Market Evolution, Liquileaks and the Economic Benefits of Speed
Tue, 17 Jan 2012 02:57:00 GMT
An Interview with Albert J. Menkveld
HFT Review: Albert, welcome to the HFT Review. Can you provide some background on who you are and what your work involves in relation to high frequency trading and market structure?
Albert J. Menkveld: I was trained as an econometrician and I originally worked in the airline industry. The shares of my company, KLM Royal Dutch Airlines, were traded in both Amsterdam and New York and I became fascinated by the question of how the information and prices were imparted across the two market centres, and what the mechanics really were behind trading. So I eventually left the company and started a PhD, visiting the US along the way and, upon graduation, decided to pursue an academic career.
My research agenda is focused on securities trading. One of the things I get excited about is how things are changing in this “electronic age” and how the old models of the way we looked at the world don’t necessarily apply to today's markets any more. A lot of the principles still hold, but some do not. That’s where I am right now in terms of my research.
HFTR: Maybe we can start by looking at an area you covered in your paper “Electronic Trading and Market Structure” (http:/
AJM: Yes, in the past, to set up a new market and compete with an existing market was close to impossible. You could stand on a different floor and quote good prices for a particular security, but since everybody else was on the other floor, the costs to gather information from what was out there on other markets, to really check the prices, were prohibitively high.
When markets started turning electronic and the floors started to disappear in the mid-80's/90's and the matching of buyers and sellers was done essentially through a server, search costs became lower for everybody in the market. This benefited new markets because now, instead of creating a trading floor, one could just set up a server making it really easy for others to check and see what better prices might be available at the other market, on the other server.
When I started to work on this topic with Professor Thierry Foucault (http:/
HFTR: What sort of conclusions did you draw from that research?
AJM: There are two basic dimensions to how these new MTF-type markets are competing with the existing ones. One is simply trading fees. If you start a new, fully automatic market from scratch, you don't have the overhead costs of human resources, CPU power is cheap, etcetera. So you can get market share by simply having lower fees.
The other dimension is that by designing and building the new computer system from scratch, you can make sure that it is super fast so that people (or computers) who want to interact with your market get a quick response.
Also, since floors disappeared, the “upstairs” markets in the past, where buyers and sellers would meet outside of the floor to trade large quantities at advantageous prices to both is also disappearing. Although to some extent, that has been picked up by dark pools, which I see as a manifestation of upstairs trading in an electronic environment.
Finally, given the heterogeneity in trading needs of investors, a new landscape arises where different markets are tailored to the different needs of these investors. And these markets communicate with each other through high frequency traders, or modern arbitrageurs, whose trading keeps prices in these markets in sync.
HFTR: You discuss this in some depth in your paper “High Frequency Trading and the New-Market Makers” (http:/
Click here to access the full text of this interview (registered, logged-in users only)
If you are not a registered user, click here to register for free
