•
•
•
The normalized results are used in the following stages of the analysis.
LIMITATIONS OF NORMALIZATION
There are several significant limitations to this approach. First, it does not determine the exact amount of impact achieved from a given tranche. Determining the exact contribution of an investment requires further analysis of all investor contributions, including financial and non-financial factors and investment and impact management processes, as well as analysis of impacts that would likely have occurred independently of that investment. Second, normalization seeks to unlock an understanding of performance at the investment level, with particular attention to the proportions and timing of capital injections. However, it does not intend to downplay the fundamental role of investees in advancing social impact achievements.
3.2. OUTCOME EVALUATION
Project outcome evaluation is based on the "theory of change”. Its essence is that normalized outcomes reflect progress towards long-term outcomes.
The full range of outputs and outcomes is a continuum with varying levels of investor influence on the activities and conditions necessary for longterm effects to occur. In order to manage the impact of the project effectively, the degree of certainty in the relationship between activities, outputs and level of control must be identified. Clearly, this certainty cannot be 100 % clear, if only because there are always external factors that could affect the impact created by the investment. For example, an entity may have complete control over the outputs and services that result in some intermediate output, but it has no ability to affect the outcomes created by those intermediate outputs. For example, an organization can easily control the number of mosquito nets distributed to people in malaria-prone regions, but it is much harder to control that people use mosquito nets every night to reduce the incidence of malaria.
It is important that each link between intermediate, short-term and long-term outcomes is supported by the evidence base. Such links should also consider the applicability of the evidence base to investments in different geographic and demographic settings.
Conducting experimental or quasi-experimental evaluations after investment, such as randomized control trials, to check whether the intended impact has been achieved is not always possible, ethical or feasible for most investors. Therefore, data on project outcomes are not always fully available and therefore standardized and accurate evaluation may be difficult. However, in cases where significant literature and evidence already exists, investors can reasonably rely on the results of assessments carried out by external evaluators, NGOs or other development organizations. This methodology uses such knowledge and applies actual indicators of intermediate outcomes as proxies for final outcomes.
It is not always possible to assess the long-term results of social projects. In some cases, intermediate outcomes are sufficient, as they are often well scaled and accurately indicate immediate direct impacts. Furthermore, the final results assessed using this methodological approach do not need to be confirmed by more rigorous impact assessments. Experimental or quasi-experimental analyses can subsequently be conducted, but such studies are not — and should not be — common practice in the socially transformative investment industry. This approach assumes that causal relationships cannot be established and instead examines outcomes that can be linked to investments. Finally, intermediate and final outcomes are interrelated and multifaceted. This analysis provides some illustrative examples, but the data available to investors does not always reveal the full theory of change. For example, improving access to education and training can be assessed using multiple outcome measures, such as average student test scores or the number of occupied jobs. But spillovers may not be fully accounted for in this model.
It is not always possible to assess the long-term results of social projects.
3.3. CLUSTERING THE FINDINGS