How much should you trust third party evaluation of your SRI portfolio companies? As a SRI interested investor or researcher you might want to have some their party have a look at the companies that you are investigating but how much can you really trust their evaluations? This question comes up more and more frequently as third part evaluations become much more freely available.
I have compiled a short list of issues that you might want to take a look at when evaluating if the intelligence can be used in your investigation of companies.
- Transparency is of cause a major concern and just because a report or paper is made from somebody outside the company it does not necessary make it more useful. It is not that third-party investigations are necessary biased but you need approach their reporting in much the same way you would corporate self-evaluation and reporting, with a fair amount of scepticism. Take a look at how they present their data and look for their source of information is if the raw material is available for your own analysis it is the best if they only present the conclusions then take it with a grain of salt.
- To what extend does the analysis base its conclusions on Corporate data? A lot of analysis only have corporate self-reporting as it source but might come up with very different conclusions that internal analysis came up with. What you need to do is look for triangulation of data sources. Were interviews or questioners conducted during the process, were experts included and what was their background (if it is just people from the analyst own group they might not be as reliable as other experts might be) or were data verified by other means. I recommend that you at least look for one other source of aw data other than corporate information. This will show you that some efforts were put into doing the report and some level of forensics were conducted.
- The analyst themselves can have a big impact on how reliable the data is. Some analysis are doing the analysis on a part time basis while others does it for a living. This does not mean that the work of the amateur cant be very valid it just means that when you evaluate technical data you might place more reliance on what the professional have to say. The person who sits with data on finances, CO2-emissions, supply and value chain data have a great deal of routine in looking at these numbers and have a good idea when they are off the mark so to speak.
- Standards are a big issue within CSR not just because they are evaluated against but also because they are seen as normative truths. We all know standards like the Global Compact (GC) and reporting according to Global Reporting Index (GRI) or the new ISO 26000 but just because standards are used they o not constitutes the truth. Look at them as a way to present data not as a factor for goodness. A A+ rating just not represent a higher degree of goodness than a C rating in GRI it just shows you what corporate data should be available to you for analysis.
- The Scope of data within CSR is constantly being debated. Some would include everything others limit their scope to just include the 10 principles in the GC. When you do your analysis you should scope what you perceive as valid data and collect sources of material that fits this scope. If you find evidence that is outside the scope but have a impact on your analysis you should put it aside for further analysis later one when the rest of your investigations have been included. This exercise will help you not only evaluate your ability to create a correct scope but also understand the complexity of your target company.
- Academics or not just academics. I would whish that academics could be seen as beacons of truth but the fact is that many have to make a living too and some do this by making reports for different institutions. This does not mean that it is not quality work that is being produced it just means that you can look at the reporting as you would a academic paper which have been evaluated by peers. Do not judge a report by the name on the cover.