Enron and WorldCom have made governance a topic the public actually cares about, which is to say the most important conversations are now possible to have in public for the first time in a long while. This letter is about the slower, less visible work of trust.
On the Year of Confessions
The most consequential financial fact of 2002 is not the bankruptcies themselves but the restatements that preceded them. Auditors, regulators, and analysts were given, in a single year, dozens of cases in which the official numbers turned out to be a fiction maintained by people we were supposed to be able to trust. The lesson is not that fraud exists; the lesson is that the verification systems we built around finance assumed a level of cooperation that we no longer feel comfortable assuming.
We have re-read, this year, every quarterly board package from every company in our portfolio. The exercise was not driven by suspicion of any specific company; it was driven by the recognition that, if a company we backed were systematically misleading us, the materials they showed us would likely look indistinguishable from the materials they would show us if they were not. We did not find anything we considered material. We did find a small number of patterns that warranted clarifying conversations. The clarifying conversations were, in two cases, more useful than the materials themselves had been.
The exercise has become an annual practice. We will repeat it every year, regardless of macro environment, because the value of the practice is highest in the years when we do not expect to find anything.
On the Boards We Sit On (and Have Resigned From)
We resigned from two boards this year. We will not describe the circumstances in detail. The general pattern was the same in both cases: information that should have been visible to the board was being deliberately filtered, and our objections were treated as procedural rather than substantive.
A board seat is not a privilege; it is a responsibility to people who are not in the room. When we cannot discharge that responsibility, the only honest action is to leave. We are aware that resignations carry signaling weight, and that the act of leaving a board can damage the company's prospects more than the underlying issue might have. We considered the trade-off in both cases. We concluded, in both cases, that the alternative — remaining on a board where our presence served as institutional cover for governance we did not endorse — was worse.
One of the two companies has since reorganized its board materially. The reorganization is encouraging. We are not seeking re-appointment; the act of leaving is meaningful only if it is not reversible on convenience.
On What Sarbanes-Oxley Solves and What It Cannot
The new legislation, signed in July, will, we think, prevent some of the more flagrant accounting abuses of the last cycle. It will require executives to certify financials personally, raise penalties for misconduct, and prohibit auditors from selling consulting services to the same companies they audit. Each of these is a useful structural change. None of them, taken alone or together, is sufficient.
The relationship between management and investors is repaired one quarter at a time, by founders who tell the truth even when no rule requires them to. Legislation can punish lies. It cannot produce truth. The companies we want to back are the ones whose financial disclosures would have looked the same with or without Sarbanes-Oxley, because the founders' disposition toward their investors has always been to tell them more rather than less. The companies that will produce the next cycle's restatements are the ones that disclose only what is required and not a sentence more. Both kinds exist. The act of distinguishing them is the substance of our work.
On the Companies That Will Survive Their Own Mistakes
Several companies in our portfolio have made meaningful mistakes this year. None of them have hidden any of those mistakes. We have, in two cases, rewarded the candor with additional capital. We do not have a formula for this; we have only a heuristic — that founders who acknowledge errors quickly are founders who learn from them quickly.
The opposite heuristic is also true and worth recording. Founders who frame their errors in language designed to obscure responsibility are founders whose subsequent decisions will repeat the errors in different forms. We have seen this pattern in three companies adjacent to our portfolio this year. The pattern is consistent enough that we now treat the framing of bad news, in board meetings, as a more reliable signal than the bad news itself. A founder who says "the market did not respond as we expected" is a different founder from one who says "we mispriced this product and lost six months." The second founder will recover. The first will compound the error.
A Closing Note
The work of trust is slow and unimpressive. It does not produce headlines. It does, in our experience, produce returns over horizons that the headlines cannot see. We will be back next year with a letter that, we hope, has fewer crises to discuss. If the past three letters are any guide, we should not count on it.
The Partners
Winzheng Family Investment Fund · December 2002