Moldova: the highest relative budgetary effort among the countries compared

According to the World Bank (indicator SE.XPD.TOTL.GD.ZS), public expenditure on education in the Republic of Moldova amounted to 6.557% of GDP in 2023. This is the second-highest value in the 2014–2023 series, after the peak of 6.601% in 2020.

For a fair comparison, the common reference year is 2022, the latest year for which data are available for the EU, Romania, Poland and Bulgaria. In that year, Moldova allocated 6.104% of GDP to education. This was 1.4 percentage points (pp) above the EU average (4.657%), 1.6 pp above Bulgaria (4.501%) and 1.8 pp above Poland (4.314%). The gap was 2.3 pp relative to Georgia (3.755%) and 2.8 pp relative to Romania (3.279%).

Romania ranks last in the comparison group, below the world average (3.861% in 2022). The ratio of Moldova's share to Romania's was 1.86 in 2022. When the latest available year for each country is compared (Moldova 2023, Romania 2022), the ratio reaches 2.0.

Ukraine is the only country in the region with a share close to Moldova's: 5.14% of GDP in 2021, the latest year for which the World Bank publishes data. Ukrainian data for the period after the 2022 invasion are unavailable, so the comparison with Ukraine cannot be updated.

Public expenditure on education, 2022 (Ukraine: 2021)

Source: World Bank, World Development Indicators (SE.XPD.TOTL.GD.ZS)Unit: % of GDP
View data table
% of GDP
Moldova6.1
Ukraine (2021)5.1
European Union4.7
Bulgaria4.5
Poland4.3
World3.9
Georgia3.8
Romania3.3

Trends 2014–2023: Moldova fluctuates, the EU and Poland decline

The trajectories differ markedly. In the EU, the share of education in GDP fell from 4.882% in 2014 to 4.657% in 2022 (−0.2 pp). In Poland it declined from 4.939% to 4.314% (−0.6 pp). Both recorded a temporary peak in the pandemic year of 2020 (4.931% and 4.801% respectively). Part of this jump may reflect the contraction in GDP (a denominator effect), but the available data do not allow this effect to be separated from a genuine increase in allocations.

Romania remained within a narrow band, between 3.026% (2016) and 3.58% (2020). It ended 2022 at 3.279%, only 0.2 pp above its 2014 level. Bulgaria rose from 4.194% (2014) to 4.501% (2022), with a peak of 4.709% in 2021.

Moldova followed a fluctuating path. From 6.338% in 2014, the share fell to 5.539% in 2018. It then rose to 6.226% in 2019 and 6.601% in 2020. After dropping to 5.824% in 2021, it recovered to 6.104% in 2022 and 6.557% in 2023. The cumulative increase between 2018 and 2023 is 1.0 pp, yet the 2023 level is only 0.2 pp higher than in 2014.

Georgia recorded the largest increase over the whole period among the countries compared: from 2.941% in 2014 to 3.98% in 2024 (+1.0 pp). Even so, it remained 2.6 pp below Moldova in each country's latest available year.

Public expenditure on education, 2014–2023

Source: World Bank, World Development Indicators (SE.XPD.TOTL.GD.ZS)Unit: % of GDP
View data table
MoldovaEuropean UnionPolandBulgariaRomania
20146.34.94.94.23.1
20155.84.84.83.93.1
20165.74.84.63.83.0
20175.74.64.54.13.1
20185.54.64.64.13.3
20196.24.64.64.23.6
20206.64.94.84.53.6
20215.84.84.64.73.3
20226.14.74.34.53.3
20236.6––––

Need an analysis on this topic tailored to your programme or organisation?

Request a proposal

A high percentage of a small base: spending per capita

The share of GDP measures relative effort, not actual resources. Socium estimated public expenditure on education per capita by combining this share with GDP per capita at purchasing power parity (World Bank). In 2022, Moldova (GDP per capita of PPP USD 16,442 × 6.104%) spent approximately PPP USD 1,004 per inhabitant.

The same estimate for 2022 yields around PPP USD 2,728 in the EU, 2,018 in Poland, 1,635 in Bulgaria and 1,377 in Romania. Moldova thus stands at approximately 37% of the EU level, 50% of Poland's and 73% of Romania's, despite allocating, as a share of GDP, 1.3, 1.4 and 1.9 times more than these respectively.

Among the countries compared, Moldova exceeds only Ukraine (around PPP USD 917 in 2021) and Georgia (around PPP USD 843 in 2022) on a per capita basis. In 2023, GDP per capita rose to PPP USD 17,751 and the share to 6.557%. The estimate for Moldova therefore rises to around PPP USD 1,164, or approximately 85% of Romania's 2022 level.

This estimate is per inhabitant, not per pupil. The ratio between spending per pupil and spending per inhabitant depends on the share of the school-age population, which is not included in the dataset. The available data therefore do not allow a direct per-pupil calculation.

Public expenditure on education per capita, 2022 estimate (Ukraine: 2021)

Source: Socium calculations based on World Bank data (SE.XPD.TOTL.GD.ZS × NY.GDP.PCAP.PP.CD)Unit: PPP USD per capita
View data table
PPP USD per capita
European Union2,727.8
Poland2,018
Bulgaria1,634.8
Romania1,376.5
Moldova1,003.6
Ukraine (2021)917.3
Georgia843.3

Budgetary effort and outcomes: not an automatic relationship

A high relative effort does not, in itself, guarantee proportionate resources or outcomes. The dataset used does not include indicators of educational outcomes. Any assessment of spending efficiency would require additional data on pupil performance and on how funds are allocated.

Part of the high share may be a denominator effect. With GDP per capita at 30.5% of the EU average in 2025, the same costs of running the school network represent a larger fraction of GDP. At the same time, Moldova's population is declining rapidly (−2.843% in 2023, −1.754% in 2025, according to the World Bank). The same network may thus be serving ever smaller cohorts.

The relationship between the share of GDP and educational performance cannot be interpreted causally on the basis of these data. The available dataset lacks the structure of expenditure (salaries, infrastructure, maintenance), spending per pupil and coverage indicators. Without these, efficiency cannot be assessed.

Implications for Moldova

The scope for further increasing the share of education in GDP appears limited. At 6.557% in 2023, Moldova already exceeds the EU average (4.657% in 2022) by 1.9 pp and the world average (3.569% in 2023) by 3.0 pp. Improving real funding depends largely on growth in GDP per capita, which remains at 39% of Romania's level (2025).

Moldova's per capita spending in 2022 was around 27% below Romania's, even though its relative effort was almost twice as high. This suggests that public debate would gain in precision if it shifted from 'what percentage of GDP' to 'how much per pupil' and 'for what'. Systematic publication of spending per pupil, by level of education and by urban/rural area, would enable more meaningful comparisons.

Demographic decline (−1.75% in 2025) and an already high share of education in GDP make the sizing of the school network a key variable. Maintaining fixed costs for shrinking cohorts raises spending per pupil without necessarily improving learning conditions. The available data do not, however, indicate which of the options (consolidation, school transport, distance learning) is the most efficient.

The comparison with Georgia and Ukraine shows that Moldova stands out in the region for its combination of the highest relative effort in the group and a small economic base. Annual monitoring of the per capita indicator alongside the share of GDP would give policymakers a more complete picture of convergence with the EU.

Methodology

The main indicator is 'Government expenditure on education, total (% of GDP)' (SE.XPD.TOTL.GD.ZS) from the World Bank's World Development Indicators. The latest available year differs between countries: Moldova 2023, Georgia 2024, Ukraine 2021, and the EU, Romania, Poland and Bulgaria 2022. Direct comparisons use the common year 2022, and Ukraine's 2021 value is explicitly flagged. Where different years are compared (for example, Moldova 2023 with Romania 2022), the years are stated in the text. Per capita spending is a Socium estimate obtained by multiplying the share of GDP by GDP per capita at purchasing power parity (NY.GDP.PCAP.PP.CD, current international dollars) for the same year. Results are rounded to one decimal place and refer to the total population, not to pupils. They are approximate, since the economy-wide PPP conversion factor may differ from the price level in the education sector. Differences are expressed in percentage points (pp) for shares and in per cent for ratios. The 'European Union' aggregate is that calculated by the World Bank. The data do not allow an assessment of spending per pupil, the structure of expenditure, educational outcomes or allocative efficiency. Relationships between funding and other variables are presented as associations, not causal relationships.

Sources