Using a dynamic model of unionism and wage determination we find that the unobserved factors that influence union membership also affect wages. The estimates suggest that UK trade unions still play a non-negligible, albeit diminishing, role in wage formation. It appears that the greater impact of un observables in determining individual union propensity concerning the second period under analysis, versus past unionisation experience, implies that those remaining in unions during (1997-2002) gain most from their sorting decision. The significant contribution of unobserved heterogeneity renders the total union wage differential highly variable across individuals. The endogeneity correction procedure employed yields a discernible pattern of the estimated union wage effect relative to OLS and Fixed effects. This is in line with Robinson (1989a) and Vella and Verbeek (1998) and refutes the pessimistic conclusions reached by Freeman and Medoff (1982) and Lewis (1986) that endogeneity correction methodologies do not contribute to our understanding of the union wage effect puzzle.