Introduction

Several Global leading financial institutions (FI) & long-term investor, including UBS, Blackrock and pension funds, are accounting climate risk into their investment decision as to protect their client asset, as well as limit the risk related to climate change and other ESG issues. Moreover, companies that operating with more sustainable guidance may face less unforeseeable risk due to policy changes and other uncertainties, therefore, it is an opportunities for the company to take step to operational efficiency and be more sustainable in the longer term. The product will be an aid and guidance for FI during making investing decision. Promoting ESG investing can provide the benefit to inspire the firm to modify the product design, procurement, energy/water usage and prioritize the competitive advantage, and also change the office culture.
THis project is to focus on visualizing the relationship between ESG rating and financial statistic.

Data Selection

I have selected MSCI World index as a benchmark that consisted of 1645 companies around the world which representing 85% of free floated market capital. For simplicity, statics weight of the index will be implemented, and only 70 companies that contribute most to MSCI world index in term of portfolio weight will be selected for this project, which is around 50% of weight of the MSCI World Index. The weight of the 70 companies that used in this project would be based on the weighting on 11/16/2018.

Final Data selection:

  1. Daily Price of the 70 stock
  2. Sustainalytics ESG Rank (Monthly)


Challenge:

• None-standardized date: o Since different ESG indicators have different update period (e.g. CDP Climate Score update every year, Sustainalytics Rank is updated every month), the unit needed to be adjusted before making any comparison.

• Available Data: o Some of the Mid- Cap companies may have different available data-set comparing with Large–Cap • No existing public data-set that has already done aggregating and data clearing that can be used for this project o Some potential good sources of ESG metrics (Arabesque’s S-ray) is only able to view but not able to download from Bloomberg.


Example

This is an unprocessed data sample of Microsoft, and only average age of board member and stock price data are available since the intercept of the data-set. Sustainalytics Rank is available since mid-2014, while other variables that related to environment are yet available.

DT::datatable(apple$`MSFT US EQUITY `[1:5])


Sample of processed data

The following two variables have been selected for the exploration 1. Stock Price 2. Sustainalytics Rank

The following technique has been used during the cleaning process:

  1. Scale: Every data-points are subtracted by the mean of its own sample and divided by the standard deviation
  2. Row Mean: Compute Monthly average for stock price
  3. Remove data-point that without value

Spread (different of scaled closed price and scaled sustainability rank) is also computed for visually comparison. formula: Monthly Average price - Sustainalytics Rank

DT::datatable(data_meanprice[1:5]) # Close Price of stock that has been converted to monthly average price
DT::datatable(sustain_rank_NA[1:5]) #Sustainalytics Rank that updated monthly


Explanation of Sustainalytics Rank Methodology

Sustainalytics, a leading sustainability research company, examine public traded companies sustainability performance based on the assessment issues of the environmental, social, and governance that the company faces in its business. Companies that do great on this evaluation are considered to be proactively managing the ESG issues, which are the material to their business.
Sustainalytics Rank are based on the following indicator

• Thematic and overall Environmental, Social and Governance scores, • Momentum indicators, which reflect ESG trends scores over time, • Controversy assessments, identifying high-profile environmental, social or governance incidents involving the company, • Preparedness, Disclosure, and Performance assessments which offer investors insights into a company’s ESG management and risk exposure, • Product involvement indicators highlighting company exposure to a list of 11 product lines, such as tobacco, nuclear power generation or military contracting. (From Website of Sustainalytics )


Exploratory Data Visualization

The following chart is plot with daily/ monthly IBM stock price against Sustainalytis Rank. The plot shows that the Sustainalytis Rank is the lagging indicator of the stock price, the reason behind will be determinate with further research.

For comparing relationship of changes of stock price and sustainalytics rank, unit standardization is necessary to make a accurate judgment. Following two graph was the spread ( different of scaled stock price and sustainlytics) of the selected companies.

The Spread plot indicated that there is a upward trend on the spread for most of the companies. Ideally, the spread should be flat, one of the explanation is because the most recent stock price upward momentum is exceed the Sustainalytics Rank’s upward momentum. It would be wise to look for the autocorrelation with different time lag is test to find out is there a autocorrelation between those stock price and ESG metrics.


Correlation

Finding the correlation between the time series of each company stock price and sustainalytics Rank. The higher the correlation represents the stock price goes in the same way as ESG ranking.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.


Interpretation: The correlation of Sustainalytics Rank and Stock price of all 70 companies has range (-0.3,0.3), with mean slightly less than zero, which indicated that there has not one clear answer to whether is there a correlation between those two variables for all companies.

Therefore it shows there is no linked between stock price and ESG ranking.

## Registered S3 methods overwritten by 'ggalt':
##   method                  from   
##   grid.draw.absoluteGrob  ggplot2
##   grobHeight.absoluteGrob ggplot2
##   grobWidth.absoluteGrob  ggplot2
##   grobX.absoluteGrob      ggplot2
##   grobY.absoluteGrob      ggplot2


Interpretation: The graph above shows the change in Sustainalytics Rank from Year 2014 to Year 2018. On the top of the chart indicated the company has made substantial change in the ESG factor and vice versa. Facebook has the highest change in ESG ranking, and United Technologies Corporation has the largest negative change in ESG ranking.


Calculating Value at Risk

Value at risk is a common sense fact that everyone can be understand, which is referring to the “worse case scenario” of the stock/ portfolio with in certain time frame, confident interval and loss percentage. The following calculating will be using the historical method, which is reorganize all historical return and find the worse x % of the return from the time series.

The following calculating use the data from 2014 to 2018 of the 70 companies that is selected for the project. Historical method is used during the calculation process.


The following Chart will answer whether a higher ESG rank will decrease the volatility of the stock. The underlying theory is the company scores higher on ESG ranking, the company that is active on innovating on the environment technology to reduce emission, treating employee well will tends to be innovative on their business model and recognized their risk on negative news and hence less volatile on average.

Interpretation: Value at risk is represents by the intensity of the color. The darker blue represents the higher value at risk. The x-axis represents the normalized Sustainalytics ESG Rank, the higher the ranking, the position in term of x-axis will be on to the right. The company list on the y axis is sorted by descending order of the Sustainalytics ESG rank.


Conclusion:

There is no solid confirmation that there is a statistical relationship on the ESG ranking and stock performance by implementing the chart above. However further statistical test should be conduct.