impact analysis

our initial impact analysis highlights that vulnerability varies considerably by wealth group

Very Poor & Poor households, who represent 60% of the population, are vulnerable to drought. Wealthier households have enough assets, primarily livestock, to provide an effective coping mechanism that makes them resilient to moderate drought

The key difference between Very Poor & Poor household is due primarily to income from cattle sales rather than crop production — over time the lower risk profile of sorghum could enable poorer household to re-build this crucial asset base if they can avoid selling their livestock to cope with drought

to ensure impact it is essential to proactively target poorer households — otherwise the benefits will accrue primarily to wealthier households who supply 80% of the grain sold into local markets

it is also essential to provide enough support to ensure that poorer households can cultivate at least 2 ha of sorghum — anything less and they will still remain vulnerable to drought

A typical smallholder farmer in Southern Zambia produces 3,500kg of maize a year and owns 8 cattle & 8 goats. Livestock sales are their largest source of income & Maize provides most of their food - but Maize also provides their second largest source of cash

But not all farmers are the same and these averages mask significant differences. Poorer farmers, who represent about 60% of the population, typically produce 1,500kg of maize and own 3 cows.

The difference between Very Poor & Poor farmers comes down to the number of cattle they own, rather than how much maize they grow. Two cows makes all the difference; Very Poor farmers own two cows while Poor farmers own four. This means that Poor farmers can afford to sell one cow every year, but Very Poor farmers cannot.

Wealthier farmers buy significantly more fertilizer than other farmers and as a result have a much larger maize harvest. These differences in production mean that wealthier farmers, who represent 40% of the population, produce 80% of the surplus maize that is sold into local markets.

Moderate drought halves farmers maize production, reducing both their food & cash income. Farmers sell more cattle as a coping strategy. But poorer households, who own few livestock, are unable meet their basic needs during a moderate drought - and have to rely on humanitarian assistance

In the six year between 2014 and 2019 Zambia experienced four droughts: one year of severe drought, with over 2 million people facing food insecurity, and three years of moderate drought, with almost 1 million people food insecure.

In order to model the impact of moderate drought we compare crop production & food price data for the two moderate drought years (2015/16 & 2018/19) with the HEA baseline reference year (2014/15). We found that maize yield reduced 40% during the drought years and food prices increased 25%. We also assumed that livestock prices would decrease 30% as a result of deterioration in livestock body condition & increased livestock sales and that self-employment would be reduced due to reduced demand. When we enter these parameters into the HEA model the results show that:

  • Poor households were projected to have with a small livelihood protection deficit, equivalent to almost 150kg of maize or 6% of their minimum food, livelihood, health & education expenditure requirements. This deficit that would be equivalent to IPC Phase 2 which would not typically require a humanitarian response but a slight deterioration in the situation would quickly push Poor households into IPC Phase 3.

  • Very Poor households were projected to have a large livelihood protection deficit, equivalent to almost 500kg of maize or 20% of their minimum food, livelihood, health & education expenditure requirement. A large deficit such as this would be classified as a humanitarian crisis (IPC Phase 3) and would require an emergency response.

Farmers that grow sorghum secure higher yields and larger profits when there’s drought. Crucially, they produce enough to enable even poorer farmers to meet their immediate needs - and do not have to rely on humanitarian assistance.

Sorghum is naturally adapted to the irregular rainfall and poor soils that characterise Southern Zambia. During a moderate drought, sorghum yields only reduce by a quarter compared to halving for maize. Not only can Sorghum produce a good yield when rains are poor, but it can also produce a good yield without artificial fertilizers. Reduced costs mean larger profits. This is especially important for poorer farmers, who cannot afford to fertilizers. Lower inputs also translate into lower risks.

To model sorghum production we remove first removed maize and then replaced it with sorghum. We set sorghum’s yield reduction during drought to 20% compared to a 40% reduction for maize - and reduced expenditure on fertilizer (but maintained expenditure on seeds & pesticides). The results showed that if they switched to growing sorghum, during a moderate drought:

  • Poor household no longer face a deficit. After meeting their minimum food, livelihood, health & education expenditures they have a surplus income equivalent to 14% of their minimum expenditure requirement.

  • Very Poor households have a small livelihood protection deficit, equivalent to 150kg of maize or 6% of their minimum food, livelihood, health & education expenditure requirements. This deficit is significantly smaller than when they were cultivating maize and it is small enough that a humanitarian response would not be required i.e. the HEA indicator would be in IPC Phase 2

We use Household Economy Analysis and its inherent modelling capacity, to understand hunger & vulnerability in Southern Zambia and quantify the impact of our work.

Household Economy Analysis is a well established method, widely used in national early warning systems by Governments across Africa as well as FEWS Net & the UN. HEA has two key components: the baseline & outcome analysis. The baseline provides a static ‘snapshot’ for a specific year - while the outcome analysis enables that baseline to be quickly updated in future years based on changes to key “key parameters”. Because the outcome analysis provides an annual analysis, by running the analysis early in the year it is possible to project the impact on household food security later in the year - which is why HEA is such a powerful tool for early warning.

HEA outcome analysis has traditionally been used to model the negative impact of shocks & hazards on household food, income and expenditure - but it can also be used to model positives changes associated with project interventions. By combining a hazard analysis & a positive project intervention it is possible to generate an analysis of household economic resilience. This HEA resilience analysis process is explained in the video at the bottom of this page.

Household Economy Analysis baselines quantify food, income & expenditure for households from different wealth groups living within a specific livelihood zones. There are three steps to preparing an HEA baseline:

  1. The livelihood zone map: A livelihood zone is an area within which people share broadly the same means of production and similar access to markets. We used WFP’s 2014/15 HEA baseline for the Commercial Rail Line Maize, Livestock and Cotton (ZM08) livelihood zone which covers much of Southern Province, including parts of Kalomo. Note that the plateau area of Kalomo is covered by a separate livelihood zone (ZM09).

  2. The wealth breakdown: This is a division of the livelihood zone population into 4 locally defined wealth groups (very poor, poor, middle and better-off), based primarily upon the ownership of productive assets (e.g. land & livestock).

  3. Food, Income & Expenditure analysis: This is a detailed analysis of sources and amounts of food, income and expenditure, for a defined or reference year. Knowing where households obtain their food and income, and what they need to spend money on, plus a quantification of these, provides the starting point for understanding how they will be affected by a shock.

The baseline analysis relates to a specific reference year (e.g. 2014-15). For agricultural livelihood zones the reference year usually starts with one harvest and ends 12 months later. For example, if crops are harvested in April, then the reference year might run from April 2014 - March 2015. The reference year will generally be a year that was neither especially good nor especially bad, but somewhere in the middle. The most important point about the reference year is not that it should be an average year, but that it should provide a good starting point for understanding how livelihoods will vary from one year to the next in relation to changes in factors such as crop production and market prices. Once a baseline has been prepared, it can be used repeatedly over a number of years (generally between 5 and 10), until significant changes in the underlying economy render it invalid.

HEA outcome analysis builds on this baseline data and models the economic impact of a specific hazard, such as drought or food price shocks, to determine whether household are able to meet their minimum expenditure requirements for food, health, education & livelihoods. Outcome analysis consists of three steps:

  1. Problem Specification: the translation of a shock or hazard such as drought into economic consequences at household level (such as a percentage fall in crop production or increase in food prices compared with the baseline)

  2. Coping Strategy Analysis: the assessment of the capacity of households in different wealth groups to cope themselves with the hazard

  3. Projected Outcome: access to food and income at household level is predicted for a defined future period and compared to two critical thresholds – the survival and livelihood protection thresholds - to determine whether there is a gap or deficit.

Baseline + Hazard + Coping = Outcome

The survival threshold provides a measure of a household’s ability to cover the bare minimum requirements for survival – to obtain and prepare basic food and, if necessary, purchase water. The livelihoods protection threshold provides a broader measure of a household’s ability to sustain local patterns of livelihood, including covering the costs of productive inputs (seeds, livestock drugs, etc.) and basic expenditure on health and education.

A key feature of outcome analysis is that it is not an analysis of behaviour. Rather, it provides an estimate of what the deficit might be given certain conditions. This is especially important in relation to coping, and which coping strategies are included in the analysis. The most damaging negative strategies are always excluded from the analysis (e.g. sale of all livestock, mortgaging or sale of land). Including such strategies would have the effect of reducing the calculated deficit, effectively delaying any intervention until after that strategy has been fully utilised. Since we want to intervene before that stage is reached, we need to know what the deficit will be if these strategies are not used, i.e. if they are excluded from the analysis.