Unbalanced liquidity management in commercial bank

<p>Unbalanced liquidity management allows not only to reduce the risk of losses, but also to increase profits.<br />In this paper the term &quot;management&quot; is primarily considered. The existing key structure of Russian banks&rsquo; management has been analyzed. We emphasize weaknesses and shortcomings in the organization of the liquidity management and suggest the ways to improve the structure according with Lavrushina O.I. and Karpova E.I. paper.</p><p>Firstly, in addition to fundamental terms, methods of liquidity management are discussed in the theoretical part of the paper. Secondly, there is provided a justification for using the term &ldquo;unbalanced&rdquo;: it allows to distinguish between liquidity management at it&rsquo;s lack, and managing in the situation of it&rsquo;s excess. Among others liquidity ratios, the Russian Central Bank&rsquo;s &ldquo;H3&rdquo; ratio has been separately allocated and there has been analyzed possible affect of introduction of Basel LCR and NSFR ratios on the banks&rsquo; balance sheet.</p><p>Empirical part of research provides more intuitive way to analyze the situation in the bank, using the expression of lost profits value (or potential loss) in absolute values. Next, we consider the hypothesis about the relationship between the size of the lost profits and intensity of management of net liquidity position of the bank. As a proxy for lost profits, we use directional distance function (like in Shawna Grosskopf (2004) and Domrachev and Norkin (2001)). For banks that have got into the first percentile of the ranking in terms of assets the definition of intensity based on calculation of standard deviation of the ratio between operating assets and total liabilities. For the analysis of other banks with a smaller volume of assets there have been used the standard deviation of the ratio between liquid assets and total liabilities (data have been taken from 2009 to 2012 year). To test this hypothesis we have used regression analysis. In the first case there has been explored a linear model, which at the 5% significance level confirms the hypothesis. In the second case, a U-shaped relationship have been used, which also adequately describes the data and confirms the studied dependency.</p>