It's Rishi

Alpha, in the world of asset management, is the measure of the difference between a portfolio’s actual returns and its expected performance, given its level of risk as measured by beta (beta is basically a measure of a portfolio’s volatility as compared to the market i.e. the S&P 500 index). For alternative investment funds, such as hedge funds, alpha is how one fund is compared to another. Furthermore, a significant chunk of the fees that a fund manager earns is dependent on the alpha that his fund returns. (Let’s ignore the fact that much of the “alpha” was probably unaccounted for beta and fund managers were likely overcompensated). Thus, alpha generation, is critical to a fund manager’s success.

So..where does alpha come from? Benjamin Graham, the father of value investing, found excess returns by buying attractive stocks that were very underpriced — priced far below their tangible book value with no warning signs to justify it. Back then (in the early to mid 20th century), analyzing securities was a challenge. Access to financial statements and timely market news was difficult. Thus for people like Graham — and his disciples — who 1) put in the effort to obtain the relevant data 2) crunched the numbers (remember no Excel!) 3) had the proper analytical framework (courtesy of Graham) and 4) were patient there were bonafide bargains to be had.

Over the course of the 20th century, access to market information improved substantially and the number of “sophisticated” investors skyrocketed. This trend continued until the point where toward the end of the century, all investors both big and small were essentially on an even playing field. Obvious bargains in the equity markets dried up. Given that classic Graham-esque bargains were sparse and that his assets under management was ever increasing, Buffett describes how his investing philosophy evolved over time to focus on concepts such as intrinsic value and economic goodwill.

Today, with ubiquitous access to real-time market information, markets are efficient. An important note here is that I define efficient to mean that the price of a security on the open market is an accurate, without-time-lag reflection of the collective sentiment of investors’ opinion on where the price should be based on news and forecasts as well as emotions/irrationalities. I am not using the word efficient to imply that market prices are always rational. Humans, for reasons that only some of which are scientifically understood, are susceptible to biases and irrational behavior. The bottom line here is that investing, even for the most disciplined strategies, has become difficult. Wouldn’t it be nice to rewind the clock a hundred years back to the Benjamin Graham’s day when access to information was difficult?

I think so.. and this leads me to wonder that maybe investors need to dig deeper to find new sources of information that aren’t obvious to other investors. As I’ve discussed many times before on this blog, there has been an explosion of news publishing because of the Web in many forms, the most obvious perhaps being blogs, both professional and personal, and the less obvious mediums such as twitter and message boards.

One concrete example that comes to mind is a blog post from January 2008 by Markus Frind (founder of plentyoffish.com, one of the largest dating web sites in the world). The post talked about how because of a subtle design change by Google, his AdSense CTR (click-thru rate) dropped by 60%. The change was that in late ‘07, Google changed AdSense so that only the link in the ad is clickable, not the whole ad area. Google had presumably done this to reduce accidental clicks. In the long term this is a good thing for everyone because it reduces click-fraud issues. However, near-term CTR drops a bit, thus # of clicks drops and thus ad revenue drops. Now, ultimately, advertisers should see a rise in their conversion rates (since the quality of clicks goes up) and thus be willing to pay more for each click, evening out the ad revenue. However, there is a lead time for that to happen. Near-term ad revenue drops and there should, in theory, be a negative impact on Google’s quarter. TechCrunch had actually reported on this change on November 11, 2007 but the post came and went without much drama.

Now, fast forward to February 26, 2008, comScore publishes a report indicating that Google’s CTR may be dropping and GOOG drops 4% to close at $471..down from the mid $600 range in January. On February 29, 2008, TechCrunch posts Google CTR Down Due to Click Area Changes referencing the ealrier post in January by Markus Frind.

I remember thinking in January that I should take a short position in GOOG. Of course, I didn’t really take myself seriously, but when the comScore report came out and GOOG dropped sharply, my jaw dropped. At that moment I realized that my idea of finding these nuggets of gold on the Web isn’t crazy. In fact, I’m not the only one with the idea. Roger Ehrenberg, a wall street veteran, had co-founded a company named Monitor110. The company has since went under, but here’s an excerpt from a TechCrunch article on the company:

Monitor110 gathers information from 40 million sources of various types (100 million by the end of next year they say), ranked by financial market knowledge through a proprietary algorithm that takes 50 factors into account – inbound links being just one reputation metric. Users can chose between top sources preselected for their market sector and subscribe to sources of their own. Static sites can be monitored for changes with good granularity. Premium subscription and other deep web sources, blogs, forums, news and regulatory filings are among the sources included.

Here’s an image that used to be on Monitor110’s homepage. It elucidates the concept very well:

Think about an engineer who blogs about how he’s working like crazy because his project at work is behind schedule. If we know that he works at THQ, this information could be valuable to an investor. That’s precisely what Monitor110’s business model was. Sell to hedge funds and other folks desperate for alpha. Is this scalable? I’m not sure. Probably? Maybe too early for it’s time?…it was for Monitor110.