Companies’ GICS classifications are reviewed at least annually, and are also under constant surveillance for corporate actions potentially impacting classification. In general, a GICS code changes whenever there is a major corporate action that redefines a company's primary business activity. At a minimum, annual reviews are conducted to ensure that a company has not redefined its primary business activity through a series of smaller events. S&P and MSCI review the GICS structure on an annual basis.

All information regarding the GICS has been derived from publicly available information, without independent verification. This information reflects the policies of, and is subject to change by, S&P and MSCI.

Expert Judgement

Each of the Index Administrator and the Index Calculation Agent, as relevant, will exercise any discretion and make any determination in respect of the Index by using a standard of judgement (“Expert Judgement”) which shall consist of (1) acting in good faith and in a commercially reasonable manner; (2) to the extent practicable, reflecting the commercial objective of the Index and market practice; and (3) to the extent practicable, promoting consistency in the exercise of discretions and the making of determinations in respect of both the Index and other indices in respect of which it acts, as relevant, as index administrator or index calculation agent.

MSCI Barra Open Optimizer Events

If, on any weekday, the MSCI Barra Open Optimizer is temporarily unavailable or inaccessible by the Index Administrator or Index Calculation Agent, the Index Calculation Agent may:

If, on any weekday, the functionality of the MSCI Barra Open Optimizer is discontinued in whole or in part and is replaced by a successor optimizer (or combination of successor optimizers) (the “Successor Optimizer”) which is acceptable to the Index Administrator and the Index Calculation Agent and, in the joint determination of the Index Administrator and the Index Calculation Agent, uses the same or substantially similar methodology in calculation as the MSCI Barra Open Optimizer, then that Successor Optimizer will be deemed to be the relevant source for the calculations required to be carried out using the MSCI Barra Open Optimizer with effect from the date determined by the Index Calculation Agent.

If the Index Calculation Agent determines that a Successor Optimizer shall be used, it may make any adjustment(s) to the Index Rules as it determines appropriate to account for such change.

If: (1) MSCI makes or announces that it will make a material change in the formula for or method of calculation used in the MSCI Barra Open Optimizer (an “Optimizer Modification”), or (2) the MSCI Barra Open Optimizer ceases to exist or MSCI announces that it will cease to exist and no Successor Optimizer exists (an “Optimizer Cancellation”), then the Index Calculation Agent may:

be published for such day(s) but will instead be deemed to be the same as that of the immediately preceding Index Business Day;

Events Relating to the MSCI World Index

If the MSCI World Index is (1) not calculated and announced by MSCI but is calculated and announced by a successor sponsor acceptable to the Index Calculation Agent; or (2) replaced by a successor index using, in the determination of the Index Calculation Agent, the same or a substantially similar formula for or method of calculation as used in the calculation of such index, then in each case that index will be substituted for the MSCI World Index with effect from the date determined by the Index Calculation Agent, who may (but is not obliged to) make such adjustment(s) to the Index Rules as it determines appropriate to account for such change.

If a Reference Index Cancellation, a Reference Index Licence Termination or a Reference Index Modification occurs in respect of the MSCI World Index, then:

(1) the Index Calculation Agent may suspend the calculation, publication and dissemination of the Index and the Index Level until the first succeeding Index Business Day on which such event does not occur or continue to occur; and/or

(2) the Index Calculation Agent may (exercising Expert Judgement) replace the MSCI World Index with a replacement reference index which has substantially similar characteristics to the MSCI World Index, having regard to the manner in which the MSCI World Index is used in the calculation of the Index, in which case the Index Calculation Agent will:

(a) determine the effective date of such replacement; and

(b) make such adjustment(s) to the Index Rules as it determines appropriate to account for the effect on the Index of such replacement; and/or

(3) the Index Administrator may discontinue and cancel the Index.

“Reference Index Cancellation” means that MSCI permanently cancels the MSCI World Index.

“Reference Index Licence Termination” means that a license granted (if required) to the Index Administrator and/or the Index Calculation Agent and/or any of their respective affiliates to use the MSCI World Index in connection with the Index is terminated, or any such entity’s right to use the MSCI World Index in connection with calculating the Index is otherwise disputed, impaired or ceases for any reason.

“Reference Index Modification” means that MSCI announces that it will make a material change in the formula for or method of calculating the MSCI World Index or in any other way materially modifies the MSCI World Index (other than a modification prescribed in that formula or method to maintain the MSCI World Index in the event of routine events).

Cancellation, Termination or Modification of ESG Score

If an ESG Data Cancellation, an ESG Data Licence Termination or an ESG Data Modification occurs in respect of the ESG Score, then:

“ESG Data Cancellation” means that Arabesque permanently cancels the provision of the ESG Score.

“ESG Data Licence Termination” means that a licence granted (if required) to the Index Administrator and/or the Index Calculation Agent and/or any of their respective affiliates to use the ESG Score in connection with the Index is terminated, or any such entity’s right to use the such data in connection with calculating the Index is otherwise disputed, impaired or ceases for any reason.

“ESG Data Modification” means that Arabesque announces that it will make a material change in the method of calculating or providing the ESG Score or in any other way materially modifies the calculation or provision of the ESG Score (other than a modification prescribed in that method to maintain such data in the event of routine events).

Regulatory Event

Following the occurrence of a Regulatory Event, the Index Calculation Agent will determine (using Expert Judgement) whether or not such occurrence has a material effect on the Index.

(1)    If the Index Calculation Agent determines that such occurrence has a material effect on the Index, then the Index Calculation Agent will inform the Index Administrator of such determination and either:

(a) the Index Administrator will follow procedures in order to amend the Index Rules; or

(b) the Index Administrator may discontinue and cancel the Index.

(2)    If the Index Calculation Agent determines that such occurrence does not have a material effect on the Index, then with effect from (and including) a date designated by the Index Administrator (in which case the Index Administrator will notify the relevant date to the Index Calculation Agent) or the Index Calculation Agent:

(a) the constituent stock affected by such Regulatory Event (the “Removed Constituent”) will be removed from the Index; and

(b)  either (i) the weight of the remaining constituent stocks will be scaled up such that the weight of the Removed Constituent is proportionately redistributed to the remaining constituents or (ii) the Index Calculation Agent may replace the Removed Constituent with a replacement constituent which has substantially similar characteristics to the Removed Constituent, having regard to the manner in which the Removed Constituent is used in the calculation of the Index, in which case the Index Calculation Agent will determine the effective date of such replacement and make such adjustment(s) to the Index Rules as it determines appropriate to account for the effect on the Index of such replacement.

A “Regulatory Event” means that, owing to any change in applicable law or regulation, the Index Calculation Agent determines that either:

Index Governance

The Index Administrator has ultimate control over the development, the operation and the publication of the Index, including the performance of any calculation, the exercise of any discretion, the making of any determination, and all administrative processes required to perform these functions (together, the “Index Activity”). Notwithstanding that certain parts of the Index Activity may be performed by persons other than the Index Administrator, the Index Administrator has overall responsibility for all parts of the Index Activity.

The Index Administrator maintains oversight over the Index Activity through its Index Governance Committee. The Index Governance Committee fulfils its role of ensuring accountability and providing oversight through (1) reviewing and challenging all parts of the Index Activity, in accordance with its charter and its written policies and procedures; and (2) conducting an annual review of the Index to determine whether it continues to be an accurate and reliable representation of the economic realities of the relevant interest or market.

Notices

MSCI Indices notice

The MSCI World Index is used by Citigroup Global Markets Limited (“CGML”) as the selection universe for the selection of the companies included in the Index.

MSCI does not in any way sponsor, support, promote or endorse the Index. MSCI was not and is not involved in any way in the creation, calculation, maintenance or review of the Index. The MSCI World Index was provided on an “as is” basis. MSCI, each of its affiliates and each other person involved in or related to compiling, computing or creating the MSCI World Index (collectively, the “MSCI Parties”) expressly disclaim all warranties (including, without limitation, any warranties of originality, accuracy, completeness, timeliness, non-infringement, merchantability and fitness for a particular purpose). Without limiting any of the foregoing, in no event shall any of the MSCI Parties have any liability for any direct, indirect, special, incidental, punitive, consequential (including, without limitation, lost profits) or any other damages in connection with the MSCI World Index or the Index.

MSCI Barra Open Optimizer notice

The MSCI Barra Open Optimizer is used in the calculation of the Index. MSCI Barra Open Optimizer and Barra Optimizer are registered trademarks of Barra International LLC (“Barra”) and MSCI; these trademarks have been licensed to Citigroup Global Markets Limited.

Neither MSCI nor Barra in any way sponsors, supports, promotes or endorse the Index. Neither MSCI nor Barra was or is involved in any way in the creation, calculation, maintenance or review of the Index. The MSCI Barra Open Optimizer was provided on an “as is” basis. MSCI, Barra each of their affiliates and each other person involved in or related to compiling, computing or creating the MSCI Barra Open Optimizer (collectively, the “MSCI Parties”) expressly disclaim all warranties (including, without limitation, any warranties of originality, accuracy, completeness, timeliness, non-infringement, merchantability and fitness for a particular purpose). Without limiting any of the foregoing, in no event shall any of the MSCI Parties have any liability for any direct, indirect, special, incidental, punitive, consequential (including, without limitation, lost profits) or any other damages in connection with the use of the MSCI Barra Open Optimizer and/or the Index.

GICS notice

The Global Industry Classification Standard (“GICS”) was developed by and is the exclusive property and a service mark of MSCI and Standard & Poor's Financial Services LLC (“S&P”) and is licensed for use by the Index Administrator. Neither MSCI, S&P nor any other party involved in making or compiling the GICS or any GICS classifications makes any express or implied warranties or representations with respect to such standard or

classification (or the results to be obtained by the use thereof), and all such parties hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability or fitness for a particular purpose with respect to any of such standard or classification. Without limiting any of the foregoing, in no event shall MSCI, S&P, any of their affiliates or any third party involved in making or compiling the GICS or any GICS classifications have any liability for any direct, indirect, special, punitive, consequential or any other damages (including lost profits) even if notified of the possibility of such damages.

Arabesque Notice

We or one of our affiliates have entered into a license agreement with Arabesque pursuant to which we or our affiliates are entitled to use the ESG Score in connection with the Index in exchange for a fee.

The ESG Score does not constitute investment advice from Arabesque. The ESG Score is not a statement of fact or recommendations to purchase, sell or hold any securities. Provision of the ESG Score is not an asset management activity by Arabesque, and is not a solicitation or an offer to buy any security or instrument or to participate in any trading.

Hypothetical Back-Tested Index Performance Information

All Index performance information prior to December 15, 2020 is hypothetical and back-tested, as the Index did not exist prior to that date. Hypothetical back-tested Index performance information is subject to significant limitations. The Index Administrator developed the Index Rules with the benefit of hindsight—that is, with the benefit of being able to evaluate how the Index Rules would have caused the Index to perform had it existed during the hypothetical back-tested period. The fact that the Index generally appreciated over the hypothetical back-tested period may not therefore be an accurate or reliable indication of any fundamental aspect of the Index methodology. Furthermore, the hypothetical back-tested performance of the Index might look different if it covered a different historical period. The market conditions that existed during the hypothetical back-tested period may not be representative of market conditions that will exist in the future.

The hypothetical back-tested Index performance information has been calculated by the Index Administrator by applying the Index methodology substantially as described above to historical data and hypothetical back-tested ESG Scores calculated by Arabesque. Arabesque has published the version of the ESG Score currently used by the Index since March 2020. For periods prior to that date, Arabesque has calculated hypothetical back-tested ESG Scores. Arabesque has stated that it calculated the hypothetical back-tested ESG Scores by applying the methodology used to calculate the version of the ESG Score currently used by the Index to historical data. However, it is important to understand that the historical data used to calculate the hypothetical back-tested ESG Scores are subject to significant limitations. Specifically:

For these reasons, the hypothetical back-tested ESG Scores calculated by Arabesque may differ from the ESG Scores that would have been calculated had all of the Metrics been available throughout the back-tested period. Moreover, the companies selected for inclusion in the Index for purposes of calculating the hypothetical back-tested Index performance may differ from the companies that would have been selected if hypothetical back-tested ESG Scores were available throughout the back-tested period based on all of the Metrics and for the same universe of companies for which ESG Scores are available today. As a result, the hypothetical back-tested performance of the Index may differ from the performance the Index would have achieved had it actually existed during the back-tested period and been calculated based on the same data that are available today.

It is impossible to predict whether the Index will rise or fall. By providing hypothetical back-tested Index performance information in connection with any offering of any investment linked to the Index, we are not representing that the Index is likely to achieve gains or losses similar to those shown. In fact, there are frequently sharp differences between hypothetical performance results and the actual results subsequently achieved by any particular investment. One of the limitations of hypothetical performance information is that it did not involve financial risk and cannot account for all factors that would affect actual performance. The actual future performance of the Index may bear no relation to the hypothetical back-tested performance of the Index.

Description of the Arabesque S-Ray ESG Score

Overview

The Arabesque S-Ray ESG Score (the “ESG Score”) is intended to quantify a company’s performance on a range of environmental, social and governance (“ESG”) metrics. The ESG score is calculated by Arabesque S-Ray GmbH (“Arabesque”), a provider of ESG metrics and data services headquartered in Frankfurt, Germany. Arabesque has published an ESG Score since 2017 and has published the version of the ESG Score currently used by the Index since March 2020.

The ESG Score aggregates a company’s performance in 22 different ESG-related areas (referred to as “Features”) into a single score, where each Feature is weighted in a manner intended to reflect the financial materiality of that Feature to the sector and industry group to which the company belongs. See “—Materiality Weights” below for a discussion of how financial materiality is measured for purposes of the ESG Score.

A company’s performance with respect to each Feature is quantified by a Feature Score for that Feature. The Feature Score for each Feature measures a company’s performance on multiple data points relevant to that Feature relative to other companies. A company’s Feature Score based on these data points is subject to adjustment downward for negative news articles related to the relevant Feature, and subject to adjustment upward or downward for positive or negative non-governmental organization (“NGO”) reports about the company related to that Feature. See “—Features” below for information about the Features and “—Feature Scores” below for information about how the Feature Score for each Feature is calculated.

Arabesque currently calculates ESG Scores for over 7,000 publicly listed companies in both developed and emerging markets. We refer to the universe of companies for which Arabesque calculates an ESG Score as the “S-Ray Universe”.

There is no single accepted methodology for quantifying a company’s ESG performance. The ESG Score is just one approach. No assurance can be given that the ESG Score will accurately capture any given company’s performance in terms of the ESG factors that are important to any investor.

The description of the ESG Score contained herein is based on information provided by Arabesque. We have not independently verified that information.

Although the version of the ESG Score utilized by the Index on the date of this index supplement is the latest public version of Arabesque’s ESG Score, that may cease to be the case if Arabesque makes changes to the methodology by which it calculates the latest public version of its ESG Score that are not reflected in the version of the ESG Score utilized by the Index. See “—Versions” below for more information. For this reason, any ESG Scores that Arabesque may publish on its website may differ from the ESG Scores that are utilized by the Index. Arabesque will provide the current ESG Scores for each company in the S-Ray Universe, calculated based on the version of the ESG Score methodology then utilized by the Index, free of charge to any person who requests them from Arabesque at any time.

Calculation of ESG Score

The ESG Score for a given company is a value between 0 and 100 and is calculated as the weighted sum of the Feature Scores with respect to that company, where the weight given to each Feature is equal to its “Materiality Weight”, determined as described under “—Materiality Weights” below. The Materiality Weights of all Features for any given company sum to 100%.

The table below provides a hypothetical illustration of the calculation of the ESG Score for a hypothetical company. The ESG Score is the sum of the products of the Feature Score and Materiality Weight for each Feature.

Feature Score

The table below lists the 22 Features that are taken into account in the ESG Score.

A “Feature Score” for each Feature is determined for each company based on its Long-Term Feature Score for that Feature, as adjusted by its Short-Term Feature Score for that Feature. Each Feature Score is a number between 0 and 100.

The formula for calculating the Feature Score for each Feature is as follows:

Feature Score = Long-Term Feature Score × (1 + Short-Term Feature Score)

The Long-Term Feature Score for each Feature is a number between 0 and 100. The Short-Term Feature Score is a percentage that may range from -50% to +25%.

For example, if a company’s Long-Term Feature Score for a Feature were 50, and if its Short-Term Feature Score for that Feature were -25%, then that company’s Feature Score for that Feature would be calculated as follows:

Feature Score = 50 × (1 + -25%) = 37.50

Long-Term Feature Score

A company’s “Long-Term Feature Score” for each Feature is determined based on the company’s performance on one or more data points, or “Metrics”, relevant to that Feature relative to other companies. Arabesque obtains the Metrics used to calculate the Long-Term Feature Scores from third-party data providers, which in turn obtain information from company reports, filings and disclosures. The Long-Term Feature Score is referred to as “long-term” because the Metrics are published infrequently (e.g., annually or semi-annually). The Metrics used to calculate the Long-Term Feature Score for each Feature are listed in Annex D.

The Long-Term Feature Score for a given Feature and a given company is the weighted sum of the Metric Scores for that Feature. The Metric Scores for a given Feature with more than one Metric are weighted so that Metrics with a “Preparation” Focus make up 50% of the Long-Term Feature Score and Metrics with an “Outcome” Focus make up 50% of the Long-Term Feature Score. For Features with only one Metric, the Metric Score for that Metric is the Long-Term Feature Score for that Feature.

Metric Scores

The Metric Score for a given company and a given Metric generally reflects the performance of that company on the Metric in question relative to other companies, with a higher Metric Score reflecting more favorable performance, and vice versa.

The Metrics come in two different data types:

Arabesque obtains nearly all of the Metrics from third-party data providers IdealRatings, Institutional Shareholder Services (“ISS”) and Refinitiv. Arabesque obtains one Metric each from each of FactSet, Accenture and CFRA.

The Long-Term Feature Score for each Feature is calculated as the weighted sum of the Metric Scores for the Metrics that make up that Feature. The “Metric Score” for each Metric is intended to reflect the company’s performance on that Metric relative to other companies. The calculation of the Metric Score depends on the data provider, as follows:

The percentile rank of a company for a given Metric Score generally represents the percentage of companies in the region, sector or S-Ray Universe, as applicable, with less favorable performance on the applicable Metric than the company in question. To convert a percentile rank into a Metric Score, the relevant percentage is multiplied by 100.

For each Binary Metric, which will be either a “yes” or a “no” answer, the percentile rank of each company will be determined as follows:

For example, consider a hypothetical Binary Metric provided by IdealRatings. Suppose that there are 100 companies in the S-Ray Universe and that 50 have a “yes” answer for this Metric and 50 have a “no” answer. In this case, each company with a “no” answer would be ranked from 1 to 50, and each company with a “yes” answer would be ranked from 51 to 100. The average rank of each company with a “no” answer would be the sum of the rankings from 1 to 50, divided by 50, which would be 25.5. The average rank of each company with a “yes” answer would be the sum of the rankings from 51 to 100, divided by 50, which would be 75.5. Each of these results would be divided by the total number of companies in the dataset (i.e., 100), and then multiplied by 100, resulting in a Metric Score of 25.5 for each company with a “no” answer and 75.5 for each company with a “yes” answer.

Arabesque calculates Metric Scores for Metrics provided by Refinitiv using the statistical concept of a “cumulative distribution function”, or “CDF”. This method of calculating Metric Scores is similar to the percentile rank approach in that it produces a Metric Score that reflects how a given company compares to other companies in the S-Ray Universe in terms of the relevant Metric – under the CDF approach, just as with a percentile rank approach, Metric Scores range from 0 to 100 and a given company will have a higher Metric Score than all companies that rank below it in terms of the relevant Metric. However, whereas the percentile rank approach is a simple indication of the percentage of companies in the dataset that performed more poorly on the given Metric than the company in question, the CDF approach is an indication of the probability that a randomly selected company from the dataset would have a measurement in terms of the relevant Metric that is lower than that of the company in question. For this purpose, probability is determined using the statistical concepts of mean and standard deviation and assuming a normal distribution.

For example, a company whose performance is exactly equal to the mean of the performance of all companies in the dataset in terms of the relevant Metric would have a Metric Score of 50 under the CDF approach, because under a normal distribution there is a 50% chance that a randomly selected company would have a measurement that is less than the mean. A company whose performance in terms of the relevant Metric is one standard deviation above the mean would have a Metric Score of approximately 84, because under a normal distribution there is an approximately 84% probability that a randomly selected company would have a measurement that is less than one standard deviation above the mean. Similarly, a company whose performance in terms of the relevant Metric is one standard deviation below the mean would have a Metric Score of approximately 16, because under a normal distribution there is an approximately 16% probability that a randomly selected company would have a measurement

that is less than the value that is one standard deviation below the mean.

Metric Weights

Each Metric is assigned an attribute called “Focus”, which may be either “Preparation” or “Outcome”. Metrics with a “Preparation” Focus reflect data points which describe a company’s intentions with respect to a given topic. For example, the existence of an emissions reduction policy is taken as an indication that the company intends to implement certain environmental initiatives relating to emissions. However, this data point does not provide any information as to whether the company has been effective at implementing that policy. In contrast, Metrics with an “Outcome” Focus reflect data points which measure or quantify a company’s actual performance with respect to a given topic. For example, an “Outcome” Metric may measure a company’s actual CO₂ emissions. Annex D indicates the Focus of each Metric.

In calculating the Long-Term Feature Score for each Feature that has more than one Metric, Metrics with a “Preparation” Focus make up 50% of the Long-Term Feature Score, and Metrics with an “Outcome” Focus make up the other 50%. Within the group of Metrics with a “Preparation” Focus, each Metric is equally weighted with each other Metric with a “Preparation” Focus, and within the group of Metrics with an “Outcome” Focus, each Metric is equally weighted with each other Metric with an “Outcome” Focus. Accordingly, in calculating the Long-Term Feature Score of each Feature, each Metric with a “Preparation” Focus has a “Metric Weight” equal to 50% divided by the total number of Metrics with a “Preparation” Focus for that Feature, and each Metric with an “Outcome” Focus has a “Metric Weight” equal to 50% divided by the total number of Metrics with an “Outcome” Focus for that Feature.

For example, if for a given Feature there are five Metrics with a “Preparation” Focus and ten Metrics with an “Outcome” Focus, then the Metric Weight of each Metric with a “Preparation” Focus will be 10% (i.e., 50% divided by five) and the Metric Weight of each Metric with an “Outcome” Focus will be 5% (i.e., 50% divided by ten).

Adjustments

After summing the products of the Metric Scores and Metric Weights for each Metric, a normalization adjustment is added to determine the Long-Term Feature Score (except where the Long-Term Feature Score would exceed 100). The normalization adjustment is a fixed value of 5 and is intended to cause the Feature Scores to have a median of approximately 50, offsetting the reduction resulting from the News-Based Correction Factor, which Arabesque has determined tends to represent approximately 5 points.

For a given company, if no data is available with respect to a given Metric, then that Metric will be ignored (i.e., treated as if it were not one of the Metrics) for purposes of calculating the Long-Term Feature Score for that company.

Hypothetical Calculation of Long-Term Feature Score

The table below provides a hypothetical illustration of the calculation of the Long-Term Feature Score for the Emissions Feature for a hypothetical company. The Long-Term Feature Score for each Feature is the sum of the products of the Metric Score and Metric Weight of each Metric for that Feature, plus the normalization adjustment.

Note that not all Metrics that relate to the Emissions Feature are included in this calculation, as it is common that companies do not have data available for all Metrics for a given Feature.

Short-Term Feature Score

The Short-Term Feature Score for each company and Feature is intended to reflect negative news articles and positive or negative NGO reports about that company and related to that Feature within the last year. The Short-Term Feature Score is referred to as “short-term” because it is intended to reflect information on a more current basis than the Long-Term Feature Score, although it may reflect information that is up to a year old.

The Short-Term Feature Score is calculated based on a News-Based Correction Factor and an NGO-Based Correction Factor. The News-Based Correction Factor may range in value between -50% and 0%. The NGO-Based Correction Factor may range in value between -25% and +25%.

As a result, negative news stories about a company that relate to a given Feature may reduce that company’s Long-Term Feature Score for that Feature by up to 50%. Positive or negative reports by NGOs about a company that relate to a given Feature may either reduce or enhance that company’s Long-Term Feature Score for that Feature by up to 25%.

If the News-Based Correction Factor and the NGO-Based Correction Factor are both negative, then the Short-Term Feature Score is equal to whichever factor is most negative. If the NGO-Based Correction Factor is positive, then the Short-Term Feature Score is the sum of the News-Based Correction Factor and the NGO-Based Correction Factor.

As a result, the Short-Term Feature Score is always between -50% and +25%.

News-Based Correction Factor

The News-Based Correction Factor is determined based on data provided to Arabesque by Covalence SA, a provider of ESG-related data based in Geneva, Switzerland.

Covalence identifies and classifies news articles on a daily basis. For each article that Covalence identifies as having ESG-related content with respect to a company, Covalence provides the following data to Arabesque:

Arabesque takes the data provided by Covalence and uses a pre-defined set of internal rules to produce:

For each negative article, Arabesque calculates its Present News Value by multiplying its Impact Score by its

Magnitude Score. The Present News Value of each negative article may range from 1 to 9.

The News-Based Correction Factor for a given Feature and a given company is calculated based on the news article published within the past year with respect to that Feature and that company that has the highest Present News Value. To determine the News-Based Correction Factor, the highest Present News Value is divided by 9, and then the result is multiplied by -50%. For example, if the highest Present News Value of all negative articles relating to a given Feature and a given company over the past year was 9, then the News-Based Correction Factor for that company and that Feature would be -50%, and if the highest Present News Value of all negative articles relating to that Feature and company over the past year were 1, then the News-Based Correction Factor for that company and that Feature would be -5.56% (1/9 times -50%). If there were no negative articles relating to that Feature and company over the past year, then there would be no News-Based Correction Factor.

Covalence’s news monitoring process relies on an in-house web scraper and crawler, on automated extraction techniques, on an algorithm and database, machine learning, sentiment analysis, natural language processing, as well as on human analysis. All information herein regarding Covalence and its services is based on publicly available information published by Covalence. We have not independently verified any of that information, and we do not make any representation about the effectiveness of Covalence at identifying and classifying news articles.

NGO-Based Correction Factor

The NGO-Based Correction Factor is determined based on data provided to Arabesque by Sigwatch, a UK-based provider of global NGO and ESG issue tracking and reputational impact data.

For each NGO campaign that it reviews, SigWatch maps that campaign to a company and a Feature and provides the following scores to Arabesque:

For each NGO campaign identified by Sigwatch, Arabesque calculates a “Present Campaign Value” according to the following formula:

Present Campaign Value = [(Sentiment Score × Prominence Score) + Partnership Score] × Impact Score

The highest possible positive Present Campaign Value of a NGO campaign by a single NGO sponsor is 24.75, and the lowest possible negative Present Campaign Value of a NGO campaign by a single NGO sponsor is -22. If there is more than one NGO sponsor of a NGO campaign, the Impact Scores of up to five NGO sponsors are summed, and as a result the highest possible positive Present Campaign Value is 123.75, and the lowest possible negative Present Campaign Value is -110.

The NGO-Based Correction Factor for a given Feature and a given company is calculated based on the NGO campaign from the past year relating to that Feature and company with the highest positive Present Campaign Value (if any) and the NGO campaign from the past year relating to that Feature and company with the lowest negative Present Campaign Value (if any). Those two Present Campaign Values are then added up, and the sum is divided by 110. The result is then multiplied by 25%, subject to a minimum NGO-Based Correction Factor of -25% and a maximum NGO-Based Correction Factor of 25%. As a result:

25%);

All information herein regarding SigWatch and its services is based on publicly available information published by SigWatch. We have not independently verified any of that information, and we do not make any representation about the effectiveness of SigWatch at identifying and scoring NGO campaigns.

Materiality Weight

In calculating the ESG Score for a given company, each Feature Score is weighted according to a determination of the “financial materiality” of the applicable Feature to that company, based on the sector, industry and industry group to which the company belongs. The rationale for weighting the Feature Scores in this manner is to recognize that the relevance of a given Feature can vary across sectors, industries and industry groups. For example, emissions may be a more relevant, or more material, factor to companies in the energy sector, but less material for companies in the financial or technology sectors. On a quarterly basis, Arabesque calculates a Materiality Weight for each Feature for each company in its data set as described below.

The Materiality Weight assigned to a given Feature for a given company is a percentage value, ranging between 0% and 100%, determined based on a combination of two different approaches to assessing materiality:

SASB’s materiality determinations make up 20% of each Materiality Weight. The remaining 80% is determined based on statistical regression.

As further described below, for each Feature, a company will be given a SASB-based materiality score of either 0 or 3, and a regression-based materiality score of between 0 and 12. The sum of the SASB-based materiality scores and regression-based materiality scores (which sum will be between 0 and 15) will then be divided by 15, resulting in a preliminary Material Weight. Once the preliminary Materiality Weights have been determined for all Features, they are scaled proportionately so as to sum for all Features to 100%.

SASB Materiality Determination

SASB is a nonprofit standard-setting organization that develops and maintains reporting standards to enable businesses to identify, manage and communicate financially material sustainability information to their investors.

SASB publishes a “Materiality Map”, which identifies sustainability issues that SASB has determined are likely to affect the financial condition or operating performance of companies within an industry. The SASB Materiality Map indicates, for each of 26 sustainability issues and each of 77 different industries, SASB’s determination as to whether that sustainability issue is likely a material issue for companies in that industry. SASB makes the Materiality Map publicly available on its website. We are not incorporating any information on SASB’s website into this document.

For purposes of calculating the Materiality Weights, if the SASB Materiality Map indicates that a given Feature is likely material to companies in a given industry, then the SASB-based materiality score for that Feature for all companies in that industry will be 3, and if the SASB Materiality Map indicates that a given Feature is not likely material to companies in a given industry, then the SASB-based materiality score for that Feature for all companies in that industry will be 0. The 26 sustainability issues covered by the Materiality Map do not correspond exactly to the 22 Features. Therefore, in determining whether the SASB Materiality Map indicates that a given Feature is

likely material to companies in a given industry, Arabesque has made a determination as to which SASB sustainability issues correspond to each Feature.

SASB classifies companies within 77 different industries under its proprietary Sustainability Industry Classification System (“SICS”). The SICS classification system differs from traditional industry classification systems. Unlike other industry classification systems—which use common financial and market characteristics to group companies within industries—the SICS classification system uses sustainability profiles to group similar companies within industries. In the SICS classification system, a company’s sustainability risks and opportunities are more important for its classification than other traditional factors, such as economic cycles and revenue streams. SASB determines a company’s SICS classification by overlaying SASB’s sustainability framework to other industry taxonomies.

All information herein regarding SASB and the SASB Materiality Map is based on publicly available information published by SASB. We have not independently verified any of that information.

Regression-Based Materiality Determinations

Arabesque conducts a statistical analysis in an attempt to identify those Features that have historically had a statistically significant relationship with the stock price performance of companies in a given sector and industry group. Arabesque conducts this analysis for each company and Feature with respect to 12 different historical timeseries, as described in more detail below. If, based on this analysis, Arabesque determines that there has been a statistically significant relationship between a given Feature and the stock price performance of companies in a given sector or industry group for a given historical timeseries, then that Feature will be given a regression-based materiality score of 1 for companies in that sector or industry group for that historical timeseries. Alternatively, if Arabesque determines that there has not been a statistically significant relationship between a given Feature and the stock price performance of companies in a given sector or industry group for a given historical timeseries, then that Feature will be given a regression-based materiality score of 0 for companies in that sector or industry group for that historical timeseries. Because Arabesque conducts this analysis with respect to 12 historical timeseries for each company and Feature, the sum of the regression-based materiality scores for a given Feature for all historical timeseries for a given company may range from 0 to 12.

For purposes of this analysis, a timeseries is a combination of three parameters – a Lookback Period, a Scope and a Weighting. The “Lookback Period” may be 1 year, 3 years or 5 years. The “Scope” may be either an industry group or a sector (as defined by FactSet’s RBICS Focus product). The “Weighting” may be either market capitalization or equal. For example, one timeseries consists of the monthly returns of a particular industry group over the prior 1 year, where the stocks in that group are weighted by market capitalization, and another timeseries consists of the monthly returns of a particular sector over the prior 5 years, where the stocks in that sector are equally weighted. There are 12 possible combinations of these three parameters, and therefore 12 historical timeseries.

For purposes of this analysis, Arabesque uses FactSet’s Revere Business and Industry Classification System (RBICS) Focus product to classify companies within sectors and industry groups. The RBICS Focus system classifies approximately 45,000 global public companies within 32 sectors and 317 industry groups. Companies are classified within sectors based on stock co-movement and within industry groups based on the products and services they offer. All information herein regarding FactSet and its RBICS Focus system is based on publicly available information published by FactSet. We have not independently verified any of that information.

To determine whether a given Feature had a statistically significant relationship with the stock price returns of a given sector or industry group for a given historical timeseries, Arabesque conducts a statistical analysis on the relationship between:

to as the “residual” return.

The return of the long-short portfolio represents the difference in the return of the companies in the relevant sector or industry group with the highest Feature Scores for the relevant Feature from the return of the companies in that sector or industry group with the lowest Feature Scores for that Feature. The statistical approach used by Arabesque assumes that this difference is in some way attributable to the relevant Feature. This may or may not be the case, and if it is not the case, the output of the statistical analysis may not provide meaningful information.

The return of the long-short portfolio is then regressed against the residual return of the relevant timeseries from the 3-Factor Model. The 3-Factor Model, proposed by Professors Eugene Fama and Kenneth French, describes the historical returns of a group of stocks as a function of the sensitivity of the returns of that group of stocks to each of 3 factors:

In other words, the 3-Factor Model proposes that, for any given group of stocks and any given time period, the return of those stocks over that time period can be explained to a certain degree by the sensitivity of the returns of that group of stocks to these 3 factors. The portion of the return that is not explained by the three factors in the 3-Factor Model is referred to as the residual.

The statistical analysis utilized by the Arabesque methodology seeks to determine whether there is a statistically significant relationship between the returns of the long-short portfolio in respect of a particular timeseries and Feature, on the one hand, and the residual returns of the companies included in that timeseries, on the other – i.e., the portion of the historical returns of those companies that is not explained by the 3 factors in the 3-Factor Model. This approach is premised on the idea that a given Feature is financially material to the companies in a particular sector or industry group if the return of the relevant long-short portfolio explains (in a statistically significant way) a portion of the historical returns of that sector or industry group that are not explained by the 3 factors in the 3-Factor Model.

The statistical tool that Arabesque uses for this analysis is called regression. A regression is a statistical technique to find the line that best fits the data points plotted on a graph, where the x-axis represents monthly returns of the long-short portfolio for the relevant timeseries and the y-axis represents the monthly residual returns for that timeseries. The line of best fit is the line drawn through the graph that minimizes the distances between each data point and the line. The more closely the line fits the data points, the greater the statistical significance of the relationship between the long-short portfolio returns and the residual returns, and the more explanatory power the relevant Feature is deemed to have with respect to the residual returns.

The 3-Factor Model is just one theoretical model of stock price returns. It is based on certain assumptions, including that small capitalization stocks tend to outperform large capitalization stocks and that value stocks tend to outperform growth stocks over the long term. These assumptions may or may not prove true over any time period in the future. Moreover, other theoretical models exist, including a five-factor model proposed by Professors Fama and French. There can be no assurance that the 3-Factor Model will accurately identify the factors that explain stock price returns or that the 3-Factor Model will be as successful as other models that could have been used.

Hypothetical Example of Calculation of Materiality Weight

The table below presents a hypothetical calculation of the Materiality Weight of each Feature for a hypothetical company based on the hypothetical SASB-based materiality scores and regression-based materiality scores for each of 12 historical timeseries indicated below.

Data Availability

In order for a company to receive an ESG Score from Arabesque, the company must have a Feature Score for at least one environmental Feature, one social Feature and one governance Feature. In order for a company to receive a Feature Score with respect to a given Feature, data must be available with respect to at least two Metrics relating to that Feature (if the Feature has more than one Metric).

It is important to note that this means that at any point in time many companies that receive an ESG Score will not have Feature Scores for all Features. Moreover, many companies will have Feature Scores for a given Feature even where not all Metrics are available with respect to that company for that Feature.

Versions

We expect that Arabesque will update the methodology by which it calculates the ESG Score from time to time. Arabesque has advised us that it intends to update the ESG Score methodology in 2021. We can provide no assurance as to whether or when Arabesque may update the ESG Score methodology or as to the nature of the changes Arabesque may make.

The current version of the ESG Score is version 2.6.2, which is the latest public version of Arabesque’s ESG Score methodology as of the date of this index supplement. If Arabesque updates its ESG score methodology in the future, it is possible that the version of the ESG Score that is used by the Index will diverge from the latest public version that Arabesque uses for other purposes. The version of the ESG Score used by the Index may be updated in the future only to reflect changes to the Features, the Metrics or the manner in which the Long-Term Feature Score is calculated. If Arabesque wishes to make other changes to its methodology, it may implement those in the public version of its ESG score, but not in the version used by the Index. As a result, it is possible that the Index will cease to be calculated based on the latest public version of Arabesque’s ESG score.

The version number of the version of the ESG Score used by the Index will always start with the number 2. If Arabesque updates its ESG Score methodology and the new version has a version number that starts with 2, that version will become the version of the ESG Score used by the Index. If a new version of the ESG score starts with a number other than 2, that version will not be used by the Index. If Arabesque updates the ESG Score to a new version that will be used by the Index, there is likely to be a delay between the launch of the new version by Arabesque and its incorporation in the Index to permit the Index Administrator to prepare for the incorporation of the new version.

At the time of any offering of securities linked to the Index, it is possible that Arabesque may have announced or disclosed to us changes that it intends to make to the methodology by which it calculates the ESG Score and that any such changes are not reflected in this index supplement.

It is important to understand both that the ESG Score methodology may change in the future in unanticipated ways, and that the ESG Score used by the Index may not keep up-to-date with the latest public version of Arabesque’s ESG score. The Index may perform less favorably than it would have if the ESG Score methodology did not change, or if the ESG Score methodology kept up with changes made to Arabesque’s latest public version.

Description of the MSCI World Index

Overview

The MSCI World Index^SM is a free float-adjusted market capitalization index that is designed to measure the equity market performance of the large- and mid-cap segments of certain developed markets. The MSCI World Index^SM currently consists of the following 23 developed market country indices: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Hong Kong, Ireland, Israel, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Singapore, Spain, Sweden, Switzerland, the United Kingdom and the United States. The MSCI World Index^SM covers approximately 85% of the free float-adjusted market capitalization in each country. The U.S. dollar price return version of the MSCI World Index^SM is reported by Bloomberg L.P. under the ticker symbol “MXWO.”

The MSCI World Index^SM is composed of 23 individual developed market country indices, each of which is a Standard Index calculated pursuant to the MSCI Global Investable Market Index methodology described below.

Constructing the MSCI Global Investable Market Indices

The MSCI Global Investable Market Indices are constructed and maintained at an individual market level. MSCI undertakes an index construction process that involves: (i) defining the Equity Universe for each market; (ii) determining the Market Investable Equity Universe for each market; (iii) determining market capitalization size-segments for each market; and (iv) applying Index Continuity Rules for the Standard Index.

Defining the Equity Universe

Plan of Distribution; Conflicts of Interest

 

The terms and conditions set forth in the Amended and Restated Global Selling Agency Agreement dated April 7, 2017 among Citigroup Global Markets Holdings Inc., Citigroup Inc. and the Agents listed on Schedule I thereto, including CGMI, govern the sale and purchase of the Securities.

 

The Securities will not be listed on a securities exchange, unless otherwise specified in the applicable pricing supplement.

 

Unless otherwise specified in the applicable pricing supplement, CGMI, an affiliate of Citigroup Global Markets Holdings Inc., will be the underwriter of the sale of the Securities and will purchase the Securities as principal from Citigroup Global Markets Holdings Inc. at the public offering price less an underwriting discount specified in the applicable pricing supplement. CGMI may offer the Securities directly to the public at the public offering price specified in the applicable pricing supplement. CGMI may also offer the Securities to selected dealers, which may include dealers affiliated with Citigroup Global Markets Holdings Inc., at the public offering price less a selling concession specified in the applicable pricing supplement.

 

A portion of the net proceeds from the sale of the Securities will be used to hedge Citigroup Global Markets Holdings Inc.’s obligations under the Securities. Citigroup Global Markets Holdings Inc. may hedge its obligations under the Securities through an affiliate of Citigroup Global Markets Holdings Inc. and CGMI or through unaffiliated counterparties, and CGMI or such counterparties may profit from such expected hedging activity even if the value of the Securities declines. This hedging activity could affect the value of the Index and, therefore, the value of and your return on the Securities. You should refer to the section “Risk Factors Relating to the Securities—Risk Factors Relating to All Securities—The value of the Index may be affected by our or our affiliates’ hedging and other trading activities” in this index supplement, the section “Plan of Distribution; Conflicts of Interest” in the accompanying prospectus supplement and the section “Use of Proceeds and Hedging” in the accompanying prospectus.

 

CGMI is an affiliate of Citigroup Global Markets Holdings Inc. Accordingly, each offering will conform with the requirements addressing conflicts of interest when distributing the securities of an affiliate set forth in Rule 5121 of the Financial Industry Regulatory Authority. Client accounts over which Citigroup Inc. or its subsidiaries have investment discretion will not be permitted to purchase the Securities, either directly or indirectly, without the prior written consent of the client.

 

Citigroup Global Markets Holdings Inc. has agreed to indemnify CGMI against liabilities relating to material misstatements and omissions with respect to the Securities, or to contribute to payments that CGMI may be required to make relating to these liabilities. Citigroup Global Markets Holdings Inc. will reimburse CGMI for customary legal and other expenses incurred by CGMI in connection with the offer and sale of the Securities.

 

Secondary market sales of securities typically settle two business days after the date on which the parties agree to the sale. If the issue date for the Securities is more than two Business Days after the Pricing Date, investors who wish to sell the Securities at any time prior to the second Business Day preceding the issue date will be required to specify an alternative settlement date for the secondary market sale to prevent a failed settlement. Investors should consult their own investment advisors in this regard.

 

Certain Selling Restrictions

 

Prohibition of Sales to EEA Retail Investors

 

The Securities may not be offered, sold or otherwise made available to any retail investor in the European Economic Area.  For the purposes of this provision:

 

(a)

the expression “retail investor” means a person who is one (or more) of the following:

 

(i)

a retail client as defined in point (11) of Article 4(1) of Directive 2014/65/EU (as amended, “MiFID II”); or

 

IS-62 

 

(ii)

a customer within the meaning of Directive 2002/92/EC, where that customer would not qualify as a professional client as defined in point (10) of Article 4(1) of MiFID II; or

 

(iii)

not a qualified investor as defined in Directive 2003/71/EC; and

 

(b)

the expression “offer” includes the communication in any form and by any means of sufficient information on the terms of the offer and the Securities offered so as to enable an investor to decide to purchase or subscribe the Securities.

 

IS-63 

 

Benefit Plan Investor Considerations

 

A fiduciary of a pension, profit-sharing or other employee benefit plan subject to the Employee Retirement Income Security Act of 1974, as amended (“ERISA”), including entities such as collective investment funds, partnerships and separate accounts whose underlying assets include the assets of such plans (collectively, “ERISA Plans”), should consider the fiduciary standards of ERISA in the context of the ERISA Plan’s particular circumstances before authorizing an investment in the Securities. Among other factors, the fiduciary should consider whether the investment would satisfy the prudence and diversification requirements of ERISA and would be consistent with the documents and instruments governing the ERISA Plan.

 

Section 406 of ERISA and Section 4975 of the Internal Revenue Code of 1986, as amended, (the “Code”) prohibit ERISA Plans, as well as plans (including individual retirement accounts and Keogh plans) subject to Section 4975 of the Code (together with ERISA Plans, “Plans”), from engaging in certain transactions involving the “plan assets” with persons who are “parties in interest” under ERISA or “disqualified persons” under Section 4975 of the Code (in either case, “Parties in Interest”) with respect to such Plans. As a result of our business, we, and our current and future affiliates, may be Parties in Interest with respect to many Plans. Where we (or our affiliate) are a Party in Interest with respect to a Plan (either directly or by reason of our ownership interests in our directly or indirectly owned subsidiaries), the purchase and holding of the Securities by or on behalf of the Plan could be a prohibited transaction under Section 406 of ERISA and/or Section 4975 of the Code, unless exemptive relief were available under an applicable exemption (as described below).

 

Certain prohibited transaction class exemptions (“PTCEs”) issued by the U.S. Department of Labor may provide exemptive relief for direct or indirect prohibited transactions resulting from the purchase or holding of the Securities. Those class exemptions are PTCE 96-23 (for certain transactions determined by in-house asset managers), PTCE 95-60 (for certain transactions involving insurance company general accounts), PTCE 91-38 (for certain transactions involving bank collective investment funds), PTCE 90-1 (for certain transactions involving insurance company separate accounts) and PTCE 84-14 (for certain transactions determined by independent qualified asset managers). In addition, ERISA Section 408(b)(17) and Section 4975(d)(20) of the Code may provide a limited exemption for the purchase and sale of the Securities and related lending transactions, provided that neither the issuer of the Securities nor any of its affiliates have or exercise any discretionary authority or control or render any investment advice with respect to the assets of the Plan involved in the transaction and provided further that the Plan pays no more, and receives no less, than adequate consideration in connection with the transaction (the so-called “service provider exemption”). There can be no assurance that any of these statutory or class exemptions will be available with respect to transactions involving the Securities.

 

Accordingly, the Securities may not be purchased or held by any Plan, any entity whose underlying assets include “plan assets” by reason of any Plan’s investment in the entity (a “Plan Asset Entity”) or any person investing “plan assets” of any Plan, unless such purchaser or holder is eligible for the exemptive relief available under PTCE 96-23, 95-60, 91-38, 90-1 or 84-14 or the service provider exemption or there is some other basis on which the purchase and holding of the Securities will not constitute a non-exempt prohibited transaction under ERISA or Section 4975 of the Code. Each purchaser or holder of the Securities or any interest therein will be deemed to have represented by its purchase or holding of the Securities that (a) it is not a Plan and its purchase and holding of the Securities is not made on behalf of or with “plan assets” of any Plan or (b) its purchase and holding of the Securities will not result in a non-exempt prohibited transaction under Section 406 of ERISA or Section 4975 of the Code.

 

Certain governmental plans (as defined in Section 3(32) of ERISA), church plans (as defined in Section 3(33) of ERISA) and non-U.S. plans (as described in Section 4(b)(4) of ERISA) (“Non-ERISA Arrangements”) are not subject to these “prohibited transaction” rules of ERISA or Section 4975 of the Code, but may be subject to similar rules under other applicable laws or regulations (“Similar Laws”). Accordingly, each such purchaser or holder of the Securities shall be required to represent (and deemed to have represented by its purchase of the Securities) that such purchase and holding is not prohibited under applicable Similar Laws.

 

Due to the complexity of these rules, it is particularly important that fiduciaries or other persons considering purchasing the Securities on behalf of or with “plan assets” of any Plan consult with their counsel regarding the relevant provisions of ERISA, the Code or any Similar Laws and the availability of exemptive relief under PTCE 96-23, 95-60, 91-38, 90-1, 84-14, the service provider exemption or some other basis on which the acquisition and

 

IS-64 

 

holding will not constitute a non-exempt prohibited transaction under ERISA or Section 4975 of the Code or a violation of any applicable Similar Laws.

 

The Securities are contractual financial instruments. The financial exposure provided by the Securities is not a substitute or proxy for, and is not intended as a substitute or proxy for, individualized investment management or advice for the benefit of any purchaser or holder of the Securities. The Securities have not been designed and will not be administered in a manner intended to reflect the individualized needs and objectives of any purchaser or holder of the Securities.

 

Each purchaser or holder of any Securities acknowledges and agrees that:

 

(i)       the purchaser or holder or its fiduciary has made and shall make all investment decisions for the purchaser or holder and the purchaser or holder has not relied and shall not rely in any way upon us or our affiliates to act as a fiduciary or adviser of the purchaser or holder with respect to (A) the design and terms of the Securities, (B) the purchaser or holder’s investment in the Securities, or (C) the exercise of or failure to exercise any rights we have under or with respect to the Securities;

 

(ii)       we and our affiliates have acted and will act solely for our own account in connection with (A) all transactions relating to the Securities and (B) all hedging transactions in connection with our obligations under the Securities;

 

(iii)       any and all assets and positions relating to hedging transactions by us or our affiliates are assets and positions of those entities and are not assets and positions held for the benefit of the purchaser or holder;

 

(iv)       our interests are adverse to the interests of the purchaser or holder; and

 

(v)       neither we nor any of our affiliates is a fiduciary or adviser of the purchaser or holder in connection with any such assets, positions or transactions, and any information that we or any of our affiliates may provide is not intended to be impartial investment advice.

 

Each purchaser and holder of the Securities has exclusive responsibility for ensuring that its purchase, holding and subsequent disposition of the Securities does not violate the fiduciary or prohibited transaction rules of ERISA, the Code or any applicable Similar Laws. The sale of any Securities to any Plan is in no respect a representation by us or any of our affiliates or representatives that such an investment meets all relevant legal requirements with respect to investments by Plans or Non-ERISA Arrangements generally or any particular Plan or Non-ERISA Arrangement, or that such an investment is appropriate for Plans or Non-ERISA Arrangements generally or any particular Plan or Non-ERISA Arrangement.

 

However, individual retirement accounts, individual retirement annuities and Keogh plans, as well as employee benefit plans that permit participants to direct the investment of their accounts, will not be permitted to purchase or hold the Securities if the account, plan or annuity is for the benefit of an employee of CGMI or a family member and the employee receives any compensation (such as, for example, an addition to bonus) based on the purchase of Securities by the account, plan or annuity.

 

IS-65 

 

Annex A
Index Conditions (Framework)

 

 

 

 

 

 

 

 

 

IS-66 

 

 

Citi World ESG Indices

Index Conditions (Framework)

Citi Investment Strategies

21 January 2021

IS-67