Paper: Stablecoins under Stress in a National Economy: Transaction-Level Evidence from Austrian Crypto-Asset Service Providers
Authors: Pietro Saggese, Michael Sigmund, Burkhard Raunig, Esther Segalla, Bernhard Haslhofer, Christos A. Makridis
Date: 2026-07-09
Estimated Reading Time: 45 minutes
The paper examines how cryptoasset activity intermediated by licensed Austrian crypto-asset service providers (CASPs) responds to major market shocks using a regulatory registry of provider-controlled blockchain addresses rather than heuristic address attribution. The dataset covers all 12 Austrian CASPs registered at the end of 2024 and reconstructs nearly 12 million on-chain asset transfers across Bitcoin, Ether, USDT, and USDC through May 2025. The analysis shows that Austrian CASPs are primarily integrated with global counterparties rather than other domestic CASPs, while a substantial share of observed on-chain volume consists of internal wallet management rather than economic transactions. The authors distinguish retail-like from institutionally mediated activity using wallet-tier interactions and transaction activity, allowing event studies that separate behavioral responses across participant groups. Across the Terra-Luna collapse, the FTX bankruptcy, and the Silicon Valley Bank failure, retail-like and institutional participants respond through different transaction patterns that aggregate statistics would obscure. Stablecoins do not consistently function as safe havens during stress, with responses varying by event and participant type rather than exhibiting uniform inflows. The paper concludes that registry-based transaction measurement provides a reproducible framework for monitoring crypto market risks within national financial systems.
Core insights
- Registry-based identification The study replaces heuristic blockchain attribution with a regulatory registry containing blockchain addresses reported by licensed Austrian CASPs. This allows direct observation of provider-controlled wallets and reduces uncertainty associated with traditional clustering methods.
- Global integration of Austrian CASPs Austrian CASPs intermediate roughly $30 billion of external on-chain activity while transactions between Austrian providers remain economically negligible. The results indicate that domestic providers participate primarily in international rather than domestic crypto networks.
- Retail and institutional separation The paper classifies counterparties using warm or cold wallet interaction and overall chain activity instead of relying only on transaction size. Retail-like addresses dominate transaction counts, while institutional participants account for most transferred value.
- Shock-specific behavioral responses The Terra-Luna collapse, FTX bankruptcy, and Silicon Valley Bank failure each generate different transaction patterns across assets and participant groups. Retail-like and institutional actors consistently adjust through different flow directions and asset choices.
- Stablecoin safe-haven evidence The paper does not find consistent support for the hypothesis that stablecoins serve as a safe haven during market stress. Instead, stablecoin flows depend on the nature of the shock and institutional redemption mechanisms, particularly for USDC during the Silicon Valley Bank episode.
The paper has direct implications for token demand by showing that stablecoin demand cannot be interpreted as a single defensive response during market stress. Instead, transaction-level evidence demonstrates that demand differs across participant groups and depends on the institutional structure governing redemption. This raises an important question: if redemption access differs between retail and institutional participants, how should aggregate stablecoin demand be interpreted during future market disruptions?
On the supply side, the study does not analyze token issuance schedules or protocol-level tokenomics. Rather, it examines the effective circulating movement of existing assets through licensed intermediaries. Internal transfers account for approximately 41% of observed transaction value, implying that aggregate blockchain volume may substantially overstate economically meaningful transfers. Observed transaction volume therefore cannot be treated as a direct proxy for market activity without separating operational wallet movements. The reward structure examined in the paper arises through market access rather than protocol incentives. During the Silicon Valley Bank event, institutional participants appear capable of exploiting direct USDC redemption while retail participants rely on secondary markets. Could differences in redemption rights create recurring incentives that shape participant behavior during future periods of financial stress? The paper presents this mechanism as consistent with observed transaction flows rather than direct observation of redemption activity. The distinction between retail-like and institutional participants also changes how liquidity should be interpreted. Retail participants dominate transaction counts, whereas institutional participants dominate transferred value. As a result, aggregate flow statistics combine behaviors that often move in opposite directions, masking underlying market dynamics observed at the transaction level.
From a tokenomics perspective, the findings suggest that demand, liquidity, and custody behavior cannot be evaluated solely through aggregate blockchain metrics. The registry-based methodology demonstrates that participant segmentation materially changes conclusions about market responses, while the absence of consistent stablecoin safe-haven behavior limits interpretations based only on aggregate inflows. The paper assumes that registry disclosures accurately identify provider-controlled wallets, while acknowledging that undisclosed addresses could introduce some incompleteness into the dataset.
