Sweden Spends More on Research Than Romania Mainly Because Its Companies Do

A target set in 2002 and still not met

The European Union has committed to spending 3% of GDP on research and development for more than two decades. In 2024 it spent 2.3%.

The shortfall is a familiar headline. What the same dataset shows, and what the headline omits, is where the shortfall comes from — because research spending is not one thing, and the Member States that reach the target do not reach it the same way as those that miss it fall short.

Who reaches 3%

Six Member States are at or above the target. Sweden leads at 3.5%, followed by Austria at 3.3%, Belgium at 3.3%, Finland at 3.2%, Denmark at 3.1% and Germany at 3.1%.

Below them the drop is steady rather than sharp: the Netherlands at 2.4%, France at 2.2%, Slovenia at 2.1%, Estonia at 2.0%, Czechia at 1.8%.

At the bottom, Romania spends 0.5% of GDP on research, Malta and Cyprus 0.6% each and Bulgaria 0.8%. Sweden spends close to eight times the share Romania does.

Where the gap actually sits

Research is performed by four sectors: business, government, higher education and private non-profit. Separating business research from the total changes what the ranking is about.

Business enterprise research accounts for 1.5 percentage points of the EU's 2.3%, or roughly two thirds of the total.

Between Sweden at the top and Romania at the bottom, the total gap is 3.1 percentage points. The business research gap between the same two countries is 2.3 points — Sweden 2.6%, Romania 0.3%. Three quarters of the distance between the Union's strongest and weakest research economies is private sector research.

The same pattern holds across the leading group. Every Member State above the 3% target has business research of 1.9% of GDP or more. No Member State reaches 3% on public and academic research alone, and none comes close to trying.

What that implies about the target

The 3% target is usually discussed as a public spending commitment, and public budgets are the instrument governments control directly. The composition of the gap suggests the target is not principally a public spending question.

Take each of the twenty-one Member States below the target and double the research performed outside business — government, universities and non-profits together. One of them reaches 3%: the Netherlands, which goes from 2.4% to 3.1%. The other twenty do not. The largest non-business research effort among the twenty-one is Estonia's, at 0.8 points of GDP, so the doubling is worth less than a point wherever it is applied, and every one of the twenty-one except the Netherlands starts further below the target than that. Reaching the target requires firms that do research, which is a matter of industrial structure — what a country's companies make, and whether the sectors they operate in are research-intensive — rather than of appropriations.

This study establishes the composition. It does not establish how a country acquires research-intensive firms, which is a question the data cannot answer.