Saturday, February 6, 2010
Thoughts on Arbitrage Pricing Theory by Roll and Ross - Earl Malvar
As an alternative pricing model to the CAPM, I think the Arbitrage Pricing Theory works on a very plausible framework that attempts to identify and understand the factors that influence movements in asset prices. For one, formulating an approach wherein determinants are segregated as systematic or idiosyncratic factors helps investors visualize where the theory is going. Supported by empirical tests, the four systematic factors that were identified by Ross and Rolls strengthens its case over the CAPM as a pricing model that is sensible and realistic. I also think that the model intuitively recognizes the price-auctioning fundamentals that the market functions and exists by; wherein buyers and sellers make bids and offers based on perceptions and expectations. Furthermore, it highlights how the market, as if like a self-regulating entity, attempts to price-in all material information to identify undervalued and overvalued assets in order to make the “best deal”. I think what makes the APT special, more than anything, is the flexibility it offers in valuing securities, whereas the CAPM requires one to identify an efficient market portfolio based on maximizing utility via returns and risk, this model moves beyond the “revered” model portfolio and promotes the identification of factors that influence price movements. This, in my opinion, further expands the practice of asset valuation and forecasting through statistical processes such as correlation, regression, etc. Although it does not suggest absolute predictability of future prices, I believe it recognizes the role that humans play in creating market consensus and perceived price parity. After all, what is price but an agreed upon value by people based on current market and economic conditions. What I found interesting about the APT, however, is its recognition of risk-taking in a different light. Although the CAPM also gives regard to risk, it does so with the premise that investors look to minimize it. With the APT, risk is reflected as the surprise factor or the alpha that investors price-in or pay for; that although a certain security (or stock) may already be at parity to the anticipated value, unknown future events can be and are speculated upon and are even reflected through the sentiment of risk-takers via the presence of bids and offers in the market. It also suggests that although an investor can opt to make money passively by buying the market (diversifying away idiosyncratic or company-unique risks), well diversified portfolios can still perform variedly and innumerable factors will ultimately influence price. Taking equity markets as an example, if stock prices in the market can move by the seconds and minutes, does it really mean that a company’s value and business prospects can change that fast? And if information plays a part in stock prices, how fast does a newsroom, a reporter, a news editor or even a rumormonger have to work to disseminate facts, half-truths or even lies to influence prices? Encompassing these questions, in any case, I think that the “anticipated” figures are still relative. Though market players can price-in all available information, I believe that an outlook is still opaque amongst investors and never entirely consensual. Sometimes, the direction of a company’s stock (whether up or down) can be predicted on (even by those not entirely exposed to the teachings of finance), but the numbers used by analysts will vary and even conflict each other.