According to the final results of the Household Budget Survey 2022/2023, reflecting refinements to the provisional data released in December 2023, the mean annual household expenditure in 2022/2023 was € 23,900, about 2/3 of which were concentrated in housing (39.3%), food (12.9%) and transport (12.1%).
Compared to the rest of the territory, the results show a smaller share of expenditure allocated to culture, recreation, sports and leisure and education services in predominantly rural areas. At the NUTS 2 level regions, the mean annual expenditure was highest in the Área Metropolitana de Lisboa (€ 26,891), but the Algarve also exceeded the national average. On the contrary, the minimum regional mean expenditure was observed in the Região Autónoma dos Açores (€ 19,431), which also has the regional expenditure profile that most differs from the national average.
The results suggest that, considering the household type, households with dependent children spend annually on average € 9,731 more than households without dependent children, which leads to a mean monthly expenditure € 811 higher. This difference extends to all COICOP divisions (expenditure classes according to the Classification of Individual Consumption by Purpose). Households in the top income quintile (20% of households with the highest incomes) spent more than twice as much as households in the bottom income quintile (20% of households with lower incomes).
The proportion of health expenditure above 10% of monetary income was mainly observed in households without dependent children (14.1%) (in households with dependent children, the proportion was 7.7%), in households with elderly people, especially elderly people living alone (20.4%), and in households with two or more adults in which at least one is elderly (16.9%).
This edition of the survey accommodates methodological improvements, compared to 2015/2016 edition, described in the methodological note included in this document. They include changes in: 1) COICOP (COICOP-2018 was implemented for the first time); 2) the annualisation factors (in that the questionnaire now collects data about how often expenditures are incurred by each household, allowing more accurate information to be obtained); 3) sample sizing (the survey sample included a new methodology for reinforcing dwelling units in order to mitigate, albeit partially, the problems resulting from the asymmetry in income distribution and the higher incidence of non-response among households with greater resources); 4) weighting (the calibration includes monetary variables in the adjustment model for the first time).
Even though these improvements indicate a break in the series, based on a simplified exercise to reconcile the time series, the data suggest that the relative importance of housing in the structure of household expenditure has increased in recent decades and that, on the other hand, the amount spent on food, clothing and footwear lost weight in the households’ mean expenditure.
See the Publication