Household Expenditure

The indicator measures the average amount of money the target households spent in a given period on goods and services. It is commonly used as a proxy indicator for measuring income.

There are two main options for collecting household expenditure data:

A) THE RECALL APPROACH

During the recall approach, respondents are asked to report how much they spent on different categories of consumption goods and services in a certain period. It involves the following steps:

1) Define the recall period – i.e. the time period for which respondents are asked to report their household's consumption (it is recommended to use a maximum of 7 days recall period).

2) Define the reference period – i.e. the period your survey aims to assess (for example, expenditures during the past 3 weeks - in such a case, you will have to interview the respondent 3 times, every 8th, 15th and 22nd day – always using a 7-day recall period).

3) Define different expenditure categories – it is recommended to use four main types of categories: i) food items (any types of purchased food and drinks); ii) non-food items and services (any types of frequently purchased items/ services, including health and education-related expenses); iii) consumer durables (usually more expensive items that are used over an extended period of time, such as mobile phone or cooking stove); iv) housing (rent, utilities, repairs).

4) Prepare a questionnaire enquiring about households' expenditures - see an example of one part of the questionnaire:

i) Food items

Q1: In the past 7 days, how much money did your household spend on purchasing food (not in the restaurants)?

Q2: In the past 7 days, how much money did your household spend on purchasing any types of drinks (not in the restaurants)?

Q3: In the past 7 days, how much money did your household spend on purchasing food and drinks in the food stalls/ restaurants/ bars?

5) Interview the respondents.

6) Calculate the indicator's value by summing up the expenditures of all surveyed households and dividing them by the total number of households.

B) THE DIARY APPROACH

In the diary approach, respondents are asked to keep a diary over a certain period of time in which they record every single expense (= the approach is suitable for literate respondents only). This approach is often viewed as more precise as it involves a lower risk of people forgetting the different expenses they made during the recall period; on the other hand, people often forget or are not motivated to record their expenses. It involves the following steps:

1) Define the reference period – the longer the period you use, the more likely it is that people will regularly stop (or completely) recording their expenses; a maximum of 2 weeks reference period is recommended.

2) Prepare a simple, easy-to-use recording form whose rows list all the different expenses and the columns list specific dates; do not use the form before you pre-test it and you are sure that the respondents can use it well.

3) Explain to the respondents how to use the form, fill out the previous (= first) day with them and request them to use it for the duration of the reference period (consider proposing an incentive where those respondents who regularly (daily) fill the form will, at the end of the survey, receive a payment; respondents' compliance must be controlled through spot-check visits).

4) Calculate the indicator's value by summing up the expenditures of all surveyed households and dividing them by the total number of households.

Disaggregate the data by wealth.

1) It is important to make the distinction between consumption, which refers to the goods and services people actually use, and expenditures – the goods and services they buy. Many households make purchases in bulk (e.g. buying a sack of rice), resulting in a situation where in one day they have extremely high spending and in the following weeks they have no or very limited spending for the given item. If you want to understand the value of the goods/ services the households actually used in the given reference period, you have to use the Household Consumption indicator.

2) In order to interpret the data, you must understand the context to know whether the (for example) increasing consumption represents a worsening or improvement in the given situation. For example, while increasing food expenditures might mean an improvement of the socio-economic situation, it might also be a sign of poor harvest, forcing families to increase their spending on food (financed by using negative coping strategies).

3) Expenditures are often prone to significant changes in time (determined by the income availability, festivals and other factors). To ensure that the baseline and endline data is comparable, do your best to collect it in the same period of a year (unless you work in an emergency context where it is not possible to wait with an endline survey until the time of a year when the baseline survey was conducted). Avoid collecting data during the times of unusual expenditures (e.g. festivals).

4) Be aware that if you interview only one household member, s/he might not have a complete overview of the other members' expenditures.

5) Do not use long recall periods – although it might seem that they capture more expenditures and therefore provide better data, in reality, they are very imprecise as people simply do not remember all the expenses they made and usually provide incorrect data.

6) Households' expenditures are not necessarily a good indicator of their well-being. A rural household that spends an average of 100 USD per month, does not pay any rent and produces most of its food might be much better off than an urban household spending 250 USD per month. Compare data from households with similar characteristics only.