6. Discussion and conclusions
To date, almost all research on the effect of risk on public service innovation is based on conceptual theorizing or on a limited number of case studies (Brown and Osborne, 2013). To the best of our knowledge, this is the first study to use a large survey of several thousand public administration agencies to investigate the relationship between a culture of risk aversion and the benefits agencies obtain from service innovations. Using a holistic and multiple contingencies approach grounded in configurational theory, this study applies fuzzy-set QCA to explore how different levels of risk aversion combine with other factors and innovation support strategies to produce high and low levels ofinnovation benefits. The analyses identify four equifinal configurations of factors leading to high innovation benefits and six equifinal configurations for low innovation benefits. The level of risk aversion is a relevant but not deterministic condition for the successful development of service innovations that elicit high benefits. Rather, the ability of an innovative agency to deploy and properly combine different strategies, based on the context in which it operates, is a fundamental driver for innovation success.