While scholars of local service delivery arrangements are fully aware the process is dynamic, research has tended to take the form of cross-sectional studies that are inherently static in nature. In this article, the authors model the determinants of production mode accounting for past delivery decisions. They find, not surprisingly, that there are strong inertial effects; previous delivery mode is a strong predictor of the current service delivery arrangement. More interestingly, the impact of the transaction cost nature of services on production choice is conditioned on past decisions, such as the extent of contracting and the type of vendors used. There is also evidence that contract management capacity and the competitiveness of the contracting environment are influential.

No abstract

The authors argue it is time to move beyond thinking of competition in social service contracting simply in terms of whether competitive tendering procedures are used. Although the procurement process is important, other factors should be examined as well. In particular, they look at how market consolidation and accountability are related to competitive procurement. Findings indicate that demand side-driven consolidation (i.e., governments purposely choosing to go with fewer and larger contracts in which lead agencies manage vendor networks) has both competitive and noncompetitive aspects that are in need of further study and that the competition-accountability link is more complex than generally assumed. Although, as expected, there is evidence that competition, in and of itself, leads to some contractor turnover, it does not appear that competitive vendors are held to higher standards than their noncompetitive counterparts regarding performance (as measured by adherence to contract terms).

Using four consecutive International City/County Management Association alternative service delivery arrangement surveys, this study explores the determinants of service shedding by local government in the United States. Our findings indicate that service shedding is fairly common, with almost 70% of jurisdictions experiencing at least one termination between 1992 and 1997. With regards to why jurisdictions shed services, we find that prior delivery mode is very influential. Specifically, services that were contracted out in the previous time period are much more likely to be dropped than are those that were produced in-house. We also find that the behavior of neighboring jurisdictions matters—if your neighbors tend to provide a service, you tend to continue to do so as well. We find little support for the idea that either budget stress or ideology is impactful in the decision to drop services.

Objectives. In the past, research focusing on local service decisions was dominated by the exploration of make-or-buy choices (i.e., how to produce services). In this article, we extend this venue of research to explore service termination (i.e., whether to provide). In doing so, we adopt four theoretical strands developed in the fields of public policy and management to guide our research: policy termination, political economy, make-or-buy, and policy diffusion. Methods. We utilize multiple International City/County Management Association surveys for our analysis. We supplement these data with information gleaned from a variety of other sources, including Census data, to construct a binary logistic regression model examining the determinants of service termination. Results. We find that local governments are likely to terminate their services when they were previously outsourced to third-party contractors rather than produced by their own employees. Further, our results indicate that locales tend to drop services more often when they are not commonly provided by peer jurisdictions, supporting the idea of diffusion. Conclusions. Our findings suggest that, despite a dearth of research undertaken, service termination is surprisingly common and widespread across service areas.