Doctoral conference 2012
Thursday, 10th May
9:45am - 10am Welcome Address
Environmental Economics
(10am - 12:20pm session will be chaired by Hormoz Ramian)
10am - 10:30am Natina Yaduma
"The Environmental Kuznets Curve at Different Levels of Economic Development and Distributions of Income"
The Environmental Kuznets Curve (EKC) theory posits that the early stages of a country's developmental process are associated with increasing environmental damage. However, after the attainment of a threshold level of income, progress leads to greening. Advocates of this theory prescribe that economic growth is the cause and panacea to environmental degradation; thereby undermining the role of environmental policies in mitigating pollution. Methodologically, the conventional techniques used in investigating the hypothesis focus on estimating the rate of change in the mean distribution of emissions as a function of income, thus, being incapable of capturing country heterogeneity within the (panel) sample investigated. We employ a relatively new technique used in Kuznets Curve estimations - the quantile fixed effects method - in our exploration of the CO2 Kuznets Curve. Further, we extend decomposition methods largely employed in labour economics research to the Kuznets Curve framework to explain the most important factors accounting for the OECD-Non-OECD emissions gap. Our study confirms the existence of the EKC in the global, OECD, Western, Latin American and East Asian samples. Additionally, we find that the OECD countries polluted about 60 to 369 percent more CO2 than their Non-OECD counterparts. In sum, we suggest that all hands must be on deck to mitigate CO2 pollution and policies promoting economic progress should move in tandem with those promoting greening.
10:30am - 11:00am Jonathan Gibson
"Health, Labor Supply and Productivity: An Econometric Study of Cambodian Farm Households and Arsenic Consumption"
The consumption of arsenic through contaminated groundwater and agricultural crops in South-East Asia is a major public health concern, exacerbated by the encouragement of groundwater well construction. Whilst the mortality and morbidity attributable to exposure to geogenic arsenic through drinking, cooking or consumption of rice and other foods is well known and increasingly studied, the economic impacts of geogenic arsenic exposure have been less well investigated. We present the framework and results of an econometric study of the impacts of arsenic consumption in rural Cambodia on household decision making, particularly labour supply choices, and farm productivity. The research analyses how the health impacts of arsenic consumption are manifested in economic decision making and agricultural production, utilizing household data from the Cambodian Socio-Economic Survey (CSES) along with village level arsenic estimates.
11:00am - 11:20am Tea Break
11:20am - 11:50am Paul Balcombe
"Energy from Microgeneration: Sustainability and Perceptions in the UK"
The drive for climate change mitigation and more secure energy supplies has led to government support of the microgeneration industry in the UK. Microgeneration is the small scale production of heat and/ or electricity from a low carbon source such as with solar panels, wind turbines and heat pumps. Despite government targets and incentive schemes, microgeneration uptake in the UK remains low with only 180,000 homes, out of over 25 million, having microgeneration installations. To achieve greater microgeneration uptake, significant demand-side barriers must be reduced, including high capital costs, long payback times, large space requirement within the home, personal effort required and perceived bureaucracy associated with planning and installation. The main research question is: can microgeneration contribute to meeting UK climate change, energy security and fuel poverty targets and, if so, how can greater uptake be achieved? The research is framed into three sub-questions: how sustainable is microgeneration; what are the motivations and barriers associated with microgeneration adoption in the UK; and, if microgeneration is sustainable, where does the greatest potential lie in increasing uptake and maximising the benefits? Methodological tools to be used within this research are life cycle analysis (LCA), best-worst scaling and choice experiments. Thus far research has focussed on determining the environmental sustainability of microgeneration and understanding consumer motivations and barriers associated with microgeneration adoption. The presentation will illustrate the work carried out so far and the design of a best-worst scaling survey due to be implemented in the coming months.
Development Economics
11:50am - 12:20pm Swati Virmani
"Is India's Outward FDI Consistent with Dunning's Investment Development Path Sequence?"
The purpose of the study is to examine whether India's Outward Foreign Direct Investment (OFDI) pattern suggests consistency with the Dunning's Investment Development Path (IDP) Sequence or is a refinement to the established theories. The question is addressed using macro level data on India's growth and investment position over the period 1980-2010. We test whether the level of development proxied by GDP per capita is the main factor explaining OFDI, and also extend the hypothesis to examine other major determinants of OFDI — Exports, GDP per person (proxy for labour productivity), Inward FDI (IFDI) and R&D. We adopt the Cointegration and Error Correction Model technique to carry out the analysis. The issues raised here have an important policy implication for developing nations. Should the countries seeking internationalization wait for their per capita incomes to grow to undertake OFDI or should they invest at an early stage of their development, forming an exception to the IDP theory? The latter emphasizes the importance of other contributing factors, apart from the income level of a country. The paper also contributes towards finding if there is a two-way causal relationship between OFDI and the explanatory factors, following the Granger Causality test. An interesting finding is that the OFDI granger causes R&D, suggesting a possibility of reverse technology spillover effect. It also calls for a further research to find whether R&D spillovers have a positive impact on domestic productivity, the so called Feedback Effect.
12:20pm - 1:10pm Lunch
Microeconomics
(13:10pm - 16:00pm session will be chaired by Natina Yaduma)
1:10pm - 1:40pm Angelos Angelopoulos
"The Maximin Value Allocation: Interpretation and Properties"
There are three main (Walrasian general equilibrium related) reasons to employ maximin-non Bayesian preferences, instead of the standard-Bayesian ones, for the agents of a Radner's (1968) partition type differential information exchange economy. All of them are originated by the (inherent in such an economy) private asymmetric information measurability requirement. All of them are jointly discussed and formally presented in a recent paper of Castro-Yannelis (2010), who, taking this measurability condition for granted, assert that: (i) the Bayesian agents have incomplete preferences because they are only able to compare private information measurable individual allocations, (ii) the efficiency of the general equilibrium out- comes is reduced since we lose the non private information measurable Pareto optimal allocations of this economy and (iii) its seems "unreasonable" to assume that the agents are Bayesian in the first place since they cannot possibly assign a probability of occurrence to the states between which they cannot distinguish. By adopting the Gilboa Schmeilder's (1989) maximin preferences for the agents of this economy, the aforementioned issues are naturally resolved. On these grounds, we pursue in an "appropriately" well defined exchange economy of this type, able to accommodate the agents' maximin preferences, a cooperative Walrasian equilibrium allocation, the maximin (Shapley, 1969) value allocation. We do this by extending the (Bayesian) private value allocation of Krasa-Yannelis (1994). This maximin value allocation concept was introduced by Castro-Yannelis (2010) as well, but it was not axiomatized. By imposing, therefore, a set of assumptions on the maximin expected utility of the agents, trading (i.e., writing contracts) in such an economy, we prove existence of the maximin value allocation (i.e., contract). We also establish all its other normative maximin properties: Pareto optimality, individual rationality and incentive compatibility.
1:40pm - 2:10pm Jinrui Pan
"Liminal Exponential Discounting"
A decision maker's propensity to forgo current utility for future utility is known as their discount rate. The classical model of decision making over time, exponential discounting, assumes that the discount rate is constant. This paper introduces a new model we call Liminal Discounting. This model generalizes exponential discounting model in a simple way, yet the model can accommodate preferences exhibiting decreasing or increasing impatience. An individual with such preferences has a constant rate of time preference up to some threshold point in time. After this threshold the rate may change, but will then remain constant at the new rate. Such preferences are stationary before and after the threshold. These long periods of stationarity make the model especially tractable for economic applications. Violations of stationarity, such as the present bias, may occur when comparing the near and distant future. Our main theorem provides a preference foundation for the Liminal Discounting model. The theorem is proved within the standard framework, so is a genuine generalisation of the exponential discounting model. In particular, the threshold time arises as a consequence of our preference axioms.
2:10pm - 2:30pm Tea Break
2:30pm - 3pm Katarzyna Warner
"Foundations for Prospect Theory through Probability Midpoint Consistency"
For the famous prospect theory model there is hitherto no preference foundation for general sets of outcomes. All existing models assume a rich structure for the set of outcomes and propose preference conditions that hinge upon that structure. Yet in many important applications where prospect theory is assumed, like health or insurance, the set of outcomes is degenerate. In these more general settings it is unclear what preference conditions are required, beyond the standard assumptions, to pin down prospect theory. This paper proposes a consistency principle for elicited probability midpoints that requires a consistent treatment of probabilities of gains and similarly a consistent treatment of probabilities of losses. We show that, in the presence of the other standard preference conditions, this consistency principle implies prospect theory.
3pm - 3:30pm Mikhail Zhitlukhin
"The von Neumann-Gale Model of Financial Markets"
We consider a model of a stochastic financial market that generalizes the classical model by including transaction costs, portfolio constraints and a new notion of hedging. Our model is based on the framework of von Neumann-Gale dynamical systems - the theory that was originally related to the problem of optimal redistribution of resources in a growing economy. The central question of the model consists in determining whether it is possible to hedge a contract (i.e. to make certain payments at certain dates) starting from a given initial endowment. We provide a criterion of hedging that generalizes the celebrated Fundamental Theorem of Asset Pricing and the Risk-Neutral Pricing Principle. As an example, we apply the criterion to a specific model of a stock market.
Macroeconomics
3:30pm - 4pm Roy Silverman
"Bank Capital Regulation, Credit Friction and the Cyclical Behaviour of Interest Rates"
This paper examines the macroeconomic effects of bank capital regulation in a simple Dynamic Stochastic General Equilibrium (DSGE) model with credit market imperfections. A key feature of this model is the derivation of the bank loan rate and the endogenous probability of default from break even conditions. We show that in a model which accounts for bank capital and bank capital regulation in the form of the Basel Accords, the endogenous probability of default impacts the lending rate through multiple channels. We also define the Basel I and Basel II regulatory regimes in terms of the calculation of the risk weight attached to loans with a distinction made between the Foundation Internal Ratings Based (IRB) and Standardized approaches of Basel II. Our simulation results suggest that the Basel II regulatory regime amplifies the response of macroeconomic variables following both supply and monetary shocks when compared to Basel I, while a comparison between the two variants of Basel II depends on the nature of the shock.
Friday, 11th May
Macroeconomics
(10am - 12:50pm session will be chaired by Katarzyna Warner)
10am - 10:30am Jaqueson Galimberti
"A Comparative Assessment of Adaptive Learning Algorithms as Representative of Macroeconomic Expectations Formation"
Adaptive learning algorithms have been proposed to provide a bounded rationality view on agents process of expectation formation and as the means through which expectations shocks provide another source for business cycle fluctuations. Despite of the preeminence of the Least Squares (LS) algorithm as the representative of agents learning, it is now understood that the dynamic properties resulting from adaptive learning depend on the chosen algorithm. We evaluate the empirical plausibility of assuming the LS algorithm as representative of agents macroeconomic expectations formation by comparing it with a computationally simpler alternative, namely, the Stochastic Gradient (SG) algorithm. This latter has been proposed as a more stringent alternative to the LS in terms of bounded rationality. We assess this hypothesis with an empirical exercise comparing the performance of forecasts provided by these algorithms, as well as their resemblance to forecasts obtained from surveys. This is done within the context of vector autoregressions with time-varying coefficients for inflation and growth using US real-time data. We find mixed evidence on the comparative of forecasting performance, though with scarce support for statistically significant differences between these algorithms. Compared to the survey forecasts, however, we find that the forecasts from the LS algorithm provide a closer resemblance to the survey forecasts than the SG algorithm does. A thorough analysis on the initialization of these algorithms is also covered as a byproduct of our study.
10:30am - 11am Maria Paola Rana
"Organized Crime, Corruption and Economic Growth: ""An Empirical Analysis for the Case of Italy"
The paper examines the impact of corruption on economic growth in the presence of organized criminal activities. Using a panel data of 20 Italian regions for the period 1961-2009, the analysis reveals (i) a growth-inhibiting effect of both corruption and organized crime, and (ii) that in the presence of organized crime the impact of corruption is less severe. This finding offers support to the argument that with organized corruption arrangements and better coordination in the bureaucrat's rent-seeking behavior, corruption is less distorting for economic growth. The results are robust to different specifications and different estimation methods.
11am - 11:20am Tea Break
11:20am - 11:50am Xin Hao
"The Role of the State: Taxation and the Composition of Public Spending"
The paper responds to a prevalent empirical observation that as the economy grows, the share of the government expenditure spent on transfer payment also increases. In my model, the state decides policies that involve setting a tax rate and the composition of how to spend its tax revenue, specifically, it chooses the ratio between public goods provision and transfer payments. In essence, public goods propel growth and transfer payments mitigate the utility inequality between agents that are heterogeneous in wealth. I will argue that there exists a habit formation process on transfer payments that cause the state to choose a policy that harms the growth of the society. This habit induces a welfare loss in the economy and poses financial challenge to the state.
11:50am - 12:20pm Serena Masino
"Instability and Human Capital Formation"
This paper focuses on the impact that instability of the macroeconomic and political environment has on human capital formation. The transmission mechanism is analysed in the context of an OLG model. With the help of some calibration exercises, instability is shown to have a positive effect on the ratio between unskilled and skilled wages. A decreased skill premium discourages education decisions and leads to low levels of human capital accumulation. These predictions are subsequently tested empirically and confirmed by the results of a cross section panel data analysis.
12:20pm - 12:50pm David Chivers
"Inequality without Imperfection: The Role of Loss Aversion"
We construct a stochastic model in order to demonstrate how individual's loss averse preferences can result in multiple equilibria without the need for capital market imperfections. Individuals face a decision of whether to invest in a risky wealth enhancing project. We find that an increase in uncertainty, or in the preference to avoid losses, increases the critical value of wealth needed for individuals to invest in these projects. Hence, if initial wealth is distributed heterogeneously, initial inequality may persist and increase over time.
12:50pm - 2pm Lunch
Visiting scholars
(2pm - 3:30pm session will be chaired by Jonathan Gibson)
2pm - 2:30pm Emma Apps (University of Liverpool)
"Contagion in VaR amongst UK Financial Institutions"
My discussion focuses on the use of "Value-at-Risk" in defining a financial institution's exposure to systemic market risk. Specifically, its application results in the estimation of potential maximum losses in times of sustained market turbulence. Furthermore, I suggest that the current methodologies are, perhaps, at best, basic but necessary and ultimately failed to adequately quantify the huge financial losses sustained post 2007. The objective of the research is to introduce the concept of "Conditional Value-at-Risk"- where this measure attempts to identify how the risk of one institution may increase when a second institution falls into distress. This suggests that, rather than quantifying your risk in isolation, you should consider the negative "risk-spillover" effects of other financial institutions i.e. the contribution of individual financial institutions on the risk of the whole financial system. Consequently, I develop the theory that time varying Co-VaR and forward delta Co-VaR are more appropriate bases for the setting of regulatory risk constraints on Financial Institutions.
2:30pm - 3pm Anwen Zhang (University of Lancaster)
"Returns to Education for the Self-Employed in England and Wales: An Attempt to Correct Income Underreporting"
The paper attempts to correct income underreporting of self-employed workers in two UK surveys by adopting an expenditure-based approach. A functional form of Engel curve is first estimated for employee households, then applied to self-employed households to infer their true earnings from self-employment by looking at their food and fuel expenditures. Further, inferred self-employment earnings are used to estimate rates of return to schooling and vocational qualifications. The results show evidence of self-employment income under-reporting and further suggest that income underreporting largely affects the estimates of rate of return to education for the self-employed.
3pm - 3:30pm Vasileios Pappas (University of Lancaster)
"Analysing Default Risk in Islamic Banks"
This paper compares the level and determinants of default risk in Islamic banks versus commercial banks located in 20 Middle and Far Eastern countries. Survivor and hazard functions are estimated conditional on annual firm-level accounting data pertaining to 421 banks and country-level macroeconomic fundamentals over the period 1995 to 2010. Islamic banks are shown to be better positioned in terms of failure risk than commercial banks suggesting that they contribute favourably to the overall stability of the financial system. The survival rates are shown to be driven by firm-level balance sheet, income statement and financial ratios inter alios, but with different sensititivies for the two banking systems. Latent factors raise the chance of within country co-default for commercial banks but not so for Islamic banks.
3:30pm - 4pm Tea Break
Econometrics
(4pm - 6:30pm session will be chaired by Swati Virmani)
4pm - 4:30pm Gantunalag Altansukh
"Structural Breaks in International Inflation Linkages for OECD countries"
The paper proposes an iterative structural break testing methodology, which aims to provide more reliable inference for structural break tests applied to conditional mean and conditional variance parameters by iterating between these two components, while also taking account of outliers. This iterative testing procedure is applied to both univariate and bivariate inflation models; the first is employed to examine the stability of domestic inflation, and the second is to investigate changes in a linkage between global and domestic inflation over time. The empirical analysis uses monthly Consumer Price Inflation (CPI) for 19 OECD countries, over the period between 1970M01 and 2010M09, and following key results emerge. First, univariate models yield less but broadly consistent mean breaks to the existing literature. We also document clusters of variance breaks occurring around mid 1970s, early 1980s and early 1990s, while only clusters of mean breaks have been documented widely so far. Second and more importantly, we find positive and increasing contemporaneous relationship between domestic and country specific global inflation. These results may be informative with regard to a co-movement of inflation across countries.
4:30pm - 5pm Hormoz Ramian
"Daily Volatility of Large-Dimensional Portfolios in Financial Markets with Asynchronous Trading""and Microstructure Noise"
Volatility plays an important role in financial econometric models. We use the high-frequency data to study the daily variance-covariance-matrix of assets returns in large-dimensional portfolios in financial markets with market microstructure noise and non-synchronous trading. Our focus is on portfolios where the number of assets is larger than the number of synchronised intra-daily observed returns. We first outline the theoretical shortcomings of conventional daily variance-covariance-matrix estimators using intra-daily returns which motivates us to provide an introduction and evaluation on the performance of the recently suggested estimators that address these shortcomings. We use the multivariate realised kernel estimator which addresses the consequences of microstructure noise and use the blocking and regularisation method (Hautsch et al., 2008) which reduces the loss of the data due to the non-synchronous trades, then we discuss the implications of allocating the assets into irregularly sized blocks based on the trading frequencies. We analyse the performance and applications of this estimator in Value at Risk context and forecasting the covariance matrix. We use the Conditional Autoregressive Value at Risk by Regression Quantiles method as a measure to evaluate the performance of blocking and regularisation method.
5pm - 5:20pm Tea Break
5:20pm - 5:50pm Rabeya Khatoon
"First Order and Second Order Asymptotic Analysis of GEL Estimators for Grouped Data Model"
The existing estimation techniques for grouped data models can be analysed as a class of estimators of IV-GMM type with matrix of group indicators being the set of instruments. Econometric literature (e.g. Smith 1997, Newey and Smith 2004) show that in some cases of empirical relevance, GMM can have shortcomings in terms of the large sample behavior of the estimator being different from the finite sample properties. GEL estimators are developed that are not sensitive to the nature and number of instruments and possess improved finite sample properties compared to GMM estimators. In this paper, with the assumption that the data vector is iid within group, but inid across groups, we developed GEL estimators for grouped data model having population moment conditions of zero mean of errors in each group. First order asymptotic analysis of the estimators show that they are 
consistent and normally distributed. The paper explores second order bias properties that demonstrate sources of bias and differences between choices of GEL estimators.
5:50pm - 6:20pm Mohammad Rahman
"Is the Retirement-Consumption Puzzle Solved in the UK?"
Empirical evidence in some developed countries has found that the life cycle hypothesis cannot fully explain why consumption falls at the retirement age for some individuals. This suggests that the individual's wealth at the retirement age is below the optimal level. However, the buffer-stock saving model where income is the driving force to change consumption can provide an explanation for this puzzle as named in the literature. On the other hand, the consumption growth is independent of income in the life cycle hypothesis. Applying Fuzzy Regression Discontinuity design on household survey data- Family Expenditure Survey (1968-2001), Expenditure and Food Survey (2002-2007), and Living Cost and Food Survey (2008-2009) in the UK, I find that consumption substantially fell at the retirement age before 1980s. This fall is less severe after the 1980s, because of tax policies designed to boost pension income. However, throughout the data period, consumption falls at the retirement age are fully explained by the income falls, which contradicts the life cycle hypothesis but supports the buffer-stock model.
6:20pm - 6:30pm Endnote/Thanks