The Robustness of SOX

In our paper, The Robustness of the Sarbanes Oxley Effect on the U.S. Capital Market, which was recently accepted for publication in the Review of Accounting Studies, we re-evaluate the effects of the Sarbanes Oxley Act (SOX) on the number of exchange-traded firms in the U.S. capital market, taking into account general market conditions and firm fundamentals, especially size and financial performance.

We start by developing a model to predict new listings based on conditions in U.S. markets for products and capital, and then invert that model to predict de-listings over time. We analyze a comprehensive dataset of new listings and de-listings from U.S. national stock exchanges over 44 years. We find that market factors that explain the frequency of new listings on U.S. exchanges from 1962 through 2005 also explain most of the variance in the frequency of de-listings over the same time period. The peak in the number of delisted firms and the rate of de-listings was in 2001. In addition, the frequency of de-listings has continued at high levels relative to the historical average subsequent to the passage and implementation of SOX. Our time series analysis uses an indicator variable as a proxy for the passage of SOX, and the SOX variable is not associated with abnormal levels of de-listings in general, conditional on market factors. However, our proxy for the implementation of SOX Section 404 in 2004 is associated with a significant decline in de-listings, contrary to the prior finding of SOX effects resulting in increased de-registrations based on shorter time series and more restrictive definitions of firms exiting the U.S. public capital market.

Next we look for changes in the propensity to delist by estimating firm-specific models including the general market conditions from our first set of tests, firm fundamentals, and the SOX proxies. We use a sample composed of all U.S. firms listed on national stock exchanges during the period 1962 to 2005. We find that the passage of SOX is associated with increased probability of delisting for publicly traded U.S. firms, but only when proxies for general market conditions are not included in the model. The implementation of SOX section 404 (for the largest firms in the U.S. capital market) in 2004 is associated with a significant increase in the probability of delisting, even after controlling for general market conditions and firm fundamentals. This result is not inconsistent with recent literature documenting an increase in the probability of delisting during the SOX time period, especially for small firms (although small firms have been granted an extension on applying Section 404 that has already rolled over several times).

Finally, we disaggregate the sample of U.S. exchange-traded firms by size to replicate earlier results suggesting that small firms were especially pushed out of the public equity market. We find that prior stock return and size are significantly negatively associated with the probability of delisting for firms in all five size quintiles, meaning that smaller firms and poorly performing firms were more likely to delist across the whole time period. We also find that, for the smallest quintile of firms in the sample, the passage of SOX was only marginally associated with an increase in the probability of delisting, and the implementation of Section 404 was not associated with the probability of delisting (consistent with the delayed applicability of Section 404 to small firms). On the other hand, the passage of SOX is either not associated with or significantly negatively associated with the probability of delisting for the largest 80% of the sample, but the implementation of Section 404 is associated with an increase in the probability of delisting for all but the smallest 20% of the sample firms.

We conclude from this evidence that there is little basis for associating the increased incidence of firms delisting from U.S. markets in the first half of this decade with the passage or implementation of SOX rather than with the general worsening of market conditions. On the other hand, the implementation of SOX Section 404 has been associated with increased probability of larger firms exiting U.S. public capital markets, especially if they are performing poorly. Increasing the probability of poorly performing firms exiting the U.S. public capital markets is a different regulatory implication of SOX than is pushing small firms out of the market.

The full paper is available for download here.