Banking vs dealing capital as a priced risk

Recent papers have found that intermediary capital can explain prices across a number of asset classes. I will be testing intermediaries' explanatory power during the period of Glass-Steagall restrictions, in which commercial banks were ineligible to trade many categories of assets. So far, I have not found that the capital or assets of dealers eligible to trade asset classes explain those prices better than the ineligible banks. Instead, the ineligible commercial banks appear to explain prices better in the time series and at least as well in the cross-section. These findings provide some support for the idea that the apparent explanatory power of intermediaries comes from correlation with time shifting risk preferences rather than their interaction with markets.