Information Cascades in Financial Markets
A recent paper coauthored by faculty in the Department of Economics and ELSE from University College London, namely Antonio Guarino and Steffen Huck, offers insight to how the flow of information in financial markets fluctuates about equillibria that, if flows a little too far from the stability line, ends in chaotic and extremely different ending market scenarios. The article, titled “Aggregate Information Cascades,” can be found here.
The course ties to the topics that are discussed in this paper through the topic of information cascades, which we covered two weeks ago. Information cascades are usually depicted as the result of independent agents making sequential, equivalent decisions as a result of exposure to the information that previous agents had made those same decisions.
Guarino and Huck argue that entrepreneurial success and failure is subject to the ability of the agent representing the business to provide possible investors with the appropriate information, as to start an information cascade in favor of the business. For example, if an entrepreneur needs to raise money by selling stock in his or her company to investors, then it would make sense for the business owner to persuade investors that such a decision would be beneficial to them. If previously, the entrepreneur had gone to 100 well-known venture capitalists and convinced only 3 to invest in the company, the entrepreneur could use that data to provide a distortion of the available information to favor the business. If he tells a new investor that those 3 venture capitalists had invested in his company and refuses to inform the investor of the 97 that had not, the investor would be more inclined to give the entrepreneur his money.
Now, this is how an information cascade could arise. Once the equilibrium point has been surpassed and word spreads that several agents are buying stock in the company, the entrepreneur will find it easy to generate the goal amount of financing. However, this could also be devastating for investors. Say hundreds of individual businesses in a single market, such as the housing market, use this method to convince investors to invest in their businesses. Investors such as venture capitalists, creditors, and investment banks would soon be informed that everyone else is purchasing stock or the assets from this market, and demand in this particular market only goes up. However, the demand is not built on actual demand of consumers, but on the demand of investors after being informed that other investors have increased demand. All of the sudden, the majority of the value of the entire market is based off of superficial nothingness and as soon as the real figures of sales and quarterly reports do not match the expectations of the market cap, the entire industry, along with the superficial value, disappears. This is how to start a financial crisis of 2008.
What we can learn from such publications as this, is that the value and demand in a financial market could be subject to an information cascade that is a result of a sheer lack of appropriate information. At all times, it is made public who has purchased what, but it is hardly ever known what has not been purchased by whom. So, as a result, financial success through investment and “reading the market” is more an effect of sheer randomness than as a result of sheer genius.
This concept is appropriately highlighted in Paul Lutus’ original publication which can be found here , under the section titled “investment genius.” It is a curiosity at the very least why some investors seem to make billions and then publish their analytical secrets only to leave readers’ wallets $24.95 short, even after investing, on average. Why is this? Is it because they lied about their true analytical skills and simply wrote entire books filled with fake investment hints and tricks? Is it because once the hints become known to everyone they are no longer an advantage? Lutus argues, and with good reason, that the investors’ analysis’s of stock and other possible investments are merely right a majority of the time because of sheer luck.
A computer-generated random sample of 100,000 investors in a accurately modeled financial market was simulated and found that about half of the investors had positive returns at the end of the simulation, while the other half had negative returns. Only a very small percentage (.22%) of the investors made millions from their original portfolio, which equates very comparably to the real-world population of investors. From the computer-generated, entirely random model, a real-world situation was reproduced, without the nitty-gritty “analysis” that investors usually make before investing. The distribution of successful and not-so-successful investors showed that the belief that the market can be played by genius is simply a complicated case of the Gambler’s dilemma, and financial success in investing is more likely an effect of random probability than it is of analytical intelligence.
Because random probability is more of a factor in the financial markets than is information that is supplied to investors, it can be derived that the information is not as helpful as investors believe it to be. As discussed before, the information is not helpful in many cases, because it is a result of information cascades that build the value on possible investments off of a lack of unbiased information, falsifying the true value of a company, and negating the reliability of the market in all aspects.
References:
http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=7&ved=0CFMQFjAG&url=http%3A%2F%2Felse.econ.ucl.ac.uk%2Fpapers%2Fuploaded%2F380.pdf&ei=cK3BTtnVGITW0QHkg9nBBA&usg=AFQjCNEUBHRG8OIFiSDFE2lyEKt5ef9WCQ
http://arachnoid.com/randomness/index.html