Essays on financial reporting

In this dissertation I study the informational and efficiency implications of financial reporting. Chapter 1 provides a multi period model of financial reporting in which firms' internal control systems (ICS) are uncertain to investors and managers are allowed to misreport information when firms' ICS are ineffective. I use this model to understand the notion of earnings quality, provide a new measure of earnings quality based on the likelihood that firms' ICS are effective, and evaluate four commonly used measures of earnings quality. I find that---unlike the Smoothness and Predictability of reports---the Volatility of Prices induced by the reports and the Persistence of reports are valid measures of earnings quality.;Chapter 2 extends the previous model and considers the interaction between financial reporting and trading. In this context I address the following question: is the presence of large frauds a sign that firms' ICS are bad, i.e., that firms' ICS fail with a high probability? I establish, on the contrary, that the presence of few but very large frauds indicates that firms' ICS are good, i.e., they are effective with high probability. By contrast, the extended presence of small manipulations reveals that firms' ICS are effective with low probability. Next, I examine the effects of allowing managers to include non verifiable disclosures (such as projections) as part of firms' financial statements. I show that even though this may trigger some fraudulent an inefficient transactions, on average, both the market's liquidity and its efficiency improve when non verifiable disclosures are allowed.;In Chapter 3, I consider the allocative efficiency of three accounting methods that can alternatively be used to record the value of an acquisition: the Purchase Price, Exit Value, and Value-in-Use methods. I rank the efficiency of these three valuation standards, modelling the acquisition as a first price auction with reserve price, and assuming that bidders are partially concerned with the income statement implications of the transaction. The main result shows that Value-in-Use is always the most efficient method and Purchase Price is the least efficient method whenever the gains from trade are sufficiently large.