In a fascinating thought piece, Joe Fuller argues reporting models are confusing the puzzle of a business model with unexplainable mysteries. He argues that a rethink of what we require for good governance and understanding firms are reset boundaries rather than simply increased regulation.

Here we go again. Over seven years since the collapse of Enron, we find ourselves in the midst of an even more widespread accounting debacle swirling around the mortgage and financial industries, the Madoff affair and even as far afield as India with Satyam.

In the aftermath of Enron we became subject to the rules of the Sarbanes-Oxley Act, which supposedly tightened accounting standards for all public company boards, management, and public accounting firms.

One element of Sarbanes-Oxley (SOX) was that it was supposed to enhance financial disclosure, including off-balance sheet transactions. Public companies have responded by publishing mammoth quarterly reports. For example, in the third quarter of 2007 UBS published a financial report that totalled more than 500 pages of data, and yet anyone reading that massive report would have had no inkling that the company was about to write off billions in losses that would jeopardise its future as a viable business entity. The same could be said for the financial reports of the late Lehman Brothers and numerous other companies caught up in the sub-prime securitised mortgage meltdown.

What went wrong? How could public companies provide all this data and yet still fail to give investors and regulators enough information to spot these enormous financial landmines before they blew up?

Playing detective

The answer may rest within a distinction that Dr. Gregory Treverton illuminated in his 2003 book, 'Reshaping National Intelligence for an Age of Information'. Like investors and analysts, intelligence officers try to understand the world by collecting and studying information and data. However, Treverton suggests that intelligence officers must recognise that they face two different types of uncertainty-based challenges: puzzles and mysteries.

Puzzles have straightforward answers. Collecting more data always helps you solve a puzzle, just as collecting more pieces of a real jigsaw puzzle gets you closer to assembling the picture. You know when you have solved a puzzle because all the pieces fit and the picture becomes completely clear.

Mysteries are different beasts entirely: sometimes adding more data to a mystery obscures the truth. In a murder mystery, such misleading bits of data are known as ‘red herrings'. Often mysteries do not have simple answers. Solving a mystery depends not simply on amassing data, but on developing an intelligent hypothesis and applying judgement to determine whether the hypothesis is correct. Mysteries contain inherent uncertainty. You have to live with the knowledge that your answer to the mystery may not be correct.

Legislators want to treat financial reporting as a puzzle. They believe that forcing companies to collect and report more data will reveal an indisputable answer, some immutable truth about the company's operations and financial health. However, the dynamism of capital markets, the originality of financial engineers and the complexity of the markets suggest that financial reports should be treated as mysteries. The financial markets need to exercise judgement over a company's operations, rather than seeking to reduce and simplify complex instruments that resist such simplification.

Assembling the clues

By thinking of Enron's operations as a puzzle, we assumed that Arthur Andersen should have been able to dig through the 3,000 special purpose entities (SPEs), each described by a 40-50 page single spaced document, and come up with a definitive answer about Enron's financial health. But what if Enron was not a puzzle, but a mystery? What if all the data fed to Arthur Andersen obscured the truth rather than revealing it? If this is the case, then all the subsequent rules compelling companies to provide even more data for analysis have only added fuel to a dangerous fire.

To examine the dilemma from another angle, consider the problem of Bonini's paradox, which deals with the problem of modelling complex systems. According to Bonini, the more completely and accurately the model describes the system, the more complicated the model itself becomes. The great Italian journalist and fiction writer Italo Calvino described this modelling conundrum using the analogy of the ultimate city map. In order to accurately capture all the detail of a city, a map would have to be as large and detailed as the city itself. Such a map would not only be impossible to create, it would be as useless to the traveller as no map at all.

Similarly, the usefulness of a financial report may decrease as its complexity grows. Which leads to the question: what information should a financial report really contain? What information do we need to make good decisions about a company's fiscal health? This much is clear: we need new ways to articulate the nature of risk in a business.

Solving the mystery

In late 2007, Merrill Lynch had approximately $60bn in assets and $5 trillion of credit exposure. Yet the sheer volume of transactions that Merrill Lynch had been reporting had obscured this hugely leveraged risk profile. No one had been able to cut through this mountain of data and dig deep enough to find out the details of the transactions.

In the cases of Merrill Lynch, Lehman Brothers and so many other fallen financial institutions, data only deepened the mystery rather than solving any puzzles. The instinctive response to the current credit crisis will be to push for more regulation, more reporting and yet more information, but this will only make matters worse.

Instead, we should think of new and innovative ways to make the mysteries of financial reporting less bewildering. How can we get a clear picture of the cumulative exposure that companies have against certain types of risks and a timeframe for when those risks will become clear? If you want to bet on the future of a British bank, it would be helpful to have a schedule with across the board, high level information on its exposure to X amount of sub-prime mortgages, Y amount of sub-prime loans and Z amount of foreign exchange with rough timeframes for risk exposure. Yet no schedule like this currently exists in any financial report.

We have outgrown the ways in which we measure risks. Now that financial markets operate more as mysteries than as puzzles, we need new integrated and synthetic methods of articulating risk positions. If the markets think that the mystery of valuing companies has grown too difficult to figure out, this virus of uncertainty could have devastating consequences. Indeed, it has already spelled the doom of Lehman Brothers and Bear Stearns.

Therefore, we must solve the mystery of valuation, not by adding more detail (which would be appropriate for a puzzle), but rather by stepping back and taking a high-level view. Our goal should not be absolute understanding of the markets or a particular company's transactions. Instead, we should seek new ways of articulating risk profiles so that people can exercise their judgement and apply heuristic methods to find plausible solutions to valuation mysteries.

Joe Fuller is the CEO of Monitor Group [1], a privately-owned global management consulting firm founded in 1983 by a group of six entrepreneurs with ties to the Harvard Business School