ارتباط توام رشد اقتصادی و جریان پول نقد در بانکداری

Prior empirical research indicates that loan growth in the banking industry is positively related to cash flow. I offer an alternative methodology that is better able to capture the effect of cash flow on loan growth while controlling for the potentially coincident effect of loan growth on cash flow. Using a sample of 171,389 observations on banks, 1986–2007, I find that causality runs more consistently from growth to cash flow than from cash flow to growth. This extends prior empirical research by Houston and James (1998) and Campello (2002) on cash flow sensitivities in the banking industry.

Previous research indicates that cash flow in the banking industry is positively associated with loan growth and that, furthermore, the association is stronger among banks that are unaffiliated with a multi-bank holding company. This is consistent with a hypothesis that bank holding companies operate internal markets to allocate capital among subsidiaries that are cash-flow constrained. Companion studies by Houston et al. (1997) and Houston and James (1998), along with a more recent study by Campello (2002), constitute the core of empirical evidence. The evidence must be interpreted in the context of industry-wide correlations of loan growth and cash flow that can be observed in data obtained annually, 1986–2007, from the “Call Reports” of condition and income.1 Although the correlations vary anomalously in sign and level of statistical significance depending on year, definition of cash flow and procedure for deleting outlying observations, they are usually negative. The apparently paradoxical observation of negative correlations of cash flow and loan growth during part of an era in which previous research found the opposite could be related to endogeneity (see Gatchev et al., 2010). In competitive markets, banks may achieve growth by lowering interest rates on loans, increasing interest rates on deposits or increasing branches, personnel, advertising or other elements of production capacity. All of these would tend to reduce cash flow—and at the same time, potentially, that banks may be cash flow constrained. I address the issue of endogeneity using a model in which loan growth and cash flow are determined simultaneously. Using a sample of 171,389 observations on banks, 1986–2007, I find evidence that loan growth exerts a negative influence on cash flow. I also find that cash flow exerts a positive influence on loan growth under some circumstances. Supplemental analysis on subsamples partitioned by year underscores a causality that runs more consistently from growth to cash flow than from cash flow to growth. This is particularly so after changes to regulatory capital adequacy that were introduced in the early 1990s. My results are applicable to the analysis of internal capital markets in the banking industry. They also extend research, outside the banking industry, that questions hypothesized relationships between investment and internally generated funds (Gatchev et al., 2010). This is important because research on investment–cash flow sensitivities constitutes “one of the largest empirical literatures in corporate finance (Brown and Petersen, 2009)”. The paper is organized as follows. In the next section, I describe my methodology in the context of prior research. My sample is presented in Section 3, with particular attention given to the process for deleting outlying observations. The main results of the tests are in Section 4. Section 5 has ancillary tests of robustness and Section 6 concludes.

I analyze the relationship between cash flow and loan growth in banking using a methodology that is intended to capture simultaneous relationships between them. I find evidence of a negative influence of loan growth on cash flow that is more consistently observed than is associated evidence of a positive influence of cash flow on loan growth. This bridges research on cash flow sensitivities in the banking industry (Houston and James, 1998; Campello, 2002) and research on the effect of loan growth on bank risk (Foos et al., 2010). With respect to the latter relationship, in particular, my findings suggest that cash flow may be an important mechanism through which growth influences capital and liquidity. It appears relevant to regulators in monitoring bank risk and condition. The nuanced nature of relationships between cash flow and growth that I have identified can be viewed in the context of ongoing research into internal capital markets. It offers complementary evidence concerning the conclusion of Campello (2002) that small banks are insulated from Fed actions when they operate jointly with large unconstrained banks under the same conglomerate. It also offers complementary evidence concerning the conclusion of Houston and James (1998) that banks are typically capital constrained. My findings suggest that the operation of internal capital markets may be more consistently observed in studies that use deposits, rather than cash flow, as a constraint on growth (Jayaratne and Morgan, 2000 and Cremers et al., 2010).