| dc.description.abstract | This dissertation consists of three chapters. The first chapter - ``Modelling Weights and Dependence Parameters in Mixed Copulas using Penalized Likelihood - Various Applications'', relates to modelling of the weights and dependence parameters of copulas and using data driven copula selection methods to select the correct copula structure underlying the data. The first part of the paper uses a parametric weight specification to model the dependence parameters, as a regression equation of exogenous variables. The model is simulated using a modified EM algorithm and the non-linear dependence structure is studied in a relatively new area namely the offshore exchange markets in India. In the second application with parametric weights, the dependence structure in US housing markets is studied. Results show that while the Gaussian copula is enough to characterize the dependence structure in offshore markets in India, housing markets in New York and Kansas exhibit negative dependence. The second part of the paper designates weights as non-parametric (unknown) functions of structural variables leading to functional index coefficient model structure coupled with selection of mixed copula. The model has been simulated and applied to measure the dependence structure of money markets in Denmark which have recently taken to negative interest rates. Results show that the shares of normal and frank copulas have been steadily declining while the share of copulas with extreme dependence has been rising with a prominent rise of clayton copula in the bivariate distribution of short and long run interest rates post financial crisis. The second chapter of the dissertation - ``Asset Price Volatility and Optimal Policy Mix in Overlapping Generations Model'', revisits a perennially unresolved issue – the question of central bank intervention in managing asset bubbles through the Taylor rule or through (calibrated or un-calibrated) macroprudential policy intervention in credit disbursement. The chapter lays down the theoretical foundations of macroprudential policies like capital adequacy by extending Gali (2014) overlapping generations economy with credit frictions. While the results of the model with financial frictions vindicate Gali (2014) that a leaning against the wind monetary policy generates a larger volatility in the bubble than a policy of benign neglect, the paper finds that minimisation of bubble volatility requires an active macro-prudential policy. It is also observed that stronger interest rate response of monetary policy to the bubble necessitates a stronger macroprudential response possibly to absorb the excess volatility generated by the monetary policy. However, the paper also finds that tightening macroprudential policy parameter beyond a threshold value may encourage banks to take more risks and increase credit supply, aggravating the bubble in the process. With respect to macroprudential policy, there is no conflict between stabilization of current aggregate demand and stabilization of future aggregate demand and both call for a strong macroprudential response, at least until the macroprudential parameter attains the threshold value, although the conflict between the two objectives persists with respect to monetary policy as in Gali (2014). The second paper has been empirically verified through a vector autoregression with sign restrictions approach of (i) Mountford and Uhlig (2005) penalty function approach (PFA); and (ii) Andrew Binning’s underidentified SVAR approach. Using US data, this paper finds that provisioning shock in general and dynamic loan loss provisioning shock, in particular reduces stock prices and dampens output growth as opposed to monetary policy. This result is substantiated both under recursive SVAR specifications as well as sign restrictions imposed according to Mountford and Uhlig (2005) specification. However, by combining sign and zero restrictions in an underidentified SVAR model proposed by Binning(2013),the paper finds that stock prices decline when variables are assumed to be responding to provisions shock only in the long run and the provisions shock is exactly identified. However, when sign restrictions are imposed on inflation, impact response of stock price to provisions shock as well as monetary policy shock is ambiguous. However, stock prices decline during the second quarter in response to a provisions shock. Chapter 3 - ``Financial Inclusion, Financial Stability and Credit Cycles in EMEs'' examines the role of monetary and macroprudential policy in enhancing financial stability in the backdrop of bank-based financial inclusion in emerging market economies like India. A new Keynesian DSGE model of heterogeneous households characterised by various financial frictions typical of an economy adopting bank-based financial inclusion measures is constructed. Results from the DSGE model show that increased financial inclusion may accelerate credit flow to all sectors and raise output if the banking correspondent agents serve as business facilitators involved through the life cycle of the loan and not merely intermediaries involved in cash in and cash out type of transactions. The non-performing asset to gross loan ratio declines indicating the model promotes financial stability. The result confirms how appropriate involvement of BCAs can attenuate the problem of adverse selection for the bank raising overall credit supply in the economy. With respect to standard shocks, financial inclusion seems to delay the impact of monetary tightening, while contractionary macro-prudential shocks have the desired impact but lead to a net transfer of wealth from impatient borrowers to entrepreneurs. Using data from 29 countries spanning Asia, Africa, Latin America and peripheral Europe and twin methods - (i) cointegration and VECM techniques and (ii) VAR with long run restrictions using instrumental variables, the paper also investigated the long run relationship between financial inclusion and financial and banking stability and the significance of permanent shocks to financial inclusion. The cointegration analysis confirms the presence of a common stochastic trend running through credit to domestic sectors, non-performing loans (NPLs) as a proportion of total loans and bank regulatory provisions as a per centage of total assets in 24 of the 29 countries surveyed. In 14 of the 24 countries, there is clear evidence of a permanent decline in NPLs following an inclusion shock. All sample countries that implement bank based financial inclusion exhibit this trend with the exception of Turkey and Peru. The VAR analysis with long run restrictions does not confirm a statistically significant reduction of NPLs following a permanent shock to financial inclusion. However, crucial evidence is found confirming the reduction in NPLs in the long run following a positive productivity shock and a rise in credit following a permanent financial inclusion shock. | |
