Since the development of modern portfolio theory (MPT) in the late 1950s and early 1960s, academics have offered numerous competing theories. MPT's simplicity is appealing: The expected return on an asset is simply a function of the return of the market portfolio and the asset's beta to the market portfolio. Arbitrage pricing theory (APT) suggests that factors over and above the market portfolio may be relevant; empirical tests somewhat validate APT, though APT does not address specific factors. Prospect theory or behavioral finance (PT) finds that investors do not behave as predicted by MPT, and that MPT seems to lack a reference point. The adaptive markets hypothesis (AMH) lacks a quantitative theory but suggests that applying principles of evolution, competition, adaptation, and natural selection to financial interactions may explain some observed market phenomena. This paper boldly claims that it is possible to capture many aspects of MPT, APT, PT, and AMH in a simple paradigm, the relative asset pricing model (RAPM), which is anchored in how investors behave and act. In RAPM, at the highest level of application, assets are invested to maximize surplus or funded status, not to maximize wealth, thereby providing a reference point — liabilities — as desired by PT. At a secondary level, asset management is delegated to agents and they make decisions relative to subcomponents of the liability proxy, with specific constraints because delegation leads to certain challenges. Depending on how liabilities have been proxied, RAPM credibly rationalizes many additional factors postulated in APT. RAPM also can absorb AMH because as regulations have changed and investors have become more sophisticated and creative in delegation, investors have adapted how they have proxied liabilities/ structured portfolios. If the simple RAPM postulated synthesizes all these attributes, it could provide the basis for a unifying theory of asset pricing, demonstrating that all current branches of asset pricing are merely different perspectives of a multi-dimensional RAPM, and additional research will flesh out RAPM in more detail and greater complexity
