Modernization Hub

Modernization and Improvement
How FEWS NET Makes Projections: An Introduction to Scenario Development

How FEWS NET Makes Projections: An Introduction to Scenario Development


Imagine you are a decision maker responsible
for planning future food assistance. You need to anticipate who will be in need,
where they are located, and when they will need assistance. All of this requires you to think about the
future. You would probably consider rainfall and prospects
for the harvest. But that’s not all. Staple food prices, labor demand, livestock
health, import levels, migration patterns, crop diseases, market functioning, and conflict
– these are just a few of the many other factors that you might need to consider to
analyze the big picture. How would you make sense of all that information,
recognizing that you must project several months in the future? How would you determine priorities? At FEWS NET, we use a methodology known as
scenario development to assist in projecting food insecurity and future food assistance
needs. In the simplest terms, scenario development
is a type of “if-then” analysis. In other words: “If this event happens,
then these would be the likely consequences.” Scenarios are rooted in assumptions about
critical events, based on the best evidence available at the time that the scenario is
built. At FEWS NET, scenario development is the methodology
we use to make projections about acute food insecurity outcomes. This approach is the basis of the analysis
presented in our food security outlooks and updates. Scenario development allows FEWS NET to meet
its core mandate of giving decision makers early warning about potential food security
crises. Before going through the details, let’s
understand the overarching principles of FEWS NET’s analytical process. First, scenario development integrates the
analysis from our main technical sectors: livelihoods, agroclimatology, markets and
trade, and nutrition. Second, FEWS NET analysis is livelihoods-based. Livelihoods are the ways that people access
food and earn income. This information is normally conveyed through
maps and descriptions of the agro-ecology, market systems, and other geographical and
political factors that determine livelihood patterns. FEWS NET relies on this information to identify
who is vulnerable to hazards, why they are vulnerable, and how those people typically
respond to shocks or unusual events. A third key point is that FEWS NET analysis
is aligned with the IPC, the Integrated Food Security Phase Classification. This tool is the international standard used
to classify the severity of food insecurity, and provides a common framework and language
to describe household food insecurity. Because of that common language, decision-makers
can compare crises across national borders, and from one part of the world to another,
and over time. A final point is that FEWS NET classifies
food security outcomes. This is for two reasons. First, outcomes are comparable across countries,
while conditions – like below-average rainfall or conflict – are not. We also focus on outcomes because we’re
interested in the impacts on people. The primary outcome we look at is food consumption. For acute food insecurity, we mostly focus
on dietary quantity. Are people getting enough food? We also look at livelihood change. Are people’s asset levels, or livelihood
strategies, changing due to a food insecurity issue? Are people turning to any atypical coping
strategies in response to current or expected food gaps? These two outcomes help us to classify household
food security. We also look at nutritional status, and mortality,
when they are driven by food insecurity. These give us an estimation of food security
for an area, since data on nutrition and mortality are usually collected at the population-level
in a given area. Now that we’ve reviewed a few fundamentals
of FEWS NET analysis, let’s look at scenario development in detail. FEWS NET uses an eight-step process for scenario
development, documenting the evidence in a scenario summary table. Step One: Set Scenario Parameters
The first step is to choose an area of focus. Usually, that is a livelihood zone. Within the livelihood zone, analysis typically
focuses on the poorest wealth group that accounts for at least 20% of the population. We focus on the poor because this group almost
always experiences the most severe food security outcomes. We apply the 20% threshold to ensure that
our analysis and mapping reflect food insecurity that is affecting a sizeable part of the population. This approach is compatible with the IPC and
is explained in more detail within the IPC manual. Analysts might also choose to focus on displaced
households, or different wealth groups, if these groups are at significant risk of food
insecurity. In this step we also set the duration of the
scenario. Normally, the FEWS NET scenario period is
for eight months. We also choose the scenario type. FEWS NET typically develops “most likely
scenarios,” as this type of scenario tends to be the most useful for decision-makers. Step Two: Describe and classify current food
security In Step 2, we examine current food security
conditions, their impact, and what this means for people’s food consumption, livelihoods,
nutrition, and mortality. With this data, we use a convergence of evidence
to classify the current phase of food insecurity. This means we consider the range of available
data and information and exercise expert judgment to arrive at the best possible conclusion. That’s important because classification
can never be based on a single piece of information – all indicators have limitations. With a convergence of evidence, the analyst
classifies the severity of acute food insecurity using the IPC reference table. Step Three: Develop Key Assumptions
Assumptions are the foundation of the scenario. They are forward-looking, and based on all
available evidence about critical factors such as: rainfall, prices, nutrition, and
humanitarian assistance. The key assumptions we make here drive the
rest of the scenario, since we’ll be making other projections in later steps based on
these assumptions. Steps 4 and 5: Describe impacts on household
income and food sources In this step, we use livelihoods information
to identify the typical sources of food and income that households rely on during the
eight months of the scenario period. Then, based on the assumptions made in Step
3, we think about how these future events might impact these sources of food and income,
and how people might respond. Step 6: Describe and classify projected household
food security After considering the impact on household
income and food sources, we describe food security outcomes. Will households be able to access an adequate
quantity of food? Will people change their livelihoods in response
to current or expected food gaps? We use livelihood information to understand
how outcomes will change with seasonal patterns, and the variations in factors such as food
availability, market access, and rainfall over the eight months. Next, we use the IPC to classify the projected
level of food insecurity for households during the next four months of the scenario period,
and for the second four months. There are two time periods because outcomes
may change in important ways due to seasonal factors or changing conditions. Step 7: Describe and classify projected area
food security In Step 7, we make assumptions about the evolution
of malnutrition and mortality. We consider typical levels of Global Acute
Malnutrition, or GAM, and mortality and examine the underlying causes. We also think about potential changes in GAM
and mortality throughout the scenario period due to factors such as conflict, disease outbreaks,
or food assistance. Then we make the IPC classification for the
area we identified in Step 1. To do this, we consider the classification
we made for the household group in the last step, along with our projections of malnutrition
and mortality. Step 8: Identify events that could change
the scenario In the final step, we consider events that
are possible and that would change the scenario significantly. This helps to communicate the uncertainty
inherent in scenario building. Scenarios are outlined in FEWS NET’s Outlook
reports, produced three times a year, and revised in Outlook Updates. To learn more about scenario development,
download the Scenario Development guidance document here.

1 comment on “How FEWS NET Makes Projections: An Introduction to Scenario Development

Leave a Reply

Your email address will not be published. Required fields are marked *