According to an article published by Kontroll on 7 November, Hungary spent 16.7 per cent of its gross domestic product on social benefits and pensions in 2025. The European Union average was 27.5 per cent. Hungary recorded the third-lowest share among EU member states.
Only two countries ranked lower
Ireland had the lowest level of spending relative to GDP, at 11.2 per cent. Malta followed with 13.8 per cent, with Hungary next in the ranking.
Kontroll linked the low figures in Ireland and Malta to differences between national pension systems. Voluntary and occupational schemes play a larger role in those two countries, reducing the weight of state pension expenditure. In Hungary, by contrast, almost all pension provision comes from the state pillar.
Austria was at the other end of the table, spending 32.4 per cent of GDP on social protection and pensions. Finland recorded 32.2 per cent and France 32.1 per cent.
Spending rose, but its share remained low
The year-on-year figures present a different picture. Spending on social protection, measured in euros, increased in every EU member state in 2025. Hungary recorded growth of 6.6 per cent compared with 2024, placing it in the middle of the field, according to Kontroll.
The figures measure different things: 6.6 per cent describes the annual rise in euro-denominated expenditure, while 16.7 per cent shows how large that spending was in relation to Hungary’s GDP.
Across the European Union, social protection expenditure reached €5,183 billion in 2025. This was 4.9 per cent higher than in the previous year.
Pensions accounted for the largest share
Pensions were the biggest item in the EU total. They accounted for €2,152 billion, or 41.5 per cent of all social protection expenditure. Healthcare benefits amounted to €1,556 billion, representing 30 per cent.
The remaining funds went to family benefits, disability pensions, unemployment and housing support, and other forms of social assistance. Kontroll presented the figures as findings from an Eurostat survey.
Source: Kontroll
