Supervisory research · Market microstructure · Financial stability

From liquidity fragility research to market monitoring

MicroDash is a research prototype that studies quote and trade-price paths to identify unusual market adjustment patterns and support further investigation of potential liquidity fragility. It is designed to complement, not replace, conventional liquidity indicators and supervisory judgement.

QV · Quote-path variation

Tracks instability in quoted midpoint paths relative to net price displacement. Elevated QV can indicate intensive quote revisions without a comparable directional move.

PM · Trade-price excursions

Measures the largest trade-price excursion relative to net movement. Read alongside QV to distinguish market adjustment patterns.

Supervisory interpretation

Compare futures markets, examine corroborating spread and depth conditions, and prioritize questions for human review. Flags are not findings of misconduct or proof of impairment.

Evidence and limitations. Research results and scenario estimates require out-of-sample validation, false-alarm assessment and market-specific calibration before operational supervisory use. Cross-market scenario estimates are predictive associations, not established causal spillovers.
MicroDash · Market liquidity fragility monitor

Market monitor: what is flagged, where stress may spread, and what to check

MicroDash reads quote and trade-price paths in futures markets to flag liquidity fragility early. This page reproduces one snapshot for the core contracts discussed in the talk: S&P 500 E-mini, 10-year Treasury, WTI crude and gold futures, plus international index futures.

Research-universe liquidity status NORMAL
Systemic score · 17 assets0.09
Fragile · 17-asset universe0 / 17
Futures shown that are stressed1 / 9

A low systemic score suggests stress is not synchronised across the model's full research universe; this does not establish that all markets are liquid.

Scope: The table below displays nine futures contracts. The systemic score and fragile count refer to the original 17-instrument model universe and are not recalculated for the displayed subset. No individual-company equities are shown. Data are illustrative, not live.
1 · What is flagged, and why

S&P 500 E-mini is the only stressed contract in this illustrative futures view

Oil, gold and Treasuries are normal on all four indicators. Select a contract to see its card.

ContractRegimeQVPMSpreadDepth
NormalStressedFragile Flags relative to the 17-instrument universe
Reading the signal

E-mini: elevated quote-path variation, limited price movement

Fragile liquidity Crisis state Stable Emerging stress high QV, low PM high QV, high PM low QV, low PM low QV, high PM −10+1+2+3 +3+2+10−1 Price momentum, PM z-score → Quote volatility, QV z-score → Treasuries, gilts, FTSE,Nikkei, gas, VIX E-mini

High QV, low PM: dealers are repricing quotes intensively while trade prices barely move. This can motivate further investigation; it is not proof of deteriorating liquidity.

QV, quote volatilityHow unstable the prices are at which liquidity is offered.
PM, price momentumHow persistently prices keep moving one way.

Research states from Karam (2026). Boundaries drawn at the universe mean for illustration; contracts with reported z-scores shown. Dashboard labels (Stressed, Fragile) use their own operational thresholds.

2 · Where could stress travel next?

Treasuries and Europe within 2–4 hours; Asia within 4–8 hours

Estimated rise in each market’s probability of becoming stressed after a severe shock to S&P 500 E-mini futures. Which market moves first offers clues about where stress started.

Shock sourceE-miniS&P 500 futures, severe liquidity shock

All six channels rated Medium risk. Model-specified scenario estimates: predictive associations, not measured or causal spillovers.

3 · What should the supervisor check?

Monitoring prompts and built-in checks

01Watch

S&P 500 E-mini and 10-year Treasury futures, more often.

02Confirm

False-alarm filter: act only if Treasury spreads widen or depth falls.

03Reassess

Refresh the cross-asset state after 30 minutes.

04Escalate

If stress persists beyond 1 hour, prepare an escalation note.

An alert is a prompt to investigate, not an automatic decision to intervene. The supervisor decides, not the model.