Liquidity Risk, Firm Risk, and Issue Risk Premium Effects on the Abnormal Returns to New Issues of Convertible Bonds

Abstract

I. INTRODUCTION A large body of empirical evidence demonstrates that the issuance of new CB's is associated with negative abnormal returns of the underlying shares (1). This is puzzling since: (1) the market normally reacts positively to the straight corporate bond issuance; (2) CB's have payoff structures entail a straight bond component and an equity option component; and (3) in spite of the negative equity market reaction to CB issuance, the market for CB's has grown rapidly over time (2). One possible resolution of the puzzle that has been adduced is that due to dilution effects, investors perceive CB's as equity from the outset, rather than debt (3). In this vein, according to the pecking order theory, the market reaction should be negative. However, the evidence that we provide in this paper demonstrates that dilution effects are negligible. What then explains the abnormal equity performance of firms that issue CBs? This paper serves to provide new evidence on this score. In particular, we examine the underlying firm characteristics that serve as drivers of the abnormal returns when CB's are issued. The focus is on the relationship between the short-term wealth effect (4) around the issuance of CBs and the characteristics of issuer firms and the features embodied in the issues themselves. In particular, we examine the impact of three factors suggested in Liu and Switzer (2009): liquidity risk (logarithm of issue size), issue risk (Vega, which measures the sensitivity of CB value to the volatility of underlying stocks), and firm volatility risk (standard deviation of beta (5)) on the abnormal returns to convertible bond issues. We find that all of these three risk factors serve as significant drivers for the abnormal returns around the CB issue date. The market responds favourably to the issuance of convertible bonds by issuers with mild level of firm volatility risk. However, liquidity risk and issue risk are significantly negatively related to performance. The latter two risk components serve to offset the risk management benefits of convertibles for firms. These findings are robust to different grouping criteria and estimation methods. The remainder of this paper is organized as follows. Section II describes the data used in this paper. Section III explains the methodology and proxies employed in this study. Results are reported in Section IV. The paper concludes with a summary in Section V. II. DATA DESCRIPTION The sample consists of all CB offerings from January 1, 1986 to December 31, 2005 for which the underlying shares are traded on the New York Stock Exchange (NYSE), the American Exchange (AMEX), or the over-the-counter (NASDAQ) market from the SDC Platinum database. The basic CB data, including conversion price, coupon rate, expiry date, issue date, ratings, and issue size are obtained from SDC. Missing observations from SDC are replaced with data from the Convertible Bond Database that was provided to us from Morgan Stanley. The underlying stock prices of the above identified firms are from CRSP. During the observation period, stock prices are adjusted for stock dividends or splits. The market index returns are also from CRSP. We employ three market proxies in our tests: returns on the value-weighted market portfolio, returns on an equally-weighted market portfolio, and returns on the level of the Standard & Poor's 500 Composite Index. Company financial data are obtained from the Standard & Poor's Compustat database. Firms are included in the analyses if they have complete data on cash flow, working capital, investment in fixed assets, the real tax rate, growth rates in assets and sales, and various size and risk measures. Market expectations for CB issuers are proxied by analysts' opinions, as reported in IBES, which includes the estimation of earnings per share, cash flow per share, sales, or operating profit, and the Buy/Hold/Sell recommendations. …