Market microstructure, a review of classic foundations

We look at the basics around limit order books (LOB) more closely… The written post complements and extends the attached audio. (The title picture is taken from Bouchaud, Farmer & Lillo, one of the papers we’ll look at in this post.) As ever, please let me know if you spot mistakes or see things differently.

We’ve been looking at platforms around the European Intra-Day Markets, thinking about CLOB “plumbing,” the role of ISVs, and how exchange technology enables interaction. Let’s now spend more time with the structure generated by limit order books (LOB).

Last week’s Lunch with the FT featured Jean-Philippe Bouchaud—hedge fund manager, physicist, part-time theatre director—reflecting on markets, Camus, and the risks associated to misguided financial theory.

The interview article made me think how interesting it would be to connect classical theory on market microstructure to our context in the short-term energy markets, and how well this would fit into our series on platform-thinking in the energy industry. The study of market microstructure is fundamentally the study of interactions on a market platform.

Platforms take many forms, from stock exchanges to dating clubs, as Evans & Schmalensee point out. What unites them is that they set rules and infrastructure for interactions.

In the CLOB

In finance, the LOB has been dissected for decades; in power markets, analysis is still comparatively young. Nevertheless, core concepts like spread, depth, impact, resiliency, link classical order book theory to the power trading context.

Today we will look at ‘five classics’ in the literature surrounding LOBs. These are landmark contributions each offering intuition that translates into the energy market context.

It’s always best, I think, to start with the classic papers, where the development of ideas tends to be more direct and elegant. The classics provide a reference point and centre of gravity for explorations of more recent work and heavier mathematical or ‘data-driven’ machinery.

We may be ‘data-driven’ but it’s crucial to keep a close eye/AI on the classics in the background.

For each of the ‘five classics’ we’ll outline the core idea(s) and then translate to the modern power markets.