Exposing Ideas to the Envelope of Serendipity

Contact moo at: bluechipbulldog@gmail.com

Wednesday, June 6, 2012

Fooled by Randomness

I've become completely immersed in a research report I discovered from Credit Suisse called "The New Normal Investing". I came upon it when I was looking into how others are trading volatility with all the new products being made available and at a rate that is raising many investor/trader's eyebrows. 

Written in April of 2012, they have proposed that we are currently trading in a "new" environment that is becoming increasingly and exponentially distinct from the financial landscape "that once was".  Here are the main points I took away...

1. A new environment is here to stay. With what now has been the product of perpetual anemic economic growth spawned from unemployment, low interest rates, etc...

2. New ways, that are really needs, are now evolving for institutional investors that center on searching for ways to make money, yields, and to manage what they call "fat-tail" events - the end point areas on what use to be a standard bell curve distribution for most risk analysis. To understand this more simply, if you took all the daily %changes in price for Google, for instance, for the last say 5 years and created a histogram of prices, you would NOT see a "normal" distribution of prices. The distribution for most stocks would be skewed negative but lately, and much more frequently,  more instances on the negative extreme sides of the curve are appearing. Those "negative" extremes are called Fat-Tails. This makes sense since we are in the "Great Recession" still...right?

Here is a diagram to help the imagination:

3. This "skewing" has gotten so bad now, that heads of investment strategy at Credit Suisse are telling clients how to modify their "risk" in their "new" playground. It's called "incorporating higher moments" of a return distribution. A "moment" is a statistical term. The first 2 moments, I'm sure you have heard at some point.  They are called the "mean" and "variance" . However, it seems that's not good enough anymore. Mean and variance are inputs to calculating Standard Deviation. When you go to a banker and given them your money for x,y and z, you want to know what to expect as a return. Right? Standard Deviation used to answer that question quite well. Not anymore. At least not enough. We now need to consider the 3rd and 4th orders of how things are distributed in the measure of the risk based curve to minimize these "fat-tails". The "New Normal" is a term that was canonized by Pacific Investment Management (PIMCO) in March of 2009 by the way. 

4. See how steep 3 sigma events are now taking place in the 2000-present years?
These "rare events" are not so rare anymore. This is a game changer with persistence. It's not just Equities that are at RISK. 

It's also FIXED INCOME and CURRENCIES that are getting skewed. 
Given the current facts, there will be the need to increase focus on "diversification potential" and "uncorrelated returns" because of the increase in volatility at times. The new jargon is just beginning. But how else to mitigate against lower returns?  

5. Areas of highlight to my eye (and raised eyebrow) included the following:
  • "We then identify low-growth years, reflective of the US economy’s current growth potential (e.g., 1.5% to 2.5% GDP growth) to proxy the new normal environment, and compare them against average growth years (3% to 4% GDP growth)."
  • "In the new normal, increasing equity correlations and volatility can lead to an opaque market with little visibility in underlying fundamentals."
  • "Furthermore, equity premiums can be further pressured by other prevailing trends including corporate deleveraging, moderate growth and high levels of unemployment."
  • "Recent declines are, in fact, emblematic of a longer term trend—yields that were in the 12% to 14% range in the early 1980s have fallen to current levels of 0% to 2%"
  • "...this yield narrowing..."
  • "globalization and the transfer of labor and production to lower-cost nations have also contributed to falling yields"
  • "US government and corporate bond markets may be facing low yields as the new normal for the next several years."
  • "In the new normal scenario where traditional asset classes and long-only strategies may provide lower returns, investors may look to strategies that can withstand or even benefit from challenging conditions. Going forward, we believe the emphasis will shift from returns that accrue from long-only positions in strong markets (beta) to alternative strategies that can extract different sources of risk premia"

  • "We also presented the reasons behind what we think is a significant and enduring consequence of the new environment: Distribution of outcomes have become flatter and the tails are fatter."

  • "In addition,investors can help optimize their portfolios by using extreme value theory and expected shortfall analysis (also known as CVaR optimization) to structure the portfolio."

  • "Our results show that the only strategy that exhibited normal distributions in the analyzed period was managed futures."

  • "On a historical basis, we see that the return and risk are approximately 8% and 16%, respectively. Over that same period, the maximum drawdown is about 51%."

  • "The key to our analysis is this: By combining these strategies with traditional beta, investors can minimize the drawdown risk, and reduce the skewness and kurtosis associated with the traditional asset class in a portfolio."

  • "market volatility is hedged with derivatives"

  • "Equity option strategies (such as traditional put options, or option collars that combine puts and calls)."

  • "we believe investors should re-assess whether standard risk techniques that employ normal distributions are sufficient to protect portfolio assets,"

6. But more importantly this will likely change the way traders will be trading. The reliance on current tools to measure, judge and give clues to the market are slowly ( if not already ) becoming antiquated. For instance, our current toolset is based on simple moving averages, least-square methods, using variance as a measure of dispersion, etc. These increasing incidences of "fat-tail" rare events cannot be seen by conventional methods - that are used by all of us. There is very little correlation with  a standard volume histogram - the basis for many trading tools. Instead we will be trading on "clues" that occur at infrequent intervals. Everything else is noise. Instead we will need to find ways (if not already) to trade off of these types of "new" signals rather than be screwed by them.  

I'm not sure I can post the report here...but look for it.

Its called "New Normal Investing: Is the (Fat) Tail Wagging Your Portfolio?" by Credit Suisse's Thambiah & Foscari, strategic managing investment directors. 
blog comments powered by Disqus