Economics - Revision history

Yutang.Xiong: Yutang.Xiong moved page IFs Economic Model Documentation v2023 03 25 to Economics: title change so that it can be linked to the main page

2023-08-01T17:49:38Z

Yutang.Xiong moved page IFs Economic Model Documentation v2023 03 25 to Economics: title change so that it can be linked to the main page

← Older revision	Revision as of 17:49, 1 August 2023
(No difference)

Julia.Detar: /* Appendix A: Data and Initial Conditions */

2023-07-31T16:55:01Z

Appendix A: Data and Initial Conditions

← Older revision		Revision as of 16:55, 31 July 2023
Line 4,830:		Line 4,830:
	Zhang, Jie, and Junsen Zhang. (2009). "Longevity, Retirement, and Capital Accumulation in a Recursive Model with an Application to Mandatory Retirement." ''Macroeconomic Dynamics'' 13: 327-348.		Zhang, Jie, and Junsen Zhang. (2009). "Longevity, Retirement, and Capital Accumulation in a Recursive Model with an Application to Mandatory Retirement." ''Macroeconomic Dynamics'' 13: 327-348.

	== Appendix A: Data and Initial Conditions ==		== 12 Appendix A: Data and Initial Conditions ==

	=== 12.1 Initializing the Country and Global SAMs: The Preprocessor ===		=== 12.1 Initializing the Country and Global SAMs: The Preprocessor ===

Julia.Detar: /* 2.12 Methodological Note: Algorithmic Adjustment of Forecasts in IFS */

2023-07-27T21:59:33Z

2.12 Methodological Note: Algorithmic Adjustment of Forecasts in IFS

← Older revision		Revision as of 21:59, 27 July 2023
Line 578:		Line 578:
	In an algorithmic procedure the system also computes a ratio of the GDP per capita (MER) in the countries that by average GDP per capita make up the richest 10 percent of global population to those that populate the bottom 10 percent (NSGAPRTB).		In an algorithmic procedure the system also computes a ratio of the GDP per capita (MER) in the countries that by average GDP per capita make up the richest 10 percent of global population to those that populate the bottom 10 percent (NSGAPRTB).

	=== 2.12 Methodological Note: Algorithmic Adjustment of Forecasts in IFS ===		=== 2.13 Methodological Note: Algorithmic Adjustment of Forecasts in IFS ===
	Economic growth rate in the goods and services market is only one area in IFs where specification of initial values for variables using formulations (statistically estimated or based on standard computational structures) often produces values that differ from historical data in the base year. That is the first of three challenges for forecasting across the time horizons supported by IFs. The second is that computations using such formulations can generate patterns of change in early years of forecasting that differ substantially from patterns evident in the data from recent years. Third, it is very common in the long term that formulations generate problems including exceeding reasonable limits (e.g. percentage values below zero or above 100%), failing to saturate smoothly to such limits, or saturating at values that exhibit unreasonable patterns across the 188 countries of the IFs system.		Economic growth rate in the goods and services market is only one area in IFs where specification of initial values for variables using formulations (statistically estimated or based on standard computational structures) often produces values that differ from historical data in the base year. That is the first of three challenges for forecasting across the time horizons supported by IFs. The second is that computations using such formulations can generate patterns of change in early years of forecasting that differ substantially from patterns evident in the data from recent years. Third, it is very common in the long term that formulations generate problems including exceeding reasonable limits (e.g. percentage values below zero or above 100%), failing to saturate smoothly to such limits, or saturating at values that exhibit unreasonable patterns across the 188 countries of the IFs system.

	There are many reasons for such forecasting challenges. There are also a variety of approaches for addressing such issues, but in general terms they require building ''algorithmic structural wrappers'' around even the best estimated and/or most strongly grounded theoretical specifications. The subsections below consider the three challenges sequentially by time horizon.		There are many reasons for such forecasting challenges. There are also a variety of approaches for addressing such issues, but in general terms they require building ''algorithmic structural wrappers'' around even the best estimated and/or most strongly grounded theoretical specifications. The subsections below consider the three challenges sequentially by time horizon.

	==== 2.12.1 Initial Condition-Related Forecast Challenges ====		==== 2.13.1 Initial Condition-Related Forecast Challenges ====
	Calculations by IFs in the model’s base year, even using formulations estimated with high statistical significance, will almost invariably produce values that differ from data. These are essentially the country-specific residuals in the estimation or more theoretically founded specification. There are a variety of explanations for and treatment of those residuals in correction of those discrepancies in the first year and forecasting over time.		Calculations by IFs in the model’s base year, even using formulations estimated with high statistical significance, will almost invariably produce values that differ from data. These are essentially the country-specific residuals in the estimation or more theoretically founded specification. There are a variety of explanations for and treatment of those residuals in correction of those discrepancies in the first year and forecasting over time.

Line 606:		Line 606:
	In all such instances, formulations, whatever the combination of structural logic and estimation they are built upon, are likely to produce patterns of forecasted change over time that are inconsistent with historical patterns in the data unless there is some use in early model forecast years of information about those historical data. The forecasting system can benefit greatly from algorithmic structural wrappers that combine information about individual driver and/or forecast variables with others and with that historical information.		In all such instances, formulations, whatever the combination of structural logic and estimation they are built upon, are likely to produce patterns of forecasted change over time that are inconsistent with historical patterns in the data unless there is some use in early model forecast years of information about those historical data. The forecasting system can benefit greatly from algorithmic structural wrappers that combine information about individual driver and/or forecast variables with others and with that historical information.

	==== 2.12.3 Longer Term Forecast Behavior Issues ====		==== 2.13.3 Longer Term Forecast Behavior Issues ====
	Looking beyond issues around the base year of the model and the early forecast years, the longer-term poses another set of temporal model issues. Some of these are very simple, including the assurance that percentage variables do not fall below zero or rise above 100 percent. Such boundary conditions are used throughout IFs; in many cases hard boundaries have been replaced with algorithmic functions that smooth final approaches to those boundaries.		Looking beyond issues around the base year of the model and the early forecast years, the longer-term poses another set of temporal model issues. Some of these are very simple, including the assurance that percentage variables do not fall below zero or rise above 100 percent. Such boundary conditions are used throughout IFs; in many cases hard boundaries have been replaced with algorithmic functions that smooth final approaches to those boundaries.

Julia.Detar: /* 2.11 Indicators */

2023-07-27T21:59:17Z

2.11 Indicators

← Older revision		Revision as of 21:59, 27 July 2023
Line 573:		Line 573:
	GDP (and its value-added shares) take a third form in IFs called “relative-price adjusted” GDP or GDPRPA. The SAM section of this documentation explains and uses this form.		GDP (and its value-added shares) take a third form in IFs called “relative-price adjusted” GDP or GDPRPA. The SAM section of this documentation explains and uses this form.

	=== 2.11 Indicators ===		=== 2.12 Indicators ===
	The economic model computes several indicators of interest to many IFs users. These include global product (WGDP), global product per capita (WGDPPC), an absolute measure of the per capita GDP gap between more developed (D) and less economically developed (L) regions (NSGAPA), and a ratio measure of the same gap (NSGAPR). A parameter ('''''nsdiv''''') establishes the GDP per capita at MER dividing point in thousand constant dollars between more and less developed.		The economic model computes several indicators of interest to many IFs users. These include global product (WGDP), global product per capita (WGDPPC), an absolute measure of the per capita GDP gap between more developed (D) and less economically developed (L) regions (NSGAPA), and a ratio measure of the same gap (NSGAPR). A parameter ('''''nsdiv''''') establishes the GDP per capita at MER dividing point in thousand constant dollars between more and less developed.
	[[File:ED-eqn2.11.png\|center\|frameless\|425x425px]]		[[File:ED-eqn2.11.png\|center\|frameless\|425x425px]]

Julia.Detar: /* 2.10 GDP at MER and PPP */

2023-07-27T21:58:35Z

2.10 GDP at MER and PPP

← Older revision		Revision as of 21:58, 27 July 2023
Line 554:		Line 554:
	The IFs system remains long-term oriented, but the use of '''''tfpshockadjshr''''' is an important augmentation of that orientation. Use of it in future shorter-term analysis would require attention to the hard-coded setting of the COVIDShock variable and of the years of that shock.		The IFs system remains long-term oriented, but the use of '''''tfpshockadjshr''''' is an important augmentation of that orientation. Use of it in future shorter-term analysis would require attention to the hard-coded setting of the COVIDShock variable and of the years of that shock.

	=== 2.10 GDP at MER and PPP ===		=== 2.11 GDP at MER and PPP ===
	Gross regional or national domestic product (GDP) is the sum of value added across sectors, which needs to equal the sum of production for final demand across sectors (see the later discussion of the social accounting matrix).		Gross regional or national domestic product (GDP) is the sum of value added across sectors, which needs to equal the sum of production for final demand across sectors (see the later discussion of the social accounting matrix).

Julia.Detar: /* 2.9 Short Term Shocks to Growth, as with COVID-19 */

2023-07-27T21:58:20Z

2.9 Short Term Shocks to Growth, as with COVID-19

← Older revision		Revision as of 21:58, 27 July 2023
Line 531:		Line 531:
	A second parameter, '''''ifsexomodeltype''''' specifies whether the data series provided for exogenous use contain absolute values (=1) or annual percentage change (=2).		A second parameter, '''''ifsexomodeltype''''' specifies whether the data series provided for exogenous use contain absolute values (=1) or annual percentage change (=2).

	=== 2.9 Short Term Shocks to Growth, as with COVID-19 ===		=== 2.10 Short Term Shocks to Growth, as with COVID-19 ===
	The economic model of IFs was structured for long-term analysis. For that reason, the supply side is especially well developed via the Cobb-Douglas function and the elaboration of multifactor productivity change across time.		The economic model of IFs was structured for long-term analysis. For that reason, the supply side is especially well developed via the Cobb-Douglas function and the elaboration of multifactor productivity change across time.

Julia.Detar: /* 2.8.2 Long Term Scenario Analysis Including SSPs */

2023-07-27T21:58:06Z

2.8.2 Long Term Scenario Analysis Including SSPs

← Older revision		Revision as of 21:58, 27 July 2023
Line 524:		Line 524:
	The Pardee Center routinely draws upon releases by the IMF of the ''World Economic Outlook'' to pull in its database estimates of GDP growth rates in recent years and for a few years into the future (generally 3-5 years). The data preprocessor of IFs uses those estimates to specify values of '''''gdprext''''' for those years. Unless scenarios like the SSPs specify values for GDP and/or its growth for years beyond the horizon of the IMF, the model then resumes computing economic growth (GDPR) endogenously		The Pardee Center routinely draws upon releases by the IMF of the ''World Economic Outlook'' to pull in its database estimates of GDP growth rates in recent years and for a few years into the future (generally 3-5 years). The data preprocessor of IFs uses those estimates to specify values of '''''gdprext''''' for those years. Unless scenarios like the SSPs specify values for GDP and/or its growth for years beyond the horizon of the IMF, the model then resumes computing economic growth (GDPR) endogenously

	==== 2.8.2 Long Term Scenario Analysis Including SSPs ====		==== 2.9.2 Long Term Scenario Analysis Including SSPs ====
	In the same manner that IMF growth estimates can be put into IFs to exogenously specify GDP growth forecasts for the early years of scenario runs with IFs, other scenario sets can determine growth for longer periods. Using the ENV-Linkages model, the OECD has provided five sets of projections for the Shared Socioeconomic Pathway scenarios (SSPs) that extend through the 2100 horizon of IFs. Prepackaged .sce files (see SSP1.sce, etc.) exist in the IFs structure (a scenario folder) to run those projections in association also with demographic, education, income distribution, and other operationalizations of the SSPs for optional use.		In the same manner that IMF growth estimates can be put into IFs to exogenously specify GDP growth forecasts for the early years of scenario runs with IFs, other scenario sets can determine growth for longer periods. Using the ENV-Linkages model, the OECD has provided five sets of projections for the Shared Socioeconomic Pathway scenarios (SSPs) that extend through the 2100 horizon of IFs. Prepackaged .sce files (see SSP1.sce, etc.) exist in the IFs structure (a scenario folder) to run those projections in association also with demographic, education, income distribution, and other operationalizations of the SSPs for optional use.

Julia.Detar: /* 2.8.1 Early Years */

2023-07-27T21:57:53Z

2.8.1 Early Years

← Older revision		Revision as of 21:57, 27 July 2023
Line 521:		Line 521:
	''s = 3,4 ... nsector, if '''agon''' and '''enon''' = 1; s = 1,2 ... nsector if '''agon''' and '''econ'''''		''s = 3,4 ... nsector, if '''agon''' and '''enon''' = 1; s = 1,2 ... nsector if '''agon''' and '''econ'''''

	==== 2.8.1 Early Years ====		==== 2.9.1 Early Years ====
	The Pardee Center routinely draws upon releases by the IMF of the ''World Economic Outlook'' to pull in its database estimates of GDP growth rates in recent years and for a few years into the future (generally 3-5 years). The data preprocessor of IFs uses those estimates to specify values of '''''gdprext''''' for those years. Unless scenarios like the SSPs specify values for GDP and/or its growth for years beyond the horizon of the IMF, the model then resumes computing economic growth (GDPR) endogenously		The Pardee Center routinely draws upon releases by the IMF of the ''World Economic Outlook'' to pull in its database estimates of GDP growth rates in recent years and for a few years into the future (generally 3-5 years). The data preprocessor of IFs uses those estimates to specify values of '''''gdprext''''' for those years. Unless scenarios like the SSPs specify values for GDP and/or its growth for years beyond the horizon of the IMF, the model then resumes computing economic growth (GDPR) endogenously

Julia.Detar: /* 2.8 Exogenous Control of Economic Growth */

2023-07-27T21:57:19Z

2.8 Exogenous Control of Economic Growth

← Older revision		Revision as of 21:57, 27 July 2023
Line 506:		Line 506:
	The variable TEFF is a multiplicative term in the production function and the computation of value added and GDP. Thus, any change in TEFF feeds through to GDP in the same proportion.		The variable TEFF is a multiplicative term in the production function and the computation of value added and GDP. Thus, any change in TEFF feeds through to GDP in the same proportion.

	=== 2.8 Exogenous Control of Economic Growth ===		=== 2.9 Exogenous Control of Economic Growth ===
	While IFs can endogenously compute economic growth for all countries and all years, there are at least two circumstances that call out for exogenous overrides of such computation: (1) when data exist for economic growth in the early years after the base year of the model; while the Pardee Center regularly advances the base year as new data become available, the IMF, other organizations, or national governments often have released data or estimates for some subsequent years; (2) analysts wish exogenously to specify GDP growth rates for future years and/or countries in order to explore the secondary effects of those; for instance, in exploration of the five Shared Socioeconomic Pathways (SSPs), which are often used in analysis of climate change, sets of exogenous GDP series exist for them across all countries through 2100.		While IFs can endogenously compute economic growth for all countries and all years, there are at least two circumstances that call out for exogenous overrides of such computation: (1) when data exist for economic growth in the early years after the base year of the model; while the Pardee Center regularly advances the base year as new data become available, the IMF, other organizations, or national governments often have released data or estimates for some subsequent years; (2) analysts wish exogenously to specify GDP growth rates for future years and/or countries in order to explore the secondary effects of those; for instance, in exploration of the five Shared Socioeconomic Pathways (SSPs), which are often used in analysis of climate change, sets of exogenous GDP series exist for them across all countries through 2100.

Julia.Detar: /* 2.7.4 The Economic Impacts of the Informal Economy */

2023-07-27T21:56:54Z

2.7.4 The Economic Impacts of the Informal Economy

← Older revision		Revision as of 21:56, 27 July 2023
Line 439:		Line 439:
	With respect to time horizons, most analyses of climate impact on the economy stop at 2100 as does that in IFs. But atmospheric carbon increases will persist much longer. It is for this reason that analysis with FUND can extend through 3000 and that with PAGE through 2200. A central purpose of those models is to determine the social cost of greenhouse gas emissions, often conceptualized as the long-term cumulative cost per ton of current carbon dioxide equivalent emission, with final values often computed by applying a discount rate to future costs. The purpose of the analysis of economic impacts in IFs is to close the loops that connect the economy, climate change, and all other models in the IFs system.		With respect to time horizons, most analyses of climate impact on the economy stop at 2100 as does that in IFs. But atmospheric carbon increases will persist much longer. It is for this reason that analysis with FUND can extend through 3000 and that with PAGE through 2200. A central purpose of those models is to determine the social cost of greenhouse gas emissions, often conceptualized as the long-term cumulative cost per ton of current carbon dioxide equivalent emission, with final values often computed by applying a discount rate to future costs. The purpose of the analysis of economic impacts in IFs is to close the loops that connect the economy, climate change, and all other models in the IFs system.

	==== 2.7.4 The Economic Impacts of the Informal Economy ====		==== 2.8.4 The Economic Impacts of the Informal Economy ====
	Section 7 of this documentation details the representation of drivers of the informal economy. Here we focus on the effects of that informality on economic productivity and therefore growth. Whereas climate change is progressing from a relatively low base of impact and therefore is affecting both immediate GDP and underlying productivity patterns that historically were less or even little affected by it (for instance, little impact on '''''igdprcor'''''), economic informality has long been with us and is already baked into historical levels and growth rates of GDP. Thus, the representation of climate change economic impact in IFs significantly builds on absolute levels of it and represents the impact of increase in those for both GDP and productivity, whereas the representation of informality impact looks much more to changes (and often reductions) of that impact on productivity alone, letting those impacts then pass through to GDP.		Section 7 of this documentation details the representation of drivers of the informal economy. Here we focus on the effects of that informality on economic productivity and therefore growth. Whereas climate change is progressing from a relatively low base of impact and therefore is affecting both immediate GDP and underlying productivity patterns that historically were less or even little affected by it (for instance, little impact on '''''igdprcor'''''), economic informality has long been with us and is already baked into historical levels and growth rates of GDP. Thus, the representation of climate change economic impact in IFs significantly builds on absolute levels of it and represents the impact of increase in those for both GDP and productivity, whereas the representation of informality impact looks much more to changes (and often reductions) of that impact on productivity alone, letting those impacts then pass through to GDP.