AbstractThis paper presents a simple model of the provision of higher educational services that considers and exemplifies nonlinear, stochastic, and potentially chaotic processes. I use the methods of system dynamics to simulate these processes in the context of a particular sociologically interesting case, namely that of the Turkish higher education sector during the recent global economic crisis. I undertake simulations for two variables, namely for the quality-adjusted performance and the technology-induced expected employment. Simulations for the quality-adjusted performance display stochastic fluctuations around a modestly rising deterministic trend. Chaotic performance trajectories are shown to be possible for certain parameter values; however, given the historical experience and observed current tendencies in higher education in Turkey, they appear to be highly unlikely. Simulations for expected employment indicate that increases in the levels of technology utilization in education could lead to improved perceptions of employment prospects and potentially help to increase the level of employment in the economy.Keywords: Stochastic simulation * Chaotic patterns * Education * TurkeyJEL codes: I21 * I23(ProQuest: ... denotes formulae omitted.)Education has a number of complexity-generating, interrelated features that present considerable challenges to the orthodox economic theory. First, it creates, for individuals in particular and society in general, positive dynamic externalities that are too enormous to ignore and too difficult to measure and internalize. In the presence of such externalities, underproduction of output or services is a likely intertemporal outcome, leading to inefficiency at the societal level. Second, education creates interdependencies for the strategically interacting agents that make atomistically optimizing behavior problematic, producing a wide range of possibilities for cooperation and competition with complicated patterns of stability and optimality. Third, it involves a wide range of nonlinearities that are prone to potentially chaotic behavior. A small change in the initial conditions could produce significant differences in outcomes over time. Fourth, it has, in part, certain indeterminate, stochastic, and/or fundamentally uncertain dimensions that render the educational outcomes open-ended and only partially predictable.There are a number of works in the literature that explore the selected dimensions of these complexities. Among these are Barlas, Diker, and Polat (1997); Barlas and Diker (2000); Hazlett (2000); De Fraja (2002); Marginson (2007); Kara (2007; 2013); Viaene and Zilcha (2009); Loomis and Rodrigues (2009); Eckwert and Zilcha (2012); Budge, Beale, and Lynas (2013); Velasco (2014); Aasen, Proitz, and Sandberg (2014); Schumacher, Dias, and Tebaldi (2014); and Kuziemko (2014).The topics covered by these works range from the dynamic and strategic issues in education to the design of optimal education policies. Needless to say, each work was certainly limited in scope and focused a particular aspect of education. Barlas and Diker (2000) focused, for instance, on the longterm, dynamic, strategic management problems, such as growing student/faculty ratios, poor teaching quality, and low research productivity (p. 331). The analysis they present could be considerably enriched by a proper account of positive externalities associated with education. Such externalities are indicated in Hazlett (2000). The creation of such education-induced spillover benefits has a lot to do with the public and private investment in education, the public and private organization of education, and creative designs of optimal educational policies, which are explored, however partially, in many works including De Fraja (2000), Marginson (2007) and Eckwert and Zilcha (2012).Within this general stream of research education, there are many saddle issues that are delicately dealt with. …