budget financing consistent with macroeconomic stability. - Default fiscal rule assumption is ‘none’. If debt-to-GDP is not credible, the user can apply a ‘debt ceiling’ assumption to stabilize debt at a chosen percent of GDP by cutting primary expenditure. - Mechanics when debt ceiling (floor) is used: - Suppose the debt ceiling (floor) is reached at time T; - Q-CRAFT calculates the fiscal gap needed to stabilize debt relative to GDP at time T; - Primary expenditure at time T+1 is reduced (increased) by the fiscal gap (space) calculated; - The process is repeated for the projection horizon. - Because fiscal adjustment involves lags, the debt ceiling target is never precisely achieved. - Q-CRAFT does not include feedback from fiscal consolidation on GDP growth. ### Interest rates and debt dynamics - Interest-rate considerations: - Long-term fiscal projections are highly sensitive to interest rate assumptions; there is considerable uncertainty. - Advanced-economy interest rates have been declining for decades; despite recent post-pandemic increases, there are grounds to expect them to remain low into the future. - For developing economies, convergence to OECD productivity implies reduced access to concessional lending—upward pressure on interest rates—while more developed financial markets could increase domestic savings and lower domestic interest rates. - In the long run, real interest rates tend to track productivity growth; convergence to slower productivity growth could imply declining interest rates. - Q-CRAFT interest-rate input options: - Users choose between constant nominal interest rate; (nominal) interest growth differential; or real interest rate. - Debt-dynamics equation used by Q-CRAFT: - D_{t+1} = D_t * [(1+ i) / (1+g)] – pb_t - Variables: nominal GDP growth g; nominal interest rate i; primary balance (percent of nominal GDP) pb; debt-to-GDP ratio D. - Key intuitions and corollaries: - All else equal: larger primary surplus or smaller primary deficit lowers debt-to-GDP; higher initial debt raises debt-to-GDP; higher interest rates raise debt-to-GDP; faster GDP growth lowers debt-to-GDP. - The primary balance is the variable governments can control in the near- to medium-term. - The interest rate–growth differential (i – g) represents “automatic debt dynamics”; for i > g, dynamics are unfavorable; for i < g, favorable. - Debt-stabilizing primary balance: - Q-CRAFT calculates the debt-stabilizing primary balance each year. - If projected primary balance is weaker than the debt-stabilizing primary balance, the difference is the fiscal adjustment task to stabilize debt. - If the debt-stabilizing primary deficit (surplus) is larger (smaller) than the projected primary balance, the country has fiscal space to loosen policy without debt rising. ### Climate scenarios (overview included in the same Q-CRAFT text) - Q-CRAFT focuses on five key IPCC-based climate scenarios (with an additional “Hot adapted” and “Hot Un-Adapted” distinction), projecting global mean temperature increase between 0.7 and 3.5°C by the end of the century with respect to present levels. - Scenario descriptions: - Paris: based on SSP1-2.6, assumes international commitments from the 2015 Paris summit are met; keeps global temperature increase above pre-industrial below 2°C at the end of the century. - Moderate: based on SSP2-4.5; emissions increase in line with current trends and stabilize at the end of the century. - High: based on SSP3-7.0; high-emissions scenario with scaled-back mitigation. - Hot: same emissions as High but uses the 90th percentile of temperature increase among climate models using SSP3-7.0. - Hot adapted: same temperature increases as Hot, but countries adapt more quickly. - Hot Un-Adapted: same temperature increases as Hot, but countries adapt more slowly. - Macroeconomic effects of climate change in Q-CRAFT: - Q-CRAFT uses updated estimates from Kahn et al (2021). - Kahn et al (2021) estimate that a persistent increase in average global temperature by 0.04 degrees Celsius per year, in the absence of mitigation policies, would reduce world real GDP per capita by 7 to 13 percent by 2100. - Kahn et al (2021)’s counterfactuals have been updated for 171 countries for the five scenarios. - Q-CRAFT captures the effect of climate change on real GDP as a slowdown in productivity growth. - Inflation is assumed unchanged across climate scenarios; the decline in labor productivity directly reduces nominal GDP. - Monetary and external sectors are exogenous and not endogenized; climate-driven inflation spikes from natural disasters or supply shocks are not endogenized in the baseline. *Source: Q-CRAFT user guide (IMF); Amaglobeli, D., and Shi, W. (2016).* ### Box 4. Summary of Kahn et al (2021) ### Box 4. Summary of Kahn et al (2021) ### Methodology and scope - Kahn et al (2021) estimate long-term effects of weather patterns transformed by climate change on economic activity across countries. - The analysis: - Begins with a theoretical growth model linking “deviations” of temperature and precipitation (weather) from their long-term moving-average historical norms (climate) to growth in real GDP per capita. - Is estimated using data from 174 countries over 1960-2014. - Employs an econometric technique that allows for dynamics, nonlinearity, an implicit model for adaptation to climate change, and accounts for effects on economic activity of changes in the distribution of weather patterns (both averages and variability of temperature and precipitation). - Explores the efficacy of adaptation by tracking the elasticity of per capita GDP to climate variables over time. ### Key empirical findings (Kahn et al (2021)) - Per-capita real output growth is adversely affected by persistent changes in the temperature above or below its historical norm. - A persistent increase in average global temperature by 0.04 degrees Celsius per year, in the absence of mitigation policies, reduces world real GDP per capita by around 7 percent by 2100 compared to a baseline in which temperatures increase according to their 1960-2014 trends. - Estimated losses would increase to 13 percent globally if country-specific variability of climate conditions were to rise commensurate with annual temperature increases of 0.04 degrees Celsius. - Abiding by the Paris Agreement goals, thereby limiting the temperature increase to 0.01 degrees Celsius per year, reduces the loss substantially to about 1 percent. - While adaptation to climate change can reduce these negative long-run growth effects, it is highly unlikely to offset them entirely. - The authors do not estimate country-specific impact of temperature on GDP; rather, temperature projections vary between countries, leading to different GDP impacts. ### Fiscal effects of climate change in Q-CRAFT - Q-CRAFT links macroeconomic impacts from climate scenarios to fiscal variables under specific behavioral assumptions: - Revenue: - As nominal GDP decreases, tax revenues decline correspondingly. - Under each climate change scenario, revenue is assumed to decline in line with nominal GDP, so the revenue-to-GDP ratio remains unchanged from the baseline. - The underlying intuition: in the face of slow-building climate change, governments do not adjust tax policy and the effective tax rate (revenue-to-GDP ratio) remains constant. - Primary expenditure: - Primary expenditure is assumed rigid and held unchanged from the baseline, reflecting unchanged policy settings in the face of slow-building climate change. - With unchanged primary expenditure and declining nominal GDP, the ratio of primary expenditure to GDP rises in climate change scenarios. - Expenditure flexibility parameter: - Q-CRAFT includes a parameter that allows the expenditure rigidity assumption to be relaxed; this parameter varies between 0 (fully flexible) and 1 (completely rigid). - As the parameter approaches 0 (more flexible), real primary expenditure per capita reduces until the primary expenditure-to-GDP ratio matches the baseline, and primary expenditure in local currency terms is significantly lower than the baseline. - Fiscal balances and debt: - Declining revenue combined with unchanged primary expenditure (rigid setting) results in worsening primary and overall balance in climate change scenarios. - Primary balance deteriorates in both local currency unit terms and relative to GDP. - Worsening primary balance raises the debt-to-GDP ratio, increases interest expenditure, and worsens the overall balance. - Additional fiscal costs: - Q-CRAFT users can manually include additional fiscal costs of climate change risks (for example, costs of natural disasters) under different climate change scenarios. - The model assumes such ‘fiscal shocks’ either raise primary expenditure or lower revenue upon materialization, and they will occur more frequently and with greater magnitude with climate change. - Fiscal space: - Q-CRAFT results show climate change can reduce fiscal space, potentially leading to unfavorable or unsustainable fiscal trajectories for severely affected countries. - Even if a country’s debt-to-GDP ratio is stable in the baseline, under climate change scenarios slower economic growth and worsening primary balance can cause debt to spiral out of control relative to GDP and make the fiscal task of stabilizing debt significantly more onerous. ### Adaptation assumptions and scenarios in Q-CRAFT - Kahn et al (2021) embed an adaptation parameter (m) that assumes countries will adapt to higher temperatures over the course of 30 years; this parameter can be adjusted in Q-CRAFT. - Adaptation and model limits: - Lower m reflects faster adaptation and reduces the macroeconomic impact of persistent temperature increases on growth. - The dataset does not have any estimates of climate adaptation spending. - From a modeling standpoint, projected adaptation spending will not exceed expenditure levels anticipated in the baseline for any country. - Indirect or second-order effects—such as reallocation of funds from education to adaptation, subsequent reductions in productivity, and resulting impacts on GDP growth—are not accounted for in the model. - Scenario parameterizations: - ‘Hot Adapted’ scenario: m is set to 20 (countries adapt to higher temperature in 20 years rather than 30). Macroeconomic effects are less severe than in the Hot scenario but remain significant. Faster adaptation likely requires significant public and private investment; there is no established econometric link between adaptation investment and macroeconomic/fiscal outcomes applicable to Q-CRAFT. - ‘Hot Un-Adapted’ scenario: m is set to 50 (countries adapt to higher temperatures in 50 years rather than 30). Macroeconomic effects are more severe than in the Hot scenario. ### Policy-relevant implications - Mitigation reduces long-run GDP losses: limiting temperature increase to 0.01 degrees Celsius per year (Paris Agreement goals) substantially reduces global GDP losses compared to a 0.04 degrees Celsius per year pathway. - Adaptation matters but is not a full offset: adaptation can ameliorate long-term macroeconomic and fiscal effects but is highly unlikely to fully offset damages from persistent temperature increases. - Fiscal planning and flexibility: - Maintaining rigid primary expenditure in the face of GDP declines raises primary expenditure-to-GDP ratios and worsens fiscal balances and debt dynamics. - Introducing expenditure flexibility (parameter toward 0) can restore primary expenditure-to-GDP ratios but implies lower real primary expenditure per capita and may require difficult reallocations. - Need to account for discrete fiscal shocks: climate change may increase frequency and magnitude of natural disasters and other discrete fiscal risks (power purchasing agreements, PPPs, SOEs, subnational governments), which can raise primary expenditure or lower revenue beyond econometric scenario estimates. --- _Source: https://www.imf.org/-/media/files/topics/fiscal/fiscal-risks/tool/qcraft-user-guidev10.pdf_