This paper argues for an efficiency account of secured debt and against an externalization account. It describes the results of an investigation into firms' use of secured debt. I interviewed over twenty lawyers, bankers, and business people who are experts in this area and made use of my own knowledge as a former corporate lawyer. My investigation suggests the following pattern of secured debt and explanation therefor: At lower-quality levels, firms very often secure all their assets whereas at higher-quality levels, some firms secure some of their assets. At lower-quality levels, the divergence of interest between the firm and its lenders apparently dominates its financing choice. Granting the lender a security interest in all a firm's assets effectively prevents the firm from taking certain actions that would benefit the firm or the firm's owners but adversely affect the lenders. At higher-quality levels, there is still a divergence of interest between the lenders and the firm, but the divergence is not nearly as large. Other factors also influence a firm's financing choice. One such factor is the nature of the firm's assets: higher-quality firms tend only to secure assets that are easy to repossess and resell into a liquid secondary market. Other relevant factors include the availability of economies of scope, information cost savings, and regulatory and other arbitrage benefits. My investigation also suggests that most firms have limited ability to externalize liabilities using secured debt. Most voluntary creditors will adjust. And many firms will not have appreciable liabilities to involuntary creditors. My investigation thus casts doubt on some of the arguments for limiting the priority of secured creditors. The next step is empirical work to systematically appraise the results of my investigation.