Economic Cycles and Firm Boundary Decisions in Hospitality: Extending the Governance-Choice Model

Abstract

This dissertation examines how economic cycles and firm-level factors together shape boundary decisions in the hospitality industry, focusing on firms' choices among acquisitions, alliances, and divestitures. Building on the governance-choice model of Villalonga and McGahan (2005), this study draws on transaction cost economics, the resource-based view, agency theory, relational governance theory, real options theory, and organizational learning theory to analyze how structural, strategic, and environmental factors influence governance decisions. The sample includes 1,979 firm-year observations from 78 publicly traded hospitality firms over the period 1991 to 2025. Transaction data were obtained from the Securities Data Corporation (SDC) Platinum database, patent data from the United States Patent and Trademark Office (USPTO), and macroeconomic cycle data from the National Bureau of Economic Research (NBER. A hierarchical multinomial logistic regression is used across three model specifications: a baseline model that replicates and extends Villalonga and McGahan (2005) in the hospitality context; an extended model that adds macroeconomic cycles; and a sub-sector analysis that estimates the full model separately for Hotels and Casinos, Restaurants, Airlines, Amusement Parks, Golf Courses, and Cruises. An important departure from Villalonga and McGahan (2005) is the use of multinomial rather than ordered logistic regression, because the three governance forms do not follow a single integration continuum in the hospitality context, alliance and divestiture represent fundamentally different strategic orientations, not just different points on a scale. The baseline model fits well (R² = .558) and confirms that governance choice in hospitality is strongly path-dependent. Governance specialization, the degree to which a firm has concentrated its boundary activity in a single form, is the strongest and most consistent predictor across all specifications and sub-sectors (B = −2.136, p < .001 for the acquisition versus alliance comparison), reducing the odds of switching governance forms by 88.2%. Market relatedness and cross-border status are the strongest transaction-level predictors. Insider ownership significantly increases acquisition propensity (B = 1.979, p < .001), consistent with entrenchment theory, while institutional ownership increases divestiture probability (B = 1.013, p = .002), consistent with exit-voice theory. The extended model confirms the main contribution of this dissertation: macroeconomic cycles significantly predict governance form choice beyond all firm-level factors. During recessions, hospitality firms move toward alliance governance, the most flexible and reversible form, consistent with real options theory. During expansions, firms move toward acquisition. An unexpected finding is that expansions also increase divestiture relative to alliance (B = .485, p = .040), suggesting that firms use favorable economic conditions to sell assets strategically at peak valuations (Igosheva et al., 2024; Sutton, 1998), a pattern this study calls opportunistic divestiture. The sub-sector analysis reveals that the economic cycle effect is concentrated almost entirely in Hotels and Casinos (Div vs. All: B = 1.458, p < .001; Acq vs. All: B = 1.889, p = .002). Airlines show governance choices dominated by path dependence and institutional constraints, with no significant cycle sensitivity.