¶ … Duchin, Ozbas & Sensoy (2010) is to study "how shocks to the supply of external capital affect the real economy" (p. 419), using archival data from the 2007-08 recession. They use conventional models to study change within firms over time before and after the initial onset of the financial crisis, in a "differences-in-differences approach" (Duchin, Ozbas & Sensoy, 2010, p. 419) that compares firms' investment before and after the crisis sorted for particular factors, and then compares that difference between a large sample of firms. Firms are compared based on measures of "internal financial resources (cash reserves and net debt), external financing constraints, and dependence on external finance" (Duchin, Ozbas & Sensoy, 2010, p. 419) after controlling for cash flow and a variable that models investment opportunity, "Q." The goal is to rule out differences between firms based on whether they are dependent on and have access to external finance, and then measure the change in investment based on the dependent variables, to see if firms that held more cash at the onset of the crisis, firms in industries more or less dependent on external finance, and firms with higher information asymmetry all else held as equal as possible, had higher ability to invest in desired capital projects than firms with more short-term debt and less cash. They are also interested in testing to see if and when reduced demand for credit overtook credit supply shortage as the primary driver for reduced investment. Other studies have considered the costs and benefits to 'precautionary savings' but most of these studies focused on agency costs of holding what is typically considered excess cash (Duchin, Ozbas & Sensoy, 2010, p. 421). This study uses archival data to compare firms for share of cash to total assets but also for relative short-term debt proportion, whereas most other studies have focused on long-term debt relative to total assets. The significance of this paper is robust and exhaustively tested evidence that suggests the generally-held view that holding too much cash and foregoing short-term borrowing are generally signs of "managerial abuse due to agency problems" (Duchin, Ozbas & Sensoy, 2010, p. 432), should be revised if these factors become assets that help smooth out volatility in times of supply-side credit shocks.

Duchin, Ozbas & Sensoy's (2010) introduction describes the scope and objectives of this research, the difference between the new and existing knowledge, a general outline of methods and results, and general conclusions described in more detail in later sections. The second section is an overview of relevant literature, which locates this work within, and differentiates the contribution to, the development of the current theory of capital formation going back to the early twentieth century. The third section identifies and limits data parameters and sets out methodology. Section four reports the results of this empirical study, from which conclusions are drawn in the final Section five.

Duchin, Ozbas & Sensoy (2010) trace the theory that firms hold cash in order to counter volatility in financial markets back to J.M. Keynes' 1936 "precautionary savings theory" (p. 423), which has been widely discussed since then, most notably by Modigliani and Miller (1958) and Stiglitz and Weiss (1981) among others (Duchin, Ozbas & Sensoy, 2010, p. 419). Keynes himself sets this theory out in Book IV, chapter 13 (1953: 1991, p. 170) and then traces it back through Marshall and the neoclassicals to the Classical theory of interest proper (ibid, p. 175). Most of the existing work focused on explaining what role supply of finance had in the crisis and establishing whether / how much excessive risk-taking and/or innovative securitization "contributed to the problem" (Duchin, Ozbas & Sensoy, 2010, p. 423) or not. This research looks beyond the general "causes and consequences of the financial crisis" (ibid.) to investigate the real effects on the corporate sector; the authors claim that as far as they can tell, they are the "first to study the impact of the financial crisis on corporate investment using archival data" (Duchin, Ozbas & Sensoy, 2010, p. 423).

Duchin, Ozbas & Sensoy (2010) add to an existing body of work that is so extensive the authors often have to provide notable examples, as indicated by 'e.g.' (for example), rather than describe the entire body of work. They locate their contribution within dozens and dozens of other papers that support the generally accepted model where holding too much cash on the books or not maximizing possible current debt, is generally considered a cost, specifically an opportunity cost, if those assets could generate higher return in other capital projects. The optimal capital structure holds just enough...

313). Duchin, Ozbas & Sensoy (2010), challenge this canonical view by demonstrating that firms with more current liquidity, industries with less information asymmetry, and firms and industries with less frictional finance cost, had higher investment returns after the 2007 financial supply shock, and thus financial managers may want to increase reserve liquidity over what is generally considered the appropriate capital structure to hedge against such shock in the future.
Previous authors have demonstrated that stock price and return on investment after the 2007-era financial supply shock were higher for firms with more access to liquidity and that many firms failed to act on investment opportunities they formerly would have taken during the financial supply shock, but those studies generally considered long-term debt or other variables that limited the generalizeability of the results. Duchin, Ozbas & Sensoy (2010) argue their sample generalizes to more firms because they consider short-term debt, which includes long-term debt becoming current, and also smooths for and excludes potential outliers from the shoulders of their sample that other studies may have included. Since Duchin, Ozbas & Sensoy (2010) confront previous consensus, this contribution would actually cause controversy where the question had previously been considered answered. "If some firms take these risks into account and consequently hold more cash," they explain (Duchin, Ozbas & Sensoy, 2010) "existing empirical models of optimal cash may incorrectly classify them as holding cash in excess of what is optimal. Therefore, we hypothesize that seemingly excess cash may allow firms to fund investment during the crisis that they otherwise would not be able to fund" (p. 431).

Duchin, Ozbas & Sensoy (2010) describe their "main hypotheses" as being "that the tightened supply of external finance following the onset of the [2007 financial] crisis hurt investment mainly in firms lacking sufficient short-term liquidity, either because of small cash reserves or because of large short-term obligations" (p. 415). They seek to test the theory that

"negative shocks to the supply of external finance, together with the presence of financing frictions, might hamper investment if firms lack sufficient financial slack to fund all profitable investment opportunities internally" (Duchin, Ozbas & Sensoy, 2010, p. 419), and that this effect should be pronounced to statistical significance for such 'financially constrained' firms compared to firms that had held what, prior to this research, would have been considered disproportionately high cash given identical possible investment opportunities compared to other firms and industries. They support these main hypotheses probing for an extensive list of sub-hypotheses that result in ten data tables reporting up to five different results per table.

Duchin, Ozbas & Sensoy (2010) considered publicly traded companies using April 30, 2009-vintage Standard and Poor's Compustat results between July 1, 2006 to June 30, 2008, which they extend where available to March 2009 in order to model change driving financial friction from supply to demand as the crisis evolved (p. 423). By extending the sample beyond the crisis itself, the authors compared the recession against more normal context, and ultimately extended their sample back as far as 1994 (p. 431) to include the 9/11 World Trade Center financial freeze, and establish baselines for parameters of interest including but not limited to cash flow, investment and leverage. The authors then screened out financial firms and utilities in certain SIC classes, for market capitalization around high and low bounds; for the highest and lowest capital growth rates per year, and if they had undergone significant enough mergers or restructuring as to suggest they may be unrepresentative of typical firms in their respective industries (Duchin, Ozbas & Sensoy, 2010, p. 424). Measures of finance constraints include " the Kaplan-Zingales (1997) index, the Whited-Wu (2006) index, firm size as measured by total assets, payout ratio, and bond ratings," among others, which they describe in the Appendix (Duchin, Ozbas & Sensoy, 2010, p. 427). The result is "26,421 quarterly observations for 3,668 firms" (ibid.) with debt less than 50% of total assets, and net debt under 750% of excess cash flows, which the authors then manipulate in various ways and test for a battery of results of interest. Whited and Wu in fact argue the opposite, that constrained firms in fact earn more…