Assessing impacts of public investments has long captured the interest and attention of the development community. This paper presents the evolution of different methods and approaches used for ex ante appraisal, monitoring, project evaluation, and impact assessment over the last five decades. Among these tools, impact assessment (IA) conducted retrospectively comes closest to providing the proof of development effectiveness. It is defined as the systematic analysis of the significant or lasting changes in people's lives brought about by a given action or series of actions in relation to a counterfactual. There are three basic types of retrospective IAs: macro-level IAs that focus on the contribution of developmental efforts to an impact goal aggregated at a sector or a system level; micro-level impact evaluations (IEs) concerned with estimating the average effect of an intervention on outcomes at the beneficiary level; and micro-level ex post impact analysis concerned with total effects of a development effort after the outputs are scaled-up. Ex post IAs have evolved and expanded over the decades in both breadth and depth of analysis in response to evolving development themes and methodological advancements. The increased emphasis on learning from evaluations has also seen responses from both quantitative and qualitative camps of the evaluation community. The paper argues that generation of robust knowledge that feeds into making developmental policies and investment decisions requires a hierarchical and cumulative approach to "improving the proof" through rigorous and a variety of impact assessment methods applied incrementally at the project, program and system level. Subjecting as many development interventions as resources allow to rigorous impact assessment based on a common framework can help build a critical body of evidence on impacts of development interventions, which can then be subjected to meta-analyses to help assimilate results across different studies and build a knowledge base on what works and what does not.

The spread of modern varieties and hybrids of pearl millet and sorghum that began in the mid-1960s has had an important impact on small farmer welfare in India. The success and sustainability of these improved cultivars resulted from three types (or periods) of interventions by the Indian government: (1) increased investments in crop improvement by national and international agricultural systems during the 1970s; (2) development of efficient seed systems, with the gradual inclusion of the private sector in the 1980s; and (3) the liberalization of the Indian seed industry in the late 1990s. In addition to increased overall production levels of sorghum and millet, there have been substantial yield gains in semi-arid regions as well as improved cultivars adopted in some of the poorest areas of India. The innovations of new, hybrid technology have not been limited to the Green Revolution crops; they have also had significant impact on the productivity of small-farmer households growing dryland crops, such as millet and sorghum in India.

Humanity has made enormous progress in the past 50 years toward eliminating hunger and malnutrition. Some five billion people--more than 80 percent of the world's population--have enough food to live healthy, productive lives. Agricultural development has contributed significantly to these gains, while also fostering economic growth and poverty reduction in some of the world's poorest countries.