How many "Blue Cards" have EU countries issued?

Question

The "Blue Card" is EU-level scheme

allowing high-skilled non-EU citizens to work and live in any country within the European Union, excluding Denmark, Ireland and the United Kingdom,

According to a 2016 EU press release it didn't work too well, so it was reformed then:

The EU Blue Card scheme, adopted in 2009, has proven insufficient and unattractive so far and is therefore underused. Restrictive admission conditions and the existence of parallel rules, conditions and procedures at national level have limited the use of the EU scheme. Only 31% of highly-educated migrants to OECD countries chose the EU as a destination, meaning skilled workers are choosing other destinations which compete economically with the EU.

Today's proposal revamps the existing rules and aims to improve the EU’s ability to attract and retain highly skilled third-country nationals [...]

But neither source gives any actual numbers for how many of these Blue Cards have been issued (either before or after the reform). So are there any yearly stats of how many such cards have been granted? (It would be even better if the stats were also broken by country.)

Answer 1

I managed to find some stats in a paper but only pre-2016 (pre-reform)

After 2013, when an approximately 4-fold increase happened, over 85% of Blue Cards were issued by Germany.

There's also some extreme weirdness in the data: why did the numbers collapse in Spain? Why they're practically non-existent in Sweden? The paper doesn't have such specific answers... but overall it says:

In 2015 in Belgium, 50,085 national permits were issued compared to 19 Blue Cards; in France, 226,630 compared to 659 Blue Cards; in Sweden, 110,623 compared to two Blue Cards. Thus, in all countries, the amount of national work permits issued is much larger than the amount of Blue Cards granted. [...]
While the EU Commission has set the general rule that the salary threshold should be around 1.5 times the average gross annual salary, member states were free to determine the exact amount.

So basically, countries can "opt out" by setting some ridiculous threshold for the salary. Note that this was before the 2016 reform; the paper is alas unclear as whether countries can still do this. The paper exemplifies with Romania that set a multiplier of 4 (they average salary in Romania is much lower than in Western EU countries), but it still granted some cards... There are also harder to quantify local labor market tests (in 12 of the countries, but the paper doesn't list which precisely) that give precedence to local/nationals over foreign applicants.

Presumably Eurostat has a more complete series, which I have yet to find.