The composition of CMBS risk

This paper identifies the put-option, liquidity availability proportion, and shadow liquidity risk premia embedded within commercial mortgage backed securities (CMBS) using reduced form and structural generalization models. These risk values are then interpreted as trading signals which are tested with automated trading strategies that buy undervalued and sell overvalued CMBS from November 2007 through June 2015. All three signals generate substantial positive trading profits in testing for the reduced form model but not for the structural generalization. The risk signals constructed independently of market pricing provide more profitable automated trading insights than those constructed from interactions between modeled risk measures and market spreads. In my tests of the information content of the risk signals with respect to future macroeconomic indicators, I find statistically significant evidence in keeping with recent studies. While I cannot reject CMBS efficiency, this paper's disclosure of new risk measures, the profitability of automated strategies based on those risk measures, and the statistical significance of their forward guidance capabilities, together contributes to our understanding of CMBS risk and the credit spread puzzle debate.