You Analyze, I Analyze, We Analyze. And we disagree.
May 28, 2010 2 Comments
In messy world of finance, every single individual tries to provide their two cents worth. There many groups of people ranging from, economists, analysts, news commentators, astrologers, professors and an ever growing group of econbloggers (finance related bloggers). In addition, there also qualitative and quantitative ways to form arguments. Lets start with two of the most common groups of people who will make their judgments on the markets, economists and analysts.
Firstly analysts, these groups of people usual belong to brokerage firms, investment banks and are investment related entities. As seen from the image below from Mckinsey,
We can see that the forecast EPS, which is represented by the green line, is usually above the actual EPS for the period which is represented by the green dot. The main reason that I believe is that this is so is due to the fact that analysts belong mainly to the sell side have belong to brokerages. Since the brokerages makes money from commissions, they will make more if its clients trades more. Given the fact that you can buy any stock or investment instrument but only can sell those that you have on hand, hence analysts tend to provide a rosier image to encourage sales. This kind of optimism bias is at its most evident with credit rating agencies like moody’s, fitch and S&P which have been under great attack from both the public and the SEC for missing the sub prime mess. In addition, they also failed to forecast the debt issues in the Eurozone. Notably, Barry Ritholtz has affirmed the worthless presence of these credit analysts.
Besides an optimism bias, they are also susceptible to Behaviourial Leapfrog, which simply put is the fear to go aggressive with their targets even if they really believe that they are right. For instance if the average consensus in the market of EPS on a stock ranges mainly between 1.00 and 1.25, however as an analyst I discovered that most analyst had missed out a very important aspect like maybe the impact of a new product and did my calculations and my EPS is 1.80. Even though I have great conviction about my analysis, I choose to only present an estimate of 1.35. Why is this so? This is due to fear of sticking out their necks out, hey seriously they provide buy and sell ratings, as long as they get the direction of their analysis right they are most likely going to be credited with their findings.
Now moving on to economists who have a different set of issues but similarly they are poor forecasters were to slow to react to the crash in 2008 and slow to adjust to the rebound in 2009. Notably, Rajiv Sethi from the University of columbia pointed out a few constructive suggestions for macroeconomists. However the point below really stood out.
What I could draw from this statement was that sometimes we just need common sense and logic. Too many a times economists make use of econometrics to project relationships and how one variable will affect one another. These bloggers however take the route of rationality in order to present very compelling arguments. Falkenblog does well in highlighting the limitations in Econometrics which are
1) Economic relationships are not Linear. For instance, you buy beer in bulk at the middle of the night desperately needed for party you might be charged a premium, however if you bought in the morning maybe you get the bulk discount.
2) There is a bias in trying to remove variables that do not concur with your findings. In regression, we omit variables that are not significant, but is that always correct?
3 ) In a sense, econometrics as a science allows shoddy empirical work to hide behind pretentious techniques that try to avoid these issues. Vector auto-regressions, or natural experiments, do not obviate the need for common sense, and understanding of the subject to which the tool is being applied.
In the next part I will discuss about how to approach analysis and its different forms.