Jump to content

GMO 4Q 2008 Letter 2nd part - Value Traps


Recommended Posts


Interesting related article

"Black Swans and Market Timing: How Not to generate Alpha", Journal of Investing,Javier Estrada vol 17, no. 3



If you missed the best 100 trading days on the Dow over the last 107 years your wealth would be 99.7% lower; but if you missed the worst 100 trading days your wealth would be 43,396% higher. Similar results (lessor scale) for the most recent 17 year period, & in 15 other international markets.


It would seem that no matter how 'cheap' it is, it is better to stay in cash untill you see clear & hard evidence that XYZ coy has turned around. You will give up some return, but are likely to end up roughly 435x [43396/99.7] better off. Magnitude may vary by time period/market.





Link to comment
Share on other sites

Sharper, I don't ever recall you posting so frequently... hope all is well!


"If you missed the best 100 trading days on the Dow over the last 107 years your wealth would be 99.7% lower; but if you missed the worst 100 trading days your wealth would be 43,396% higher."


What exactly is this supposed to tell us?  Seems like mathmatical sleight of hand to me by conflating two things that are not really comparable.


First, what is the baseline for wealth increase if you just held the whole time?  Second, what was the return of those 100 worst days vs. the 100 best?


Everyone knows that crashes tend to be more dramatic on a daily basis thans rises, but I don't think this helps us somehow avoid the former and ride the latter.


No one wants to be a value investor at the bottom, because at the bottom, it's scary.  Hopefully we can pull through.  With our beliefs intact.



Link to comment
Share on other sites

Guest ericopoly




It has long been my view that the pricing of value stocks

has a folk memory of the Great Depression when many

cheap companies went bust and the expensive Coca-

Colas survived the best. Remember, you cannot regress

from bankruptcy. Using proprietary research data, we

examined one fi xed time slot: October 1929 to June 1932.

With no rebalancing, the data showed a massive “value”

wipeout in which high P/E stocks declined far less than

low P/E stocks.



Link to comment
Share on other sites


In the spirit of the Martin letter:


The authors probably didn't quite realize it, but they found some masterkeys.


At any given time the majority of the years portfolio gain/loss will be attributable to whether you were long the day a statistical outlying event occurred. IE: over the long run if you're long the index, the normal curve tail risks will pretty much determine how you do. Example: If you were long insurance coys & unhedged the day Katrina hit, that event pretty much dominated what your return was that year


Unfavourable consequences far outweigh favourable consequences, they did so in all 15 markets, & some markets are far more sensitive than others. IE: If the outlier is in the negative tail it has far more impact.


Business & economic cycles generate serial correlation and that impacts the consequences of the tails. IE: When the cycles are in downturns the positive effects are lower & the negative higher - & the longer the cycle the greater the magnitude of those consequences. Example: If you were long the index today your losses are far higher than they would have been had we allready had a mild recession a year ago.


Application ?


You dont know when these outliers will occurr but as the downside is more severe, you should hedge the downside pretty much all the time. Rule #1?


The more concentrated the portfolio the bigger the impact. IE: because you're concentrated the hit could easily miss - but if it does hit it will be devastating. Rule #2?


If you bought the day the negative outlyer occurred, you were set for life. The once in a lifetime opportunity ?







Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...