Hotels that combine AI, cognitive analytics and dynamic strategic capabilities report stronger resilience and better performance, and regulatory and peer pressures amplify these benefits.

Main Finding

Hotels’ AI capabilities, cognitive analytic capabilities, and dynamic capabilities each strengthen organizational resilience, and greater resilience improves hotel performance. Institutional pressures (coercive, normative, mimetic) amplify the positive effect of resilience on performance. In short: AI matters, but its productivity depends on complementary cognitive and dynamic capabilities and on the institutional context.

Key Points

• Theoretical framing: integrates the Dynamic Capabilities View (sensing–seizing–reconfiguring) with institutional theory (coercive, normative, mimetic pressures).
• Hypotheses tested:
  • H1: Organizational resilience → hotel performance (positive).
  • H2: AI capabilities → organizational resilience (positive).
  • H3: Cognitive analytic capabilities → organizational resilience (positive).
  • H4: Dynamic capabilities → organizational resilience (positive).
  • H5: Institutional pressures moderate (strengthen) the resilience → performance link.
• Capabilities are complementary: AI increases sensing and decision support but yields performance only when cognitive analytic readiness and dynamic reconfiguration capabilities exist.
• Institutional pressures do not just constrain firms; they can facilitate capability deployment and raise the returns to resilience.
• Managerial takeaway: align digital transformation (AI) with managerial sense‑making and reconfiguration capacity to realize performance gains.

Data & Methods

• Empirical context: hotel industry in the United Arab Emirates (selected for tourism significance, regulatory environment, and AI adoption).
• Sample: survey of 400 managerial-level employees from UAE hotels.
• Analytical approach: Partial Least Squares Structural Equation Modeling (PLS-SEM) to test the measurement model, structural paths, and moderation by institutional pressures.
• Constructs (examples): AI capability; cognitive analytic capability (information + technology); dynamic capabilities (sensing, seizing, reconfiguring); organizational resilience; institutional pressures (coercive, normative, mimetic); hotel performance (growth, survival).
• Design limitations (noted by authors / implied): cross-sectional, single-country, self-reported measures — causal inference limited; no granular temporal dynamics.

Implications for AI Economics

• Complementarities matter: Economic models and empirical studies of AI productivity should incorporate complementary organizational capabilities (managerial cognition, reconfiguration skills). Ignoring complementarities risks overstating or misestimating AI’s returns.
• Institutional context as a multiplier: Institutional factors (regulation, professional norms, imitation) alter marginal returns to AI-driven resilience. Macro and micro studies should include institutional variables to explain heterogeneity in AI payoff.
• Measurement guidance: Researchers should move beyond binary AI adoption indicators and measure AI capability intensity, integration with decision processes, and interactions with cognitive/dynamic capabilities.
• Policy relevance: Regulators and industry bodies can raise sectoral productivity from AI by setting standards, building norms, and encouraging diffusion channels that increase legitimacy and lower adoption frictions (e.g., data governance, training incentives).
• Empirical strategy recommendations:
  • Use longitudinal or quasi‑experimental designs (difference‑in‑differences, instrumental variables, panel methods) to identify causal effects of AI-capability investments on resilience and performance.
  • Estimate heterogeneous treatment effects: stratify by institutional strength, firm size, human capital, legacy systems.
  • Model spillovers and equilibrium effects: assess how imitation (mimetic pressure) and regulatory changes affect industry-wide AI adoption and aggregate productivity.
• Cost–benefit modeling: Incorporate adjustment and complementary investment costs (training, process redesign, sensing/seizing capacity) when evaluating ROI of AI in service sectors; returns will be higher where institutional pressures encourage or enforce capability alignment.
• Industry selection and external validity: Findings from high-tourism, high-regulation settings (like the UAE hotels) suggest stronger institutional multipliers — general equilibrium or cross-country analyses should test how institutional heterogeneity shapes AI’s macroeconomic contribution.

If you’d like, I can: - Draft a brief empirical research design (data needs and identification strategies) to estimate causal returns to AI under varying institutional regimes, or - Translate these implications into econometric specifications and variable lists for a replication/extension study.