The Sarbanes-Oxley Act (SOX) of 2002 was meant to strengthen corporate governance and financial reporting in the US in the wake of several corporate scandals. While it was intended to curb malfeasance and sustain investor confidence, it has also come under sharp criticism for the substantial costs it has created for issuers of equity. One group particularly affected is the more than 1,300 foreign private issuers (FPIs) that had reporting obligations to the Securities and Exchange Commission at the time SOX was enacted.

FPIs cross-list in both their home country and the US because they can obtain a lower cost of capital and greater access to capital markets there due to the stronger corporate governance demands. This is known among researchers as the “legal bonding hypothesis”. Xi Li used this as a starting point to examine just how SOX has affected FPIs.

He was particularly interested in whether SOX reduced the incentive for managers or controlling shareholders in FPIs to extract private benefits since this corporate governance issue has received scant attention in the past.

What he found not only calls into question the benefits of SOX, but also raises questions about the reliability of the legal bonding hypothesis.

Prof Xi looked at data from FPIs before, during and after the enactment of SOX, and in particular the factors that drove firms in all three periods to “go dark” – that is, to delist or deregister as FPIs voluntarily, so they no longer had to meet US disclosure and other requirements.

First, he found that more companies go dark after SOX, although he cautioned this alone did not explain the impact of SOX and the legal bonding hypothesis. It was more important to look at the firm’s returns after going dark and any changes in firm characteristics – information that was available from the home country returns filed by FPIs even after they went dark.

Prof Xi explained why this information was important. “In either the pre- or post-SOX period, worse post-dark governance characteristics would have been consistent with going-dark FPIs being released from legal bonding, thus providing evidence of agency problems and supporting the legal bonding hypothesis; reduced growth potentials would have predicted worse post-dark growth characteristics.”

But he found mixed changes in governance characteristics and an apparent improvement in growth potential after firms went dark. It appeared the latter had motivated firms to go dark, which suggested that compliance requirements before SOX were not discouraging firms.

However, that changed with SOX. “Going-dark FPIs in the post-SOX period exhibited more favourable changes in every characteristic – both in terms of governance and growth potential – which is consistent with the explanation of increased compliance costs due to SOX,” he said.

“In summary, the results suggest not only that SOX imposes net costs on shareholders in going-dark FPIS, but also the net costs more than offset the extant net benefits from cross-listing (including legal-bonding benefits) for these FPIs.

“Going dark in the pre- and post-SOX periods appears to be driven by reduced growth potentials and increased compliance costs, respectively” – leading to the additional conclusion that there is weak support for the legal bonding hypothesis for FPIs in general, and that this needs to be examined further in future.