3.1.2. Household Size and Population Density

The countries with lower average household sizes—Belgium, Germany, the United Kingdom and Finland—are also some of the most densely populated (Figure 4). At the same time, countries with larger average household sizes are more sparsely populated—e.g., Slovakia, Croatia and Poland.

Notable exceptions are Malta (with high average household size and a predominantly urban sample (92%) and Denmark (with low average household size and a largely rural sample, with as much as 43% of the sample living in sparsely populated areas). Denmark has a long tradition of a social-democratic welfare regime [46] with more liberal attitudes to family relationships and lower levels of religiosity, which may explain the relatively lower household sizes at lower population density. Compared to Western Europe, there is higher religious participation in Malta, attaching great importance to teachings regarding family life, the morality of abortion, divorce and other matters [47], which may explain the relatively large average household size. In addition, there may also be geographical reasons for the relatively high population density, with Malta being a small island.

Similar to other studies [23], we find that population density is important for per capita carbon and energy footprints (see SM2). Descriptive statistics should be interpreted with caution as they do not control for the differences in income levels and other relevant factors, which tend to vary substantially between urban and rural areas.

**Figure 3.** Mean per capita carbon (**a**) and energy (**b**) footprints by EU country by household size. Households with household sizes >5 have been aggregated in the same group. The carbon footprints are measured in tCO2eq/cap and energy footprints in TJ/cap. Household weights provided by the HBS have been applied.

**Figure 4.** Association between average household size and share of households living in densely (**a**) and sparsely (**b**) populated areas in the EU. In cases where the shares do not add up to one, the difference amounts to the share of households living in intermediately populated areas. Household weights provided by the HBS have been applied.
