| These discussions would be improved if more of the participants understood the problem of transacting in large blocks of tradable instruments. The impression HN trading discussions create is that there is a universe in which block trades are frictionless or even remotely predictable. In fact, moving large blocks across the market isn't just an annoying detail of the markets; it's one of the basic fundamental problems of institutional trading, and a large part of the rationale for the existence of brokers. Transacting in blocks of stock is to professional trading what the CAP theorem is to distributed software development. One of the most famous and approachable books about market structure is Larry Harris' _Trading And Exchanges_. The book is like the TCP/IP Illustrated of money. It is supremely readable and written in a style that software developers in particular will find congenial. You can get a Kindle version of it right now. I feel extremely comfortable recommending it. It is a great read. The example of trying to move a large block of (fictitious) Smithsonian Industries is one of the opening, motivating examples the book uses to outline the challenges of trading. The inheritor of a huge chunk of Smithsonian Industries needs to sell 900,000 shares of thinly-traded stock. The example continues: Goldman's block brokers face the following predicament. If nobody knows that they have stock to sell, they will not be able to sell it. However, if too many people know that a large block of stack is hanging over the market, speculators will push the price down. The Goldman brokers thus must be selective when approaching potential buyers. The motivating example Lewis gives in his book is of a trader at a large investment bank who, based on their $2MM/year salary, is presumably being paid handsomely for the service of figuring out how to move blocks like that without having the market shift out from under them. In Harris' example, the Goldman traders research other owners of Smithsonian Industries and approaches them privately and individually in the hopes of placing much of the block privately at a small discount. In Lewis' example, the handsomely-paid trader sees a spot price in their blotter screen, expects to push a single button (no, really, that's how Lewis frames it) to sell at that price, and is outraged when the price moves. There's an interesting debate to be had about HFT and, particularly, the conflicts of interest between broker-dealers and exchanges and dark pools. But it's hard to have that discussion if you start from the belief that institutional trading is supposed to be easy. The opposite is true. |