Any treatment of ESG and impact investing has got to measure impact measurement. In part because the measurement of specifically what someone cares about is organically challenging. It is often potentially subjective. And even though there are some areas where we can actually make scientific measurements, for example, the amount of sulfur dioxide produced by all company operations. Those are often expensive and very challenging especially, for companies that are perhaps not as open to releasing information. So impact measurement is needed to build indexes, to analyze investments in portfolios, and to quantify and to qualify the expression of investor values. From an investor's perspective, impact measurement is a key consideration. And just like with the number of index providers and indexes themselves, there's been an explosion of sources of ESG information. It's important for investors through their investing lenses to examine impact and ESG investments from the dual perspective, financial returns and social impacts. It may be the case that for certain investors, a successful investment can have a financial return lower than other investments with comparable risks. If its social impact, if it's ESG impact compensates the investor for the foregone financial return component in a rational world. For financial only fiduciaries, this might represent a challenge. It's also possible for successful investments to have potentially greater investment returns. If it is the case like with some endowment perspectives, alpha or outsized performance may be a function or measurable from the ESG characteristics and data. As investors, we will use these ideas to frequently evaluate which among a range of possible social impacts and ESG impact to generates the most utility for us. Not all investors are alike in their preferences. And that's an important component that lies beneath the multiple kinds of measurements that exists and the many issues that could be examined. Here's an example, which has more utility? A small business successfully launched in the developing world and employing people in a sustainable way and with fair business practices and compensation? Or a 1,000 tonne reduction in carbon emissions? Well, the answer is obviously, it depends where you stand. Measuring social impacts is fraught with problems and the tonne challenges. One first-order challenges quantifying the amount of impact ESG impact needed to offset a lower risk adjusted return. Even assuming the social impact can be precisely measured, the tradeoff may or may not be desirable. Depending upon the investment goals and social values of the decision maker or the ultimate beneficiary. Another first-order problem is what we call apples versus oranges or the comparison problem. There's no one set of agreed upon definitions, standards, methodologies or even units of measurement. And the variance of values and missions means standardization may in fact never be possible. Although there are a number of efforts afoot that in fact, a lot has been accomplished via a kind of normalization of these consideration. The baseline is that even units are different, as we saw with the apples versus oranges idea. A potential path to standardization is the largest actors in the measurement community using scale and market power to enforce their own approach. For example, MSCI has been one dominant player with market power that might be attempting this. However, barriers to entry are quite low in the impact measurement business. And with the AB dent of computer technology, machine learning, natural language processing. A number of upstarts have arisen, a number of disruptors have arisen to challenge the dominant incumbents. A uniform approach, in any case, appears to be a distant goal. However, there are a number of protagonists who are attempting to present a standardization that could be adopted by the world. For example, the United Nations Sustainable Development Goals or SAS bees materiality matrix may form the undergirding structure that investors are looking for when they seek standardization. All that having been said, were far away from a single provider or a single source or a single methodology that would provide such standardization. Now, there are numerous causes of bias and distortion in impact measurement. For example, as plagues many retrospective historical studies in finance and investing. Survivorship bias can be present, which is the idea of reporting, in which the data being provided are skewed because failures simply don't report. For example, in some private areas of investing hedge funds, for example, certain hedge funds might not be in your data set because they've died. This notion of survivorship bias is especially important for smaller entities again, in private markets. Subjectivity is rampant and it may be necessary in understanding ESG. Many impact measurement metrics are simplistic ratings for example, a 1 to 5 scale, and there's often a necessary subjective element. A subjective determination is not the same thing of course, as a scientific measurement. Again, underscoring the units issue we talked about. We could potentially need to be aware of conflicts of interest that can arise when a rater earns fees, creating an inflation of assessment issue. Especially, when a commercial provider of ratings is getting paid by companies, to provide the rating. And we saw that after the 2008 and 2009 financial crisis began. Some say it actually began in 2005, where some of the credit rating agencies were accused of such conflicts, especially in rating credit derivatives. The impact investing community is still relatively small in collegial and connected, and there are now a number of centralized. Let's call them, services or authorities that may in fact, have made progress against this notion of a conflict of interest. On the other hand, there maybe in some contexts our reluctance to assign poor ratings to a popular or influential entity. Outputs are not necessarily the same as outcomes. A widely discussed measurement problem in US medicare. We can perhaps measure or assess outputs, but outcomes ultimately, maybe what we need to think about. Choices of units are diverse and different. With ratings, intra-rating consistency and intra-rating consistency is a challenge. The definitions of ENS and G might actually change across time and their sub components, certainly. And of course, ENS and G as we've mentioned elsewhere overlap. There's a component of S inherent in G and a component of G inherent in S, and the same thing with the for example, consider sustainability reports. All of the S&P 500 firms currently produce so called sustainability reports. But you have to ask yourself, is that an environmental idea? Or is it a governance idea related to openness and transparency? Well, some would argue that it's both. And that means that ENS and G are what we call non orthogonal or uncorrelated, meaning that they don't represent unique ideas to measure. And that perhaps provides one more level of challenge.
