In recent times, a large body of literature has emerged that asserts the role of financial intermediation in the macroeconomic models. 1 The significance of financial institutions, mainly banks, lies in the following activities:(i) banks accept deposits of household savings and lend to a large number of agents;(ii) banks hold liquid reserves against predictable withdrawal demand;(iii) banks issue liabilities that are more liquid than their primary assets;(iv) banks reduce the need for self-financing of investment.

The implication of the above is that holding savings in bank deposits is safe in respect of returns compared to equities or direct lending to firms that have uncertain returns. The risk-averse agents would hold more of their savings in bank deposits than in equities or direct lending. The funds from deposit mobilization are lent to entrepreneurs to finance investment projects. Asymmetric information about the investment projects require ex ante evaluation and ex post monitoring which, in turn, require skill, as well as cost. An individual investor usually does not have the necessary skill and the cost is also prohibitive, while banks can do the job efficiently. 2 In the process, banks can exploit the law of large numbers to forecast the number of unsuccessful projects and, as a result, the expected returns of the loans advanced. The savers can be assured of a safe return. In short, the bank is the institution through which savings are channelled into investment in the absence of a perfect insurance market for loans. Thus the process is conducive to growth in the real economy. Levine (2004) gives an excellent survey of this literature. On the other hand, however, many noted economists3 hold a diametrically opposed view. For example, Robinson (1952) argues that the