Chapter 1. While financial crises tend to be preceded by credit booms, most credit booms do not end up in crises. Crises typically occur when there is also a persistent slowdown in productivity growth. I develop a model in which risk of crisis emerges endogenously during boom because of increased fragility of the banking sector. Banks raise financing from households to invest in long-term projects, but their ability to do so is limited by moral hazard. Demandable deposits create discipline by exposing misbehaving banks to runs, and thus help them increase external financing. Normally, banks finance themselves with a mix of equity and deposits that maximizes discipline, but ensures that they always remain solvent. When growth prospects become sufficiently strong, however, worsening moral hazard induces banks to rely exclusively on deposit financing, which enables a boom in credit, asset prices, and investment. If the anticipated growth fails to materialise, though, the excessive deposit financing leads to a systemic banking crisis. I also study policy implications of the model.
Chapter 2. I document that future returns on real assets in dry bulk shipping are predictable and negatively related to ship earnings, prices, and industry investment during global economic recessions, but are not forecastable in expansions. The predictability in recessions is statistically significant and economically large. The empirical evidence points against standard risk-based explanations of return predictability, and is also inconsistent with previously suggested behavioural explanations. I propose and formalise a theory of countercyclical return predictability arising due to liquidity constraints faced by the industry participants in downturns. When cash flows dry up in recessions, financial troubles force firms to sell assets to their liquidity constrained peers, resulting in plunges in ship prices and high expected returns to the buyers. Liquidity shortages also lead to slumps in industry investment.
Chapter 3. I study managerial compensation in financially constrained firms with high growth opportunities. Such firms can often profitably reinvest their operating incomes and pay no dividends. Absence of reliable performance indicators together with asymmetric information create acute agency problems. Stock-based compensation schemes, notoriously popular during the tech bubble of the 1990s, provide the necessary incentives for the manager to exert costly effort to pursue further growth. However, in line with the previous literature, I find that they may also induce the manager to conceal bad news about the future prospects of the firm, in order to maintain a high stock market valuation. This leads to a substantial overinvestment in the firm's assets, and expands a bubble-like surge in the firm's stock price, followed by an inevitable crash. I find that a stock-based compensation contract that delays payments by several years effectively deters the manager from concealing, while still providing incentives to work hard.
