While I was at NASD Economic Research, one of the primary responsibilities of our group was to conduct market microstructure research -- the analysis of the basic functioning of the market, generally to determine or predict the effect of changes on the marketplace. Examples of such changes include implementation of exchange rules or policies, changes in the composition and behavior of market participants, and alterations in market-related conditions (e.g., economic factors, etc.). Variables of interest include measures of "market quality," such as bid-ask spread (representative of the cost of retail trades), market depth, and liquidity, among many others. Typically, microstructure research requires massive amounts of data, including audit trail databases of trades and quotation updates, which frequently constitute over a million observations per trading day. At present, I'm interested in three research topics along these lines, which are briefly discussed below.

• Additional study of the effect of order flow "preferencing" -- essentially the absence of market-wide secondary priority -- on trading costs to investors. Bob Jennings (Indiana), Robert Battalio (Georgia State), and I have written two papers that empirically illustrate that preferencing is associated with reduced trading costs and is a natural consequence of the ability of dealers to differentiate informed and uninformed order flow, respectively.
• Research on market impact, specifically for large institutional orders, as determined by market structure. Is capital commitment greater in traditional dealer markets? To what extent can innovative alternative trading systems -- such as electronic trading systems, or "ECNs" -- reduce costs?
• Design of effective opening processes for marketplaces. Currently, the opening processes used by financial markets are flawed -- the discontinuities induced by periods of market closure are increasingly severe in a global marketplace in which information costs continue to fall. Accordingly, "gapping" at the open -- possibly exacerbated by intermediaries -- translates to risks and costs that must be borne by investors. Barring 24-hour trading, opening processes should be designed to minimize the cost component.

Papers

I've co-authored the following papers (available in Adobe Acrobat):

• RLP
• Case study
• ASR
• Nasdawg
• Data paper

Useful Research Links