Dominant shareholder incentives affect the cost of corporate borrowing

A paper by Paul Malatesta, a professor of finance at the University of Washington Foster School of Business, on the cost of corporate borrowing has won the Jensen Prize for Corporate Finance and Organizations. The prize recognizes the best corporate finance paper published last year in the Journal of Financial Economics.

For the paper, “Ownership structure and the cost of corporate borrowing,” Malatesta collaborated with Chen Lin of the Chinese University of Hong Kong, Yue Ma of Hong Kong’s Lingnan University, and Yuhai Xuan of the Harvard Business School.

Their analysis of nearly 3,500 firms in 22 countries over the period from 1996-2008 revealed that the cost of borrowing money is significantly higher for companies that have a wider divergence between the largest owner’s control rights and cash-flow rights.

To better understand this finding, we asked Professor Malatesta to explain.

What’s the difference between control rights and cash-flow rights?

Previous studies have shown that in many countries around the world corporate ownership structures are often highly complex. In many cases such ownership arrangements result in dominant shareholders who have greater control rights than cash flow rights. Consider, for example, a simple pyramid structure where you own 25% of firm A which, in turn, owns 20% of firm B. If 20% of the votes is enough to secure effective control (many think 10% would usually be enough) then you control firm A directly and firm B indirectly. Control rights are measured as the weakest link in the control chain, so your control rights in B are 20%. Your share of B’s cash flow (or “cash flow rights”), though, is the product of ownership stakes along the control chain: .25 x .20 = 5%. We say that your excess control rights in B, the “wedge” between control and cash flow rights, are .20 – .05 = 15%.

How does such a scenario affect that dominant shareholder’s incentives?

Excess control rights create incentives for the controlling shareholder that are likely to impair Firm B’s value to non-controlling security holders. As controlling shareholder would you rather have Firm B earn an extra dollar or Firm A? Clearly you prefer that Firm A earn the dollar because you would own 25% of the dollar (compared to just 5% of the dollar that Firm B earns). Hence, you would be inclined to manipulate transactions between A and B to transfer cash earnings from B to A. For example, you can have B sell goods or services to A at bargain prices or have B purchase them from A at premium prices. You could have B make a sweetheart loan to A. We call such moral hazard transactions “tunneling” and we say that the controlling shareholder extracts “private benefits of control,” not shared with other stockholders pro rata.

What is the relationship between excess control rights and corporate borrowing?

Earlier empirical work has shown that measures of firm stock value (Tobin’s Q ratio) are negatively related to the excess control rights of the ultimate controlling shareholder along the ownership chain. Our paper is the first to investigate the impact of excess control rights in bank credit markets. Using a large sample of syndicated bank loans made to many firms in 22 different countries, we show that firm borrowing costs are positively related to excess control rights, after controlling for other factors that also affect borrowing costs. This relationship is robust to several different econometric analyses. It appears that excess control rights cause lenders to anticipate tunneling activities that impair their collateral and borrower creditworthiness. As a consequence, lenders charge higher interest rates to borrowers whose ultimate controlling shareholders possess excess control rights.

How big is the impact on the cost of corporate credit?

The impact of excess control rights is both statistically and economically significant.  A one standard deviation increase in excess control rights is associated with an increase in borrowing costs of 27 to 38 basis points.