Economic Apocalipse blamed on math and physics

[giann_tee]

(actual headline may differ)

After the Crash: How Software Models Doomed the Markets
Overreliance on financial software crafted by physics and math PhDs helped to precipitate the Wall Street collapse
http://www.sciam.com/article.cfm?id=after-the-crash

 

[Evo]

Blaming the software instead of the people that used it wrongly, typical of finger pointing when things go wrong.

You've heard the term PEBKAC? Problem Exists Between Keyboard And Chair

 

[Averagesupernova]

Yeah when a company puts you off for a day or so because 'computers are down' I always figure that is a convenient excuse.

 

[shoehorn]

"Economic apocalypse"?

Sheesh. We're not eating our dead quite yet.

 

[giann_tee]

I sincerely, honestly don't believe in problems such as global economic recession. It sounds unreal at least because people can rearrange things and make new agreements. When I saw the title in newspaper I knew it was perfect for physics-forums. Big names in business can't explain these things because they are (not around here to complain) temporary phenomena that exist at various scales. Simply I would bet on more of network effects rather than doubt people and ambition.

 

[D H]

SciAm is a bit late in the "blame the physicists" game. 60 Minutes and Physics World played this game over a month ago. We discussed this exact topic in the https://www.physicsforums.com/showthread.php?t=256784"" thread starting around post 362.

This is akin to blaming, for example, a chemical plant explosion on the chemical engineers who designed the plant. While those chemical engineers may have some culpability, the fault more properly falls on management. Why is nobody blaming the management of these financial institutions? The media is instead blaming the government and the quants. This is wrong-headed. Every other industry where risk is a large factor and failures have a significant societal impact have learned that internal oversight is very important. The financial industry instead made they very quants who developed these models also responsible for the risk analyses. The corporate management set themselves up for failure.

 

[BWV]

The models were crap, so they were at fault as well as the people using them. Most of them assumed liquidity was always available, normal distributions of returns, and neglected the systemic risks of what would happen when everyone followed what the models suggested.

 

[Ivan Seeking]

Already posted in P&WA, but perhaps applicable here as well.

What role did the credit rating agencies play in the current economic crisis? This week, a former managing director at Standard & Poor's speaks out on U.S. television for the first time about how he was pressured to compromise standards in a push for profits. Frank Raiter reveals what was really going on behind closed doors at the credit rating agencies the public relies on to evaluate the safety of their investments.

"During this period, profit was primary; analytics were secondary," Raiter tells NOW Senior Correspondent Maria Hinojosa.

Who was watching the watchers? Surprising new revelations about the economic debacle, this week on NOW...

http://www.pbs.org/now/shows/446/index.html

 

[franznietzsche]

The models were crap, so they were at fault as well as the people using them. Most of them assumed liquidity was always available, normal distributions of returns, and neglected the systemic risks of what would happen when everyone followed what the models suggested.

This gets more to the heart of the matter. It's the problem of the spherical cow approach to economics.

The simple fact was that the models were based on assumptions, and then the people using the models ignored that. This isn't the fault of the model makers, it's the fault of the people using the models not knowing that the models are inherently fallible (or not caring).

 

[SW VandeCarr]

While it may be unfair to blame the mathematicians employed by banks and security credit rating agencies, they influenced the decisions of the executives who believed their models (based on estimates of probability of default) could somehow transform BBB rated securities to AAA ratings when properly packaged. Both the creators and users of these models wanted to believe in their reliability. It was a very profitable game for a while.

 

[John Creighto]

The models were crap, so they were at fault as well as the people using them. Most of them assumed liquidity was always available, normal distributions of returns, and neglected the systemic risks of what would happen when everyone followed what the models suggested.

If someone puts 100% of their faith in a computer model of something as complex as a global economy which is a measure of actions of people (highly irrational by nature) then they deserve to look like fools. But seriously, anyone who thinks that computers are the only problem in the markets has a lot to learn with regards to recent events.

I recommend this blog:
http://www.creditwritedowns.com/

 

[jreelawg]

I watched Freddy and Fanny get interrogated on C-span. I can't remember which of the two it was, whos risk manager was warning them the whole, while, and instead of listening to him, they fired him. Turns out that what the risk manager warned of is exactly what happened.

I do think there is a problem with the technological advances nowdays though, because the people can't answer simple questions without being both econ and math majors. You can't ask how it works, because they use special formulas and algorithims that only a few people in the world understand. People are being too dependant on computers. A settlement capitol corp., won't stop harrassing me because I asked for a quote. Now the computer has me in it's memory, and it doesn't matter who I tell to leave me alone, the computer keeps generating mass memos to solicit everyone and their grandma. With the new mess of personal information shairing, loan sharking, soliciting, and scamming, things are way out of hand. I get so many recorded message calls offering me life insurence deals, telling me I could have won a free vacation, trying to get me to sign up for a credit card, shop at gothschalks, etc, etc. It is really getting to be a problem.

Where the complex math comes in, is to help scam us is the fact that we can't understand how it works. If it was a clear contract and clearly showed that you are being screwed, then people would stand up and have it abolished. But, because even congress doesn't really get how it works, it is easy for the institutions to manipulate people. It is like a complex version of, "you give me 5 dollars, I'll give you 10, then you give me 15, I'll give you 10, you give me 65, I'll give you 100, you give me 400.", type of thing.", made confusing and deceptive enought to actually work.

 

[John Creighto]

I watched Freddy and Fanny get interrogated on C-span. I can't remember which of the two it was, whos risk manager was warning them the whole, while, and instead of listening to him, they fired him. Turns out that what the risk manager warned of is exactly what happened.

If I understand correctly, the assets of Freddy and Fanny were not sufficiently liquid to deal increased delinquency. For a bank 70:1 would be considered under capatalized (See: http://wfhummel.cnchost.com/capitalrequirements.html ).

Over 98 percent of Fannie's loans were paying timely during 2008.[48] Both Fannie and Freddie had positive net worth as of the date of the takeover, meaning the value of their assets exceeded their liabilities. However, Fannie's total assets to capital (leverage ratio) was about 20:1, while Freddie's was about 70:1.[49][50] These numbers increase significantly if one includes all of the mortgage-backed assets they guaranteed. These ratios are considerably higher than investment banks, which leverage around 30:1.[51][52]

However, there was concern that the GSEs' liquidity was insufficient to handle growing delinquency rates, such that although viable in September 2008, the scale of loss in the future would be sufficient that insolvency would occur and that knowledge of this future failure would induce immediate or near-immediate failure due to buyers refusing to buy debt. Both GSEs roll-over large amounts of debt on a quarterly basis and failure to sell debt would lead to failure due to lack of liquidity. A slower form of failure would be the issuing of debt at high cost (to compensate buyers for risk) and this would greatly diminish the earning power of both GSEs, rendering them unable to earn the money they would need to handle expected future losses. Both GSEs counted large amounts of deferred tax assets towards their regulatory capital, which were considered by some to be of "low quality" and not truly available capital. The deferred tax assets would only have value if the companies were profitable and could use the assets to offset future taxes. Both companies had experienced significant losses and were likely to face more over the next year or longer.[53]

http://en.wikipedia.org/wiki/Federal_takeover_of_Fannie_Mae_and_Freddie_Mac#Credit_default_swaps

The BIS leverage requirement is that your total Tier 1 capital not be less then 3% of liabilities. 3% is equivalent to leveraging (1/0.03)=33.3 times capital. Freddie was leveraging at 70 times capital ignoring the potential liabilities of their guaranteeing of mortgage backed assets (A.K.A credit default swaps). Moreover, if regulators consider the assets dodgy they can increase the leverage requirement to 6%.

I've heard that the rules in place were sufficient and just not enforced properly. I'm not sure though because if that is the case then why was Freddie allowed to be so under capitalized prior to the finical crisis.

 

[harborsparrow]

"The economy" is an imagined thing, not a real thing. If everybody imagines that it is working, people spend resources (and it works). If a vast majority believe it is not working, people horde resources (and it doesn't seem to be working well). A key factor in "getting it going again" is to quit picking at the scabs of things that failed and focus instead on things that are, well, working. Sounds too simple to be true, doesn't it?

 

[John Creighto]

"The economy" is an imagined thing, not a real thing. If everybody imagines that it is working, people spend resources (and it works). If a vast majority believe it is not working, people horde resources (and it doesn't seem to be working well). A key factor in "getting it going again" is to quit picking at the scabs of things that failed and focus instead on things that are, well, working. Sounds too simple to be true, doesn't it?

Psychology plays a part in the market but it's not just about psychology. If people are in too much debt they can't spend regardless of weather they feel the economy is doing well. If investment grows faster then demand then eventually their will be large market imbalances, leading to large surplus of goods, leading to dramatic drops in prices, eroding earnings and eventually leading to recession or depression.

 

[Phrak]

(actual headline may differ)

After the Crash: How Software Models Doomed the Markets
Overreliance on financial software crafted by physics and math PhDs helped to precipitate the Wall Street collapse
http://www.sciam.com/article.cfm?id=after-the-crash

Pffft.

Who done it? It's the blame game, and they is us! The villains:

Physics grads scooped up by firms to work on market modeling.
The investment firms themselves, the execs and their wifes who supported them.
Anyone who supported reducing the capital gains tax.
Anyone with a 401K.
Anyone who traded tech stocks.
Anyone who traded on the internet.
Anyone who started, invested in, or was employed by a 'leading edge' tech company.
Anyone riding 'the wave'.
The guy at the bar with the hat who gave you a tip.
Everyone who bought stock based on expected future earnings.
Every CEO.
Anyone you heard that was excited about 'the market'.
Each and everyone who said anything remotely like "we've broken the old business model."
Anyone who advertized on the internet.
Anyone who clicked on an internet ad--even once.
Anyone who had anything remotely to do with building the internet, wireless, sat coms, fibre, personal computers or some goofy 'green' energy scheme.
People with cell phones.
Users of personal computers (got you all ).

 

[JakeA]

Bad models. For instance if you take a house price that has been jumping 15% per year, and you assume that it will appreciate 15% per year indefinitely, well that's a bad model. Time to fire the modelers and get better ones.

 

[GlobalRoad]

I am an economist and a mathematician turned business studies lecturer because my subject was invaded by failed physicists and mathematicians who new and understood nothing about the economy, economics or its traditions. For a well crafted detailed indictment of what they have done and why they were always wrong and why they have all failed now read Mirkowski's "More Heat than Light".

All these so called economist from Debreu, Nobel for his 'The Theory of Value', new nothing whatsoever about the economy or economics. Debreu himslef said on frequent occasion that he was not an economist. His nobel prize winning work is a well crafted piece of differential topology, an existence theorem, to prove a sollution exists to Walras' hypothesis, of a hundred years previously that prices could adjust by a process of trial and error, 'tattonement', to converge to a General Eqilibrium - a set of prices such that that all markets cleared, i.e. supply equalled demand in every market.

His work is a beautiful demonstration of the righteness of General Equilibrium Theory and the Neo-Classical Economics that has so catestophicaly failed but a proof of what utter nonsens it all is. He shows that for General Equilibrium to be possible. There needs to be a set of markets defined over a complete, differentiable measure space defined over all time from the day of Genesis, the Big Bang to the Present for all goods existing at all points of continuous time and between all points of time for this result to hold. His mathematics is beautiful but what arrant nonsense. For General Equilibrium to Exist even as an aspiration the universe has to be in stasis fo eternity. Beautiful math that proves that economics based on the idea of tattoement leading in the limit to General Equilibrium is complete nonsense.

 

[JakeA]

LOL, Debreu's "Theory of Value" is so important that nobody on Wikipedia has bothered to write about it. It's got to be one of the stupidest pieces of mathematical masturbation in history.

Economics needs to dumb its math down a lot. You don't need calculus or topological theory, you need arithmetic. While everybody was running complex and idiotic asset valuation models, nobody bothered to ask the simple question of how a person making $3k per month could afford a $4k mortgage payment after the teaser period ran out.

Math is often used to confuse and obscure. Hide the true numbers behind PhDs and pompus formulas.

 

[changeseeker]

The models are only as good as their assumptions. If you use historical data, and history is not a good predictor of future trades, you are bound to be in trouble as LTCM found out. Also, you can't argue that informatization did not have an effect. It's not the same thing to have a human being trading on instinct than a system trading on pre-programmed variables and game theoretic assumptions. Ultimately, the speed of the transactions is going to grow very quickly, humans may be unable to keep up with them, and there are going to be unanticipated consequences, sometimes positive, sometimes negative.