A pox on Sarbox: the unintended and rotten consequences of the Sarbanes-Oxley law.

AS a very young teenager on my first unsupervised afternoon in New York City, I promptly lost $10 playing three-card monte. Afterwards, I vowed that I'd never get pulled into this kind of rigged game again. I had learned an important lesson: Ignore what the dealer is saying and keep your eye on the money card. This can be a good strategy in other areas of life, too. Take, for example, the Sarbanes-Oxley law ("Sarbox")--a measure that claims to make the stock market safer, but actually has the effect of taking money from small investors and giving it to multimillionaires.

In 2002, on the heels of business scandals at Enron and WorldCom, Congress passed Sarbox to restore investor confidence in the integrity of U.S. public markets. Sarbox imposed a grab-bag of requirements on public companies--everything from criminal penalties for executives who willfully misreport financial results to restrictions on insider trading. Much of the energy behind this legislation was generated by the market meltdown that had eliminated about 50 percent of the value of the S&P 500 stock-market index in the 28 months leading up to the passage of the law. Asset bubbles are almost always followed by stricter regulation of markets. Human nature being what it is, bubbles in their later stages typically invite an abnormally high amount of fraud, and the resulting regulations are often excessively focused on preventing whatever particular flavors of malfeasance were most recently prevalent. Targeting wrongdoing is easy to understand, plays to our need to believe that somebody must be to blame for our sudden loss of wealth, and allows politicians to be seen doing something. Such regulations are usually produced very quickly, and often have effects that the authors never anticipated. The drafting and implementation of Sarbox is an excellent example of this dynamic.

NOT QUITE AS PLANNED

The most consequential component of Sarbox has turned out to be something that was not the subject of deep inquiry at the time: the "Section 404" requirement that companies specify, execute, and externally certify procedures for tracking and reporting financial results. The impact of this blandly written 168-word requirement, as subsequently interpreted by regulatory agencies and the accounting profession, is a classic example of the law of unintended consequences. According to the independent, bipartisan Committee on Capital Markets Regulation, Section 404 has driven the first-year cost of implementing Sarbox to about $15 to $20 billion, or more than 35 times what was estimated when the law was passed.

What are we getting for all that money? All else equal, more reliable financial reporting should lead to greater investor confidence and tend to increase stock values; on the other hand, all else equal, higher costs and the partial loss of management attention should tend to depress profits and therefore reduce values. Estimating the net effect is extremely difficult since it implicitly requires us to be able to know how investors would have valued companies in the absence of Sarbox. Partisans on both sides of the issue point to studies that purport to quantify the overall impact of the law, often with risible levels of asserted precision. Sarbox opponents often cite a University of Rochester study that estimates Sarbox has reduced total market capitalization by about $1.4 trillion due to net negative investor reaction. To put it in context, this is close to 10 percent of total U.S. GDP, which is pretty hard to imagine. Sarbox supporters cite an MIT study indicating that complying with Sarbox can reduce cost of capital by about a full percentage point. The author of this study, Ryan LaFond, was honest enough to admit: "I'm very comfortable saying there are some benefits, and they are economically meaningful. Whether they exceed the costs, I have no idea."

The root problem with all of these studies is that there is no objective, scientific way to know what stock prices should be. That's why we spend huge sums of money running stock markets. Of course, if you just ignore all of the complexities of real capital markets, Sarbox can be seen as a straightforward regulation that guarantees that "a dollar of net income is a dollar of net income." In this view, the argument that we should let the market sort out what is the right level of spending on internal controls and audits is like arguing that we should not prosecute a business that sells 15 rancid ounces of meat as a pound, and instead just let consumers decide which butchers they want to patronize.

It is true that there is a practical role in many spheres of life for regulations that reduce the costs that consumers bear to find what they want. Generally they are most appropriate for quantities, like weights, that can be defined unambiguously and assessed relatively cheaply, and for matters on which the result of misinformation can be catastrophic to health or physical safety. But this is nothing like the case with the evaluation of corporate performance. The appropriate measures for this are constantly debated; they are difficult and expensive; and they vary widely from one industry, time period, and company to another. In many of the most innovative sectors of the economy, the very accounting values, such as Net Income, that Sarbox is focused on validating are so useless that managers and owners simply ignore them when operating and valuing companies.

Former senator Paul Sarbanes, the Maryland Democrat who coauthored Sarbox, spent effectively his entire career in government, and apparently considers dealing with these issues way beneath his pay grade. His take on the law is simple: "The bill is really about ensuring that public companies have a legitimate system of internal financial controls. To me that is a worthwhile cost." Well, thanks. But how about you invest your money only in companies that have to bear these compliance costs, and I'll choose for myself?

'GOOD FOR US, BAD FOR AMERICA'

While it's hard to measure the net impact of the law on overall market value, one relative effect is very clear: Sarbox is much worse for small public companies than large ones. For very large companies Sarbox is just another unpleasant cost of being publicly listed in America, but for smaller public companies it can make public ownership prohibitively expensive.

A group of professors at Berkeley and USC demonstrated analytically that Sarbox has increased the rate at which small public companies have exited the stock market, disproportionately by selling themselves to private-equity firms that are not subject to Sarbox. No such effect is evident for larger public companies. This analysis is entirely consistent with my experiences as founder and CEO of a profitable, high-growth software company. We recently sold a significant slice of the shares in our company to a large private-equity firm. In contrast, five or ten years ago the obvious decision for us would have been to go public through an Initial Public Offering (IPO), but now the costs of being public are so much higher that it is just not worth it. Because Sarbox compliance costs are so much higher in relation to company size for small companies than for large ones, the break-even size at which it is sensible for a company to become public has risen.

This creates an enormous arbitrage opportunity for private investors to purchase the equity of smaller high-growth companies if they don't believe all of the extra compliance work required by Sarbox is really economically valuable. As it turns out, many private-equity firms have created practice areas to exploit exactly this opportunity, and my company was able to get the capital that we wanted at an acceptable price. The fact that these firms generate much higher sustainable returns than public markets indicates that their judgment that good investments can be made even in companies with "inadequate" financial controls just might be accurate.

So if successful smaller companies still make out fine and private-equity investors have been granted a new source of excellent profits, then who gets hurt? You do--unless you happen to have several spare millions of dollars lying around to buy your way into a large private-equity fund. In the long run, higher investment returns require greater risk. Private-equity and similar investment vehicles consistently generate better returns than the public markets by accepting the risks of, for example, earlier-stage companies or more mature companies that require radical turnarounds. Public markets are designed to allow smaller investors to participate in equity financing by efficiently pooling their capital. With Sarbox, Congress has expressed the view that many of these demonstrated opportunities to earn higher returns don't belong in the public markets because small investors aren't smart enough to evaluate them.

Therefore, the primary effect of Sarbox to date has actually been to rope off some of the opportunities to invest in smaller or riskier companies and make them available only to private investors, which is to say, very rich people. This is precisely why Steve Schwarzman, CEO of private-equity titan Blackstone, has described Sarbox as "probably the best thing that's happened to our business and one of the worst things that has happened to America."

Ironically, Blackstone has now announced that it intends to conduct an IPO. If that happens, many other large private-equity firms will very likely follow. This would allow small investors to buy shares in private-equity firms in order to buy equity in the companies that they otherwise could have invested in directly anyway. In theory, the private-equity firms will provide value-added investment selection and supervision, but small investors will no longer have the choice of whether to accept this help or simply make decisions on their own.

One way or another, market intermediaries will emerge to try to connect the huge pools of small-investor savings in 401(k) and individual accounts with the higher-return opportunities that Sarbox has pushed out of the public markets. It's like water flowing downhill. Ultimately, Congress must either absolutely deny small investors the right to invest in these situations, or else allow these investments to occur in even-less-regulated environments with higher transaction costs than the public markets. What Congress can't do is legislate away the relationship between risk and return.

RACE TO REFORM

Republicans in the last Congress put forward a bill that would ameliorate the worst effects of Sarbox by making it optional for companies with less than $700 million of market value, and limiting the costs of audits. It might be natural to assume that prospects for reform have dimmed in the new Congress, but--on this issue at least--many Democrats are sounding like grownups. Speaker Nancy Pelosi's Innovation Agenda calls for requiring "specifically tailored guidelines for small public companies to ensure Sarbanes-Oxley requirements are not overly burdensome." New York senator Chuck Schumer has called for Sarbox to be "re-examined." In real man-bites-dog news, New York governor Eliot Spitzer recently said that Sarbox "has created an unbelievable burden for small companies and may be preventing some initial public offerings." Letting Nancy Pelosi, Chuck Schumer, and Eliot Spitzer get to your right on securities regulation can't be a good idea. Republicans should push for action soon, and should focus on making the market-value threshold for opt-out as high as possible.

More broadly, reform of Sarbox should have been seen in the context of overall securities-market regulation. Some amount of market regulation is required for markets to function, but too much chokes off growth. While it is hard to evaluate fully the impact of any one regulation, U.S. securities markets have been subject to a raft of recent regulations, lawsuits, and aggressive government prosecution at precisely the same time as real international competition has started to emerge. While some erosion in the dominant position held by U.S. equity markets is probably inevitable based on global economic trends, it's hard to explain how the U.S. share of global IPO value has declined from 50 percent to 5 percent since 2000--and the U.S. share of total equity capital raised in the world's top ten economies has declined from 41 percent to 28 percent since 1995--unless the recent increases in oversight costs had something to do with it.

It's a commonplace to say that we live in an information economy, and that increasingly we are knowledge workers. What is often overlooked about this statement is that knowledge workers are workers. Just as factory workers in an industrial economy were subject to constant downward wage pressure, so are knowledge workers in an information economy as their tasks become commoditized. This is why standard software engineering, often seen in popular imagination as the acme of high-wage knowledge work, is increasingly being outsourced to lower-wage economies.

Market-making, in the broad sense of seeing and connecting otherwise disjointed problems and solutions, is the high ground in an information economy because it resists commoditization. The operation of markets has become as central to the knowledge economy as the machine-tool sector was to the industrial economy. Losing our leading position in global equity markets would be disastrous for the U.S., and not just for big investors on Wall Street. Congress should approach reform of Sarbox as one battle in a larger, ongoing struggle to maintain the international competitiveness of U.S. equity markets.

Mr. Manzi is the CEO of an applied-artificial-intelligence software company.