pith. sign in

arxiv: 2605.29689 · v1 · pith:MEU3TERMnew · submitted 2026-05-28 · 💻 cs.CE · q-fin.CP

Beyond TVL: An Explainable Risk Scoring Framework for Tokenized Real-World Assets

Rischan Mafrur , Khadijah This is my paper

Pith reviewed 2026-06-29 00:08 UTC · model grok-4.3

classification 💻 cs.CE q-fin.CP
keywords tokenized real-world assetsrisk scoring frameworkTVLliquidity riskconcentration riskon-chain metricsHerfindahl indexRWA markets
0
0 comments X

The pith

A multi-dimensional risk framework shows high-TVL tokenized RWAs can still carry substantial liquidity and concentration risks.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops an empirical and explainable risk scoring framework for tokenized real-world assets that moves beyond headline TVL or on-chain asset value. It constructs three risk dimensions—liquidity risk L, concentration risk C, and market-quality risk M—from public on-chain indicators including turnover, holder distribution, active-address activity, transfer frequency, and Herfindahl indices. Analysis of RWA.xyz data finds that several tokens with large on-chain values show high empirical risk due to limited transfer activity, low turnover, and concentrated ownership. Assets with broader participation and stronger on-chain activity instead display lower liquidity and concentration risk even when their headline values are smaller. This demonstrates that TVL alone can obscure important risks in tokenized asset markets.

Core claim

The framework evaluates three dimensions of risk—liquidity risk L, concentration risk C, and market-quality risk M—using observable indicators from public data. Several RWA tokens with substantial on-chain value exhibit high empirical risk because they combine limited transfer activity, low turnover, and concentrated ownership structures. In contrast, assets with broader participation and stronger on-chain activity display lower liquidity and concentration risk, even when their headline asset values are smaller.

What carries the argument

The explainable risk scoring framework that evaluates liquidity risk L, concentration risk C, and market-quality risk M from observable on-chain indicators including turnover, holder distribution, active-address activity, transfer frequency, and Herfindahl indices.

If this is right

  • High-TVL tokens can receive high risk scores when they show low turnover and concentrated holders.
  • Tokens with broader holder participation and higher transfer activity receive lower risk scores regardless of their TVL.
  • The framework supplies a transparent, data-driven method for comparing tokenized assets beyond headline valuation.
  • Risk scores can serve as a practical basis for liquidity assessment in RWA markets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Portfolio managers could incorporate the three risk dimensions when screening RWA exposures to reduce hidden concentration.
  • The same indicator set could be applied to non-RWA tokens to test whether TVL similarly misleads in other segments.
  • Repeated scoring over time would reveal whether risk profiles improve as markets mature or as new participants enter.

Load-bearing premise

The three risk dimensions constructed from on-chain indicators are assumed to provide a sufficient and representative assessment of risk without external validation or additional data sources.

What would settle it

A dataset of actual realized losses or market events on RWA tokens that shows no correlation between the computed risk scores and observed outcomes would falsify the framework.

read the original abstract

Tokenized real-world assets (RWAs) are often evaluated through headline indicators such as total value locked (TVL) or on-chain asset value. However, a large asset base does not necessarily imply low risk, since tokenized assets may remain illiquid, weakly traded, or highly concentrated among a small number of holders. Using public data from RWA.xyz, this paper develops an empirical and explainable risk scoring framework for tokenized RWA markets. The framework evaluates three dimensions of risk: liquidity risk $L$, concentration risk $C$, and market-quality risk $M$. These risk dimensions are constructed from observable indicators, including turnover, holder distribution, active-address activity, transfer frequency, and network concentration measured through Herfindahl indices. The analysis shows that several RWA tokens with substantial on-chain value exhibit high empirical risk because they combine limited transfer activity, low turnover, and concentrated ownership structures. In contrast, assets with broader participation and stronger on-chain activity display lower liquidity and concentration risk, even when their headline asset values are smaller. The findings demonstrate that TVL alone can obscure important risks in tokenized asset markets. By providing a transparent and data-driven risk scoring approach, this paper contributes to the empirical assessment of RWA liquidity and offers a practical basis for comparing tokenized assets beyond headline valuation metrics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper develops an empirical and explainable risk scoring framework for tokenized real-world assets (RWAs) using public data from RWA.xyz. It constructs three risk dimensions—liquidity risk L, concentration risk C, and market-quality risk M—from observable on-chain indicators including turnover, holder distribution, active-address activity, transfer frequency, and Herfindahl indices. The analysis qualitatively indicates that several high-TVL RWA tokens exhibit high empirical risk due to limited transfer activity, low turnover, and concentrated ownership, while assets with broader participation show lower risk despite smaller headline values. The findings suggest that TVL alone can obscure important risks in tokenized asset markets.

Significance. If the risk scores were shown to correlate with external market outcomes, the framework could provide a transparent, data-driven alternative to TVL for assessing RWA risks, aiding investors and regulators in the tokenized asset space.

major comments (2)
  1. [Abstract] Abstract: The central claim that TVL obscures 'important risks' requires evidence that the L, C, and M scores (built from turnover, holder distribution, active addresses, transfer frequency, and Herfindahl indices) correlate with realized outcomes such as liquidity events or redemption failures, but no such external validation, correlation analysis, or out-of-sample testing is reported.
  2. [Abstract] Abstract: No quantitative results, aggregation methods for combining indicators into L/C/M scores, validation steps, or error analysis are supplied, leaving the framework construction and qualitative findings without verifiable empirical support.
minor comments (1)
  1. [Abstract] Abstract: The exact formulas, weighting schemes, or normalization procedures used to construct the L, C, and M scores from the listed indicators are not specified.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback highlighting the need for stronger empirical grounding. We address each major comment below and outline planned revisions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The central claim that TVL obscures 'important risks' requires evidence that the L, C, and M scores (built from turnover, holder distribution, active addresses, transfer frequency, and Herfindahl indices) correlate with realized outcomes such as liquidity events or redemption failures, but no such external validation, correlation analysis, or out-of-sample testing is reported.

    Authors: We agree that demonstrating correlation between the constructed L, C, and M scores and external outcomes would strengthen the central claim. The current work focuses on an explainable framework derived directly from observable on-chain indicators and presents qualitative comparisons showing divergence from TVL. No such external validation or out-of-sample testing is included in the manuscript. In revision we will add correlation analysis where suitable public data on liquidity events or redemptions can be obtained, or explicitly discuss this as a limitation if data constraints prevent it. revision: partial

  2. Referee: [Abstract] Abstract: No quantitative results, aggregation methods for combining indicators into L/C/M scores, validation steps, or error analysis are supplied, leaving the framework construction and qualitative findings without verifiable empirical support.

    Authors: We acknowledge that the manuscript as submitted does not report the specific quantitative results, aggregation formulas used to combine indicators into the L, C, and M dimensions, validation procedures, or error analysis. These elements are necessary for verifiability. We will expand the methods and results sections in the revision to include the aggregation methods, quantitative score values, and any available validation or sensitivity checks. revision: yes

Circularity Check

0 steps flagged

No circularity; empirical construction from external data

full rationale

The paper constructs L, C, and M risk dimensions directly from public on-chain observables (turnover, holder distribution, active-address activity, transfer frequency, Herfindahl indices) sourced from RWA.xyz. These scores are then contrasted with TVL to show divergence. No equations reduce a claimed prediction or result to a fitted parameter or self-referential definition; no self-citations are load-bearing; the framework is a transparent aggregation of independent external metrics without any renaming of known results or smuggling of ansatzes. The derivation chain remains self-contained against the cited public data sources.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no information on free parameters, axioms, or invented entities; the framework is described solely in terms of observable indicators without further specification.

pith-pipeline@v0.9.1-grok · 5766 in / 1097 out tokens · 44569 ms · 2026-06-29T00:08:47.165436+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

13 extracted references

  1. [1]

    The market for “lemons”: quality uncertainty and the market mechanism.Quarterly Journal of Economics84(3), 488–500

    Akerlof, G.A., 1970. The market for “lemons”: quality uncertainty and the market mechanism.Quarterly Journal of Economics84(3), 488–500

    1970

  2. [2]

    Illiquidity and stock returns: cross-section and time-series effects

    Amihud, Y., 2002. Illiquidity and stock returns: cross-section and time-series effects. Journal of Financial Markets5(1), 31–56

    2002

  3. [3]

    BIS Bulletin No

    Bank for International Settlements, 2023a.The Tokenisation Continuum. BIS Bulletin No. 72

  4. [4]

    BIS Bulletin No

    Bank for International Settlements, 2025a.The Rise of Tokenised Money Market Funds. BIS Bulletin No. 115

  5. [5]

    Bank for International Settlements, 2025b.Does Tokenisation Fill Gaps in Traditional Real Estate Markets?BIS Working Paper No. 1311

  6. [6]

    BIS/CPMI report

    Committee on Payments and Market Infrastructures, 2024.Tokenisation in the Context of Money and Other Assets. BIS/CPMI report

    2024

  7. [7]

    IMF FinTech Note 2025/001

    International Monetary Fund, 2025.Tokenization and Financial Market Inefficiencies. IMF FinTech Note 2025/001

    2025

  8. [8]

    IMF Note 2026/001

    International Monetary Fund, 2026.Tokenized Finance. IMF Note 2026/001. 10

    2026

  9. [9]

    Final report

    International Organization of Securities Commissions, 2025.Tokenization of Financial Assets. Final report

    2025

  10. [10]

    Continuous auctions and insider trading.Econometrica53(6), 1315– 1335

    Kyle, A.S., 1985. Continuous auctions and insider trading.Econometrica53(6), 1315– 1335

    1985

  11. [11]

    Is market fragmentation harming market quality?Journal of Financial Economics100(3), 459–474

    O’Hara, M., Ye, M., 2011. Is market fragmentation harming market quality?Journal of Financial Economics100(3), 459–474

    2011

  12. [12]

    Tokenize everything, but can you sell it? RWA liquidity challenges and the road ahead.arXiv preprint arXiv:2508.11651

    Mafrur, R., 2025. Tokenize everything, but can you sell it? RWA liquidity challenges and the road ahead.arXiv preprint arXiv:2508.11651

    arXiv 2025

  13. [13]

    Empirical evidence on the ownership and liquidity of real estate tokens

    Swinkels, L., 2025. Empirical evidence on the ownership and liquidity of real estate tokens. In:Blockchain, Crypto Assets, and Financial Innovation: A Decade of Insights and Advances. Springer, pp. 434–467. 11

    2025