Unemployment, living conditions and longitudinal data: some remarks arising from the papers by Bernadi and Schizzerotto and Lemmi