Not so very long ago investors and traders had little reason to worry about what goes on behind the scenes. They placed an order, got a fill, paid a commission, and that was that. They didn't have to know "how the sausage was made." But times have changed. Infrastructure malfunctions and flash crashes have unnerved the investing community -- in part because they (and often the regulators as well) didn't understand how such disruptions could happen.
Market Microstructure in Emerging and Developed Markets: Price Discovery, Information Flows, and Transactions Costs, edited by H. Kent Baker and Halil Kiymaz (Robert W. Kolb Series in Finance, Wiley, 2013) offers a detailed analysis of the market's "guts." Or, more accurately, I should make the word "market" plural because the book covers not only the equity market but bond, derivatives, and currency markets, not only exchanges but dark pools, not only developed markets but emerging markets. Both academics and practitioners contributed to the book.
I can't, of course, give a proper sense of this book as a whole since it is so wide-ranging, but here are a few topics that are analyzed. Do circuit breakers reduce volatility, enhance price discovery, interfere with the trading process, prompt a magnet effect? Why does the bid/ask spread exist? (Yes, of course, we know the easy answer.) What is the impact of high-frequency trading on liquidity and market quality? How does pre-trade transparency affect liquidity? How would you design a market from scratch?
I personally have always been intrigued by dark pools. Put most crudely: Are they good or bad? The authors of the chapter on dark trading note that "dark pools free ride on the price discovery of other markets ... This may give rise to manipulation strategies that may negatively affect both dark pools and the main market. Dark pools also influence the market quality of other markets. Two opposite forces appear. First, dark pools allow for additional risk-sharing benefits as they may cater to traders who would otherwise not participate in the trading process. This should improve market quality. Second, market quality may deteriorate when dark pools are skimming off part of the uninformed traders. These trade-offs underpin why regulators are concerned that price discovery and market quality may suffer when the market share of dark pools becomes too large." (p. 227) And you wonder why regulators have such difficulty coming up with clear guidelines.
For those who want to know where to place their limit orders, here's a finding from 2004 about quote clustering: "The proportion of quotes that are divisible by $0.05 is around 39 percent for NASDAQ and 40 percent for the NYSE. The proportion of quotes divisible by $0.10 is 22 percent for NASDAQ and 24 percent for the NYSE." (p. 208)
Market Microstructure doesn't read like a novel, but it's an excellent reference for those who would like to delve deeper into how markets do and don't work.