One complexity surrounding the meter reading folks is the expected delay in the information transmission. But essentially if a portfolio manager is making several millions of dollars in salary and bonuses while ordinary people... then there is something wrong with the incentive systems.

What first comes to mind while reading this article is Double-loop Organizational Learning. This topic is so widely discussed among academics and practitioners. The main issue here is whether it is possible to build up effective learning routines and, if yes, then how. I can see only one solution to this and it is leadership. In this context I very much agree with Bryan and Rumelt that motivation becomes a priority. We all need to think of the motivations which would make managers become true leaders who do the right things and not just do the things right.

You could also extend the principal-agent problem to politics: the agent being the elected representative of the principal, the public representing the common wellbeing.

The talented agent can easily deceive the naive principal (as does certain management with stockholders) to vote in favor of policies for which a negative outcome is only visible several years later when the politician is not in office anymore. Ironically (sic), the most spectacular negative impacts are just being found in this space, for example with the pay-as-you go ponzi-pension schemes. Since it is a multi-generational problem, the agents who sold the idea are not around anymore.

Anybody care to calculate the impact of this on the future evolution of the economy?

The economic meltdown in America was not too much unexpected and I agree with Lowell Bryan and Richard Rumelt that doing too much meter reading and engaging in predicting the future using trend analysis and modeling may take you away from reality. The world has become so virtualized that even a simple problem needs to be explained away and represented using complex mathematical or econometric models, thanks to the number-crunching abilities of the computers and software we have today.

The fertile imaginations of MBAs and PhDs have brought virtual reality, which has invaded the world markets—including Africa, a continent that contributes only 2% of world trade and more than 20% in value of tangible resources. In Nairobi Stock Exchange, for example, at the height of the folly of meter reading and predicting the future, you could find companies whose value in the stock market was ten times more than their total tangible assets and yet such companies never made any profit. It was common, too, to find companies that made good profit and speculators who did not bother to invest in them because they were more interested in virtuality.

The effects of the economic meltdown reached Africa late mainly because those to whom they sold tangible resources no longer have enough money to buy them with. The majority of companies which survive (and they are many) in Africa today did not deal in products of virtual reality such as derivatives and futures.

I would submit that just “reading the meters” is not the real issue here, it’s that we were reading the wrong meters. Highly Reliable Organization theory teaches us that most disasters are the result of a series of compounding failures to identify, comprehend, and act on a series of small but related occurrences. The Challenger and Columbia disasters, the recent bridge collapse in Minnesota, deaths on Everest, all of these, when teased apart, were preventable if we had identified and acted on the weak signals available to us. I suspect that an autoposy on the Hindenburg disaster would reveal the same chain. Query: Would applying such a a methodology help us to avoid financial disasters in the future?

I believe many of us would have been much happier if this kind of analysis and discussion took place a few years ago. After the fact, it is easy for many to jump on the bandwagon and say what we should have done before. To use the Hindenburg analogy, unfortunately, even management gurus and experts get seduced when the sailing is smooth and forget that lighting can strike any moment.

I have a particular resonance with the language that Richard is using—the idea that there are design flaws which get masked because of a focus on “meters” rather than on reality.

I propose, though, that it go a step deeper. I suspect, based on my own sampling of executives, that the skill set required to manage in a different fashion is lacking at the executive level and has been systematically driven out of the curricula and experience set of management school and businesses for the last 20 years.

The focus on “what gets measured gets managed” and the continuing premise of “individual accountability” for results versus an understanding of system design, and real analysis of data to evaluate system design, appears to have left a huge pool of executives that view “diving into the details” as looking at the next level of a pareto chart rather than understanding the nature of the system creating the issues.

At a very core level, I interpret what both Richard and Lowell are talking about, tests the notion of “professional managers”. People who do not have core competencies in the industries of their choice, but rather in something called “professional management”.

Fundamentally, I don’t believe, and I think at least Richard agrees, that we can manage something we don’t understand. For example, the idea of independence of risk is a simplifing approximation to make types of statistical models work. No one with even a basic understanding of business could possibly believe it were really true. In fact, within the math, we know it isn’t true. Yet, people making the assumption don’t know how “untrue” it is to give them a sense of how close or far away from reality the model is. With each generation of model the knowledge of that assumption gets further and further away, and the executives understand less and less of what their “meters” are really telling them.

Few business schools teach system dynamics. Few business schools teach how to use statistics well, beyond the hype of six sigma or the claims of the black swan. Engineering and the sciences for years have understood the idea of having a model that isn’t right but is better than nothing, and knowing how to test the difference between model and reality. Yet, this is tremendously foreign to other disciplines.

I agree with a great deal of what both Richard and Lowell have said. I guess I lack confidence that there is much of a pool of accepted talent today that is actually capable of driving a correction in leadership. Further, at the moment I am skeptical that the current generation of executives are being trained any differently than their predecessors.

Yet, the knowledge exists. Tools to help in a different model of leadership exist. Perhaps if companies of this sort truly commit to a different approach a rapid change in skill sets at the top of such companies can occur, but the faces will look entirely different I think.

If the Fed, the rating agencies, and the banks had asked what was actually happening, they would have found that what they were doing was violating the pizza theory. Securitization offers the opportunity to transform below-investment-grade assets (the investment or collateral pool) into triple-A and investment-grade liabilities. The equity holders of an asset-backed trust (ABS) would only perform securitization if the process generated a positive net present value. This could occur only if the tranches were mispriced. The investment banks repackaging cash flows, the investors buying tranches and the rating agencies had all but forgotten Corp. Fin. 101 and the Modigliani Miller theorem. Merton H. Miller had a simple explanation for the first Modigliani-Miller proposition. It’s after the ball game, and the pizza man comes up to Yogi Berra and he says, ‘Yogi, how do you want me to cut this pizza, into quarters?’ Yogi says, ‘No, cut it into eight pieces, I’m feeling hungry tonight.’ Now when I tell that story the usual reaction is, ‘And you mean to say that they gave you a [Nobel] prize for that?’

The pool of subprime mortgage bonds is like a pizza. The way the tranching is done between high-grade CDOs or mezzanine CDOs is irrelevant. If the CDO trusts could finance their purchase by issuing triple-A rated CDO bonds paying lower yields, the pizza theory states that they would have to pay much higher yield on the lower rated CDO bonds. The CDO equity holders should not generate a positive net present value investment from just tranching the pizza.

This article would be very funny, were it not for the sad hit our economy has taken, and people have lost jobs.

Over the last 28 years I have asked why the folks in financial services got such outsize rewards, compared to say a engineer, or a doctor, who provides greater, tangible, societal good. And the answer I got was that in America, you need to control how capital is deployed. For this reason, up until last year, hordes of smart engineers, physicists, and others went to Wall Street, all with the idea of making tons of money by moving money around.

Now the same folks want to know what they can do in a manufacturing company. And the answer is ‘not much’.

To learn how to design and manufacture cost-effective products that customers like takes time. To build a successful business takes even more time. I tell recent MBA’s to forget their Powerpoints, and begin to get their hands dirty. There are no companies where people will throw scads of money at you, as Wall Street did. And stop thinking that an Ivy league (or other) MBA, makes you an expert in building a business, or knowing an industry, in a couple of years. Take the German and Japanese auto companies, for example, where the CEO’s often have an engineering background, unlike at GM!

Interesting discussion about design flaws in the system, and of the ubiquity of what is a patently self defeating incentive system due to the principal-agency issue. There seems to be an obvious fallacy at work here—when joint stock ownership of private assets forces the individual owners to rely on the judgement of equity analysts or fund managers who take home enormous bonuses during the boom, and find themselves out of work during recessions; when CEOs have their multi-million dollar call options linked to the price calculations thrown up by excel spreadsheets developed by the same analysts, when bank managers are paid or promoted for selling more loans than the other bank down the road thus risking our deposits, and most of all, when most of their rewards are quarterly or annual, even as the decisions they take will either reward or penalize their trusting investors in the ‘long, long, run.’ Boom and bust cycles are inevitable in this construct, both are built into the design.

The only solution I can see is to somehow design each of these reward and deterrence systems around a different axiom which does not reward or penalize anyone instantly. I can only think of a kind of ‘Reward / Penalty Pension Fund’ which builds up or draws down as the outcomes of their investment or management decisions become clear over time. A fund manager, for example, would be rewarded with additions to his ‘bonus pension fund’, or penalized with subtractions depending on the performance of his portfolio over the past five years. A CEO would build up his bonus fund if the company performs better, or draw it down if the company performs worse. This way, their decisions taken two years ago can come back to haunt them three, or even five years later. They have more skin in the long term game, a sort of surrogate ownership of the company by the CEO / CXO’s.

ESOPs are designed to act somewhat in this manner. Only, there must be no call options, the employee sells his options partly when he leaves, and partly a fixed interval after leaving the company, say two or three or even five years. I wonder if something like this can work, or would it contradict any fundamental laws protecting individual liberties in the major countries of the world?

This links to conversations we have been having recently about the need for a new role to become common in the corporate hierarchy: the “shareholders’ representative.”

While, in theory, that is the CEO, we have seen that in practice due to incentive structures, corporate cultures, and individual psychology the CEO is unlikely to be acting entirely on the shareholders’ behalf - especially if the “shareholders” are defined as those who are making long term investment into the company.

The shareholder’s representative or SR (which could be an individual, or a body) would act almost like an upper house—but in a more engaged and day-to-day way than a traditional Board.

The executive team are offered incentives (as Rumelt and Bryan point out) to share in upside, with very limited real downside risk. The shareholders’ representative must ensure adequate consideration of the downside risk carried by investors, as well as assist the executive team in finding ways to maximize sustainable returns on that investment.

SRs would essentially become a live and interactive incarnation of the ‘shareholder’s agreement,’ facilitating, shaping, communicating and enforcing (for instance, via veto) whatever principles, vision, and priorities the shareholders want to impose.

This would become a significant new force in every aspect of the company’s executive processes, and would remind CEOs and their teams of the (inconvenient) truth that for all their power and prestige, they have actually been hired to do a job, by and for somebody else.

My experience as researcher and professor is that managers do not perceive design flaws because of two reasons: they are part of the design and they are not designers, but riders. When the design provides a smooth ride, managers will ride at high speed without considering how they are interacting with their environment and creating the conditions for the subsequent crash. There are important flaws in the education of managers: too much focus on riding and less on designing. However, if managers learn to be designers, there will be a second issue to address: values. Since the knowledge of the design can be used for good or bad purposes, it is important to avoid free riding. Can we avoid free riding through regulations? Maybe, but there is always the intention to exploit their holes. Now, there are clear indications that the consequences of free riding are appearing sooner than later: financial crisis, global climate change, social tensions, and so on. Therefore, it is important to teach how to self-regulate if we want to achieve a design that can withstand future turbulence.

Risk is defined as an outcome which is not expected. This has 2 sides: the loss side and the profit side (lottery winnings). The loss side has been researched extensively. However the unexpected profit side has not been researched enough. This is the design flaw, since the profit has resulted due to circumstances that were not envisaged, therefore not managed. Why rock the boat has been the dominant principle, avoiding the examination. The parameters that went in favor have, consequently, not been controlled by the decision nor researched to improve decision making. The outcome could have gone the other way too, leading to loss in that eventuality. Hence the risk profiling has been incomplete.

You’re talking about ‘people reading meters rather than reading reality’, but nowadays, with all sorts of market anomalies around us, what is reality and what is not?

I whole heartedly agree with Lowell and Richard. Agency theory is the key.

Another factor not mentioned is that now most investors are institutional and looking for short term returns. Large companies become obsessed with meeting the expectations of these investors for fear of them leaving. This makes structural change difficult.

Another point is that executives and investors have become addicted to high, but unsustainable returns. Most pragmatic investors in 2007 knew the bubble would burst, but didn’t know how to get out. Institutional investing in a bubble is like riding a tiger, you know you have to get off, but will probably get eaten alive when you do!

The expressed opinions of Messrs Bryan and Rumelt are valuable inputs to strategic planning, since very often planning is done with reference to metrics obtained as a result of past enterprise performance. The more appropriate way of going through the process is to look at the system not only basing our premises on external factors and tangibles, but more importantly on internal and intangible factors which could bring about higher risk and disastrous outcomes.