Are health and nutrition outcomes equitably distributed across the population?
The following section that describes the performance metric Equity in food consumption and nutrition and producers was embedded in Deliverable 6.3.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
Performance metric 3: Equity in food consumption and nutrition outcomes
With respect to the outcomes of the food system the project considers here both the classical set of indicators for FNS as they are presented in the food security literature (availability, accessibility, utilization and stability) together with indicators measuring the equity among sub-population groups in nutrition and health outcomes for the population.
Equity among sub-population groups in nutrition and health outcomes
The performance metrics for balanced diets and nutrition as presented in section 4.1 provide several options for the development of equity perspectives. These would identify disproportional and un-just differernces in nutrition and health outcomes between population sub-groups, e.g. differentiated by socioeconomic class, gender, geographical location, access to / quality of health and education services, etc. The possible range of metrics that can be quantified is presented by the limitations of the individual food intake data. In the case of SUSFANS, the data collected under the nutrition surveillance allows for stratification only by age, gender, and BMI-level. Mertens et al. 2018 reports on these differences. While education levels can be distinguished, statistical representativeness is hampered by the unequal sizes of education groups.
Based on projections of this data, through links with the MAGNET model, these data can also inform a quantification of performance metrics for plausible future intake patterns. However, sub-group detail is lost in this exercise, except for 2 gender groups and 2 age groups.
FNS indicators
For the FNS indicators the SUSFANS project bases its selection of variables on the work of the EU project FOODSECURE. A summary of the available FNS Aggregate Indicators and their associated variables is provided in Pangaribowo et al. (2016). The variables have been developed based on the FAO suite of FNS indicators (FAO, 2016) as part of the FOODSECURE project.18 The metrics presented in this section have been designed to match the data and model indicators in two models in the SUSFANS toolbox, MAGNET and GLOBIOM. While food and nutrition security outcomes are interesting outcome indicators under an equity ans global justice perspective, two caveats are important. First, in theory, the metrics should record the contribution of the EU to global FNS, rather than absolute FNS levels. This raises an immediate question on the target value that can be used to assess – from a policy perspective – the performance of the EU food system in this regard. The lack of clear policy visions and targets in this area hampers the assessment of performance in this regard.
Second, the equity and social justice perspective in the FNS indicators need tob e assessed in the perspective of unequalities in the FNS outcomes and underlying drivers. Several FNS indictors are assessed at individual and household level with the exact purpose for identifying these inequalities. Intra-household inequalities, putting women and especially young women at disadvantage, are often recorded. In micro-modellling approaches, these distributional perspectives can be taken on board (as in the section above). However, in the modeling at aggregate macro-levels with the SUSFANS toolbox, distributional perspectives are restricted: it has been attempted to include household variation in the database underlying the toolbox, but this has stumbled upon the lack of consistent data in readily available cross-European databases (see Latka et al. 2019 for a discussion). Hence, the application of FNS indicators with ditributions for sub-groups in macro-modeling of EU FNS requires substantial future work.
Availability
Derived Variable: Calorie availability
This Derived Variable gives the total amount of net (kilo)calories available per capita per day for the average consumer globally, within the EU and non-EU regions and at the member states level. The indicator will also be available for different household types in the case study countries, giving insight into variations within these countries. The indicator is defined as the total calories available for domestic human consumption and therefore excludes calories in food exports and those in animal feed and biofuel feedstocks. The unit of this indicator is kilocalories per capita per day.
Derived Variable: Share of nutritious foods
The share of nutritious foods is defined as 1 minus the share of energy (kilocalories) derived from cereals (MAGNET19) or cereals, roots and tubers (GLOBIOM), where the calories from cereals or cereals, roots and tubers is the percentage share of total calories available for consumption. The unit for this indicator is percent.
Derived Variable: Reduction in share of protein of animal origin
This indicator focuses on the share of protein from animal origin (including or excluding fish) in total protein. The indicator is constructed so that an increase represents a reduction in protein from animal sources, reflecting the average over-consumption of protein in the European diet. Caution should be taken when interpreting this variable in relation to low-income countries where (animal) protein consumption may be below optimal levels.
Derived Variable: Domestic food production
The average value of primary food production is an indicator of the strength of the domestic food producing sector. It is defined over primary food rather than including food processing as it relates to the food availability dimension of food and nutrition security. The variable is defined per capita in constant (2010) prices.
Accessibility
Derived Variable: Share of non-food expenditure in total expenditures
The share of non-food expenditure in total expenditures is 1 minus the share of food spending in total spending for the average household. Higher shares of non-food expenditure in total expenditure are associated with higher food security as changes in food prices and incomes have less of a direct impact on food consumption i.e. they are insulated by having non-food income that can be diverted to food if necessary.
This variable is also available by household type in the case study regions. The unit for this indicator is percent and excludes savings.
Derived Variable: National income per capita
National income per capita gives an indication of the wealth of a nation. It is measured as GDP per capita in constant US dollars.
Derived Variable: Food affordability
Food affordability is indicated by the real domestic primary food price index. The index can also be expanded to include processed food and weighted by consumption to reflect how price changes affect the average consumer and household types with different consumption baskets. In the latter case, the variable becomes household specific as common prices affect households different due to the different weights of food products in total food consumption.
Derived Variable: Household income per capita
Household income per capita is a further measure of wealth in a region which can be household specific due to different patterns of factor incomes in different household types. At the household level, the variable also accounts for different rates of population growth.
Derived Variable: Consumption per capita
Average and household specific food consumption per capita shows how changes in prices, incomes and population combine to affect food consumption. The variable is a Paasche volume index of food consumption which includes primary and processed foods.
Utilisation
Derived Variable: Share of calories from fruit and vegetables
The standard FAO utilisation indicators centre on measures of physical development (e.g. stunting and wasting), health outcomes (e.g. anaemia) and water and sanitation. Given the limitation of economic models to inform these indicators, we introduce the share of calories from fruit and vegetables as a crude proxy for a healthy diet and micronutrient intake. The variable includes vegetables and fruits consumed directly, via processed products and via food services. The variable is expressed as a percentage of total calories consumed.
Aggregate Indicator: Stability
Derived Variable: Cereal import dependency ratio
This variable shows the degree to which a country or region is dependent on cereal imports. The variable is bounded at zero so all net exporters take this value. The variable is expressed as a percentage.
Derived Variable: Value of food imports over total merchandise exports
This variable provides a measure of vulnerability and captures the adequacy of foreign exchange reserves to pay for food imports, which has implications for national food security depending on production and trade patterns (FAO, 2016). FAO definition excludes fish which is included here. The variable is expressed as a percentage.
Derived Variable: Market pressure index
The market pressure index is an original new metric developed within SUSFANS to measure market (in)stabilty. The volatility of agricultural markets has a profound impact on food and nutrition security (Kalkuhl, von Braun and Torero, 2017). For a given agricultural commodity, the market pressure index provides the percent difference between the actual price and the predicted price based on market fundamentals, macroeconomic and financial developments, as well as the dynamics of climatic variables (Crespo Cuarisma, Hlouskova and Obersteiner, 2017). The forecasting model used to obtain the predictions is chosen after an exhaustive scrutiny of the predictive ability of a large number of state-of-the-art multivariate time series specifications and combinations thereof. The index indicates whether the prevalent climatic and economic conditions are expected to lead to an increase or a decrease of the price of a particular agricultural commodity at a given horizon (from one to twelve months ahead) and by how much the price is expected to change.