What is the origin of the soft excess in active galactic nuclei?

We investigate the nature of the soft excess below 1 keV observed in active galactic nuclei. We use the XMM–Newton data of the low-redshift, optically bright quasar, PG 1211+143, and we compare it with the narrow-line Seyfert 1 galaxy, 1H 0707−495, which has one of the strongest soft excesses seen. We test various ideas for the origin of the soft X-ray excess, including a separate spectral component (for example, low-temperature Comptonized emission), a reflection-dominated model, or a complex absorption model. All three can give good fits to the data, and χ2-fitting criteria are not sufficient to discriminate among them. Instead, we favour the complex absorption model on the grounds that it requires less extreme parameters. In particular, the geometry appears to be more physically plausible as the reflected component in the smeared absorption model is no longer dominant, and relativistic distortions, while still clearly present, are not tremendously larger than expected for a disc around a Schwarzchild black hole.