New POSIWID Materials

POSIWID lens (edited by Richard Veryard)
POSIWID wiki (edited by Amicus Rex)

Shareholder Democracy

According to John Plender, the biggest curiosity of US corporate governance is that shareholders' votes on directors' appointments at annual meetings have no force. Shareholders can "withhold" their votes, but even if more than 50 per cent do so the gesture is purely symbolic and the company can continue on the management's chosen path. Plender comments: "This shareholder democracy is about as democratic as Cuba or the old Soviet Union."

[Financial Times, November 13th 2004]

Meanwhile, Manchester United management is outraged that Malcolm Glazier should have used his substantial minority shareholding this week to vote against the appointment of directors. Apparently this is bad form; complaints have been made to various regulators. Glazier's intentions towards the football club are currently unclear - but does this mean he should be denied a vote?

Many people seem unable to distinguish the viability of a company from the perpetuation of its current management. Shareholder democracy in Europe is far from being an adequate control mechanism - it lacks power as well as requisite variety. However, many US stakeholders are strongly resistant to even this weak control mechanism; meanwhile News International's corporate move to Delaware is widely seen as an escape from the stricter corporate governance environment in Australia.

One of the preconditions for corporate governance is transparency, and systems engineering can certainly do something about this. But transparency is not enough. We need to engineer appropriate control mechanisms, so that transparent information can be properly acted upon in the interests of the real stakeholders.

Commitment and Self-Interest

Standard investment advice is to diversify. Among other things, this means that it's not a good idea to hold significant quantities of shares in your own company, since you are doubly exposed if the company fails. (Think Enron.)

Meanwhile, it is widely supposed that the company's interests are served if the directors and employees have an investment stake in the company as well as an employment stake. This is supposed to align the personal interests of the directors and employees with the interests of the company. There is personal commitment to the success of the company, with an inflated cost of exit.

Furthermore, by having large personal shareholdings, company directors demonstrate their confidence in the company's present state and future prospects, and their belief that the shareprice undervalues the true worth of the company.

There is therefore a structural conflict of interest between company (external shareholders) and employees (especially directors). How is this resolved?

Either the individual directors take an irrational stance in respect of their personal investments, accepting an unbalanced portfolio with a sub-optimal risk/reward ratio. However, we should not expect true alignment between the interests of a director with an unbalanced investment portfolio, and the majority of shareholders whose investments are (of course) properly balanced and diversified.

Or the directors cheat. For example, holding derivatives that hedge against the excess exposure to the failure of the company. For example, manipulating information. And as a privileged class, the directors hedge against failure by awarding themselves massive termination payments. (While not illegal, this is a morally corrupt practice.)

According to this argument, directors are driven by the system towards either madness or badness. (Some manage both at once.) The answer is not to recruit a new cadre of morally upright and selfless leaders, but to change the system.

AIIM Survey

Emperor's New Clothes AIIM Survey Shows Organizations Face Internal and External Barriers in Adapting to New Compliance Environment

Press Release
Report in Business Intelligence Pipeline
Download PDF

Excuses

Once upon a time, the standard excuse for bureaucratic incompetence was the computer. Sorry sir, we’d love to help you but the computer won’t let us, the computer is down, the computer is so scary you wouldn’t want to upset it now would you?

This excuse isn’t used so much nowadays. This could mean that computer systems have gotten better, or it could simply mean there are now too many computer literate people in the general population for this excuse to wash.

In the financial services sector, the new excuse is compliance. Sorry sir, you can’t do that here. This morning, a stock broker told me I cannot open an investment account for my children, because of compliance. Compliance with what, I asked. Of course, he didn’t know. I rang another stock broker, who had no problem with my request.

Compliance Machine

Trevor has drawn our attention to the recent move by Barclays Bank, which uses a product called Raft Radar from Raft International.

Does Raft Radar count as a compliance machine? How does the Raft solution compare with Trevor's?

404 is the key

SEC Press Release http://www.sec.gov/news/press/2004-21.htm

Some internet sources suggest an alternative explanation for the delay - that there is some confusion about the role of auditors. Now where have we heard that before?!?

Chris Johnson

Bill, I assume you're talking about the Chris Johnson who is head of the Glasgow Accident Analysis Group.

There is a lot of impressive-looking Safety Critical material here, but how do you see this could transfer to the Corporate Governance domain?

Compliance Machine (Functional Requirements and Quality Constraints)

Trevor has usefully stated some of the general quality constraints for the compliance machine. Easy installation and configuration, low/remote management, non-intrusive, and so on.

What we now need is a clear summary of what the machine will actually do. Trevor proposes to build one and show us. In the meantime, perhaps the rest of us can discuss what we are hoping that Trevor will be able to demonstrate. If it's a proof of concept, what exactly is the concept?

I'm particularly keen to find some organizations to pilot Trevor's machine, and the better we understand what a general compliance machine could/should do, the easier it will be to present the pilot properly.

Compliance Machine

I like Trevor's concept of a compliance machine. This could presumably be deployed in a range of management contexts, including second-party audit (monitoring your trading partners) and third-party (new style independent audit).

What are the general requirements for such a compliance machine? Can we articulate a small list of high-level design principles?

Rebalancing Power Proximity and Interest

Agency theory says that the agent (e.g, the management) has many opportunities to behave in ways that are against the interests of the principal (e.g the shareholder). This can be explained by an uneven distribution of power (the agent has too much) and proximity (the principal has too little).

Technology potentially provides greater proximity for those with a legitimate interest. It brings stakeholders closer to what is going on. This proximity may be measured in terms of data accuracy, granularity (finer detail) and currency (more up-to-date).

Some commentators believe that companies should release financial information to investors more frequently. (For example, monthly reporting rather than quarterly). Continuous information reduces surprise, reduces risk, therefore increases the investment return for the investor and reduces the cost of capital for the company. Ultimately, such published figures are driven by the company’s internal data, and could in principle be released in near real-time, subject to appropriate (automated) policies for filtering and delay.

Collaboration

Prompted by Bill's message, I had a quick read of the OECD Principles of Corporate Governance, Draft Revised Text, Jan 2004.

“Shareholders … should be allowed to consult with each other on issues concerning their basic shareholder rights … subject to some possible exceptions regarding inappropriate collusion.”

In other words, collaboration is allowed for some purposes and not for others. In terms of system governance, this is an extremely interesting rule. It calls for discrimination by purpose, which only works if we have some way of objectively determining purpose (other than naively asking people what their purposes are).

Is it possible to produce a formal monitoring system that is sensitive to purpose? Can we automate POSIWID?