Drucker (1999), Haug (2007) mentioned about the importance of right assumptions in modelling. They opined that in natural science, wrong assumptions would not change the natural phenomenon but these could significantly alter the results of a man made system like financial system. Drucker (1999) remarked that the fundamental assumptions about reality decided the scope of a theory and the researchers, teachers, practitioners usually kept those in mind and assumed the resultant theory to be a reality. He added that as far as natural sciences were concerned, the assumptions taken by scientists did not and in fact could not change the working of nature but in social sciences like management where we dealt with humans, the situation was different as people would tend to act as the assumptions of a discipline would tell them to. In his words, "What matters most in a social discipline such as management are the basic assumptions…………………Despite their importance, the assumptions are rarely analyzed, rarely studied, rarely challenged-indeed rarely even made explicit".
Haug (2007) explained that a number of models had come in a series, and were still coming in the global finance arena. He described that the models were typically based on more fundamental underlying models, generally based on probability theory and assumptions of normal distribution, which we took for granted as if the probability theory and normal distribution worked in all situations. He added that we mistakenly forgot the assumptions behind the models or the assumptions kept hidden and we inclined to believe the applicability of the models.
Till now Basel Committee on Banking Supervision (hereinafter named as BCBS) has issued two accords for risk management practices while. Besides, one more popularly called as Basel III or Basel 3 will be implemented in due course. These are accepted in banking organisations worldwide in many countries (more than hundred countries), albeit various countries have adopted these in stages.
BCBS issued its first accord for risk management, namely Basel I, in the year 1988. As per RBI guidelines, Indian banks adopted Basel I in the year 1999. Basel I dealt mainly with credit risk and to some extent with market risk. It had no specific provisions regarding operational risk. One important feature of Basel I was that the capital requirements were more of standardised nature and all banks were following similar basis to determine regulatory capital irrespective of intensities of the risks they take. Coupled with these, the changes in economic scenario, increased competition, advancement in technology etc. rendered the accord of limited relevance.
The accord consisted of three mutually reinforcing pillars which are explained with the help of a table as follows:
Table No. 1: Three pillars in Basel II
Pillar 1
Minimum Capital Requirements
Set out the minimum capital requirements for credit, market and operational risks.
Pillar 2
Supervisory Review
Required the supervisors to monitor whether banks are assessing their capital requirements as per norms set and to take action if the risks are excessively high.
Pillar 3
Market Discipline
Set out the disclosure requirements by banks in order to improve market discipline.
After a broad based and wide ranging process of consultations (three consultative papers), a document namely "International Convergence of Capital Measurement and Capital Standards: A Revised Framework", was issued in June, 2004 (supplemented by an update of the Market Risk Amendment) (BCBS, 2004). This document, widely known as "Basel II Framework" was more comprehensive in its scope. It provided more rigorous and sophisticated methods to deal with credit risk. Operational risk was given special consideration in the accord and as such the need for extra capital was made mandatory as part of the total capital requirements in capital adequacy ratio (Mathew, 2010). Unlike Basel I, it encouraged flexibility in measuring risk exposures and managing risks faced by banks so that the banks could have different capital adequacy requirements for each risk category.
Bank for International Settlements (2010) has now come with a comprehensive set of reform measures on the updates to the previous Basel Accords, named as Basel III. This is being brought in response to the global crisis.
Fatemi and Fooladi (2006) in their survey of credit risk management practices found that a rapid pace of product innovations, diversification of financial institutions, increase in credit intermediation and in some cases, the regulatory mandates prompted the development of more sophisticated approaches to the measurement of credit risk coverage including complex hedging techniques. They found that the models derived were either internally developed models or the models bought from outside agencies, like VaR marketed by J.P. Morgan.
Gersbach and Wenzelburger (2007) used the term sophistication while investigating the relation between sophistication in risk management and banking stability. They compared a simple banking arrangement in which simple rating methods were used with a sophisticated banking arrangement.
It is pertinent to mention here that the quantitative finance has always raised doubts. The major crash of stock market in 1987 raised questions on the reliability of financial models (Los and Yalamova, 2006). The failure was highly implausible as the sophisticated statistical models claimed to measure risks with certainty. Every time a crisis occurred, it overwhelmed the academically trained quantitative people and led to the development of a new set of models by changing the assumptions of previous model or to incorporate specific reasons of a crisis that occurred in the recent past. Various statisticians, economists and financial analysts have tried many methods for measuring risk, and given a host of models for the same.
Facing a crisis becomes more difficult in times of globalisation owing to increased sophistication of financial products and interdependence of markets (Jobst, 2007).
Fehle and Tsyplakov (2005) expressed concern over the use of static models for corporate risk management. In their view, these models offered rich insight into why firms managed risk, but the models didn't provide firms anything relating to sufficient number of predictions, definite decisions on the choice of risk management tools and strategies to manage risk.
Taleb (2009) discussed that not even a single expert from the field of finance, economics, banking and even the regulatory and prime financial bodies around the world could predict the crisis of 2008, and even a single forecast did not prove useful, and could not sense the results experienced in the crisis. He challenged the sophistication being posed by experts and presented evidence, through research, that predictions through econometric techniques based on standard deviation and variance were defective and did not repeat. His study resulted that the tail events (as in a normal distribution) could not be computed.
In a discussion of five experts moderated by Champion (2009), the nature of risk management in the light of the recent crisis was deliberated upon. The different views emerged on managing risk in the coming times. Mikes, one of the experts, mentioned about quantitative enthusiasts who believed that every risk could be modelled and further who could virtually take every decision (giving of loan etc.) by using the models. She discussed the limitations of banks that give too much weightage to outcome of a model and hardly bother about the assumptions that were used to generate the model.
Taleb et al. (2009) firmly criticised the use of standard deviation and measures linked to it, for instance- regression models, R-squares and betas which are employed extensively in financial calculations. They criticised the quantitative risks models used in banks to such an extent that these were primarily responsible for the failure of banks.
Varma (2009) elaborated that in the global crisis, no major derivative clearing house around the globe faced any financial trouble and one key reason for this was that derivative exchanges did not use VaR, normal distributions and linear correlations. The findings of his paper provided recommendations for exchange traded derivatives to improve their robustness. He also advocated the use of plain models based on rational measures (like expected shortfall), fat tailed distributions and non linear dependence structures. He added that "The global financial crisis has also taught us that in risk management, robustness is more important than sophistication and that it is dangerous to use models that are over calibrated to short time series of market prices."
Danielsson (2008) stated that supervisors necessitated risk calculations in response to crises. He severely criticised statistical models in his following statement, "Model-driven mispricing produced the crisis, and risk models do not perform during crisis conditions. The belief that a really complicated statistical model must be right is merely foolish sophistication".
Froot et al. (1994) described that the growth of derivatives took place due to the innovations by financial theorists who developed novel ways to find out costs and value of these complex financial instruments. They added that the engineered products created new weapons in the hands of risk managers but, the financial engineers didn't provide any guidance to risk managers on how to use the new instruments in an effective manner thereby making instruments of limited use.
