The current system of classifying welfare programs divides them into “work first” and “mixed strategy” categories, where the former approach pushes women into the first job they can find, and the latter allows for some education or training along with work. I argue that this classificatory system tells us little about what actually goes on in welfare offices. I spent over a year conducting participant observation and interviews in two welfare-to-work offices in Contra Costa County, California. I propose a new way of examining and comparing welfare programs that looks at the combination of policies, practices, and discourses that shape participants’ access to resources, relationships, and information. I contend that welfare to work programs should be viewed through the lens of economic, social, and cultural capital. I illustrate how one welfare program transmits each of these types of capital. In addition, I add to our theoretical understanding of capital by proposing that economic and social capital, like cultural capital, have both dominant and subjugated subtypes. I argue that only by acknowledging and respecting the subjugated forms of capital held by many welfare-reliant women can welfare workers successfully transmit the dominant forms of capital that would help women move permanently toward self-sufficiency. Finally, I elaborate a new classificatory system based on the successful transmission of the three types of capital in welfare programs. I envision a continuum with Empowering programs—those that are the most successful at deploying subjugated capital in order to impart dominant capital—at one end. At the other end of the continuum are Repressive programs—those that fail to make available the three forms of capital or those that impart it in such a way that it is rejected by participants. I intend for my work to shift our focus away from an understanding of TANF implementation that is focused only on state categories of allowable participation, to one that is focused on the overall experiences of participants in the program. I hope it will illuminate some of the ways welfare-to-work programs can improve their programs within the confines of federal and state regulations.