What is called "high frequency quoting" or "quote spam" in the context of high frequency trading?
Why do some people consider that as a problem for the market?
Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. It only takes a minute to sign up.Sign up to join this community
"quote spam", "book colouring", "quote stuffing", etc encompass any mechanism to modify the shape of the orderbook by a market participant who does not intend to really buy or sell shares thanks to these orders.
It means that someone fills the bid side of the book with 10,000 shares at different levels of price and does not want to buy at all, or only 100 shares. Worst he may want to sell shares.
It is not a good practice because the orderbooks are not only the live mechanism of the PFP (price formation process) but one of the only way to disseminate information on the PFP to all market participants. If you modify this information in a way that is not compliant with what you intend to do: your are misleading the market participants. You manipulate the price formation process. This is bad.
A bit belated, but nevertheless: It's worth noting that at least some of the various visible shapes of "quote spam" shown on e.g. Nanex analyses, such as sawtooth patterns, can be explained without assuming malice on the part of the order originators.
By way of example, two poorly designed agency execution algos trading buy child orders may each have a hidden reserve price above the initial market best bid and both be reasonably aggressive and be prepared to penny-jump the best bid all the way up until their reserve price, but if they are unable to beat the best bid, instead then bid at 1 tick (==1 cent in the rest of this example) above the next best bid.
Basically these two algos can interact, each attempting to outbid the other, until say algo #1's reserve price is exceeded, whereupon it will instead bid at the next best bid +1c. Algo #2 will as a result see in the order book that it no longer needs to bid as high as Algo #1's reserve price +1c and will drop back to Algo #1's recently updated price +1c. Algo #1 will then bid at Algo #2's bid +1c, restarting the process and running through the same pattern over and over, in the process generating vast amounts of quote traffic that is superfluous to any genuine price formation process.
The above pattern I would expect to see recur from time to time.
Another related possible source that I've seen at least once (and wouldn't expect to see recur too often) is an erroneously coded algo that's tried to outbid itself!
An additional source of quote spam could be as a result of liquidity providers chasing NBBO moves (US markets), where those providers are only prepared to quote if they can be top of book (or near enough to it), but will only know if they are top of book (or thereabouts) after submitting the order, since competitors would also have been trying to submit at more or less the same time. (In fact, one of the reasons for exchanges introducing new order types was that the exchanges would get hammered with new order submissions in anticipation of NBBO changes in 1 tick wide markets.)