شکل گیری قیمت در OTC بازار اوراق قرضه شرکت های بزرگ: مطالعه بازار بین فروشنده ها

Despite its importance the market-micro structure of the secondary market for corporate bonds remains something of a mystery. The major reason for this has been the OTC inter-dealer nature of this market. As far as we are aware this paper presents the first exploratory field study of the U.S. inter-dealer OTC corporate bond market. We construct a primary data-base from the trades of a major bond dealer and document the competitive structure of the market in terms of the number of active dealers and market trading mechanism. We find that the trading mechanism closely resembles a first-price sealed bid auction. We also examine the potential differences between segments of the market and develop a measure of competition based on the theory of auctions. Our measure indicates that competition is highest in US investment grade corporate bonds and lowest in junk bonds. We also examine the effect of the size of a trade on pricing and spreads.

While the U.S. corporate bond market had a value close to $3 trillion in 1998, empirical research into price and return formation in this market has been relatively sparse compared to research on equity markets. There are two reasons for this. First, while some corporate bonds trade on the NYSE and AMEX exchanges, they tend to be odd-lots accounting for no more than 2% of market volume [see Nunn et al 1986 and Warga and Welch 1993]. Second, data quotes on OTC trades tend to be diffuse and based on matrix valuation rather than on actual trades such as those produced by IDC (in association with S & P and Moody’s). While a few studies have sought to use real quotes from dealers such as Lehman Bros. [see, Warga and Welch (1993)], and find quite surprising disparities between matrix prices and actual dealer quotes, no study, thus far, has analyzed the dynamics of price formation in the U.S. corporate bond OTC market. Specifically, questions such as how many dealers bid for large blocks of OTC-placed corporate bonds? Is there a difference in the bidding behavior for investment-grade versus non-investment grade bonds? What is the link between price determination and the number of bidders—i.e., how competitive is the market, have yet to be analyzed.1 Indeed, while we know a lot about the competitive structure of US equity markets, we know very little about the competitive structure and pricing dynamics of the US corporate bond markets. The lack of transparency in pricing has been of great concern to the SEC, especially in periods of high volatility as have recently occurred.2 As a first step in closing this knowledge gap we carried out a field study of price determination in the OTC corporate bond market with the support of a major corporate bond trader.3 In Section 2, we provide a general description of the OTC bond market. In Section 3, we describe our database and in Section 4 we provide descriptive statistics of the sample. In Section 5, we report our results on the competitive structure of the market.

This is the first paper to analyze the market micro-structure and price formation in the US corporate bond market. It is unique in that it utilizes actual transaction prices submitted for large and small blocks of corporate bonds. The institutional features of this market are characterized by: (i) a few large traders acting mostly on behalf of institutional clients, (ii) an inter-dealer market (similar to the one in government bonds) and (iii) a trading mechanism that closely resembles a first price (English) sealed bid auction. Using the trading book of a major dealer over an 11-month period in 1997, we analyzed the competitive structure of the market both in terms of the number of bids received for each trade the difference between the best and second best bids on any trade. It was found that the average number of bidders is quite small, with nine major counterparties accounting for the major proportion of trades and that spreads were on average 0.15% between the best and second best bids over the whole sample. Interestingly, the OTC market for emerging market bonds appeared to have lower spreads than the domestic high yield bond market. Additional bidders appeared to have a material pro-competitive effect on the observed difference between the best and second best bids. And, large trades did not seem to suffer from the disadvantages that are now well documented for large block trades in the equity market. Indeed, the difference between the best and second-best bids decreased with increased trade size.