Metadaten







Promotionsordnung



Kumulative Dissertation



Titel


Institutions in Education and the Importance of Skill Formation

Titel (englisch)



Autor/Autorin


Jaag, Christian

2. Autor/Autorin



Geburtsdatum


10.07.1976

Geburtsort


Zürich

Matrikelnummer



Schlagwörter (GND)


Bildungsökonomie; Schule; Lehrer; Schüler; Humankapital

DDC (Dewey Decimal Classification)


Wirtschaft - 330

Freie Stichwörter (deutsch)


Bildung; Schulen; Bildungsökonomik; Institutionen; Humankapital; Alterssicherung; Demographie; Volkswirtschaft

Freie Stichwörter (englisch)


Education; Schools; Institutions; Human Capital; Pension System; Demographics; Economics

Kurzfassung


Im ersten Kapitel wird Bildung als Beitrag zur Humankapitalbildung analysiert. Zunächst wird die Bedeutung von Humankapital sowohl aus individueller als auch aus sozialer Perspektive betrachtet. Es existiert bereits eine grosse Anzahl Studien über den Zusammenhang zwischen Bildungsressourcen und Bildungserfolg. Diese sind mehrheitlich empirisch ausgerichtet und kommen zu Ergebnissen, welche der Intuition oft zuwiderlaufen. Um den fehlenden empirischen Zusammenhang zwischen aufgewendeten Ressourcen und Bildungserfolg zu erklären, wird auf institutionelle Eigenschaften des Bildungssystems zurückgegriffen. Wir schlagen im zweiten Kapitel einen Modellrahmen vor, welcher es erlaubt, die empirischen Ergebnisse aus theoretischer Sicht zu interpretieren und zu verstehen. So kann zum Beispiel der fehlende Zusammenhang zwischen Klassengrösse und Bildungserfolg durch die endogene Verteilung von Schülern auf verschiedene Klassen erklärt werden. Im dritten Kapitel werden optimale Anreizverträge für Lehrer analysiert. Insbesondere die Bedeutung einer Vielzahl verschiedener Effort-Dimensionen erschwert den Abschluss effizienter Verträge. Wir zeigen unter anderem die Bedingungen, unter welchen Anreize für gesamte Teams oder Schulbesuche als Lehreranreize optimal sind. Die Existenz intrinsischer Motivation beeinflusst die Gestaltung effizienter Verträge ebenfalls. Mit freier Schulwahl und Wettbewerb zwischen verschiedenen Bildungsinstitutionen setzt sich das vierte Kapitel auseinander. Explizit berücksichtigt werden die Präferenzen der Schüler, bzw. ihrer Eltern bezüglich des Standorts einer Schule und der Ausrichtung ihres Fächerkatalogs. In einer empirischen Auswertung der Ergebnisse der PISA-Studie finden wir, dass Schulen, welche einem Wettbewerbsdruck ausgesetzt sind, einen grösseren Lernerfolg ihrer Schüler aufweisen. Im fünften Kapitel wird ein rechenbares Modell des allgemeinen Gleichgewichts vorgestellt, welches individuelle Bildungsentscheidungen in einen makroökonomischen Kontext stellt. In einer Anwendung wird gezeigt, dass eine Verlängerung der Erwerbsphase im Lebenszyklus die Humankapitalproduktion anregt. Über diesen Mechanismus können die ökonomisch problematischen Effekte im demographischen Übergang zusätzlich zum direkten Effekt der Erweiterung der Erwerbsbevölkerung gemildert werden.

Kurzfassung (englisch)


The first chapter introduces the concept of education as a process of human capital formation. It first sets out the issue, namely the economic importance of education, thus motivating the main part of the dissertation. Education is shown to not only augment individuals' human capital, it serves also as a sorting device in the labor market. The human capital view being the empirically more important aspect of education in explaining individual wage dispersion, the chapter continues analyzing the social importance of human capital, especially in growth theory. The main focus is then on the actual education process over an individual's life-cycle and in particular at school. To shed light on the missing empirical input-output relationship in schooling, it is the main goal of the dissertation to provide clues for optimum school organization and institutions, such as incentive structures and competition, in order to maximize educational attainment and achievement from a theoretical perspective. The first chapter concludes with an overview of the remaining chapters of the dissertation. There is a large body of literature on the effect of educational resources on student performance, such as teacher qualification, class size, and physical resources in school. It is dominated by empirical studies which often find ambiguous effects of resource spending on student outcomes. The unique contribution of the second chapter is the provision of a framework to study educational production with differentiated input factors which allows for closed-form solutions. We try to interpret the empirical findings on the basis of a simple theoretical model of educational production: Class size, employed school resources and student effort are endogenously determined in order to account for differences in educational achievement. We also discuss the choice of inte-grated vs. segregated classes. Optimum class size and school quality increase with higher discipline, while in equilibrium, overall classroom disruption is equal in all classes. The third chapter considers hidden teacher effort in educational production and discusses the implications of multiple teacher effort dimensions on optimum incentive contracts in a theoretical framework. The analysis of educational production in a multitask framework is a new and unique contribution of this chapter to the economics of education. We first characterize the first-best and second-best outcomes. The model is extended to address specific questions concerning teacher incentive schemes: We compare input- to output-based account-ability measures and study the implication of the level of aggregation in performance measures. Against the background of the empirical evidence on the effectiveness of teacher incentives, we argue that performance measures should be as broad as possible. Further, we present the optimum contract for motivated teachers. Finally, if education is produced in teacher teams, we establish the conditions for optimum team-based and individual incentives: The larger the spillover effects across teacher efforts and the better the measurability of educational achievement, the stronger the case for team-based incentives. The fourth chapter considers the influence of spatial competition on education and its effect on students' school choice and educational achievement by explicitly modeling the educational production process and the students' participation decision. Education at school is a function of teacher effort and class size. Students decide which school to attend on the basis of an assessment of the associated costs and prospective benefits from doing so. We analyze how competition between schools affects equilibrium resource spending and school diversity as well as the level and distribution of student attainment and welfare. The consideration of spatial aspects of school choice without recourse to vertical differentiation is a unique contribution of this chapter. We argue that schools in metropolitan areas with short ways to school and many potential students face fiercer competition which increases school productivity and student performance. Overall learning time in school is constant in the probability that students behave well if students are segregated by type. However, better behaved students have a higher achievement due to higher optimum resource spending. Finally, we support our argument by an empirical analysis of student performance in various matura schools in Switzerland. The fifth chapter deals with two issues concerning the effects of population aging on education decisions in the presence of a pay-as-you-go pension system: We first analyze the effects of an aging population per se on individual education choices and the production structure. Second, we study the implications of postponed retirement which is often proposed as a measure to cope with the economic challenges of increased longevity. Our study uses a dynamic general equilibrium framework with overlapping generations and probabilistic aging. The model allows for capital-skill complementarity in the production of final output. As a response to population aging, in a small open economy with a fixed interest rate, our first simulation shows that GDP is depressed due to an adverse effect on skill choice and labor supply. We then introduce postponed retirement as a potentially dampening policy measure due to its encouragement of human capital formation. However, since there is less private saving in this scenario, the overall effect on GDP is even worse than in the pure aging scenario.

Universität


Universität St.Gallen

Referent/Referentin


Keuschnigg, Christian (Prof. Dr.)

Korreferent/Korreferentin


Heijdra, Ben (Prof. PhD)

Erweitertes Diss. Komitee



Fachgebiet


Wirtschaftswissenschaften

Sprache


ENG

Promotionstermin (dd.mm.yyyy)


03.04.2006

Erstellungsjahr (yyyy)


2006

Dokumentart


Dissertation

Format


PDF

Dissertationsnummer


3191

Quelle



PDF-File


dis3191.pdf

Dokumentverknüpfung


Link zu diesem Dokument







letzte Änderung: 11/15/2006 - Allgemeine rechtliche Informationen - Datenschutz [ Nach oben ]