# Regulatory Volatility in Digital Supply Chains: An Information Systems Analytics Study of Decision-Maker Risk Perceptions and Sustainability-Related Outcomes

**Authors:** Kenneth David Strang (University of the Cumberlands, Williamsburg, KY, USA; W3-Research) — ORCID 0000-0002-4333-4399; Narasimha Rao Vajjhala (Department of Computer Science, American University in Bulgaria, Blagoevgrad, Bulgaria) — ORCID 0000-0002-8260-2392
**Type:** Journal article (open access)
**Source:** Sustainability, 18(18), Article 9359
**Published:** 2026-09-11
**DOI:** https://doi.org/10.3390/su18189359
**Canonical page:** https://www.narasimharao.net/research/regulatory-volatility-digital-supply-chains-esg-risk/
**Indexing:** Scopus Q1 (CiteScore 8.9) · Web of Science Q2 · Impact Factor 4.1 (JCR 2025)

## Summary

**Question.** How do manufacturing supply chain decision-makers perceive regulatory-volatility risks — labor, environmental, customs, ownership, military-logistics, and distribution — and do those perceptions predict whether a supply chain engagement succeeds or fails?

**Finding.** Analysing 1,988 anonymized decision-maker records from a multinational logistics firm, the study finds that the six risks do not form a single "regulatory volatility" construct, and that the sustainability-motivated risks (labor and environmental) were rated lowest of all — a median labor severity of 0 and environmental severity of 1 on a 0–5 scale — even during a period of record forced-labor enforcement. A logistic regression separated success from failure perfectly in-sample, but only because two near-duplicate items leaked the outcome.

**Why it matters.** The perception gap exposes firms to "sustainability leakage": sudden enforcement pushes them to exit suppliers rather than remediate. For anyone building AI decision support from organizational records, the perfect-accuracy result is a concrete warning about label leakage. The authors recommend multidimensional (not composite) risk dashboards and provenance-aware data governance for supply chain analytics.

## Key findings

1. The study analysed 1,988 anonymized decision-maker records from the digital lessons-learned repository of a multinational supply chain logistics firm.
2. The six perceived regulatory-volatility risks did not converge on a unified construct: sampling adequacy was weak (KMO = 0.470), and parallel analysis retained three single-indicator factors — distribution, environmental, and ownership risk.
3. Sustainability-motivated risks were rated lowest: median labor-risk severity was 0 and environmental-risk severity was 1 on 0–5 scales, during a period of record forced-labor enforcement.
4. This perception gap invites "sustainability leakage" — surprise enforcement provokes supplier exit and sourcing flight rather than remediation in scrutinized regions.
5. Binary logistic regression separated success from failure perfectly in-sample (accuracy = AUC = 1.000), driven by the near-collinear distribution and supply items (r = 0.915); no individual predictor effect is identified.
6. The complete-separation pattern is a warning of label leakage when digitized organizational records are mined for AI-based decision support.

## Study at a glance

| Item | Detail |
|---|---|
| Research question | Do perceived regulatory-volatility risks form one construct, and do they predict supply chain engagement outcomes? |
| Design | Retrospective analysis of archival records from an information systems perspective |
| Data | 1,988 fully anonymized decision-maker records from the digital lessons-learned repository of a multinational supply chain logistics firm |
| Variables | Perceived severity (0–5) of six risks: labor, environmental, customs, ownership, military-logistics, distribution |
| Methods | Descriptive statistics; exploratory factor analysis with parallel analysis; binary logistic regression with classification diagnostics |
| Main result | No unified construct (KMO = 0.470); sustainability risks rated lowest; perfect in-sample classification (AUC = 1.000) traced to near-collinear items (r = 0.915) |
| Implication | Use multidimensional risk dashboards and provenance-aware data governance in supply chain analytics |

## Abstract

Geopolitical conflict, tariff shocks, and rapidly evolving environmental, social, and governance (ESG) regulation have made global supply chains markedly less predictable. This study examined how manufacturing supply chain decision-makers perceive six regulatory-volatility risks, including labor, environmental, customs, ownership, military-logistics, and distribution, and whether those perceptions predict recorded engagement outcomes. Adopting an information systems perspective, we retrospectively analyzed 1988 fully anonymized decision-maker records from the digital lessons-learned repository of a multinational supply chain logistics firm, applying descriptive statistics, exploratory factor analysis with parallel analysis, and binary logistic regression with classification diagnostics. The perceived risks did not converge on a unified regulatory-volatility construct: sampling adequacy was weak (KMO = 0.470), and parallel analysis retained three single-indicator factors (distribution, environmental, and ownership risk) that were close to orthogonal apart from one moderate environment-distribution association. Critically, the sustainability-motivated risks were rated lowest, with a median labor severity of 0 and an environmental severity of 1 on 0–5 scales, during a period of record forced-labor enforcement; this perception gap invites sustainability leakage, whereby surprise enforcement provokes supplier exit and sourcing flight rather than remediation within scrutinized regions. The logistic regression separated success from failure perfectly in-sample (accuracy = AUC = 1.000), driven jointly by the near-collinear distribution and supply items—a complete-separation pattern warning of label leakage when digitized organizational records are mined for artificial intelligence-based decision support. The two items are near-duplicate measures (r = 0.915), so no individual predictor effect is identified. The findings favor multidimensional rather than composite digital risk dashboards, provenance-aware data governance for supply chain analytics pipelines, and digitally enabled ESG compliance under volatile regulation, contributing an empirically grounded information systems lens to sustainable supply chain management.

## Key terms

- **Regulatory volatility:** Rapid, hard-to-predict change in the rules that govern trade and operations — tariffs, customs regimes, labor and environmental (ESG) regulation — that makes supply chains less predictable.
- **Sustainability leakage:** As used in this study: when surprise ESG enforcement leads firms to exit suppliers or move sourcing elsewhere instead of remediating conditions in the scrutinized region.
- **Label leakage:** When a predictor variable effectively encodes the outcome being predicted, producing unrealistically perfect model accuracy that will not hold in real use.

## Limitations

- Perfect in-sample separation means no individual predictor effect can be estimated; the two driving items are near-duplicate measures.
- Data come from one multinational logistics firm’s repository, analysed retrospectively.

## How to cite

Strang, K. D., & Vajjhala, N. R. (2026). Regulatory Volatility in Digital Supply Chains: An Information Systems Analytics Study of Decision-Maker Risk Perceptions and Sustainability-Related Outcomes. Sustainability, 18(18), 9359. https://doi.org/10.3390/su18189359

```bibtex
@article{strang2026regulatory,
  title = {Regulatory Volatility in Digital Supply Chains: An Information Systems Analytics Study of Decision-Maker Risk Perceptions and Sustainability-Related Outcomes},
  author = {Strang, Kenneth David and Vajjhala, Narasimha Rao},
  journal = {Sustainability},
  volume = {18},
  number = {18},
  pages = {9359},
  year = {2026},
  publisher = {MDPI},
  doi = {10.3390/su18189359},
  url = {https://doi.org/10.3390/su18189359}
}
```