Evaluating determinants of sticky costs and operations based earnings prediction models using air transportation industry data and validation of verifiable detail as a source of credibility in customer retention strategy disclosure

Chapter 1 examines the relevance of operations based variables predicting future earnings. We use United States Air Transportation (airline) industry data to identify operations based variables that provide predictive information beyond that found in earnings components. We show that the adjusted R-squared from accounting based prediction models can be increased up to 5.1 percentage points by including operations based variables. We also find that predictive information found in operations based variables is not fully impounded in financial analysts' earnings predictions. We show that operations based variables explain 7.7 percentage points of analyst forecast error. We observe that this effect is especially large after the terroristic attacks of September 11, 2001 suggesting analysts continued using historical prediction methods, even though the airline industry went through dramatic turmoil. Chapter 2 examines determinants of sticky cost behavior (asymmetric cost changes as revenue fluctuates). Cost accounting researchers examine sticky cost behavior to gain insights about management capacity decisions. The majority of the extant literature infers that costs decrease slower than they increase with demand fluctuations because management retains unused capacity in anticipation of future demand resurgence. I use airline industry data to provide evidence that sticky costs are associated with capacity and output selling price changes as management matches capacity with sales volume. I conclude that sticky costs arise when management adjusts capacity and sales volume (through pricing) if marginal cost of adding capacity is increasing as demand grows and marginal benefit from reducing capacity is decreasing as demand falls. There is a lack of research investigating sources of credibility in nonfinancial disclosures. This is likely due to difficulty in associating expected future performance with nonfinancial variables. Chapter 3 addresses the problem using customer retention strategy disclosures. Customer retention theory provides expectations regarding economic outcome which can be used to validate sources of credibility in nonfinancial disclosure. I find that firms that provide verifiable detail in their customer retention disclosure increase the persistence of positive abnormal performance relative to firms that do not. This result contributes by providing evidence that verifiable detail is a valid source of credibility in nonfinancial disclosure.