"For the markets to work as we once might have hoped, information and power need to be much more equally distributed than is possible in the real world ... Reality is often capriciously non-linear." Marc Gerstein, "Flirting With Disaster"

    Flirting With Disaster: Why Accidents Are Rarely Accidental
By Marc Gerstein with Michael Ellsberg
Union Square Press, New York, 2008

    Have you ever failed to speak up in a business meeting when it was perfectly clear the decisions being made were not optimal? Not protested a minor injustice to which you were witness in public - a parent being abusive with a child, other people's children misbehaving - because you didn't want to make waves? Ignored a smell, a crack, a leak that later proved to portend the need for a new boiler, windshield, roof? This is the kind of incident Marc Gerstein's book (written with Michael Ellsberg, son of Daniel - who himself provides the Foreward and Afterward) describes on a massively larger scale. Yet one key element common to catastrophes is present in all these situations: bystander behavior by people who ignore or suppress the signals of an event (sometimes even when their own personal safety is involved). And if you've ever resented the woman who spoke up in that business meeting, the loudmouth who interferes in how parents treat their own children or tells children not his own they must behave at the movies, or your spouse who made a big deal about the smell, crack or leak, you've also experienced the other key social emotion often present in the lead-up to catastrophe: hostility to the whistleblower.

    This little book offers a framework from which one may read much of the news of the day - as well as closer-to-home situations. Dr. Gerstein uses a few crucial case studies: the Challenger Space Shuttle, Chernobyl, Vioxx, BP's Texas City Refinery fire, the 1994 US Air Force friendly-fire incident that destroyed two US Army Blackhawk helicopters, the collapse of Easter Island's civilization, the fall of Arthur Andersen, the 1994 collapse of the Mexican peso and the cascade of late 1990's financial failures and country bankruptcies, including Long Term Capital Management's demise and some crucial research into biases and failures in human perception, as well as game, chaos and systems design theories to illustrate and support his premise that people and organizations are inherently biased to take on more risk than they would objectively consider acceptable, including, most spectacularly, catastrophic risks.

    The Wall Street Journal's review of the book, "Free to Choose, But Often Wrong," rather predictably focuses on the flaws in individual rationality, without mentioning how these alone make a mockery of some of the tenets of market fundamentalism, as market efficiency theories totally rely on the assumptions of perfect information and rational satisfaction-maximizing decisions that Dr. Gerstein pretty well demolishes.

    This book is as essential as it is troubling: one reading really does suggest that humanity is doomed, given the disasters we are living through - the present economic meltdown and the planetary crisis due to greenhouse gas emissions - and the degree to which we humans are apparently hard-wired to misperceive, to bystand and to choose leaders who are of the go-along-to-get along ilk, but that is certainly not Dr. Gerstein's own outlook. After some weeks of email correspondence, I finally spoke with Dr. Gerstein on Sunday, January 11. Although he is extraordinarily well-informed and a hard-nosed realist about the situations obtaining in the world today, he impresses by the optimism he is able to maintain.

    Leslie Thatcher for Truthout: What was the primary motivation behind "Flirting with Disaster" and what do you hope the book will accomplish?

    Dr. Marc Gerstein, author of "Flirting with Disaster:" There are two parts to my decision to write the book; a triggering event and my hopes.

Dr. Marc Gerstein, author of "Flirting With Disaster."

    In the book, I attempt to summarize what I and others have learned, along with my sort of special view as someone who studies organizational phenomena and the distortions that arise as a result of individual cognitive failures. I conclude that many solutions to problems of risk need to be organizational in character, whether those organizations are businesses or governments, but that these solutions will be very difficult to implement precisely because of the human biases that create the problems in the first place.

    My objective is to influence policymakers - that is, those decision-makers who are in a position to change outcomes. Recent events make it very clear - and with an emotional potency these issues did not command even a few years ago - how possible, if not likely, it is in the complex world we now inhabit to stumble into disasters so extensive and so expensive that virtually any amount of prevention is worth the costs.

    In the book, I demonstrate how this was true using simple financial calculations with Hurricane Katrina. Today, in the current financial crisis, we can imagine that if we asked most people how much they would have been willing to pay to prevent the crisis, I think we'd find that they would certainly have been willing to come up with the money rather than lose their jobs, their houses, and a good deal of their life savings (at least until the markets recover).

    One of the mysteries this raises for all of us is why disaster prevention looks like such a good thing after the fact, yet there is something in our mental risk assessment process that makes us fail to take the necessary preventative steps. That's the key question that drives the book, just as it is a decision about any form of insurance payment.

    The book you suggested I read, Bob Altemeyer's "The Authoritarians" provides a very important perspective for understanding why this is so difficult. The thinking of individuals Altemeyer calls high RWAs (Right Wing Authoritarians), particularly their dogmatism, gets in the way of doing the right thing in advance. My sense is that the thinking of certain people - mostly, but not exclusively, on the political right - is that their way of looking at the world is far too easily reinforced by like-minded thinkers. If others say things they agree with, they will suspend analytical logic because the mantra supports their view of the world. In many of the issues that we face, those views (i.e., the dangers of "big government," faith in unrestrained market "freedom," an extreme view of "individual responsibility," etc.) are based upon provably limited premises, but that doesn't win the day with people who are closed-minded about proof. Distorted thinking is very dangerous in matters of risk, and Altemeyer also makes the point that such people are also unusually vulnerable to political exploitation and manipulation when their biases are pandered to.

    The other great barrier to getting people to understand the risks we face is that many of the policymakers have little or no technical education or understanding of math. I'm not talking about higher-order calculus, simply fundamental statistics and finance of the kind that I use in the book to reveal the true nature of the foolish risks we are sometimes willing to take.

    Could you pick one episode from the news of the last few weeks and show how it conforms to the analytical framework you set out In "Flirting with Disaster?"

    Last week, I watched the House hearings on the Madoff scandal (incidentally I've found, since I started to watch hearings and read original reports, that you often get a different view from original material than you get from the 800-word newspaper summaries). On this important House committee, when asked if they understood Madoff's self-described "split-strike" strategy for running his fund, nobody raised their hand. Had it been me, I think I would have done the homework to understand such a key set of facts before the hearing.

    Of course, most of these people are lawyers, and they may not know options theory, but they are more than bright enough to understand the material if they made the effort. However, it would appear that having to understand Madoff's split-strike strategy wasn't to be a part of the conversation this day, but it is nevertheless important because it lies at the root of Madoff's fraud and the SEC's failure. When you read the actual complaints against Madoff, it becomes clear that it would have been easy to verify within a few days whether he could have made the money he said he made in the way he said he made it. Since there were complaints over many years starting in the late '90s, it became clear to me with just a little research that the SEC didn't discover the fraud because they never really looked at the details. Had they done so, billions would have been saved.

    Beyond Madoff, one can examine a wide variety of incidents, wondering why people didn't ever look at the facts. For instance, data generated by the FDA's scientists about Vioxx showed that the death rates from cardiac events were abnormally high. Why didn't the FDA go after it? Today, the urgent policy question needs to focus on what is wrong with that organization that caused vital warnings to be ignored? And more generally, what are the organizational dysfunctions among regulators that cause serious concerns to be ignored or even deliberately suppressed?

    This is not a theoretical question. As we head into a new administration, we're going to have to make some very important decisions about vital issues. I question whether policymakers are in a position to make objective decisions and to put in place the procedures to verify that people are doing what they are supposed to - that is, whether they are complying with the existing rules designed to protect us. This is an issue of more compliance, not necessarily more regulation.

    In the case of Madoff, there was no need for additional rules to have revealed his fraud. What was lacking was verification that he was doing what he claimed in his monthly statements to investors in light of a number of credible complaints that Madoff's fund was a fraud.

    What role did the SEC play?

    A big role, I believe. An important issue raised by the Madoff debacle is the structure and role of the SEC, which has taken the position for a long time that it didn't want to be a law enforcement agency. The whole issue of the role of the SEC and its relationship to its regulated firms needs to be examined. How tough or gentle should the SEC be on enforcement matters? Clearly, a prime motivation that drives not only individuals in the SEC, but also bureaucrats in other regulatory agencies, is the lucrative opportunities for jobs in industry that awaits them when they leave. The revolving door is one way government gets a discount on its labor costs: talented people are willing to work for less so they can get their ticket punched and leverage their knowledge and relationships in industry.

    What level of ability and training is required to deal with a complicated industry competently? I think it is obvious that there needs to be parity between the regulators and the regulated, and this would usually lead to roughly equal pay. But right now there's an arbitrage between these two labor markets in which people parlay their low-paying but valuable government experience today into a highly paid job tomorrow. Under such conditions, while you work for a regulator, what motivation could there be to alienate those in the private sector? You would want to do your job in such a way that it doesn't cause potential employers to question whether you'd "fit in" working for them. If you're in the FDA, making it tough for Merck or Pfizer in a way they find objectionable lessens the chances of getting a big job, so it's much more "profitable" to go after the little guys and minor infractions and give the big organizations a chance to clean up their own acts on their own schedule. I think it's pretty clear what the outcome of this arrangement has been, as the recent story about medical devices underscores.

    If there is truly a role for regulation - and I think one can make a pretty good argument that self-policing is self-limiting (consider Arthur Andersen and Enron, the financial incentives that bought us the rating agencies' failures in the sub-prime crisis, and the consequences at the FDA arising from a change in its funding and politics that defined drug companies as the FDA's "clients") - in order to have independent examination, we need independent agents who are highly trained so they can't be bamboozled. In the case of malfeasance, in particular, a lot of creative effort will go into cover-ups, so you need to have the same intellectual horsepower on the other side, plus subpoena power and relative freedom from political influence.

    The other thing you need is protection for whistleblowers, because one of the things that is an absolute organizational truth, but hardly ever discussed, is that whenever there is a widespread dangerous situation or fraud, many, many people on the inside know about it. We're seeing that in the Madoff case: lots of people are saying "he couldn't have done that alone." Why is it so important to Madoff to claim he did this himself? Under securities law, when you know about certain infractions, you may well be obligated to report them. Participation in fraud and staying silent about such a fraud are separate crimes, and not all organizational bystanders to securities infractions are necessarily "innocent."

    At the moment, there's a discussion about whether Bernie Madoff's brother Peter was in violation of the law by not raising his knowledge of the fraud (after his brother's confession) to the appropriate authorities 24 hours before he did. But my guess is that a lot more people knew what was going on. They always do.

    I know from my research and personal experience that there is also fairly widespread denial and a collusion of silence when bystanders are caught in these awful situations. Denial notwithstanding, as we know from Enron and other cases, it is impossible for anything on a large scale to go on without enormous numbers of people knowing what's happening. I've started work on a new book on bystander behavior that I hope will reveal just how widespread this is and the frightful dilemmas that organizational bystanders face as they consider what to do.

    Since most people keep silent, rationalizing their inaction even in the face of grievous harm, known risks and even crimes continue for months and years before they ever erupt into the outside world. For their part, hierarchical organizations often punish people for raising unpopular truths because they want things to stay quiet while they figure out what to do. Sometimes, they also want to continue as before, especially if big profits or criminal behavior are involved. While I'm not saying that many organizations don't seek to unearth wrongdoing, it's often a lopsided contest, especially when the institution and its leaders will be embarrassed ... or worse.

    A good case is the Office of Special Counsel in the federal government that is supposed to protect whistleblowers. It is arguably one of the most morally corrupt agencies in the government, a horrible fox-guarding-the-chicken-coop story that has been widely covered. There is considerable outside pressure to clean up the weaknesses in this important investigatory support of the Whistleblower Protection Act, but little has happened. Perhaps things will change after January 20.

    My view of all of these problems is that we can create different kinds of organizations because most people - particularly Americans - really want to do the right thing, but we need good leadership and serious consequences for those who violate the rules.

    If you could totally overhaul our social organization to engineer for large catastrophe avoidance and improve our ultimate survival chances as a species, what would you do and how would you do it?

    I say in the book that willing leaders are necessary, but not sufficient. In his afterword, Daniel Ellsberg writes about when the leaders are the problem, and I urge people to read what he says. From a societal and governmental perspective, once we've seen the consequences, and if we're serious and have identified the perpetrators, we have to go after them in a way that is different from how we typically have dealt with white-collar violators in the past. Since many are motivated by money, we need to figure out how to take away their money.

    In the last hour of the Madoff hearing, Allan Goldstein, a 76-year-old man of humble beginnings who had nevertheless managed to start a modest business that allowed him to save and then invest about $2 million with Madoff Securities over 21 years, testified. Goldstein successfully put a human face on the scandal, for it is now clear that many successful, but otherwise ordinary, people - not just very rich people - were defrauded, many losing everything in the pursuit of the steady security Madoff's fund promised.

    Tamar Frankel, professor of law from BU, and my personal hero of this hearing, said she didn't really care very much whether Madoff went to jail, but it was very important that they get ALL his money and return it all to the people he had stolen it from. That would be worse for him than going to jail. When asked about whether he should be in jail now, she said that she didn't want to pay to feed and house him.

    In the sub-prime crisis that triggered our economic woes, the real culprits are those who knowingly made large numbers of questionable loans to people who couldn't afford to pay them back if the housing market stumbled. They knew what they were doing when they booked the so-called NINJA loans - but, unfortunately, many of them are likely to get away with it. If we don't find a way to get the money back from them or at least others like them, we're never going to be able to prevent people from creating the next bubble, the next scam.

    I think that it is obvious that if someone can take personal advantage of a short-term market anomaly and pass on the costs to other people, there will be a very powerful incentive to do so. We must figure out how to align the interests of the people doing the deals with the interests of the people they are ultimately working for by making fees and bonuses conditional. The AIG executives who exacerbated risks by selling undercapitalized credit default swaps will keep their individual rewards, while society as a whole bears the costs. I think we deserve better.

    In organizational terms, we know as a dead certainty that creating dysfunctional reward systems leads to inappropriate risk-taking. In the Savings and Loan crisis, many of the people who perpetrated the risky investments and asset mismatch imbalances that eventually bankrupted the industry personally profited. They got their own money out every year.

    Does this never lead you to question the absolute and relative levels of remuneration some people enjoy?

    Absolutely, but there are two levels of looking at the multiple between the top and the bottom of economic compensation. I don't think there should be caps on the earnings of people who create wealth by creating new technologies and new businesses. However, while acts of wealth creation should not be bounded, the bonuses paid to others slipstreaming on market trends are more questionable. During the sub-prime crisis, former bartenders made fortunes selling mortgages. At this point, I think that most would agree that this makes a mockery of the American Dream.

    Therefore, the part I object to is less the level of compensation (although that creates significant distortions in the community which bother me a great deal, because they affect rents, property values, restaurant prices ...), but, even leaving that aside, the issue I see is that the interests of the people who are earning these big bonuses are not aligned with one of the key groups financial intermediaries are ultimately supposed to serve: investors.

    Dirty stories inside Wall Street abound about the selling of CDOs (collateralized debt obligations), particularly the role that incentives played in downplaying risk with investors. By exploiting investors for their own gain, intermediaries can only blame themselves for bringing down the wrath of the public and Congress. Tighter supervision would have a chilling - but, I believe, salutary - effect on the industry.

    While I don't think it's easy to legislate the level of compensation, I do think it should be possible to tie compensation in some manner to the fate of the people holding the risk. Today, it's as if there is no concept of malpractice or product liability in the financial industry. Financial advisers should be responsible for the quality of their advice and their products, although obviously not for unavoidable market risk.

    Furthermore, it should be clear from recent events that investment has become just as complicated, if not more so, as medicine, law, and accounting, professions that are bound by much tighter codes of professional conduct and liability. Perhaps we should subject financial advice to the same standards.

    Going back to Madoff, there is one additional point to make. The reason so many advisers believed in what he was selling was that they believed that Madoff was cheating. It was just that he was cheating for them by using information from his market-making firm to "front run" the market. In other words, Madoff was violating the law, but was doing so in the interests of his clients, and it was other nameless investors who were ultimately losing out. The suspected con hid the real con, so on a moral basis some of the victims probably deserve what they got. Many others, however, were doubtless innocent, and true victims.

    Okay, let's take this "up" a level. I happened to read a Keynes essay last night in which he was already decrying increased trade after the First World War as a risk multiplier. Doesn't all the interconnectedness you've talked about make the system riskier?

    You're now onto what I think is the central philosophical and design issue that underlies the risks of globalization. When there is lots of interdependence, systems evolve to a different level of risk than they possess when less interconnected. Consequently, the kinds of shocks that were probably localized in simpler times tend to propagate, and may well be amplified by feedback loops. In such an environment, we need to be a lot more careful about making mistakes, and we must also be far more careful to avoid systemic design mistakes, such as neutering the regulatory early warning system, if we are to avoid being overtaken by events.

    One of the other things that strikes me is that over the last fifteen to twenty years we've seen not only the growth in interconnectedness, but greatly increased concentration of institutions - in itself more dangerous because of the consequences of failure - along with laissez faire compliance based upon the assumption that markets will self-correct. The system is now vulnerable to certain kinds of contagion that are not easy to prevent - like an outbreak of avian flu that can spread through jet travel.

    The current economic crisis might well be the first shock of this type that our truly globalized society has had to face, and it is arguable that there are no existing tools in the tool chest that work. Monetary policy has been exhausted, and nobody has ever done spending on the scale being proposed. While monetary policy is relatively easy to implement, spending is a massive organizational problem involving people, material goods and information. We know how much waste there's been in Iraq and Afghanistan, and yet the military is actually better organized to spend large amounts of money quickly than the states and the private sector, so I'm very concerned it's going to be hard to keep the wheels connected to the ground. Unfortunately, if people try to move too fast, the loss of traction will create waste on an unprecedented scale. This raises the risk of serious distortions to the real economy and far more pain than gain. To come back to the importance of avoiding disasters of any size - even if those disasters come from trying to implement solutions to a prior crisis - it's really important to pay attention to the "weak signals" of danger and to non-defensively and objectively evaluate the potential problems if things do go wrong.

    As with all risk, the question we have to ask ourselves is whether we are willing to forgo a certain amount of return so that there will be the resources to pick up the pieces in the event that low-probability events transpire. Prudent people try to calculate their risks out to 99 percent probability, but then tend to truncate the remaining risk as if it did not exist. That may be a sensible way to deal with the possibility of an asteroid wiping out life on earth (which has a tiny but finite probability), but when it comes to more mundane risks, the chances that they will occur at some point are significant. So the question is, who's going to pick up the tab?

    For people who live on a four-year cycle, such as many people in our government, or on a five-year planning horizon like many CEOs, 50- or 100-year floods don't seem to be something they have to worry about. But society does need to worry, even when no one alive can remember the last flood. When people make decisions that affect many others, but they escape from the responsibility for these decisions, the likelihood of events going badly rises dramatically.

    In the discussion in the House committee room, Leon Metzger, another of the Madoff witnesses, explained how easy it is in finance to create misleading returns by ignoring the long-term risk. Selling stock options on a stock price's doubling within a year was the example given, and this is a good metaphor as well as a finance example. Since doubling in a year is very unlikely, you can book profit from the option premium for every year it doesn't happen. But when the unlikely event does occur, you are likely to lose everything you've "earned" and a great deal more. But up until the point of loss everything looks really good, so people may be tempted to apply intuition, figuring that the future will be like the past. This seems a silly example until you realize that this is what happened with sub-prime mortgages and many credit default swaps sold by AIG.

    Let's get specific. If you could redesign the system, what would you do?

    This is very complex in a practical sense, so my answers will take the form of some of the questions we should be answering, rather than specific formulas we need to follow. One thing we can certainly do is to set out some design objectives to be clear about what we're trying to accomplish, as well as clarify some design criteria to evaluate proposed ideas because any solution stated in the abstract will be imperfect. This is a very demanding process, and unfortunately one that does not fit well within the political process. This said, here are a couple of starter thoughts:

What would it mean to make policy-makers and various advisers truly responsible and accountable for long-term consequences and low-probability risks?

Should we be able to more easily sue financial advisers for "malpractice" when they fail in basic due diligence, and should the advisers be obligated to take out insurance or otherwise prove that they are adequately capitalized to fulfill their obligations?

Should the currently invisible aspects of the financial markets, such as derivatives, be more transparent and well-regulated?

How can we staff the government with high-quality specialists who can, on the one hand, look forward to having an impact on industry without, on the other hand, being tempted to curry favor?

What would it mean for our society to seriously protect whistleblowers?

    It's my hope that the combination of my book and recent events might move decision-makers to entertain some solutions to the risks we face as a society that they might not have been willing to previously consider.

    This ideological and moral quest frankly tests my abilities as an organizational designer and agent of change. Ironically, of course, if things get better we will only see the results as an absence of disasters, and since such things occur infrequently in any case, it will be a challenge to measure our progress. Still, it is a worthy undertaking.

    But then isn't it inevitable that over time people will forget why those structures are in place and the pendulum will swing back to more risky behavior?

    Yes, that is a definite risk. Unfortunately, it's necessary to protect ourselves from ourselves when it comes to risk. We are all subject to failures in perception and prone to unacceptably risky behavior: it's part of what it means to be human. Nevertheless, whenever we give in to those tendencies, we deprive ourselves of the protections we need. We should not allow ourselves to make decisions that are not in our long-term interests, and ensuring that this happens is one of the most important roles of government. Once we realize there are human and organizational limitations we need to take into account, we must put a regime in place that protects us from the temptations arising from these sources. Unfortunately, life doesn't come with a warning label. But even if it did, we'd probably ignore it anyway. As I said, this is a difficult problem, although that is no reason to ignore it.

    While I was reading your book, I kept thinking of how its principles apply to genetically modified foodstuffs and that industry. One of the things I was most shocked to learn when I was reading about that sector is how peer-reviewed articles in reputable scientific journals have been corrupted.

    Since you've brought up this issue, I have a comment. Think about the assumptions most professionals make about how articles get into scientific journals, and what it means for the corruption of the scientific process when there are no controls for people putting their names on articles they didn't write about research they didn't conduct. These articles become unlabeled infomercials - NOT science. And if we want to think about rule-making, we didn't even know we had to worry about what was going on in scientific and medical publishing.

    And it's not just science. One of my readers with whom I've been corresponding is a naval architect who has introduced me to the issues surrounding marine safety. And if you think from recent events that the financial sector has been irresponsible, some of the stories about marine safety, such as the wreck of Erika - which is Europe's equivalent of the Exxon Valdez - and the more recent sinking of the Princess of the Stars ferry in the Philippines will make your hair stand on end.

    I am hoping that the book will help motivate a broader community of people to care about these issues, as well as influence decision-makers to act on them. Often, it takes a crisis to prompt people. Now that we have all "paid the tuition," we should make sure that we truly learn the lesson.