Home About IUP Magazines Journals Books Amicus Archives
     
A Guided Tour | Recommend | Links | Subscriber Services | Feedback | Subscribe Online
 
The IUP Journal of Applied Finance
Decomposition of the Otcei Bid-Ask Spread
:
:
:
:
:
:
:
:
:
 
 
 
 
 
 

estimates of components,different methodologies,Huang and Stoll,two methodologies,Counter Exchange of India, quote-driven market,India comprising of small and medium size companies, two methodologies, scenarios of economics, market-making in OTCEI. An in-depth study of trading policies, however, suggests that estimates of components of spread obtained from Huang and Stoll methodology are more in tune with market microstructure of OTCEI, than estimates of components of spread obtained from Stoll’s methodology. This article helps one to understand the economics of market-making in OTCEI and its trading policies despite the slightly disparate findings of the two methodologies.

The quoted Bid-ask spread forms a major constituent of market microstructure process. According to the existing literature, the bid-ask spread is made up of three components— Order Processing Cost, Inventory Holding Cost and Adverse Information Cost. Order processing cost is the cost of doing business, inventory holding cost constitutes the cost of holding inventory and adverse information cost forms the cost of ignorance. Needless to say, the three components are greatly influenced by the structure of markets and their trading policies.

The decomposition of spread into its components began with Glosten and Harris (1988). They were the first to use a trade indicator or econometric model of spread. They used the method of maximum likelihood to estimate the parameters of their model of spread and decompose spread into only two components—Adverse Information and Order Processing components. However, they used only transaction data and not quote data. Consequently, they could not model the components directly. Stoll (1989) was the first to decompose the spread into its three components using covariance model. With 3-month transaction and quote data of Nasdaq, he shows that serial covariance of first order difference of transaction returns is related to spread in efficient markets. He estimates two parameters from these serial covariances of transaction and quotes returns and consequently decomposes spread into its constituent three components. George Kaul and Nimalendran (1991) also use covariance model of spread. However, their model takes care of the bias due to time varying expected returns and get an estimate of spread, which they show is unbiased and efficient. Yet, in another article by Lin, Sanger and Booth (1994), the spread is decomposed only into two components—Adverse Information and Order Processing costs.

 
 
estimates of components,different methodologies,Huang and Stoll,two methodologies,Counter Exchange of India, quote-driven market,India comprising of small and medium size companies, two methodologies, scenarios of economics, market-making in OTCEI.