The New Yorker: PRINTABLES THE FINANCIAL PAGE SARBOXED IN? by James Surowiecki Issue of 2005-12-12 Posted 2005-12-05 In the summer of 2002, with the stock market tumbling and fraud at Enron and WorldCom dominating the headlines, there was immense political pressure on Washington to restore investor confidence by doing something about corporate crime. Scrambling to deflect charges of indifference to the plight of widows whose 401(k)s had vanished, Congress hastily wrote and passed the Sarbanes-Oxley Act (dubbed SarbOx), a tough piece of anti-fraud legislation. A Republican-dominated Congress might have been expected to oppose costly business regulations, but politics made SarbOx a thoroughly bipartisan affair. The bill passed unanimously in the Senate, and, when President Bush signed it into law, he proclaimed the end of an “era of low standards and false profits.” Washington’s pride in SarbOx, though, was not universally shared. Businesses hated the complexity of the new rules (which, among other things, required corporate executives to certify all the financial results of their companies). Economists fastened on the inefficiency of many of the law’s provisions. Stephen Moore, the founder of the Club for Growth, recently called the law “a new cancer,” and the former chief financial officer of GlaxoSmithKline deplored it as an “American nightmare.” SarbOx, the argument now goes, is a classic example of government overreaction. Its heavy costs outweigh its meagre benefits, standing in the way of the market’s efficient allocation of capital. The Securities and Exchange Commission is now talking about loosening enforcement of the regulations, while lobbyists are pushing Congress to revise the bill in the year ahead. SarbOx is decidedly flawed, most notably because the cost of compliance is too high for small companies. Initially, the S.E.C. suggested that the average company would have to spend ninety-one thousand dollars annually, but the stringency of the regulations means that the real number is well into seven figures (for a start, a company has to appoint people to police it internally), a cost that may discourage smaller firms from going public. However, although SarbOx does need to be mended, that doesn’t mean it should be ended. Congress may have passed the law in a fit of political panic, but the fraud that it was designed to deal with, far from being a matter of the proverbial few bad apples, was becoming endemic. Executives routinely engaged in “earnings management,” releasing hyped or invented numbers in order to pump up their companies’ stock price. Between 1997 and 2002, public companies reported nearly a thousand earnings re-statements—admissions that their previous statements had been inaccurate. This fraud cost investors and lenders an enormous amount of money, vaporizing hundreds of billions of dollars in shareholder value. But corporate crime also had a significant effect on people who had never thought of buying Enron stock or WorldCom bonds. In order to make investors believe that they were earning billions of dollars a year, fraudulent companies often went to great lengths to keep their sales growing, even at the expense of profits (which they were, in any case, inventing). They made foolish acquisitions and high-profile investments that destroyed value instead of creating it—studies suggest that, in the telecommunications sector alone, bad investments totalled tens of billions of dollars. And they hired lots of people whom, in the end, they probably didn’t need. A recent paper by Simi Kedia, of Rutgers, and Thomas Philippon, of N.Y.U., for instance, looked at all the companies known to have been managing earnings between 1997 and 2000. In those years, the companies boosted hiring by a full twenty-five per cent, while other companies increased hiring by less than seven per cent. As soon as the companies were forced to come clean, employees were sacked. Kedia and Philippon estimate that the re-stating companies fired between two hundred and fifty thousand and six hundred thousand people between 2000 and 2002, slashing payrolls by more than twenty-five per cent, while other companies cut them by just 1.5 per cent. All this playacting affected not just the fraudulent companies but also their competitors, with serious consequences for the American economy at large. As Gil Sadka, an accounting professor at Columbia, suggests in a recent paper, WorldCom’s lies—about its profits, about the amount of Internet traffic its network was carrying, and about the total demand for telecom capacity—made competitors like A.T.&T. and Sprint look inefficient. Trying to keep pace with WorldCom led these companies to overinvestment in new technology and to price wars, followed by cost-cutting campaigns, layoffs, and, in A.T.&T.’s case, the decision to break up the company. WorldCom’s deception had consequences that were anything but local. It led to the misallocation of billions of dollars in capital across an entire industry, and rearranged the lives of tens of thousands of workers. It’s hard to think of many things more inefficient than that. Corporate fraud, in other words, isn’t expensive just for the people who have been defrauded. It also inflicts what economists call “social costs” on the economy as a whole. If fraud were just a matter of executives ripping off shareholders, we could expect the market to come up with a cost-effective solution. Social costs, on the other hand, generally require regulation, which is where SarbOx comes in. Because it tries not merely to punish fraud but to prevent it from happening in the first place, the law’s costs are a lot more visible than its benefits. But a world without SarbOx would be costlier still.