This dissertation consists of three essays. The first essay examines a general theory of information based on informal contracting. The measurement problem—the disparity of true and measured performances—is at the core of many failures in incentive systems. Informal contracting can be a potential solution since, unlike in formal contracting, it can utilize a lot of qualitative and informative signals. However, informal contracting must be self-enforced. Given this trade-off between informativeness and self-enforcement, I show that a new source of statistical information is economically valuable in informal con- tracting if and only if it is sufficiently informative that it refines the existing pass/fail criterion. I also find that a new information is more likely valuable, as the stock of existing information is large. This information theory has implications on the measurement problem, a puzzle of relative performance evaluation and human resources management. I also provide a methodological contribution. For tractable analysis, the first-order approach (FOA) should be employed. Existing FOA-justifying conditions (e.g. the Mirrlees-Rogerson condition) are so strong that the information ranking condition can be applied only to a small set of information structures. Instead, I find a weak FOA- justifying condition, which holds in many prominent examples (with multi- variate normal or some of univariate exponential family distributions). The second essay analyzes the effectiveness of managerial punishments in mitigating moral hazard problem of government bailouts. Government bailouts of systemically important financial or industrial firms are necessary ex-post but cause moral hazard ex-ante. A seemingly perfect solution to this time-inconsistency problem is saving a firm while punishing its manager. I show that this idea does not necessarily work if ownership and management are separated. In this case, the shareholder(s) of the firm has to motivate the manager by using incentive contracts. Managerial punishments (such as Obama’s $500,000 bonus cap) could distort the incentive-contracting program. The shareholder’s ability to motivate the manager could then be reduced and thereby moral hazard could be exacerbated depending on corporate governance structures and punishment measures, which means the likelihood of future bailouts increases. As an alternative, I discuss the effectiveness of shareholder punishments. The third essay analyzes how education affect workers’ career-concerns. A person’s life consists of two important stages: the first stage as a student and the second stage as a worker. In order to address how a person chooses an education-career path, I examine an integrated model of education and career-concerns. In the first part, I analyze the welfare effect of education. In Spence’s job market signaling model, education as a sorting device improves efficiency by mitigating the lemon market problem. In my integrated model, by contrast, education as a sorting device can be detrimental to social welfare, as it eliminates the work incentive generated by career-concerns. In this regard, I suggest scholarship programs aimed at building human capital rather than sorting students. The second part provides a new perspective on education: education is job-risk hedging device (as well as human capital enhancing or sorting device). I show that highly risk-averse people take high education in order to hedge job-risk and pursue safe but medium-return work path. In contrast, lowly risk-averse people take low education, bear job-risk, and pursue high-risk high-return work path. This explains why some people finish college early and begin start-ups, whereas others take master’s or Ph.D. degrees and find safe but stable jobs. / text