||During empirical testing of the Capital Asset Pricing Model an assumption is typically made that risk is intertemporally constant. However, prior research finds that risk changes over time. A conditional dual-state cross-sectional model was empirically tested allowing risk to change through prior identification of different market and economic states. Relationships between returns and conditional market and economic-factor betas, size, book-to-market equity, and earnings-price ratios are examined. It is found that relationships shift across regimes, suggesting the importance of a conditional, as opposed to an unconditional, model. Relationships also change in January.