They have really been in the past centuries, but they are almost dead now and it is no longer worth investing in them, let me explain why.
First of all, this is especially evident for physics: last expensive experiment at CERN by looking for supersymmetries, strings and anything beyond the standard model has definitely shown that theoretical and experimental physics projects have reached the maximum extent and the limit of the knowledge they could bring.
But this is also true about math, in a more subtle way. Its capacity of abstraction is also at the boundary of its capabilities. Researchers are just reinventing new languages to artificially rediscover the same theorems in a more obscure, cryptic way. Examples? (Co)Homology/Homotopy, Category theory and similar variations.
The best wishes for new year is that young people forget this madness and lend themselves to medicine, engineering, non-profit organisation of social utility and honest professions.
Happy New Year ! ! !