One of the suggested thick disc formation mechanisms is that they were born fast and in-situ from a turbulent clumpy disc. Subsequently, thin discs formed slowly within them from left-overs of the turbulent phase and from material accreted through cold flows and minor mergers. In this letter, I propose an observational test to verify this hypothesis. By combining thick disc and total stellar masses of edge-on galaxies with galaxy stellar mass functions calculated in the redshift range z≤3.0, I derive a positive correlation between the age of the youngest stars in thick discs and the stellar mass of the host galaxy; galaxies with a present-day stellar mass M⋆(z=0)<10^10M⊙ have thick disc stars as young as 4−6Gyr, whereas the youngest stars in the thick discs of Milky-Way-like galaxies are ∼10Gyr old. I test this prediction against the scarce available thick disc age estimates, all of them of galaxies with M⋆(z=0)≳10^10M⊙ and find that field spiral galaxies seem to follow the expectation. On the other hand, my derivation predicts too low ages for the thick discs in lenticular galaxies, indicating a fast early evolution for S0 galaxies. I propose to conclusively test whether thick discs formed fast and in-situ by obtaining the ages of thick discs in field galaxies with masses M⋆(z=0)∼10^9.5M⊙ and checking whether they contain ∼5Gyr-old stars.