Can a photometric redshift code reliably determine dust extinction?

Photometric redshifts can be routinely obtained to accuracies of better than 0.1 in Delta z/(1 + z). The issue of dust extinction, however, is one that has still not been well quantified. In this paper the success of two template-fitting photometric redshift codes (ImPZ and HYPERZ) at reliably returning AV in addition to redshift is explored. New data on the 2nd Canadian Network for Observational Cosmology (CNOC2) spectroscopic sample of 0.2 < z < 0.7 galaxies are presented. These data allow us to estimate AV values from the observed Balmer decrements. We also investigate whether the empirical value of gamma = 0.44, the ratio between gas- and star-derived extinction, as determined by Calzetti, is necessarily the best value for this sample. When comparing the two codes with the Balmer-derived A(V) (Balmer[A(V)]), a correlation between the photometrically derived A(V) (Phot[A(V)]) and the Balmer[A(V)] is found. The correlation is improved when the empirical value of y = 0.44 is allowed to vary. From least-squares fitting, the minimum in the reduced chi(2) distribution is found for gamma similar to 0.25 +/- 0.2. For the sample of galaxies here, the factor of 2 difference in covering factor implied by the Calzetti ratio is found to be plausible. The CNOC2 galaxies with detected Balmer lines have some preference for an increased covering-factor difference, which perhaps implies that they are undergoing more rapid, 'bursty' star formation than the galaxies Calzetti used in her derivation.