Introducing How Markets Slowly Digest Changes in Supply and Demand for after-work quants

Takeaway. Large buys and sells do not "finish" at the print—markets often keep adjusting for a long time because order flow has long memory and impact has a propagator shape.

How Markets Slowly Digest Changes in Supply and Demand (Jean-Philippe Bouchaud, J. Doyne Farmer, Fabrizio Lillo, 2008) is a classic review of that slow-digestion picture: persistent order flow, temporary vs lasting impact, and why naive "instantaneous impact" stories fail.

Fit for learners: Review of long-memory order flow, propagator impact, and slow digestion of supply/demand for execution design.

This Hub note is a plain-language intro. Not investment advice. Related Hub reads: square-root impact (Toth–Bouchaud) and Almgren–Chriss optimal execution.

The problem

If you treat every trade as a one-shot shock to price, your backtests and execution plans will look too clean. Real books show autocorrelated order signs and impact that builds and then partly decays. After-work learners need a mental model before coding any impact simulator.

The idea / core intuition

• Order flow is not white noise: buy/sell signs often show long memory.
• Impact can be framed with a propagator: today's trade leaves a fading footprint on future mid moves.
• "Digestion" means liquidity and opposing flow gradually absorb an imbalance—sometimes slowly.
• Execution design then becomes: size and pace your child orders so you do not fight your own lingering footprint.

No formula dump here—open the PDF for figures and references. Treat it as intuition scaffolding, not a live algo.

Why after-work Python learners care

Impact and order-flow memory show up the moment you simulate VWAP-style schedules or plot signed volume. Python practice (pandas + a toy random-walk mid) is enough to feel the idea.

Tiny evening exercise: invent a signed-trade series with mild autocorrelation (e.g. AR(1) signs), add a decaying impact kernel to a midprice, and plot mid vs cumulative signed volume—label the notebook "learning only," no live data feed required.

One-evening reading checklist

1. Skim until "long memory" and "propagator" click.
2. Note temporary vs permanent impact in your own words.
3. Ask: what breaks if my simulator assumes i.i.d. order signs?
4. Write one sentence linking slow digestion to child-order pacing.

Download PDF

Download PDF

Landing: arXiv:0809.0822

Important disclaimer

Not investment advice. This Hub note is educational only. The paper is research—not a licence to trade, quote, or execute with real capital.

What this is not

• Not CEF / subsidy framing
• Not Quant or Macro course CTA
• Not a broker / signal-group pitch

Next step

Single course CTA: SoloQuant Academy's Introduction to Python Programming—get comfortable with functions, files, and plots first; then these microstructure PDFs become much less scary.