Search

Chapter 1 examines peer effects in the take up of welfare within a school setting- free school meals. Data from the Pupil Level Annual School Census, which collects information on every child attending a school in the UK, is used, defining the peer group using the child's ethnicity and language. Hypotheses about the nature of the peer effect are tested. I exploit the introduction of cashless catering systems which removes stigma. Secondly, by comparing the peer effect for those who have claimed in previous years with those who have not, information is tested. Finally, I test a role model hypothesis. All three have an impact and information is most important for the most deprived. Chapter 2 examines the casual effect of education on teen births and child health. The increase in school leaving age to 15 and 16 increased the years of completed schooling; however, there was no effect of the introduction of new examinations. The compulsory schooling law change to 16 did increase the probability of holding a qualification and reduced the probability of a teen birth. Furthermore, the impact of the increase to 16 also - lead to improvements in birth outcomes, this was primarily driven by the mother having a more educated partner and a reduction in poverty. Other measures of child health are mixed and mostly did not improve. Chapter 3 (with Michele Belot, published in Journal of Health Economics (2011)) pro- vides field evidence on the effect of diet on educational outcomes, exploiting a campaign lead by celebrity chef J amie Oliver in 2004, which introduced drastic changes in the meals offered in the schools of one Borough - Greenwich - shifting from low-budget processed meals towards healthier options. We evaluate the effect of the campaign on educational outcomes using a difference in differences approach; using the neighbouring Local Edu- cation Authorities as a control group. We find evidence that educational outcomes did improve and that authorised absences - which are most likely linked to illness and health - fell by 14%.

This thesis extends a recent vein of literature which investigates the implications of wage decisions for the optimal design of the monetary regime when the goods market is monopolistically competitive. Chapter II abstracts from the existence of stochastic shocks and examines in detail the resulting macroeconomic wage-setting externality. It points out that when the monetary instrument is set after wage determination, there exists a monetary rule which can induce non-atomistic unions to set the market-clearing nominal wage, and which therefore is fully credible. This result is shown to be implicit in (and to have been overlooked by) previous contributions, and its sensitivity to the precise formulation of the union objective function is discussed. Chapter III allows for stochastic shocks, and in particular builds upon Herrendorf and Lockwood (1997) by assuming that each union receives a (common) noisy signal of a productivity shock prior to setting its individual wage. The adverse macroeconomic externality which arises in this scenario is analysed, and the circumstances under which improvements in signal quality are detrimental to union welfare are identified. It is shown that when unions are nonatomistic, the externality's strength, and hence also both employment variability and the stochastic inflation bias, are sensitive to the specification of the authorities' monetary reaction function. This implies a candidate explanation for why greater central-bank conservatism has generally not been found to be associated with greater output variability. The optimal central-bank delegation arrangements are identified, as well as the optimal amount of transparency regarding supply shocks. Chapters V and VI then analyse the externalities appertaining to indexation decisions taken in the absence of informative signals, both for the standard scenario in which the wage is indexed only to the price level, and for multiparameter indexation scenarios in which the wage is also contingent on a second aggregate variable.

Recombinant DNA technology has allowed the cloning and expression of heterologous proteins in a host cell. Further manipulation of these foreign proteins by genetic engineering allows them to be expressed into the periplasmic space of Escherichia coli. This compartmentalisation of the cell contents exploits a number of key bioprocess advantages. There is a reduction in the process volume, the level of protein and protease contamination is lower, the oxidative environment allows the correct folding of the protein and it is protected from denaturing conditions, such as air-liquid interfaces, during production. However, release from the periplasm at large scale is not straight forward and the balance of advantages and disadvantages can only be assessed by examination of complete processes. This thesis examines some of the process options available for the recovery of selectively released recombinant proteins using a model system comprising E. coli expressing a thermostable α-amylase derived from Streptomyces thermoviolaceus CUB 74. Selective release using lysozyme has been examined with results indicating that the amounts of lysozyme used can be reduced five-fold, from those used in previous studies, to 2.8 mg lysozyme g-1 cells at cell concentrations between 30 to 70 g 1-1. Adjustment of the composition of the periplasmic lysis mixture shows that release was still achieved where no lysozyme use was possible. The possibility of recycling lysozyme for repeated use in periplasmic release was also addressed. The project examines the issues in processing periplasmically secreted recombinant proteins particularly from process liquors that incorporate periplasmic lysis mixtures. Comparison of selective release with homogenisation and secretion, from Streptomyces lividans, was carried out as process alternatives for the recovery of recombinant protein.

The results of an experimental study of phenomena which occur in the flow of a viscous fluid in closed domains with discrete symmetries are presented. The purpose is to investigate the role which ideas from low-dimensional dynamical systems have to play in describing qualitative changes that take place with variation of the governing parameters. Such a descriptive framework already exists for the case of the Taylor-Couette system, where the domain possesses a continuous azimuthal symmetry group. The present investigation is aimed at establishing the typicality of previously reported behaviour under progressive reductions of azimuthal symmetry. In the first investigations, the fixed outer circular cylinder of the standard system is replaced with one of square cross-section. Thus there is now discrete Ζ4 symmetry in the azimuthal direction. Knowledge of the two-dimensional flow field is used to establish the nature of the steady three-dimensional motion equivalent to Taylor vortex flow. It is shown that similar bifurcation sequences exist in both standard and square systems for the case of very small aspect ratio where a single Taylor cell is formed. This flow develops as the result of a bifurcation which breaks the Ζ2 symmetry that is imposed on the annulus by two solid stationary ends. The study is then extended to consider time-dependent effects in the square system. Two different oscillatory single-cell flows are identified, and it is shown that each is the result of a Hopf bifurcation. Selection of a particular dynamic mode is found to depend on the aspect ratio of the system. A low-dimensional bifurcation structure is uncovered which connects the two modes in parameter space, and involves a novel type of steady single-cell flow. Finally, observations are reported of a nontrivial type of dynamical behaviour which bears strong resemblance to motion found in a circularly symmetric Taylor-Couette system that is related to the Šilnikov mechanism for finite-dimensional chaos. A second variant on the Taylor-Couette system is considered where the outer cylinder is shaped like a stadium. The effect is to reduce further the overall symmetry of the domain to a Ζ2 × Ζ2 group. The two-dimensional flow field is investigated using both numerical and experimental techniques. Time-dependent phenomena are then investigated in the three-dimensional flow over a relatively wide range of aspect ratio. It is found that a sequence of a Hopf bifurcation followed by period-doubling bifurcations exists up to a certain aspect ratio, beyond which there is an apparently sudden and reversible transition between regular and irregular dynamical behaviour. Although this transition is not of a low-dimensional nature, the experimental results suggest that it exists as the result of a coalescence of the bifurcations which are found at lower values of aspect ratio.