if 'system accident' publicized at time of ValuJet crash, better Enron, WorldCom resp

A DC-9 flown by ValuJet crashed in the Everglades on May 11, 1996, and resulted in the loss of life of all five crew members and all 105 passengers. Almost two years later, writer and pilot Bill Langewiesche wrote a highly intriguing article building on the work of sociologist Charles Perrow.

The Lessons of ValuJet 592, the Atlantic, William Langewiesche, March 1998.

http://www.theatlantic.com/magazine/archive/1998/03/the-lessons-of-valujet-592/306534/

' . . . Flight 592 had been loaded with a potentially dangerous cargo of chemical oxygen generators . . . '
' . . . They inhabited a world of boss men and sudden firings . . . '
' . . . replacing oxygen generators . . . '

' . . "expired" . . '

' . . "expended," . . '
' . . . The two unfortunate mechanics who signed off on the nonexistent safety caps just happened to be the slowest to slip away when the supervisors needed signatures. . . '

Toward the end of the article, Langewiesche draws his long knives:

' . . . the creation of an entire pretend reality that includes unworkable chains of command, unlearnable training programs, unreadable manuals, and the fiction of regulations, checks, and controls. Such pretend realities extend even into the most self-consciously progressive large organizations, with their attempts to formalize informality, to deregulate the workplace, to share profits and responsibilities, to respect the integrity and initiative of the individual. The systems work in principle, and usually in practice as well, but the two may have little to do with each other. Paperwork floats free of the ground and obscures the murky workplaces where, in the confusion of real life, system accidents are born. . . '
So, what if the theory of system accident had been better understood at the time of Enron (2001) and WorldCom (2002)?
 
No.

You have to separate auditing from consulting services. When the people who sign off on your financial statements are the same ones you pay to MAKE the financial state,nets (or at least be heavily involved in the company) there are going to be enormous independence issues.

Overbearing and/or extremely charismatic bosses are going to happen, see Z-best, Crazy Eddie, or Sunbeam.
 
But maybe.

If the concept of system accident was more widely discussed, more people would view requiring new ethics rules as just overlaying a new layer of complexity onto an already complex system.

And separating auditing from consulting, as you suggest, might be the far cleaner solution. Might require new legislation, or a Justice Department which more aggressively uses existing legislation.
 
And how better understanding of this and similar stories might have potentially prevented the 2008 financial meltdown?

Please give me a mid to late POD.
 
Not really - Enron / Worldcom were deliberate corporate frauds.

System accident relates to incidents which occur due to unworkable procedures being created to overcome a real problem.

Enron and World com did not occur because the systems were unworkable - they occurred because the people placed in charge of the company broke the rules to apparently boost profits and returns (and hence bonuses)
 
Not really - Enron / Worldcom were deliberate corporate frauds.

System accident relates to incidents which occur due to unworkable procedures being created to overcome a real problem.

Enron and World com did not occur because the systems were unworkable - they occurred because the people placed in charge of the company broke the rules to apparently boost profits and returns (and hence bonuses)

Or see the current Volkswagen scandal.

Bosses demand impossible targets, middle managers pass it down, no one accepts 'no' for an answer.... If you DO complain, you're in danger of losing your job....

Since such behaviour continues to this day, I doubt positing the theory a year or two earlier would have any effect.
 
But maybe.

If the concept of system accident was more widely discussed, more people would view requiring new ethics rules as just overlaying a new layer of complexity onto an already complex system.

These concepts already existed, NASA had such a culture before Challenger blew up (with engineers placed under hostile and clueless managers who ignored their input).

And separating auditing from consulting, as you suggest, might be the far cleaner solution. Might require new legislation, or a Justice Department which more aggressively uses existing legislation.

And it's not like Enron or Worldcom were the first companies to engage in deliberate fraud. ZZZZ Best, Sunbeam, Phar-Mor (you know, company people thought might overtake Wal-Mart), and Crazy Eddie had all happened already. It didn't stop anyone.

Not really - Enron / Worldcom were deliberate corporate frauds.

System accident relates to incidents which occur due to unworkable procedures being created to overcome a real problem.)

Also this.
 
Enron and World com did not occur because the systems were unworkable - they occurred because the people placed in charge of the company broke the rules to apparently boost profits and returns (and hence bonuses)
But the overall regulatory system is complex, particularly the part where a public accounting firm is expected to produce solidly audited financial statements.
 
I draw two lessons from system accident theory: that it's the complexity itself, and sometimes even more specifically, the complexity of a safety system and/or safety procedures.

And I like the theory. I think it's specific enough, or can be made specific enough to really explain some stuff, both when institutions work poorly and when they work well. And I fully concede that a hundred and one things are going on in the real world and any one theory can only explain part of it at a time.

And alright, the situation with the Arthur Andersen accounting firm and Enron, a lot of it was indirect social conformity and pressure and people not doing their jobs. And conforming too much for the sake of a working relationship. It's like the Canadian guy who took on Monsanto said, doctors always say they're not going to prostitute themselves over a free meal or a free cruise. But maybe the pharmaceutical companies understand a little about reciprocity. And (my idea) they give rewards and establish themselves as credible and as in the mainstream, and then if there's a controversy and one doctor says to another, hey, let's not get carried away, at that point the pharmaceutical company has received full value for the money. I'm thinking of the pain reliever Vioxx, and even if in some final analysis it's shown to be safe enough, the practices of Merck were anything but to write home about.

=======

So, following Enron and WorldCom, there was some talk of legally separating firms like Andersen into consulting and public accounting. I don't know if it was all that seriously considered. But if system accident was considerably more highly publicized in '98 and again in '02, it may have been. And then, it could have gone either way.

And for the whole enchilada, prevent the 2008 financial meltdown. Admittedly, a long shot. Maybe system accident would be only one of three potential PODs.
 
So, following Enron and WorldCom, there was some talk of legally separating firms like Andersen into consulting and public accounting. I don't know if it was all that seriously considered. But if system accident was considerably more highly publicized in '98 and again in '02, it may have been. And then, it could have gone either way.

Sigh...ever hear of Sarbanes-Oxley? That's part of what it does. It's WHY auditing is actually profitable for firms now.

And for the whole enchilada, prevent the 2008 financial meltdown. Admittedly, a long shot. Maybe system accident would be only one of three potential PODs.

Biggest problem is, system accidents are, as the name implies, accidents. They don't involve malicious actions undertaken by the people involved. Hell, frankly if upper management is involved in fraud we don't have a snowball's chance of finding it unless there is a tip, someone gets really lucky, or the fraud is enormous and obvious. And again, it's not like fraud hadn't happened before. I keep bringing up examples of them. Some theory that barely applies isn't going to do anything to prevent more.
 
Really interesting article. The same author had a great piece also in Vanity Fair about the AF447 accident, which was also notable for a significant portion of the root cause attributable to system errors.

Should airplanes be flying themselves?

I can't comment too much on specifics re: corporate accounting/auditing processes but at least in the aviation and medical fields despite knowing about these types of errors for a long time they still occur. My understanding is that these errors to a fair extent are inheritent to the systems themselves and despite continued efforts are unable to be eliminated completely. So basically, we might come up with a POD that avoids the 2008 crisis (no repeal of Glass-Steagal?) but the underlying complexity of the financial system means sooner if later a similar systems error occurs.

I'd argue that fraud should be considered as a systems failure. Human factors play a huge role in CRM; just because the system in place to deter financial fraud is generally much less effective than that to deter a pilot ignoring his colleagues and flying the plane into the ground doesn't mean it's not a systems error.
 
. . . Sarbanes-Oxley? That's part of what it does. It's WHY auditing is actually profitable for firms now. . .
As I remember it, Sarbanes-Oxley was a more detailed code of ethics, including asking members of boards of directors to personally sign off on statements, when many of these people were generalists with a lot of time commitments. So, it might get done. It might make a difference for a while, which I'll admit is a positive step. I'm just interested in bigger positive steps. Or, I'm interested in the method of medium step, feedback, medium step, feedback.

Did Sarbanes-Oxley split accounting firms into audit-only companies and everything-else companies? And I mean, completely separate companies. I really don't think it did. And in my mind, that's the biggest aspect of a simple, straightforward solution.
 
. . . My understanding is that these errors to a fair extent are inheritent to the systems themselves and despite continued efforts are unable to be eliminated completely. . .
That's how I understand the theory, too. And this is perhaps the scariest aspect. But at least the whole approach can warn us, be careful not to make things worse.

And here's another way. We might think of transparency more as a soft skill and a feel-good approach to dealing with employees. These system accident studies can be read as saying: in addition to all that, transparency is one of the better ways to improve safety in a hard, engineering sense.

PS Thanks for the link to this other Langewiesche article.
 
As I remember it, Sarbanes-Oxley was a more detailed code of ethics, including asking members of boards of directors to personally sign off on statements, when many of these people were generalists with a lot of time commitments. So, it might get done. It might make a difference for a while, which I'll admit is a positive step. I'm just interested in bigger positive steps. Or, I'm interested in the method of medium step, feedback, medium step, feedback.

Did Sarbanes-Oxley split accounting firms into audit-only companies and everything-else companies? And I mean, completely separate companies. I really don't think it did. And in my mind, that's the biggest aspect of a simple, straightforward solution.

You remember wrong. That is part of what Sarbanes-Oxley did. The part you refer to makes it illegal to sign off on financial statements which the officer knows have material misstatements* in them.

There 11 major points, but we are focusing on just one: independence. Sarbanes Oxley requires that the company's auditors be independent** of the audited firm. This means that an accounting firm MAY NOT act in a consultant*** role for a company that it audits, and the auditors must shift every few years (okay, I forget if SOX did that, or the PCAOB, but since the latter was created by the former I'll just go with that). Way back when what happened was Enron was the major source of Arthur Anderson's revenue. Auditing was minuscule, and competition was so fierce that it often was a loss leader, or very close to it because the consulting work was so much more lucrative, and why wouldn't a company want to hire its own auditors for that?

So firms undercut each other to get the work, to the point that budgets (how much time can be spent on any one task) became minuscule. This was a big part of why Phar-More escaped notice. The firm had undercut its needs so much that it was IMPOSSIBLE to actually do an audit. So among the thousands of stores run by the retailer they would inspect 3 (IIRC, it might have been higher, but not much) and they TOLD the company which stores would be audited, and on which day, so Phar-More would shift inventory around to make the stores look full, when in reality the inventory was mostly invented.

Incidentally, I mentioned that SPX made auditing profitable again, that was because the more stringent requirements of the Act made auditing a more intensive process. One requiring higher budgets, which in turn meant more money brought in.

Then there was ZZZZ-Best, if you haven't heard of it it was a carpet cleaning company of all things in California. The thing was the carpet cleaning never made a profit, so the owner invented revenue from other things (actually loans he would use the proceeds from to pay off other loans). Here's where independence strikes again. The owner would wine the auditors, and insist they brought their WIVES, who would be wowed by his charisma, and (without meaning to) influence the auditors to trust him.

*That is to say anything incorrect which, if presented properly would make a reasonable person change their opinion of the information provided. So a ten dollar misstatement (for some companies a one hundred thousand dollar misstatement) would not be a material misstatement. If say Wal-Mart mislaid five hundred dollars and this isn't shown on the balance sheet it is immaterial, because 500 dollars isn't relevant to the financial statements as a whole.

**in both fact and in form, to the point a reasonable person would not see an ability to cast undue influence upon the auditors.

***mostly, there are exceptions, but we are getting into far more complicated territory than a simple explanation.
 
Alright, I'll take seriously and be open to the idea that Sarbanes-Oxley was better than I initially thought. But there's still the question of, What the hell happened ? ! ?

I mean, this legislation passed around 2002(?), and the subprime mortgage 'industry' just rolled on till toward a major collapse in '08 which badly affected the rest of the economy, that is, the real economy, not just the casino economy.
 
Alright, I'll take seriously and be open to the idea that Sarbanes-Oxley was better than I initially thought. But there's still the question of, What the hell happened ? ! ?

I mean, this legislation passed around 2002(?), and the subprime mortgage 'industry' just rolled on till toward a major collapse in '08 which badly affected the rest of the economy, that is, the real economy, not just the casino economy.

Because it wasn't fraud that caused the subprime mortgage crisis?

Because auditors aren't magicians who know everything about the markets the companies they have to audit are like?

Because the government decreased regulation around the causes?

Pick your poison. What you need to understand is that auditing is not a guarantee of accuracy, or even the long term financial health of the organization (though we do try to check on those, its just not always possible.) All we offer is an opinion on the financial statements. That's why we have to use test procedures. Looking at everything just isn't possible.
 
I'm all for a combo of baseline and spot-checking. I mean, what else could an auditor do? I might simply add, believe the spot-checking.

With Enron, which filed for chapter 11 bankruptcy in Dec. 2001, I understand that that the CPAs and other auditors at Arthur Andersen accounting found all kinds of reasons and excuses not to include subsidiaries on the balance sheets, including bizarrely including the revenues, but not the expenses or exposure ? ! ?

With the financial crisis which really started to go south Sept. '08, I understand that with some of the major subprime lenders such as Countrywide and Ameriquest, it was a wild west atmosphere. It was like an auto dealership of sell, baby, sell. Sell and don't ask any questions. Sell all day long, we've got to beat those numbers! Sell, Sell, Sell!!! And, that's what they did, and this part of the casino economy eventually came crashing down; unfortunately taking a good chunk of the real economy with it.
 
How the Clinton Team Thwarted Effort to Regulate Derivatives

http://www.pogo.org/blog/2014/04/how-the-clinton-team-thwarted-effort-deregulate-derivatives.html

' . . . In September 1998, Long-Term Capital Management, a large and opaque hedge fund that made highly leveraged bets using OTC derivatives and other tools, found itself on the brink of collapse. It was saved only when the Federal Reserve orchestrated a $3.6 billion private-sector bailout of the firm. . . '
So, we might have three potential PODs.

In March '98 when Langewiesche's article comes out, just through good luck it receives almost saturation coverage for the better part of a week. Let's say it just catches the fancy of both Fox and CNN (or MSNBC and CNN) (or old-school CBS and ABC, etc)

In Sept. '98 and slightly after, this 'Long-Term Capital Management' bullshit receives much more coverage, in part perhaps due to some solid investigative reporting.

And then when Enron rolls around in Dec. '01 and WorldCom in '02, they receive the sustained coverage they should have gotten in the first place. And there's wide-ranging discussion of the whole concept of system accident.

and just maybe . . .

we avoid the '08 financial crisis.

---------

* And conservatives may have missed a genuine scandal within the Clinton administration, instead pissing away their best efforts on all this Whitewater stuff.
 
Last edited:
In the early '90s(?), some people were tickled by the feud between Stephen Jay Gould and Richard Dawkins on the mechanics of evolution. And the personalities involved were enough to motivate some people to take a look at 'punctuated equilibrium.'

So, people can take an interest in high falutin' theory.

Just don't water it down and talk down to people. Give the messy details. Give a case study and be prepared to give a couple of additional case studies. Give a big arc across the theory so people can see it all at once.

=======

Charles Perrow published Normal Accidents: Living with High Risk Technologies in 1984.

So, yeah, these ideas have been floating around for a while.
 
Last edited:
Top