Friday, May 31, 2013

Hedge Hogs (not the cute prickly kind)

I just finished reading Hedge Hogs by Barbara Dreyfus.  The book recounts the rise and fall of the Amaranth hedge fund and its star trader Brian Hunter.  In the summer of 2006 Amaranth lost approximately USD 6BB trading natural gas.  I have some personal interest in this story because at the time there were two other large hedge funds Centaurus and MotherRock also trading natural gas.  The clearing firm that I worked for at the time had to deal with the blowout of MotherRock - which resulted in large part from Amaranth's enormous positions.  This is a well  written and fast paced book and it does well to capture the environment of the time as I remember it.  I recommend it.

On page 139 of the book there is a passage that every risk manager should read and re-read and re-read again.  This takes place during February of 2006.  A that time, Hunter's positions were already extremely large and growing.  He was making money.  A lot of money.

Arora went to Jones to express his concerns over the size of Hunter's positions.  A little while later he also brought it up with Maounis.  "They reacted positively and told me they appreciated my input,"  Arora later said.

"Harry has a good nose for smelling danger," says an analyst who knows him.  "Harry saw what Hunter was doing was so risky that it was dangerous."

But Hunter was minting money.  "Most people were just amazed at how much money he was making," remembers a former Amaranth trader.  "You don't want to kill the goose that lays the golden eggs.  It was good for everyone when he was making money."

Other traders, back-office staff, and executives could see Hunter's profit-and-loss statements and could guess at the amount of money he traded.  Some could see his spreadsheets.  Other portfolio managers raised an eyebrow as they watched the volatility of his book.  But there was little incentive to say anything.  Everybody could see their bonuses growing.  If Hunter was trading outsized positions, he was earning outsize profits - he earned the firm USD 320 million in February.  He must know what he's doing, people told themselves.

"Human nature is such that when someone is making a lot of money, you don't question them," says another Amaranth trader.

Nothing was done about Arora's concerns. 



Monday, May 20, 2013

What Were They Thinking?

In my spare time I have been reading to take the PRM exams.  There are four exams
  • EXAM I: Finance Theory, Financial Instruments and Markets
  • EXAM II: Mathematical Foundations of Risk Measurement
  • EXAM III: Risk Management Practices
  • EXAM IV: Case Studies, PRMIA Standards of Best Practice, Conduct and Ethics, Bylaw
I have been reading for exams III and IV recently.  Part of IV is thirteen case studies of risk management disasters.  From these we are supposed to learn good practices by observing disastrous practices.  Some of the cases (Metallgesellschaft, Barings, LTCM)  are classics of risk.  From Barings we learned we need to separate roles- a trader should not also be in charge of the office. From Metallgesellschaft we learned that futures and forwards are not the same.  From LTCM we learned that liquidity itself has a value. In hindsight it is obvious what mistakes these firms made, however I wonder if the mistakes are obvious to us now only because these spectacular blow-ups highlighted those mistakes and we have learned from them.  Perhaps those risks are obvious in hindsight but they were not obvious in foresight.  However there is one case that just throws me for a loop.  That is the case of China Aviation Oil.

China Aviation Oil (CAO) is a Singapore based company and subsidiary of China National Aviation Fuel Group.  CAO was responsible for hedging airplane fuel (ie jet/kero)  for airports in the PRC.  Initially it just acted as a middle man between the markets and PRC airports- so it bought from the market and sold to PRC airports or visa-verse.  It did not hold significant risk exposure itself.  Eventually it began speculating on the direction of oil and fuel prices using future/swaps/and options...and they were not so good at speculating, as they lost USD 550 MM within a year on their spec positions.  Obviously there was a supervisory problem there.  However one of the reasons that the CAO derivatives disaster was not caught sooner was due to the incredible manner in which they were marking their option positions.

Quick review:  (For option basics see here).

The current value of an option is typically broken into two parts; the INTRINSIC VALUE and the EXTRINSIC VALUE.  The intrinsic value represents how much you would make if you exercised the option right now.  So for a call that would be the current underlying price $S(t)$ minus the strike price$K$, and for a put option that would be the strike price $K$ minus the current underlying price $S(t)$.  The extrinsic value just equals the current option price minus the intrinsic value
  • Call:
    • Intrinsic Value = $Max(S(t)-K,0)$
    • Extrinsic Value =Option Price - Intrinsic Value
  • Put:
    • Intrinsic Value = $Max(K-S(t),0)$
    • Extrinsic Value =Option Price - Intrinsic Value
Extrinsic value is also called time value because it represents the amount of the option value which is NOT due to the amount that one could make by exercising the option today.  The extrinsic value is the amount of the option value that is due to what could happen in the future or over time.

From the definition of Extrinsic Value above you can see that
  • Option Price = Intrinsic Value + Extrinsic Value
This is all just Options 101 - the real basics.  However that is not how CAO was valuing their options positions.  At least from an accounting point of view they considered
  •  Option Price = Intrinsic Value
So if Crude Oil is currently trading at a price of USD 100 and they sold a call with a strike price of USD 120 then their accounting books would say that option had 0 value.  That is obviously incorrect.  If they were able to sell the option for more than 0 then there must have been a reason - yet CAOs accounting books said the option was worth 0.  That is just crazy wrong.  And the kicker is - this did not occur during the stone-age of derivatives.  This occurred in 2003-2004.

If in the year 2023 I were to be studying for the PRM would I be reading about the risk management disasters of 2013 and saying "what were they thinking"?  I sure hope not.  In the same way that we learned from Barings, and Metallgesellschaft, and LTCM, I hope that we learn from today's disasters - with the caveat that maybe it will be obvious in hindsight but not in foresight.  And then there was China Aviation Oil.  What were they thinking?

Monday, May 13, 2013

Tragedy of the Metaphor

Talking Points Memo:  Gallup Admits Mistakes, Plans Changes After Dismal 2012

"Gallup’s editor-in-chief Frank Newport wrote in an email published Monday by Politico that “a blue ribbon group of outside experts” is conducting a review of the firm’s “methodological issues” during the 2012 election."

No surprise there.  Gallup had the worst polling performance in the 2012 presidential election (see here)  What was a bit surprising was this metaphor later in the article.
   
"Weeks after the election, Newport wrote a post on the firm’s website in which he defended Gallup’s findings and appeared to take a thinly veiled shot at polling aggregators.  “Individual farmers can each make a perfectly rational decision to graze their cows on the town commons,” Gallup wrote. “But all of these rational decisions together mean that the commons becomes overgrazed and, in the end, there is no grass left for any cow to graze. Many individual rational decisions can end up in a collective mess.”  "

He seems to be referring to the tragedy of the commons (kind of).  From Wikipedia

"In economics, the tragedy of the commons is the depletion of a shared resource by individuals, acting independently and rationally according to each one's self-interest, despite their understanding that depleting the common resource is contrary to the group's long-term best interests...The enclosure movement in England (which led to over 5000 Inclosure Acts between 1750 and 1860) prompted the analysis of this economic principle, probably known to Adam Smith.[1] In 1833 William Forster Lloyd published a pamphlet concerning European land tenure, specifically of herders sharing a common parcel of land, on which they are each entitled to let their cows graze. By this time, the English spinning and cloth-making industry had created an increased demand for wool. In English villages (as with mountain countries in Europe), shepherds had sometimes grazed their sheep in common areas, and sheep ate grass more severely than cows. Overgrazing could result because for each additional sheep, a herder could receive benefits, while the group shared damage to the commons. If all herders made this individually rational economic decision, the common could be depleted or even destroyed, to the detriment of all."

It's not clear exactly how that metaphor applies to the Gallup situation.  A more apt metaphor for the current situation might be

"Farmer George makes a decision to plant his crops too close together.  The other farmers tell him his plants won't grow well like that.  Farmer George tells them he knows what he is doing and they are all wrong.  When harvest time comes Farmer George gets a lousy crop yield.  He then announces that he will convene a panel of agricultural experts to look into why his crop yield was poor.  He finishes his pronouncement by incorrectly stating a metaphor about sheep and cows."


Saturday, May 11, 2013

Yes THIS is News

A week ago we asked the question "but is it news"?  The question was posed after two hours of viewing CNN's non-stop coverage of the aftermath of the Boston Marathon bombing.  Since then CNN has moved their focus to the Cleveland kidnapping case.   Again I ask - is it really "news"?  Well maybe the initial story deserved a mention and certainly it is "news" for the families and friends of those involved.  But for the common American what is the implication of this story?  How different might your life be if this event had or had not occurred?   It is certainly a horrible story - but there are a lot of horrible stories that go on in the world each day.  CNN appears to be broadcasting a horror movie rather than real news story...perhaps they should have looked into how much the rights to Stephen  King's movie Malice costs and run that instead.   Again I ask - is it "news"?

There is however a news story going on today which could potentially have enormous impact on the future of American.  And that is today's election in Pakistan.  A quick review:  Pakistan has a population of 180MM people (6th largest in the world) and a per capital GDP of USD 1,410.  Their literacy rate is below 60%  They currently suffer  from a debilitating energy shortage, a separatist movement in one of their provinces (Baluchistan), rampant crime and gang wars which have killed 1000s of persons in their largest city Karachi, plus a Taliban insurgency.  Perhaps this would not rise to the level of a horror movie except that Pakistan has nuclear weapons and Islamist sympathizers within their military.

The current government is headed by President Asif Ali Zardari the widowed husband of Benazir Bhutto.  He is extremely unpopular with approval ratings in the teens.  He and his party the secularist PPP are expected to lose in this election.  The two parties who are expected to pick up votes are the nationalist PML-N party of Nawaz Sharif and the PTI party of former cricket star Imran Khan.  Both have stated that they intend to be less supportive of the US and Khan has suggested that he would be willing to bargain with the TTP - the Pakistani Taliban.
  
This article by the CBC sums the situation up pretty well.
Here is today's NY Times coverage of the election.  Turnout appears very high (which is good).

And here is a fascinating set of polls by the the Pew Center the major results of which are summarized in the following two table



This next table at first looks promising showing relatively low support for terror groups - until you realize that 13% of 180MM people is 23MM people who look favorably on Al Qaeda. 


If there has been a major foreign policy failure of the Obama administration it has not been Syria or Bengazi or even North Korea (really what were we going to do there?) - it has been the continued deterioration of Pakistan.  One wonders how much it would have cost the US to provide schooling for 10MM Pakistani children?  Or to train 1MM Pakistani teachers?  Or to subsidize natural gas imports from Qatar to alleviate their gas shortages.  Like the Cleveland kidnapping story that CNN has been obsessed with, this situation could turn into a horror movie - but this is one that could have enormous implications for our future.  And that is real "news".



Friday, May 10, 2013

US Job Market

Recent news stories (see here ) have suggested that the job market is improving.  A cursory look at a plot of the Civilian Unemployment Rate seems to support that hypothesis:


But is the labor market situation really improving that much?  Recall how the unemployment rate is defined.


Start with the entire population (POP)

The Civilian Population (CIVPOP) is defined as the Population (POP) minus persons who are either
  1. under the age of 16
  2. in a correctional facility, residential nursing facility, or mental health care facility
  3. on active military duty.
The Civilian Labor Force (CIVLF) is defined as the Civilian Population (CIVPOP) minus persons who are jobless and also are not looking for a job.  The Civilian Labor Force (CIVLF) is meant to define the pool of persons who are either employed or actively looking for work ie the active labor force.

The Civilian Labor Force (CIVLF) can then be divided into those who are currently employed (E) and those who are not employed but are looking for jobs and are available for work ie the unemployed (U).

Finally the Civilian Unemployment Rate is defined as U / (E+U)  or alternately as U / (CIVLF)
 
While this definition may correctly represent the percent of potential workers who are actively looking for jobs it is less useful as a gauge of the health of the economy.  First there is a small problem that the Civilian Labor Force starts by excluding those who are either below the age of 16, institutionalized,or on military duty.  Over long periods the proportion of the population who fit into one of those categories can vary.  That is more a long term trend problem though.  

The much larger problem however is that persons who are jobless but NOT looking for jobs are not included in the Civilian Labor Force.  Look at the definition of CIVLF again.  And the proportion of persons who fall into this category can vary greatly over the business cycle.  Below is a graph of the Labor Force Participation Rate  ie CIVLF / CIVPOP.  You can see that it fell 1% between 1999 and 2007 and an additional 2% since 2007.


But even that graph does not tell us the full story because as was explained above the denominator CIVPOP excludes a significant number of persons to start with.  So where to go for a holistic picture of the labor market?  Many economists suggest that the Employment to Population Ratio ie E / POP is a cleanest picture of the labor market. And the Employment / Population Ratio is telling us a very different story than the Civilian Unemployment Rate.


* One of the interesting points to note in this graph is the Employment / Population ratio started falling in 1999 and then experienced a major drop with the onset of the Great Recession in 2007.  Here labor economist Robert Moffitt tries to explain the decline in the Employment / Population which occurred between 1999 and 2007.  From the summary of his paper.

"The decline in the employment-population ratios for men and women during 1999–2007, before the Great Recession, represents a historic turnaround. The decline was disproportionately concentrated in the less
educated and younger groups within both the male and the female populations and, within the latter, among unmarried women, especially those without children. A standard model that emphasizes the role of wage rates and nonlabor income can explain about half the decline for men but none of it for women, whose wages rose, on average and across all subgroups, over the period. However, separate examination of trends in wages and employment for married and unmarried women, and for unmarried women with and without children, finds a more important role for wages. The decline in female employment was by far the largest for unmarried women without children, and wages for that group declined over 1999–2007. However, the
different trends in wage rates and other determinants of employment for these different demographic subgroups raise many questions that need to be explored.

Most other possible influences on the employment-population ratio also appear unlikely to have contributed to the 1999–2007 decline. Federal income tax rates fell rather than rose, other federal tax rates did not rise, and federal transfer programs did not change in structure or in patterns of growth that line up with the employment declines, although further study of the Supplemental Nutrition Assistance Program and
the Social Security disability insurance program would be worthwhile.  Changes in health status, the minimum wage, and other factors also appear to have played no role, although rising rates of incarceration among disadvantaged and younger men may have contributed. Whether changes in time use and home production accompanied the employment declines is not answerable with the available data but remains a possibility. Further analysis of possible contributors to the employment decline is clearly needed.  Most other possible influences on the employment-population ratio also appear unlikely to have contributed to the 1999–2007 decline. Federal income tax rates fell rather than rose, other federal tax rates did not rise, and federal transfer programs did not change in structure or in patterns of growth that line up with the employment declines, although further study of the Supplemental Nutrition Assistance Program and the Social Security disability insurance program would be worthwhile.  Changes in health status, the minimum wage, and other factors also appear to have played no role, although rising rates of incarceration among disadvantaged and younger men may have contributed. Whether changes in time use and home production accompanied the employment declines is not answerable with the available data but remains a possibility. Further analysis of possible contributors to the employment decline is clearly needed.  

In 2008, with the onset of the Great Recession, the employment-population ratio plummeted, falling to approximately 72 percent for men and 63 percent for women in 2009. It has exhibited a slow recovery since that time. Given the downward trend in the ratio before the Great Recession, a natural question is whether it will return to its 2007 level after the recovery is complete, or only to a lower level. The model estimated in this paper can be used to forecast employment-population ratios in 2011, the latest year for which CPS data are available. The results suggest that the ratios may fully return to their 2007 levels."

Thursday, May 09, 2013

But Is It News?

A week or so ago I had CNN on my TV for approximately two hours. Within a two hour span there were at least a half dozen breaking stories related to the Boston Marathon Bombing.  The odd part was that there was almost no news in the stories...or at least nothing that I would consider "news". here are some of the breaking news stories that they were first to report on.

(1) The surviving bomber was moved from Beth Israel Deaconess Medical Center to a prison hospital.
(2) The bomber's mother had a press conference in which she said she thought her sons were innocent.
(3) A Congressman said that giving the suspect his Miranda Rights was bad decision
(4) One of the victims decided to have her leg amputated.

I am sure I am forgetting some of the breaking "news" stories.  It was two hours of continuous coverage of the Boston Marathon Bombing situation.  Perhaps I am being a bit too demanding here, and I am sure I am not the first to mention this, but was any of this "news"?  Let's look at the first "news" story as an example. 

(1) The surviving bomber was moved from Beth Israel Deaconess Medical Center to a prison hospital.

What is the impact of this story on me, or anyone I know, or to the world in general if he was moved?  And what would be the implications if he was not moved?  The reporter is disseminating a piece of information but the information doesn't have any obvious implications.  Maybe for the doctors and police officers involved there was some implication - but for no one else.  Is this really "news" then?  I think of "news" as new information that has some impact on me, or people who I know, or the world in general.  Let's see how Merriam-Webster defines news 

Definition of NEWS
1   a : a report of recent events
1   b : previously unknown information <I've got news for you>
1   c : something having a specified influence or effect <the rain was good news for lawns and gardens — Garrison Keillor> <the virus was bad news>
2   a : material reported in a newspaper or news periodical or on a newscast
2   b : matter that is newsworthy
3      : newscast

Definition 3 just equates news to a newscast.  That definition does not help define what is "news" for us.  Likewise definition 2a defines news as any material reported from a news source.  By that definition if CNN had decided to report on what I had for lunch that would be news as well.  I don't think of what I had for lunch as "news".  Definitions 1a and 1b have the same problem.  If CNN told you what I had for lunch today that would be a report of recent events and it would be information which was previously unknown to you - but you probably don't care either.  That is not "news" in the sense that I am looking for.  Definition 1c is a different usage of the word "news" .  That leaves definition 2b.  2b equates news to a matter that is newsworthy.  So lets check the definition of newsworthy.

: interesting enough to the general public to warrant reporting  

So by this definition if the public is interested in the matter then it is news. So maybe I am wrong.  If the public is interested in what kind of shoes Lindsey Lohan wore last night or a two headed cow or even what I had for lunch today then that would be news.  Still I go back to my definition of what is news   "news = new information that has some impact on me, or people who I know, or the world in general. "  And by this definition I did not see much "news" in the CNN breaking news reports.