“A lower than expected take-up [say less than EUR500 billion] and a failure to raise the firewall would surely lead to renewed pressure on sovereign spreads,” SocGen says.

A UBS survey yielded a median estimate of 629 billion euros but their analysts forecast slightly less than 500B.

I have to check on this but there is also talk about a “dramatic” underperformance of banks (not sure if this is bonds or stocks) who participated compared to those who sat out. This points to a stigma about using the LTRO and could lead to low participation.