Guide to Scenario Analysis in International Futures (IFs)

Introduction

The purpose of this document is to facilitate the development of scenarios with the International Futures (IFs) system. This document supplements the IFs Training Manual. That manual provides a general introduction to IFs and assistance with the use of the interface (e.g., how do I create a graphic?). In turn, the broader Help system of IFs supplements this manual. It provides detailed information on the structure of IFs, including the underlying equations in the model (e.g., what does the economic production function look like?). This document should help users understand the leverage points that are available to change parameters (and in a few cases even equations) and create alternative scenarios relative to the Base Case scenario of IFs (e.g., how do I decrease fertility rates or increase agricultural production?). It proceeds across the modules of IFs, such as demographic, economic, energy, health, and infrastructure, to (1) identify some of the key variables that you might want to influence to build scenarios and (2) the parameters that you will want to manipulate to affect your variables of interest. The Training Manual will help you actually make the parameter changes in the computer program and the Help system will facilitate your understanding of the structures, equations and algorithms that constitute the model. We begin by introducing the types of parameters within IFs and then proceed to a discussion of variables and parameters within each of the IFs modules.

A Note on Parameter Names

In this manual we will provide the internal computer program names of variables and parameters, as well as their descriptions. Those names are especially important for use of the Self-Managed Display form, which provides model users with complete access to all variables and parameters in the system. Most model use, however, employs the Scenario Tree form to build scenarios and the Flexible Display form to show scenario-specific forecasts, and both of those forms rely primarily on natural language descriptions of variables and parameters. To match the names provided here with the options in those forms, you can use the Search feature from the menu. The Training Manual describes how to use features such as the Flexible Display form to see computed forecast variables in natural language. And it also describes how to use the Scenario Tree form to access parameters in something close to natural language. Nonetheless, it helps very much in the use of those features and the model generally to know the actual variable and parameter names.

Types of Parameters in IFs

Equations in IFs have the general form of a dependent or computed variable, as a function of one or more driving or independent variables. Variables, like population and GDP, are the dynamic elements of forecasts in which you are ultimately interested. For instance, total fertility rate or TFR (the number of children a woman has in her lifetime) is a function of GDP per capita at purchasing power parity (GDPPCP), education of adults 15 or more years of age (EDYRSAG15), the use of contraception within a country (CONTRUSE), and the level of infant mortality (INFMORT). In the most general terms the equation is

$TFR=F(GDPPCP,EDYRSAG15,CONTRUSE,INFMORT)$

Parameters of several kinds can alter the details of such a relationship. That is, parameters are numbers (also represented by names in IFs), that help specify the exact relationship between independent and dependent variables in equations or other formulations (including logical procedures called algorithms). For instance, the model may contains different parameters that tell us how much TFR rises or falls per unit change in GDP per 6 capita, education levels, contraception use, and infant mortality¹ and it contains still others to set bounds on the lower and/or upper values of TFR over the long run (obviously TFR should never go negative and probably we will not even want it to go, at least for a long time, to a very low level such as an average of 0.5 children per woman). Some of these parameters are more technical than others in the sense that they may significantly affect the overall stability of the model if users are not very careful with the magnitude or direction of the changes they make; we will focus heavily in this manual on parameters that are easiest to interpret and modify.

In many cases, we are more interested in using a parameter to make a direct change to a variable, rather than indirectly affecting a variable like TFR through one of its drivers. We often refer to this as the "brute force" method of changing a variable, and this can be done by multiplying the entire result of a basic equation like that above by a number, adding something to that result, or simply over-riding the result with an exogenously (externally) specified series of values. In the case of TFR we use the multiplier approach, which is described below. The strengths of this approach should be obvious: it preserves model stability, and makes the model more accessible for users. However, the weakness is that in many instances it is more realistic to affect one of the drivers of TFR rather than TFR directly.

Beyond multipliers, there are many other types of parameters that IFs uses, although we are forced to abandon TFR to provide examples. For instance, a switch parameter may turn on or off a particular formulation in preference to another. A target may specify a value towards which we want a variable to move gradually (we would need to specify both the target level and the years of convergence to it).

Overall, key parameter types are:

1. Equation Result Parameters. Most users will use these parameter types far more often than any other. The three types are:

a. Multipliers. This most common of all parameter types in scenario analysis comes into play after an equation has been calculated. They multiply the result by the value of the parameter. The default value, i.e. the value for which the parameter has no effect and to which multipliers almost invariably are set in the Base Case, is 1.0. These parameters are usually denoted with the suffix -m at the end of the parameter name. 

b. Additive factors. Like multipliers, these change the results after an equation computation, but add to the result rather than multiplying. The default value is normally 0.0. These are usually denoted with the suffix -add at the end of the parameter name.

c. Exogenous Specification. Sometimes these parameters override the computation of an equation. In other cases, they are actually substitutes for having an equation; that is, they are actually equivalent to specifying the values of a variable over time for which the model has no equation. This typically means establishing a new exogenous series. They typically will have the name of the variable that they over-ride within their own name.

2. Targets. Especially for the purposes of policy analysis, we often want to force the result of an equation toward a particular value over time (e.g. to achieve the elimination of indoor use of solid fuels). Target parameters are generally paired, one for the target level and one for the number of years to reach the target (from the initial year of the model forecast, 2010). Targets have different types:

a. Absolute targets. In this case the target value and year define the absolute value the variable should move toward and the number of years after the first model year over which the goal should be achieved. Together they determine a path in which the value for the variable moves quite directly² from the value in first year to the target value in the target year. Trgtval and trgtyr are the parameter suffixes used for this parameter type. The first of these changes the target itself, and the second alters the number of years to the target. The default value of *trgtyr parameters should normally be 10 years, but in some cases it is 0, meaning that users must set the number of years to target as well as the target value in order to use these parameters.

b. Relative (standard error) targets. In this case, the target value and year define a relative value towards which the variable should move and the number of years that will pass before the target is reached. The relative value is defined as the number of standard errors above or below the “predicted” value of the variable of interest (a prediction usually based on the country's GDP per capita). Target values less than 0 set the target below the typical or predicted (as indicated by cross-sectional estimations) value of the variable. Target values above 0 set the target above the predicted value. As with the absolute targets, the value calculated using relative targeting is compared to the default value estimated in the model. The computed value then gradually moves from the normal or default-equation based value to the target value. If, however, the computed value already is at or beyond the target (that could be above or below depending on whether the target is above or below the default or predicted value), the model will not move it toward the target. Two different parameter suffixes direct relative targeting: setar and seyrtar. The first of these changes the target itself and the second alters the number of years to the target. The default value of the *seyrtar parameters varies based on the module and even variable. Governance parameters are set to a default of 10 years from the year of model initialization, while infrastructure parameters are set to a default of 20 years. These defaults mean that users do not have to change *seyrtar as well as *setar in order to build standard error target scenarios. Changing *setar should be enough. 

3. Rates of change. Some parameters specify an annual percentage rate of change. Unfortunately, IFs does not consistently use percentage rates (5 percent per year) versus proportional rates (0.05 increase rate per year, which is equivalent to 5 percent), so the user should be attentive to definitions. There are multiple suffixes that may apply to these, including -r (changes in the rate) and -gr (changes the rate of change, growth or decline).

4. Limits. As indicated for the TFR example, long-term national rates are unlikely to fall and stay below a minimum value. Limits can be minimum or maximum values. These are typically denoted by the suffixes - min, -max, or -lim.

5. Switches. These turn off and on elements in the model. These most often affect linkages between modules, but can also change relationships within modules. They are typically denoted by the suffix -sw.

6. Other parameters in equations and algorithms. Equations within IFs can become quite complicated. The parameter types discussed to this point provide the easiest control over them for most model users. Relatively few users will proceed further with parameters, and to do so will typically require attention to the specific nature of the equation (e.g. whether independent variables are related to dependent ones via linear, logarithmic, exponential or other relationship forms). That is, one would normally need to understand the model via the Help system or other project documentation in order to use them meaningfully and without causing substantial risk of bad model behavior. The sections of this manual will provide very little information about these technical parameters.

a. Elasticities: These are relatively common within IFs and specify the percentage change in the dependent variable associate with a percentage change in the independent variable. They are typically prefixed el- or elas-.

b. Equilibration control parameters. IFs balances supply and demand for goods and services via prices, savings and investment with interest rates, and so on. These processes typically use an algorithmic controller system that responds to both the magnitude of imbalance or disequilibrium and the direction and extent of its change over time (see the Help system descriptions of the model). Although they are not typical elasticities, the two parameters that control each such process usually have the prefix el- and the suffixes -1 or -2. Parameters ending with 1 relate to disequilibrium magnitude; and parameters end with 2 relate to the direction of change.

c. Other coefficients in equations. Beyond elasticities, many other forms of parameter can manipulate an equation. When analysts in many fields think of parameters, this is what they mean. In IFs, most users will use them quite rarely because, in the absence of knowledge concerning equation forms and reasonable ranges, the parameters often have little transparent meaning—experts in a field may use them more often. Many analysts think of such parameters as having a constant value over time, and some are unchangeable over time in IFs. IFs allows almost all, however, to be entered as time series and vary with great flexibility across time. Some can be changed for each country and/or sub-dimensions of the associated variable, such as energy types, but others can only be changed globally.

d. Equation forms. Although most users will change parameters using the Scenario Tree (see again the Training Manual), the IFs model has made it possible over time to change an increasing number of functions directly (both bivariate and multivariate ones). The advantage this confers is the ability to alter the nature of the formulation (e.g. going from linear to logarithmic) and even, to a very limited degree, the independent or driver variables in the equation. Although some module discussions will occasionally suggest this option, most users will not avail themselves of it. Users who wish to make such changes can do so via the Change Selected Functions options, which can be accessed from the Scenario Analysis Menu on the main page.

7. Initial conditions for endogenous variables and convergence of initial discrepancies

a. Initial conditions are not, strictly speaking, true parameters, but should reflect data. Yet some users will believe that they have data superior to that in IFs, and the system allows the user to change most initial conditions. After the first year, the model will compute subsequent values internally (endogenously). Initial conditions don’t have a suffix; their names are, in fact, those of the variable itself (e.g., POP for population).

b. Convergence speed of initial-condition based discrepancies to forecasting functions. Because initial conditions taken from empirical data often vary from the values that are computed in the estimated equation used for forecasting, the model protects the empirically-based initial condition by computing shift factors that represent that initial discrepancy (they can be additive or multiplicative). For many variables, values rooted in initial conditions in the first model year should converge to the value of the estimated equation over time; convergence parameters control the speed of such convergence. Most model users will never change the convergence speed. These are denoted by the suffixes -cf or -conv.

In the use of all parameters, especially those other than equation result parameters, users will often be uncertain how much it is reasonable to move them—as are often even the model developers. The Scenario Tree form provides some support for judgments on this by indicating high and low alternatives to that of that Base Case. This manual will sometimes provide some additional information.

---

¹Because the nature of the equation or formulation will vary (sometimes a driving variable is linearly linked to the dependent variable, sometimes the equation uses a logarithmic, exponential, or other formulation), the coefficients in the equation cannot invariably or even regularly be interpreted as units of change linked to units of change. You may need to explore the Help system and specific equations to fully understand the relationship. This is one of the key reasons we very often turn to the multipliers and additive factors explained in the next paragraph.

²The movement normally will be linear, except that it is possible to set moving targets that create non-linear progression patterns. In some cases, the model explicitly uses non-linear convergence; e.g. to accelerate movement in early years and then to slow it as the target is approached.

Manipulating Parameters in IFs

You will typically manipulate parameters to create scenarios or internally coherent stories about the future. You may create scenarios because you wish to represent and explore the possible impact of policy interventions. Or your stories may represent views of the dynamics of global systems alternative to that in the IFs Base Case scenario. Most of the time, you will be interested in tracking the possible futures of selected variables having particular interest to you. The following sections, each covering a module of the IFs system, begin by identifying some of the variables of potentially greatest interest to you. They then provide suggestions on which parameters are likely to be of most useful in building alternative scenarios for those variables. Each section includes tables listing the most effective parameters with which to target certain outcomes. While these suggestions are intended to help you start to think about which parameters you might use to build your scenarios, it is essential that you consider seriously what the policy-based, empirical-knowledge-rooted, or theoretically informed foundations are for your changes.

Keys to Successfully Modifying Parameters in IFs

• Test all parameter changes individually before building combinations, in order to be able to identify which parameters are having specific impacts
• After changing a parameter value and running a scenario, check the impact on the most proximate or closely related variables (identified in the tables of each module section), before checking the secondary impacts of your selected parameter on more distally related variables
• Tie parameter changes to policy options, empirical knowledge, or theoretical insight identified in literature
• Bear in mind the relevant geographical level at which a parameter operates; some parameters function directly at a global level (e.g., global migration rates), while others will be most relevant at the regional, or national level
• Some parameters are only effective when used in combination with one another (such as target values and years to reach a target)
• Some parameters cancel one another out; for example, trgtval and setar parameters cannot be used together except under very limited circumstances that we attempt to note in the subsequent text
• In many cases, variables affected by certain parameters have natural maximums (e.g. 100 percent) or minimums (e.g. fertility rate), so that changes to the parameters affecting them, where countries may already be approaching such a limit, will not have a significant impact
• The IFs systems contains many equilibrating processes, such as those around prices; interventions meant to affect one side of such an equilibration (such as efforts to reduce energy demand) may have offsetting effects (such as lower prices for energy and resultant demand increase) that make it harder than you expect to push the system in the desired direction; real-world policy makers often face such difficulties and may need to push harder than anticipated

Prepackaged Scenarios

A number of alternative scenarios come prepackaged with the model. To access them, select Scenario Analysis from the main menu, and then the option labeled Quick Scenario Analysis with Tree. Once in the scenario display, select Add Scenario Component to view all of the .sce (scenario) files that are stored on your computer normally at the path C:/Users/Public/IFs/Scenario. Exploring several simple interventions contained in the folder structure should give users an overview of some of the leverage points in that they may wish to use in each module

Demographic Module

Variables of Interest

Variable	Description
POP	Total population
POPLE15	Population, age 15 or less
POP15TO65	Population, age 15 to 65
POPGT65	Population, greater than 65
POPPREWORK	Population, pre-working years
POPWORKING	Population, working years
POPRETIRED	Population, retired
YTHBULGE	% of the population between 15 and 29
POPMEDAGE	Population, median age
LAB	Labor force size
BIRTHS	Births
DEATHS	Deaths
MIGRANTS	Net migration (inward)
CBR	Crude birth rate
CDR	Crude death rate
TFR	Total fertility rate
CONTRUSE	Contraceptive usage
LIFEXP	Life expectancy
MIGRATE	Net migration rate (inward)

The IFs demographic module breaks country populations down into 21 fiveyear age groups, each one subdivided by gender. This allows the model to create an age-sex cohort structure that responds to changes in the three fundamental drivers of population: fertility, mortality, and migration. Births are calculated as a function of each country’s fertility distribution and age distribution. As children are born, they enter the lowest band of the agesex structure, the layer representing people aged 0 through 5. Each country’s population growth is reduced by deaths at each age level; like births, deaths are calculated as a function of the mortality distribution and the age distribution. Finally, migration patterns either add to, or subtract from, each country’s population, depending on the balance of immigration and emigration³ . Each of the three proximate drivers of population is influenced by deeper social processes: births are a product of fertility patterns; deaths are linked to life expectancy; and net migrants are determined by an overall global migration rate.

Total population is represented in millions of people via POP, but users may also choose to explore the age structure within society. Three variables break population down into broad age groups: POPLE15, people age 15 or younger, POP15TO65, people age 15 to age 65, and POPGT65, people older than age 65. Three additional variables provide a similar disaggregation of population: POPPREWORK, POPWORKING, POPRETIRED—as the names suggest, they measure the number of people who have yet to enter their working years, the number of people currently in their working years, and the number of people who have completed their working years. The years comprising an adult’s working life may vary from country to country, depending on education systems and retirement ages. Users can explore additional population characteristics via the variables YTHBULGE, the percent of all adults (15 and older) between the ages 15 and 29; POPMEDAGE, the median age of a country’s population; and LAB, the size of the labor force, recorded in millions of people. For any country, the complete age and sex breakdown is available under the Specialized Displays for Issues option under the Display sub-menu. From the Specialized Displays menu, select Population by Age and Sex, and click the button labeled Show Numbers. This will bring up detailed population figures for any of the countries in the IFs system. To view a population pyramid display, toggle the Distribution Type setting on the menu bar.

The three immediate drivers of population change—births, deaths and migration—are captured in the model as flows. Every year babies are born (BIRTHS), people die (DEATHS) and people leave countries to live elsewhere (MIGRANTS). These processes alter the stock of population in countries, regions and the world as a whole. The speed at which a population will grow or decline, and the attendant shift in a population’s age structure, depend on crude birth rates (CBR) and crude death rates (CDR)—the number of births and deaths per 1,000 people.

Each of the immediate drivers is linked to deeper determinants of population. For instance, fertility rates are responsive to income, education and infant mortality rates, offering points of access elsewhere in the model. Total Fertility Rate (TFR) is a variable that is essential to our understanding of populations’ reproductive behavior. TFR is, essentially, the number of children the average woman in a country can expect to have over the course of her lifetime. In order for the overall population size to remain roughly stable, TFR must meet the replacement rate for that country. For developed countries this is approximately 2.1 children per woman, but the figure may be higher in countries with high mortality rates, and is lower in many. While TFR largely determines future population growth, it is not the only behavioral variable of note: CONTRUSE captures the percent of fertile women who routinely use some method of contraception.

For a complete discussion of mortality see the Health module, where deaths are computed. They are responsive to deep or distal factors such as income, education and technological advance, as well as to more proximate ones such as levels of undernutrition and smoking. A key indicator for the population model, linked to deaths, is LIFEXP, or life expectancy, which provides a measure of the median life expectancy of a newborn in a particular year given the current mortality distribution. Although life expectancy can be calculated for any age, IFs focuses on life expectancy at birth. This variable is key to the functioning of the IFs system because many of the parameters that affect mortality do so by changing life expectancy.

The final proximate driver of population growth is migration. MIGRANTS measures net migrants in raw figures, reported in millions of people; but this variable is determined by MIGRATE, the net migration rate, reported as percent of the total population. The basic forecasts of migration in IFs are one of the very few variables that are exogenous. Nonetheless, there is parametric control of it.

The demographic module features an array of parameters that allow users to create alternative demographic scenarios by exploring uncertainty surrounding: fertility, mortality and migration, as well as the years making up people’s working lives.

---

³In IFs, the age distribution of migrants is controlled by an internal vector across age categories, not available for manipulation through the model’s front-end.

Parameters to Affect Fertility

Parameter	Variable of Interest	Description	Type
tfrm	TFR, CBR	Total fertility multiplier	Multiplier
contrusm	CONTRUSE	Contraceptive use multiplier	Multiplier
eltfrcon	TFR	Elasticity of total fertility rate to contraception use	Elasticity
tfrmin	TFR	Long term TFR convergence value	Limit

The single most powerful way for users to modify fertility rates is to manipulate tfrm, a parameter that directly alters the total fertility rate within a country or region. This parameter serves as a multiplier on the fertility rate calculated by the model—a 20% increase or decrease in the value of the parameter will result in a similar magnitude of change in the value of the associated variable, TFR. Because it is a brute force multiplier, users should justify their modifications to the parameter. When used thoughtfully, tfrm can be a powerful tool for scenario analysis. It can be used to model the impact of fertility control initiatives that extend beyond simple contraceptive use. An example would be the implementation of a program to offer public seminars on the benefits of having fewer children, which could lower the fertility rate even when overall contraceptive usage rates are low. Health care programs for women are a major contributor to fertility decline.

Users can also directly change the percentage of the population that uses contraceptives via contrusm, a parameter that indirectly affects the total fertility rate via CONTRUSE. As this is a multiplier, it works the same way as tfrm. It can be used to model the impact of an increase in the availability of family planning education, a campaign to promote the use of condoms, or any other intervention that would likely increase (or decrease) the percentage of a population using contraceptives. Additionally, the parameter eltfrcon allows users to control the elasticity of total fertility to contraceptive use. For example, a weaker relationship between the two variables might be justified if the contraceptive methods in use in a country or region are widely known to have high failure rates.

When creating alternative scenarios that span long time horizons, users may wish to modify fertility assumptions built into the demographic module. As countries grow richer and reach higher levels of educational attainment, total fertility rates tend to decrease. However, in forecast years, a minimum value prevents countries from dipping too far below replacement rate. As a default setting, the minimum parameter, tfrmin, is set to 1.9. Thus, in the Base Case, TFR in highly developed countries will converge to just below 2 children per woman. By increasing or decreasing the parameter, users can experiment with different long-term fertility patterns.

Parameters to Affect Mortality

Parameter	Variable of Interest	Description	Type
mortm	DEATHS	Mortality multiplier (not cause specific)	Multiplier
hlmortm	DEATHS	Mortality multiplier by cause	Multiplier

The health module write-up includes a full description of the drivers of mortality in the IFs system, and explains how to manipulate each one. However, one parameter affecting mortality, mortm, is worth discussing separately. 14 This parameter functions similarly to the hlmortm parameter available in the health module, but does not disaggregate by cause of death. Similar to tfrm, mortm can be used to model the impact of events that have broad impacts across the population, such as the end of an armed conflict or the implications of a plague. Usually however, if a user is building a scenario analyzing health trends, using the hlmortm multiplier will be more useful because it disaggregates mortality on the basis of cause. Because morbidity rates in IFs are linked normally to mortality rates, these parameters will affect them also.

Parameters to Affect Migration

Parameter	Variable of Interest	Description	Type
wmigrm	MIGRATE, MIGRANTS	World migration rate multiplier	Multiplier
migrater	MIGRATE, MIGRANTS	Net migration rate (inward)	Rate of change

Users interested in modifying migration patterns should bear in mind that migrant flows are subject to an accounting system that keeps the global number of net migrants equal to zero. In other words, a person leaving one country will be accounted for when they enter another country. Changing the world migration rate, wmigrm, is the easiest way to affect migration patterns in IFs. Altering this parameter will allow users to increase the overall rate at which migration occurs at a global level, enabling users to simulate large scale increases (or decreases) in migration generated by, say, reductions in visa fees, or the opening of borders as is the case in the EU’s Schengen area. The parameter migrater, on the other hand, allows users to affect the rate of migration into individual countries or regions (values can range from positive, indicating net inward migration, to negative, indicating net outward migration).

Parameters to Affect Working Age

Parameter	Variable of Interest	Description	Type
workingageentry	POPPREWORK, POPWORKING	Working age determinant	Exogenous specification
workingageretire	POPWORKING, POPRETIRED	Retirement age determinant	Exogenous specification

In addition to manipulating the rate at which populations grow, users can experiment with the effects of changing a country’s working age, something that will be fiscally important in many countries as populations age. The variables POPPREWORK, POPWORKING and POPRETIRE map the typical age structure of a country or region’s work force. Two parameters, workingageentry and workingageretire, control the age at which a person is considered eligible for work and the age at which a person is eligible for retirement. Changes in the workforce’s age configuration link forward to economic production via the size of the labor force (LAB). Raising or lowering the retirement age will additionally affect government finances via the size of population of retirement age and the level of pension support provided to households (GOVHHPENT).

Prepackaged Scenarios

An installation of IFs includes high and low population-framing scenarios. Originally created for the poverty volume of the Pardee Center’s Potential Patterns of Human Progress (PPHP) series, the two files are located in the Framing Scenarios folder under Population. Both scenarios feature the direct total fertility rate multiplier. Tfrm in the high fertility scenario is set to 1.5 globally. In the low fertility scenario, tfrm is set to .6 in non-OECD nations, and the limit parameter tfrmin is set to 1.6 globally. Although the two scenarios only feature a few interventions, the effects of such a large change in human reproductive behavior would have significant forward linkages throughout each of the model’s systems.

Four of the prepackaged scenarios located in the folder Interventions and Agent Behavior contain additional examples of the demographic module’s parameters: Non OECD Contraception Use Slowed, Non OECD Contraception Use Accelerated, World Migration High, and World Migration Low. The pair of scenarios focusing on contraceptive usage rates both utilize contrusm. In the accelerated scenario, the multiplier takes the value 1.2 in non-OECD nations; and the value 0.8 in the slowed scenario for all non-OECD nations. The two alternate migration scenarios similarly feature interventions on a single parameter: the global migration multiplier wmigrm. In the high scenario the parameter takes on a value of 2, doubling global migration flows; and in the low scenarios flows are halved, with wmigrm declining to a value of 0.5.

Health Module

Variables of Interest

Variable Name	Description
LIFEXP/LIFEXPHLM	Life Expectancy
CDR	Crude Death Rate
DEATHCAT	Deaths by Mortality Type
HLYLL	Years of Life Lost
HLYLLWORK	Years of Working Life Lost
HLYLD	Years Lived with Disability
HLDALY	Disability Adjusted Life Years Lost
INFMOR	Infant mortality rate
HLSTUNT	Percentage of population stunted
MALNCHP	Percentage of children malnourished
MALNPOPP	Percentage of population malnourished
HLBMI	Body Mass Index
HLOBESITY	Percentage of population obese
HLSMOKING	Percentage of population that smokes

The primary variables of interest in the IFs health module are those that pertain to mortality and morbidity due to a variety of causes. LIFEXP and CDR, discussed in the population module, provide basic measures of population health. DEATHCAT provides a measure of the number of deaths (in thousands) due to different categories of mortality. IFs can display health variables in the following categories of disease: Other Communicable Disease, Malignant Neoplasm, Cardiovascular, Digestive, Respiratory, Other NonCommunicable Diseases, Unintentional Injuries, Intentional Injuries, Diabetes, AIDs, Diarrhea, Malaria, Respiratory Infections, and Mental Health. Using the Flexible Display form, it is also possible to see many of these variables in the rolled-up categories of Communicable Disease, Non-Communicable Disease, and Injuries or Accidents. Because different health conditions affect age cohorts differentially, the above measure is insufficient in understanding the full impact of ill health. For this reason, it is also possible to break down the actual number of deaths accruing to each cohort, sex, and cause via the Specialized Display menu under the health heading. For example, both the Mortality by Age, Sex, and Cause and the J-Curve displays provide useful information about the health status of a country.

Three other measures help to enrich the picture: HLYLL, HLYLD and HLDALY. Like DEATHCAT, these aggregate (across age-cohort) measures are available by cause and country. HLYLL is a measure of the number of life years lost due to premature death. It differs from the DEATHCAT variable because it represents the burden of premature mortality In terms of life years lost, which allows us to account for the fact that some diseases, like HIV/AIDS, primarily affect younger people, while others, like cardiovascular disease, are primarily fatal in older adults. Although the total number of deaths may be the same between two countries for each cause, there may be significant differences between two countries’ health profiles in terms of YLLs.

HLYLD is another measure that represents the burden of ill health in terms of life years of impact. It indicates the burden of years lived with disability or disease. In calculating YLD, IFs uses the disability weights that WHO created to rank the relative severity of different conditions and their impact on productivity.

Finally, Disability Adjusted Life Years (DALYs) are a measure of morbidity (disability or infirmity due to ill health). HLDALY sums YLLs and YLDs to create a measure of the number of years of life lost to both premature mortality and morbidity due to ill health. Like the other measures discussed above, DALYs can be broken down by different disease categories within IFs. The DALY is probably the most expansive measure of ill-health within a population because it includes mortality burden by age of death and the lost quality of life for those who did not die from health events, but who are disabled by them in some way.

Other measures provide indicators of health in regard to certain specific risk factors for disease or among certain segments of the population. Infant mortality, INFMOR, can be used to assess the burden of ill health among children under one year of age. HLSTUNT, displays the percentage of the population who are stunted (have low height for age),while MALNCHP and MALNPOPP, provide information on the percentage of the child and adult population who are malnourished respectively. The variables INFMOR, HLSTUNT and MALNCHP are especially useful for assessing the burden of ill health due to communicable diseases and other conditions that primarily affect children. By contrast, the variables HLBMI, HLOBESITY, and HLSMOKING provide risk factor information on diseases that affect primarily adults. HLBMI represents the body mass index in a country while HLOBESITY and HLSMOKING provide information on the percentage of the population that is obese or smokes.

Other variables that will be useful to users interested in specific conditions or subpopulations include indicators on stunting and BMI, as well as smoking and obesity. Variables for HIV/AIDS are also available and discussed separately below in the subsection on the HIV/AIDS sub-module.

Parameters to Affect Overall Health and Burden of Disease

Parameter	Variable of Interest	Description	Type
hlmortm	DEATHCAT/HLYLL/HLDALY	Multiplier on Mortality (by cause)	Multiplier
hlmorbm	YLD	Multiplier on morbidity	Multiplier
hlstddthsw	DEATHCAT	Switches DEATHCAT from absolute numbers to deaths/1000	Switch

The above parameters provide simple ways to directly affect the burden of disease within a country. The most important parameter for modifying mortality rates is hlmortm, a parameter that allows users to increase or decrease the prevalence of deaths in any particular category of illness. IFs modifies mortality in the following categories: Other Communicable Disease, Malignant Neoplasm, Cardiovascular, Digestive, Respiratory, Other NonCommunicable Diseases, Unintentional Injuries, Intentional Injuries, diabetes, AIDs, Diarrhea, Malaria, Respiratory Infections, and Mental Health. Altering the mortality burden will affect the variables DEATHCAT, HLYLL, and HLDALYs. The parameter will indirectly affect morbidity because of its direct link to mortality. In the case of Mental Health Diseases, the parameter will not have much impact on DEATHCAT, but may have a significant impact on the number of DALY’s experienced by a population. Because hlmortm is a multiplier, increasing its value from 1 to 1.2 represents a 20% increase in the burden of mortality from a particular cause. A similar parameter, hlmorbm, allows users to affect morbidity directly through a brute force multiplicative parameter. This allows users to affect the years lost to disability in a working life and by extension multifactor productivity due to human capital (MFPHC). The hlstddthsw allows users to switch between displaying DEATHCAT in absolute numbers to deaths per thousand people.

Parameters that Affect Communicable Diseases

Parameter	Variable of Interest	Description	Type
watsafem	WATSAFE, INFMOR	Percentage of population with access to safe water	Multiplier
sanitationm	SANITATION, INFMOR	Percentage of population with access to improved sanitation	Multiplier
malnm	MALNCHPSH	Prevalence of child malnutrition	Multiplier
ylm	YL	Yield multiplier on agriculture	Multiplier
hivm	HIVCASES	Rate of HIV infection	Multiplier

Above are a number of the parameters that users may wish to use to manipulate communicable diseases (which predominantly affect children). Ylm is a multiplicative parameter in the agriculture module that can be used to change the yield of agricultural lands within a country, affecting the number of calories available for consumption, and thereby altering the rates of malnutrition and obesity. Watsafem and sanitationm, in the infrastructure module, influence the percentage of the population that has access to safe water and sanitation respectively, thus decreasing childhood exposure to diarrheal disease, malnutrition and premature death. Other parameters to control safe water and sanitation access are discussed in the infrastructure section of the model. Finally, although HIV is more thoroughly discussed in the HIV/AIDs submodule, one brute force parameter is worth noting here. Hivm allows users to directly affect the rate of infection with the HIV virus.

Parameters that Affect Non-Communicable Disease

Parameter	Variable of Interest	Description	Type
envpm2pt5m	ENVPM2PT5	Increases levels of environmental pollution	Multiplier
hlsmokingm	HLSMOKING	Increases rate of smoking	Multiplier
hlobesitym	HLOBESITY	Prevalence of obesity	Multiplier
hlbmim	HLBMI	Multiplier on body mass index	Multiplier

Hlsmokingm is a multiplicative parameter that will change the rate of smoking, which will affect the prevalence of respiratory diseases. Envpm2pt5m is a multiplicative parameter that will change the level of ambient environmental pollution in terms of parts per million; higher levels of environmental pollution are a risk factor for both communicable and non-communicable respiratory diseases.

Hlobesitym works similarly to affect the prevalence of obesity within a society in the absence of overall caloric intake changes. This parameter can be used to model the impact of changing levels of physical activity within a society. Both of the above parameters work similarly to other multiplicative parameters: increasing the value of the parameter to 1.2 from 1, represents a 20% increase in the value of the parameter over the base case. By the same token, users can use hlbmim to affect the body mass index in a country, a major risk factor for cardiovascular diseases, diabetes, and overall morbidity. Please note: hlobesitym affects only obesity rates and has no affect on health; in contrast, hlbmim will affect body mass index, obesity, and deaths from heart disease and diabetes.

Parameters that Affect Injuries and Accidents

Parameter	Variable of Interest	Description	Type
deathtrpvm	DEATHTRPV	Multiplier on traffic deaths per vehicle	Multiplier
deathtrpvsetar, deathtrpseyrtar	DEATHTRPV	Standard error target for traffic deaths per vehicle	Relative target Value/Year

Only a small set of parameters allow users to affect injuries and accidents, and these primarily revolve around reducing traffic deaths. Users may reduce traffic deaths as a ratio of the number of vehicles in a country using either a multiplier, deathtrpvm, or a pair of standard error targeting parameters, deathtrpvsetar and deathtrpseyrtar. Standard error targeting is discussed in detail in the infrastructure module. These parameters allow users to model the impact of road safety on mortality and, by extension, on economic productivity.

Parameters to Affect Technology

Parameter	Variable of Interest	Description	Type
hlmortmodsw	CDR	Reduces crude death rate in Africa, Europe, Southeast Asia, West Pacific	Switch
hltechshift	CDR	Rate of change in health technology	Additive factor
hltechlinc	CDR	Rate of change in health technology in low income countries	Additive factor
hltechssa	CDR	Rate of change in health technology in Sub-Saharan Africa	Additive factor
hltechbase	CDR	Rate of change in health technology at base	Exogenous specification

Aside from the direct and indirect parameters affecting health, the distal drivers of health include per capita GDP, years of education, and technology. Per capita GDP is an element of the economic module and can be changed in a number of ways, but especially by changing the elements that make up multifactor productivity. Years of education is an element of the education module and can be changed by altering the duration of schooling, and the completion rate.

Moving to the third distal driver of health, there are a number of parameters built into the health module that can be used to alter the rate of technological change. Hlmortmodsw is a master switch that, when set to 1 as in the Base Case default, reduces technological progress for low-income countries of Africa, Europe, Southeast Asia, and West Pacific based on geographic and income categories. There are parameters available to alter these assumptions about differentials in mortality declines in these regions, but they only have an effect in the base case; when hlmortmodsw is set to 0 these parameters have no impact.

Once hlmortmodsw is set to 1, users can manipulate mortality patterns through several parameters. Hltechshift, alters the rate of change for health technology impacts relative to GDP. The hltechshift parameter allows users to change the mortality rate using a shift parameter that alters the technology factor affecting mortality decline relative to initial GDP. Hltechlinc and hltechssa can be used to change the rate of technological advance resulting in mortality decline in low-income countries (hltechlinc) and sub-Saharan Africa (hltechssa) specifically. Meanwhile, the hltechbase parameter allows users to change the base level of technological change across the 20 world, rather than country by country as you can do using the hltechshift parameter. All of these parameters pertain to all causes of mortality except cardiovascular mortality, which uses a different regression equation.

Prepackaged Scenarios

Three major integrated scenarios on health were developed by the Pardee Center for the health volume of the Patterns of Potential Human Progress series (Hughes et al., 2011). The World Integrated Scenario Sets folder contains the scenarios that were built for this volume, of which three are worth an extended discussion. The first is the Proximate Drivers Excluding Environment folder, which contains parameters to individually alter four of the major risk factors for several causes of mortality. These are Body Mass Index which is a risk factor for cardiovascular disease; under nutrition, which is a risk factor for communicable diseases; smoking which is a risk factor for respiratory disease; and large increases in the number of cars per person coupled with poor pedestrian safety, which is a major risk factor for accidental death. This scenario also includes increased to improved water sources and piped sanitation taken from the infrastructure module, and parameters to reduce environmental exposure to poor air quality. This scenario reduces these risk factors to their theoretical minima, to simulate aggressive efforts to reduce, high BMI, the obesity rate, childhood malnutrition, smoking, and traffic mortality. Malnutrition is set to 0.01, smoking and obesity multipliers are set to 0, BMI multiplier to 0.8, vehicle fleets to 0.5, and traffic mortality to 0.

Another important pair of prepackaged scenarios contains the optimistic Luck and Enlightenment scenario, and a scenario that considers what happens when Things Go Wrong. The Luck and Enlightenment scenario includes improvements to HIV/AIDS, sanitation access, improved air quality, and reduced smoking rates which help lower the burden of NCDs. It also features changes to the burden of communicable disease designed to increase the levels of these. A variation to Luck and Enlightenment has add-ins that also increase foreign aid donations and agricultural yields, effectively modeling a situation in which increased global cooperation supports these efforts. Things Go Wrong models a world in which air quality worsens, smoking and obesity rates increase and there is little international cooperation on addressing these challenges.

HIV/AIDS Submodule

Variables of Interest

Variable Name	Description
HIVCASES	Number of HIV cases
HIVRATE	HIV infection rate
HIVTECCNTL	Rate of technical control of infection, cumulative reduction in infection rate
AIDSDTHS	Number of AIDS deaths
AIDSDRATE	Death rate from AIDS
AIDSDTHSCM	Cumulative Number of AIDS deaths since first year of model

HIV and AIDS have attracted significant interest among policy makers because of the tremendous toll that these diseases have on populations in both human and economic terms. Because of this interest, it is worth discussing the HIV/AIDS submodule separately from the rest of the health module. That submodule represents both the extent of HIV prevalence in a population (a stock variable) and the annual deaths from AIDS (a flow variable driven in substantial part by the prevalence rate, but also responsive to technological advance in the fight against AIDS). A number of key variables are available to represent the burden of HIV and AIDS within a country.

Three variables are key to understanding the progression of infection within a country. HIVCASES provides the total number of HIV cases, HIVRATE represents a flow variable showing the rate at which people are being infected with HIV, and HIVTECCNTL indicates the progress being made in reducing the rate of infection within a country.

Three other variables assess mortality due to HIV and AIDs within a country. Similar to HIV, the variables AIDSDTHS and AIDSDRATE indicate the number of AIDs deaths and the rate of mortality from AIDs respectively, while AIDSDTHSCM represents the cumulative number of deaths due to the disease.

Parameters to Affect Prevalence

Parameter	Variable of Interest	Description	Type
hivm	HIVRATE	HIV infection rate, multiplier of percent of population infected	Multiplier
hivtadvr	HIV CASES/ HIVRATE	Technical advance rate in of control of infection	Rate of change
hivmdcm	HIVRATE	HIV infection rate maximum for MDCs, multiplier	Rate of change
hivpeakr	HIVCASES/ HIVRATE	HIV infection rate at year of peak	Target value
hivpeakyr	HIVRATE	Sets year of epidemic peak	Target year
hivincr	HIVCASES	HIV increase rate, only used prior to 2000	Rate of change

Modifying the infection rate with hivm is probably the easiest way to adjust the burden of HIV infection within a country. Like hlmortm, hivm is a multiplicative parameter. In other words, increasing the value of the parameter in scenario analysis from 1 to 1.2 represents a 20% increase in the rate of infection relative to the base case. Hivtadvr allows users to change the prevalence of HIV, once the epidemic has peaked, by a certain percent annually to model different assumptions about the rate at which control technologies will improve, reducing the prevalence of the disease over time. Unlike the mortality multiplier, which takes effect once the model has calculated the base Variable Name Description HIVCASES Number of HIV cases HIVRATE HIV infection rate HIVTECCNTL Rate of technical control of infection, cumulative reduction in infection rate AIDSDTHS Number of AIDS deaths AIDSDRATE Death rate from AIDS AIDSDTHSCM Cumulative Number of AIDS deaths since first year of model 22 case, this parameter will affect the actual calculations the model makes while running. This parameter functions as additive factor to a growth rate within IFs. In other words, a 0.01 increase in the parameter represents a 0.01 increase in the growth rate for the technical advance rate in HIV infection control (hivtadvr).

The HIV submodule is designed to allow users to affect the course of the epidemic across countries and across time. The multiplier hivmdcm is a multiplicative parameter that affects the maximum infection rate in middleincome developing countries. Another way to alter the course of the epidemic is by manipulating the coefficient on hivpeakr, which is an additive parameter that will increase the peak rate of infection over the course of the epidemic. Thus a 0.01 increase in the value of the coefficient represents a 0.01 increase in the peak infection rate. An associated parameter, hivpeakyr sets the date at which the epidemic will peak before the infection rate begins to decline. Changing this parameter in the Scenario Analysis page will allow users to set any year between 2010 and 2100 as the year of peak infection rate depending on their assumptions regarding the technical rate of advance in controlling the disease. Finally, the parameter hivincr controls the increased rate in infection prior to 2000, when our knowledge of the epidemic was much less complete and control efforts were far less effective.

Parameters to Affect Mortality

Parameter	Variable of Interest	Description	Type
aidsdrtadvr	AIDSDTHS/AIDSRATE	AIDs death rate, technical annual advance rate in control	Rate of change
aidsdratem	AIDSRATE	AIDs death rate as % of HIV infection rate, multiplier	Multiplier

Just as there are a variety of parameters available to control the prevalence of HIV within a population, there are also a number of parameters that allow users to control the lethality of the epidemic. The first of these parameters allow user to change the death rate as a percentage of the infection rate via the parameter aidsdratem. This parameter directly alters the lethality of the disease; it serves as a proxy for the presence or absence of control measures within a country since the availability of anti-retroviral medications will affect the rate at which people who are HIV positive die from AIDs. Of course, new research strongly suggests that ART therapies may also significantly reduce the HIV infection rate as well, but because these are not yet linked in the model, users should be aware that a more realistic use of this parameter would alter not only the AIDs mortality rate, but the infection rate as well. The other parameter available to users to control mortality from AIDs is aidsdrtadvr, a parameter which changes the technical annual advance rate in control. This parameter simulates the annual advance in technologies to control AIDs mortality, altering the lethality of the disease.

Prepackaged Scenarios

There are several prepackaged scenarios that deal with the HIV/AIDS epidemic. The first, under the heading Hivaids in the Technological Advance section of the Framing Scenarios folder, models two scenarios around technological advance to control the epidemic. One models rapid technical advances to control HIV infection, while the other presents a scenario in which technological progress slows, slowing the resulting decline in infections.

A second set of prepackaged scenarios are available to affect HIV/AIDS are focused on altering the course of the epidemic in key countries, rather than at a global level. They are called: Intermediate HIV/AIDS, Intermediate for New School Paper, Severe HIV Aids and Total Failure to Control HIV AIDs and are located in the Surprises and Wildcards folder, under the heading AIDs. These scenarios modify the course of the AIDS epidemic in Russia, China, India, and the world at large. Each one affects parameters controlling the infection rate at the peak year of the epidemic, the peak infection rate, the initial rate of infection, the rate of advance in the infection, and the elasticity of multifactor productivity to life expectancy. They give a good example of how to modify combinations of parameters in specific countries to create different trajectories for the epidemic.

Education Module

Variables of Interest

Variable Name	Description
EDYRSAG15 EDYRSAG25 EDYRSAG15TO24	Educational attainment, adults by age group
EDPRIPER EDSECPER EDTERPER	Percent of the population completing primary, secondary and tertiary
EDPRIENRG EDSECENRG EDSECLOWRENRG EDSECUPPRENRG EDTERENRG	Gross enrollment rate in primary, secondary, lower secondary, upper secondary and tertary
EDPRIENRN EDSECENRN	Net enrollment rate in primary, secondary
EDPRIINT EDPRIINTN EDTERINT	Primary intake rates, gross and net, and tertiary intake rates, gross
EDSECLOWRTRAN EDSECUPPRTRAN	Transition rates from primary to lower secondary and from lowers secondary to upper secondary
EDPRISUR EDSECLOWRSUR EDSECUPPRSUR	Survival rates for primary, lower secondary, upper secondary
EDPRICR EDSECLOWRGRATE EDSECUPPRGRATE EDTERGRATE	Graduation rates in primary, lower, secondary, upper secondary and tertiary
EDSECLOWRVOC EDSECUPPRVOC	Vocational enrollment rates, lower and upper secondary
EDEXPERPRI EDEXPERSEC EDEXPERTER	Government spending per student as a percentage of GDP for primary, secondary and tertiary
GDSED	Spending shares by level of education

Like all of IFs, the education module is amenable to systems thinking, or conceptualizing elements as stocks and flows. Students flow through primary, lower-secondary, upper-secondary and tertiary education levels. Each time an age cohort completes a grade level, a year is added to that group’s stock of educational attainment; as adults (aged either 15 and older or 25 and older) they will have attained some number of years of formal education based on grades completed. For any given grade, the number of students enrolled is determined by the intake rate—or, at the secondary level, the transition rate from the previous level of education—and by the percentage of students passing through the last grade. Government spending influences the system by restraining the number of students that can be sustained at a given level of education.

Most analysis of the international education system focuses on educational attainment. IFs operationalizes this stock as the average years of education successfully completed by adults in all countries, following a typology designed by Harvard economists Robert Barro and Jong-Wha Lee (2000, 2001). EDYRSAG15 captures the mean years of education attained for all adults older than 15; this important variable links forward to several other sectors of the model, including economics through multifactor productivity, health through fertility rate, and even governance through state fragility indicators. Users may wish to view mean years of education by other age breakdowns, available through EDYRSAG25 and EDYRSAG15TO24 though these variables do not have the forward linkages in the model that EDYRSAG15 does. As an alternative to average years of education, EDPRIPER, EDSECPER and EDTERPER track attainment as a percentage of the population to successfully complete each level of education.

In addition to attainment measures, most users will focus their attention on enrolment rates, the most common measure of student flows. Enrolment rates answer the simple question: at a given time, how many children are in school?

Enrollment is captured in the variables EDPRIENRG and EDPRIENRN for primary,EDSECENRG,EDSECENRN,EDSECLOWERENRG and EDSECUPPRENRG for secondary and EDTERENRG for the tertiary level. Many of the indicators used to measure education systems are either gross rates or net rates. This important distinction has to do with student age. Gross Enrollment Rates (GER), for example, are calculated by dividing the total student body by the official school-age population; Angola’s primary GER would be the number of primary students enrolled in school divided by the primary age population of the country and has often exceeded 100 percent because of primary enrollment by over-aged students who enter or return to school for various reasons. Net Enrollment Rates (NER), on the other hand, measure only the students of official school age. A country’s NER is the enrolled age-appropriate students divided by only the school-age population and should not exceed 100 percent. Net enrolment data are often more difficult to find than gross, simply because it requires the age of enrollees to calculate. Intake rates are also presented in both net and gross formulations. For a more detailed description of commonly used education indicators see UNESCO’s technical guidelines: http://www.uis.unesco.org/Library/Documents/eiguide09-en.pdf.

For each of the countries in IFs, two factors ultimately control enrolment rates: intake rates and survival rates. Drawing on drivers such as GDP per capita and educational spending, the model forecasts intake rates for primary and tertiary levels (EDPRIINT, EDPRIINTN, and EDTERINT) and transition rates from the previous level of schooling for secondary education (EDECLOWRTRAN EDSECUPPRTRAN). The model also forecasts a student persistence rate for all levels (except tertiary), often referred to as survival rate, captured in the variables EDPRISUR, EDSECLOWRSUR, and EDSECUPPRSUR. Enrolment rates at each level are calculated as the combined result of these student flows. Or thought of more simply, enrolment rates track the number of students who enter school and stay in school. Graduation rates at all levels are also forecast and the corresponding variables are: EDPRICR, EDSECLOWRGRATE, EDSECUPPRGRATE and EDTERGRATE. The model also forecasts vocational shares in lower secondary and upper secondary and the variables are EDSEECLOWRVOC and EDSECUPPRVOC.

All indicators in the model that measure the stock of educational attainment or student flows can be disaggregated by gender. Thus, the user is able to evaluate each country’s progress toward eliminating gender-based disparity in the classroom. In order to disaggregate performance by gender, a second dimension of almost all education variables allows the selection of display by male, female or total. In addition to setting normative goals, separate treatment of female education allows the model to establish powerful forward linkages. For example, female educational attainment and fertility rates are negatively related, because schooling increases the opportunity cost of having and raising a child.

In addition to student flows, the model captures financial flows tied to each country’s education system. Educational finance is the result of interplay between supply-side and demand-side forces. Demand for education can be conceptualized as the product of spending per student and enrollment rates at each level of schooling—EDEXPERPRI, EDEXPERSEC and EDEXPERTER track spending per student as a percent of GDP per capita for primary, secondary and tertiary education. However, such a formulation of demand is crude, as the number of children already in the system is constrained by the preexisting structure of government spending. Financial supply, captured as a percent of GDP by the variable GDSED, is generated by the government finance submodule. Due to a lack of reliable data, IFs does not capture private educational funding, though it is important in some countries.

Parameters to Affect Intake Rates and Survival Rates: Annual Growth

Parameter	Variable of Interest	Description	Type
edpriintngr	EDPRIINTN	Primary net intake rate	Annual growth
edterintgr	EDTERINT	Tertiary intake rate	Annual growth
edseclowrtrangr	EDSECLOWRTRAN	Lower secondary transition rate	Annual growth
edsecupprtrangr	EDSECUPPRTRAN	Upper secondary transition rate	Annual growth
edprisurgr	EDPRISUR	Primary survival rate	Annual growth
edseclowrsurvgr	EDSECLOWRSUR	Lower secondary survival rate	Annual growth
edsecupprsurvgr	EDSECUPPRSUR	Upper secondary survival rate	Annual growth
edtergradgr	EDTERGRAD	Tertiary graduation rates	Annual growth

In the IFs model, improved education outcomes can be achieved most directly through changes to intake and survival rates. The number of students entering the first grade of each schooling level can be pushed up or down using annual (percentage point) growth parameters: edpriintngr for primary school, edseclowrtrangr for transfer rates between primary and secondary, edsecupprtrangr for transition between lower and upper secondary andedterintgr for intake into the tertiary level. However, any improvements to enrolment stemming from intake rate boosts are restrained when coupled with high dropout or repetition rates. Like intake rates, survival rates can be directly manipulated through annual growth parameters. Edprisurgr controls survival growth (in percentage points) at the primary level. Secondary school rates can be shifted using two parameters, edseclowrsurvgr for lower secondary and edsecupprsurvgr for upper. And finally, edtergradgr controls tertiary survival rates (also referred to as graduation rates).

 Though intake rates and survival rates can be manipulated separately, users will often wish to manipulate them in conjunction due to the presence of interaction effects between the two sets of variables. For example, a rapid increase in intake/transition rates can result in a decline in country’s survival rate. The educational system may simply not have the resources to cope with an influx of students. Furthermore, the incoming cohort, which has spent years outside the system, may represent individuals from disadvantaged backgrounds—a group prone to dropping out. On the other hand, rising intake rates may have a positive threshold effect on survival rates. In many countries, once intake rates reach a certain level, dropout and repetition have declined. At this time such relationships are not captured in IFs. Thus, the user may want to build interaction and threshold effects into education policy interventions.

Parameters to Affect Intake Rates and Survival Rates: Target Year for Universal Education

Parameter	Variable of Interest	Description	Type
edpriintntrgtyr	EDPRIINTN	Primary net intake rate	No. of years to reach 100%
edseclowrtrantrgtyr	EDSECLOWRTRAN	Lower secondary transition rate	No. of years to reach 100%
edsecupprtrantrgtyr	EDSECUPPRTRAN	Upper secondary transition rate	No. of years to reach 100%
edterinttrgtyr	EDTERINT	Tertiary Intake Rate	No. of years to reach 100%
edprisurtrgtyr	EDPRISUR	Primary survival rate	No. of years to reach 100%
edseclowrsurtrgtyr	EDSECLOWRSUR	Lower secondary survival rate	No. of years to reach 100%
edsecupprsurtrgtyr	EDSECUPPRSUR	Upper secondary survival rate	No. of years to reach 100%
edtergradtrgtyr	EDTERGRAD	Tertiary graduation rates	No. of years to reach 100%

Development goals like universal access to education with a specific target date can be explored through target year parameters on intake (transition) or survival (graduation) rates. These parameters ramp up the base year value of the variable of interest to one hundred percent within the number of years marked by the parameter. For countries with limited availability of government resources, these target parameters might need an education budget set aside through the edbudgon parameter described later in the budget sub-section .

Parameters to Affect Intake Rates and Survival Rates: Multiplier

Parameter	Variable of Interest	Description	Type
edpriintnm	EDPRIINTN	Primary net intake rate	Multiplier on base case
edterintm	EDTERINT	Tertiary intake rate	Multiplier on base case
edseclowrtranm	EDSECLOWRTRAN	Lower secondary transition rate	Multiplier on base case
edsecupprtranm	EDSECUPPRTRAN	Upper secondary transition rate	Multiplier on base case
edprisurm	EDPRISUR	Primary survival rate	Multiplier on base case
edseclowrsurvm	EDSECLOWRSUR	Lower secondary survival rate	Multiplier on base case
edsecupprsurvm	EDSECUPPRSUR	Upper secondary survival rate	Multiplier on base case
edtergradm	EDTERGRAD	Tertiary survival rates	Multiplier on base case

In addition to the growth and target parameters described above, intake (or transition) and survival (or graduation) rates can also be modified using a set of multipliers listed in the table above. Like other multipliers in the model these work by ramping up (or down) the base case forecast over a horizon chosen in the scenario design.

Parameters to Affect Education Spending

Parameter	Variable of Interest	Description	Type
gdsm (education)	GDS	Government spending on education	Multiplier
edbudgon	GDS	Education funding impact and priority	Exogenous specification
gdsedm	GDS	Education spending distribution	Multiplier
edexppconv	EDEXPERPRI	Education spending per student, primary	Convergence speed
edexpslconv	EDEXPERSEC	Education spending per student, secondary	Convergence speed
edexpsuconv	EDEXPERSEC	Education spending per student, secondary	Convergence speed
edexptconv	EDEXPERTER	Education spending per student, tertiary	Convergence speed
edqtqltrm	EDEXPERPRI, EDEXPERSEC , EDEXPERTER and GDS	Education quantity-quality balance	Multiplier

Attainment levels also respond indirectly to changes in the educational spending structure. The number of students enrolled at all grade levels is constrained by total public spending on education. In IFs, public finance allocations are distributed between transfer payments, the military, education, health and infrastructure in the government budget submodule. Bottom-up factors like demographic changes and policies targeting intake or survival will pressure the government to increase education spending. But the model features somewhat rigid top-down control of the budget—spending on education competes with other government spending and IFs maintains accounting of both total government revenues and expenditures. Most countries spend something close to the global average of 5 percent of GDP on education each year. Thus demand-side shifts may secure an increase in resources, but it is difficult to increase total education spending much in excess of historically observed values. Users can intervene in this accounting process by manipulating gdsm (education), a multiplier that directly increases the share of government expenditure accounted for by education, at the expense of spending in other arenas such as military or health. Alternatively, users can prioritize education spending over other government expenditure targets. The priority parameter edbudgon, takes on values from zero to one, with lower values representing lower priority to education funding demands and one representing maximum priority. Any non-zero value pushes education allocations away from those determined by historical spending patterns and level of economic development—as calculated in the economic module—and toward the bottom-up demand for expenditures calculated in the education model. The parameter edbudgon can also be used to completely turn off the budget impacts on student flow rates through a zero value of the parameter. The turning off of budget impacts is helpful in projecting the demand for funds for certain types of education intervention, e.g., universal primary education. However, fund demand in such a scenario is calculated in the EDTOTCOST and not in the GDS(Education) variable, which is determined, in this case, completely through the government budget allocation algorithm without using any input from the education model.    

 The government-budget submodule automatically allocates resources to the educational level most in need (although that interacts with historical patterns of educational preference), starting with primary education. As enrolment rates in a country’s primary education system reach a very high level, the model reallocates funding up the chain to secondary. Users may wish to change this pattern and create normative scenarios that prioritize higher levels of education, or balance spending between primary, secondary and tertiary. The gdsedm parameter allows for such experimentation via a second dimension that allots the spending increase to a specific level of schooling.

When developing education budgets, planners must account for trade-offs between equity and efficiency. What is more important, increasing total spending on the system in an effort to support more students, or increasing spending per student in an effort to improve the experience of being in school? In a crude way this can be thought of as the difference between increasing intake and increasing survival rates. IFs includes parameters that shift the spending focus in the direction of system efficiency. Edexppconv can be used to decrease the time it takes a country’s initial spending per primary student to converge with the model’s expected function. Edexppslconv and edexppuconv control the student expenditure function for lower and upper secondary, andedexpptconv does the same for the tertiary level. Increasing the target value used in a convergence parameter will increase per student spending for countries below the expected value, but the opposite is true in countries exhibiting higher than expected values. If a user wishes to boost per student expenditure in a high performing country, he or she should make use of the edqtqltrm parameter, a multiplier that shifts emphasis away from education quantity and towards education quality—with values less than one indicating a shift towards quality.

Parameters to Affect Gender Parity

Parameter	Variable of Interest	Description	Type
edprigndreqintn	EDPRIINTN	Gender parity for primary intake	Convergence speed
edprigndreqsur	EDPRISUR	Gender parity for primary survival	Convergence speed
edseclowrgndreqsurv	EDSECLWRSUR	Gender parity for lower secondary survival	Convergence speed
edseclowrgndreqtran	EDSECLWRTRAN	Gender parity for lower secondary transition	Convergence speed
edsecupprgndreqsurv	EDSECUPPRSUR	Gender parity for upper secondary survival	Convergence speed
edsecupprgndreqtran	EDSECUPPRTRAN	Gender parity for upper secondary transition	Convergence speed
edtergndreqint	EDTERINT	Gender parity for tertiary intake	Convergence speed
edtergndreqgrad	EDTERSUR	Gender parity for tertiary survival	Convergence speed

Some development projects focus on gender disparity in education, rather than aggregate enrolment. In recent decades male and female enrolment rates have converged rapidly across all regions of the world. In fact, primary, secondary and tertiary enrolment taken as a global aggregate has already surpassed a female-male ratio of 0.97—commonly considered an indicator of parity. However this abstracted view obscures struggles facing women in low-performing areas, especially sub-Saharan Africa. Users may explore effects linked to improvements in female-specific enrolment rates via the intake and survival parameters described above; each parameter has a dimension that applies the annual increase to males, females, or the total student population. Alternatively, users can experiment with closing the gender attainment gap existing in some countries by setting time sensitive goals. Goals around intake can be set through edprigndreqintn for primary,edseclowrgndreqsurv and edsecupprgndreqtran for secondary and edtergndreqint for the tertiary level. Gender conversion goals may also be set for the other direct driver of enrolment, survival rates: edprigndreqsur controls female survival rates at the primary level, edseclowrgndreqsurv and edsecupprgndreqsurv control rates for lower and upper secondary, andedtergndreqgrad controls tertiary rates. Each of the gender-focused parameters are convergence speed goals, meaning the user sets the number of years that will elapse before the country’s gender parity in the corresponding measure converges to the expected function.

Prepackaged Scenarios

 An installation of IFs includes scenarios used in each entry in the Pardee Center’s Potential Patterns of Human Progress (PPHP) series. Scenarios created for the education volume include: a best and worst case framing scenario; case study interventions in several countries and regions; high spending and low spending scenarios; and over twenty normative scenarios (Dickson et al., 2010). This section will briefly describe the parameters used to construct a single normative scenario titled, Norm Mar 1 2009.

Norm Mar 1 2009 simulates a worldwide improvement in educational performance through interventions to: intake rates, student persistence, gender parity and educational spending, including boosts to foreign aid. From 2010 to 2100, the parameter edpriintngr is set to 2.2, and edprisurgr is set to 1.2, setting up annual increases to primary school intake and survival rates. The scenario includes similar annual growth increases for intake and student persistence across lower secondary and upper secondary schooling. Edseclowrtrangr is set to one percent, edseclowrsurvgr to 0.8 percent, edsecupprtrangr to 0.5 percent, and edsecupprsurvgr is set to 0.3 percent. The corresponding parameters for tertiary education are not assigned a value other than the default of zero. Additional interventions speed up the narrowing of persisting gender gaps in intake, survival and transition rates. The parameters edprigndreqintn and edprigndreqsur set a ten year time goal for gender parity at the primary level; edseclowrgndreqtran and edseclowrgndreqsurv set a thirteen year goal for parity at the lower secondary level; and,edsecupprgndreqtran andedsecupprgndreqsurv set a 20 year time goal for parity at the upper secondary level. In order to simulate improved education quality, across levels, edexppconv, edexpslconv, edexpsuconv and edexptconv all take the value of 20, signifying a convergence to the expected function after 20 forecast years. Because this scenario was created as a tool to explore the potential impacts of a world in which educational goals are achieved by all nations, the budget switch, edbudgon, is set to 0, the highest prioritization of education funding.