Priniciples and Standards of CSR



September 2002

CSR and Global Business Principles: What a Mess!

By Michael Hopkins

CSR Definition

CSR is concerned with treating the stakeholders of the firm ethically or in a responsible manner. 'Ethically or responsible' means treating stakeholders in a manner deemed acceptable in civilised societies. Social includes economic responsibility. Stakeholders exist both within a firm and outside. The wider aim of social responsibility is to create higher and higher standards of living, treat while preserving the profitability of the corporation, nurse for peoples both within and outside the corporation[1].


This paper briefly reviews the hundreds of codes of conduct and principles around the world. It concludes that these codes are proliferating but rarely, clinic if ever, situate themselves within what has happened before and why the new code is different or advances on previous ones. Few, if any, have a theoretical basis for their codes and many simply cover just one or at most two stakeholders. There is a serious need of rationalization if companies are not to become even more confused about what is expected of them than is the case now.


There are an enormous number of codes, conventions, principles, standards available that have as their aim some aspect of improving the behaviour of corporations and more appear to arrive every day. An OECD report[2] identified 182 codes of practice and an ILO report[3] several hundred. The Table below lists some of the most well-known ones and categories them into four main initiatives: Governmental and Inter-Governmental initiatives; Company led; NGO led; Governance.

The case for code[4] endorsement stems from the need to:

  • · protect and enhance a company's reputation;
  • · establish the management's commitment to sound ethical behavior;
  • · raise stakeholder confidence;
  • · demonstrate corporate prevention rather than response to crises; and
  • · acknowledge corporate peer pressure for higher standards of accountability and transparency.

The case against code endorsement is:

  • · code proliferation is a counter-productive distraction that should be ignored;
  • · there is no unique, unchangeable and widely accepted code;
  • · codes do not take into account the running of a company and its major concerns to stay alive and avoid take-over or bankruptcy;
  • · codes may lead to third-party certification that is costly without added-value.

Overview of content

Codes vary according to the issue areas they cover. The OECD report categorized five main areas of conduct: 1) Fair Business Practices; 2) Observance of the Rule of Law; 3) Fair Employment and Labor Rights; 4) Environmental Stewardship; and 5) Corporate Citizenship. Codes also vary according to the sponsoring organization or partnership. According to a report[5] by the US Council for International Business (USCIB), most existing codes have been developed by individual companies for use in their own operations and management.

An ILO Report by Michael Urminsky defines 'code of conduct' as:

"a written policy, or statement of principles, intended to serve as the basis for a commitment to particular enterprise conduct".

In his study of 258 codes he found that 67% were devised by enterprises themselves, 11% by enterprise associations, 7% by NGOs, 8% by framework agreement (whatever that means), 3.5% by workers' organisations and only 0.4% by Governmental bodies. Of the codes examined, only 20 (8%) included some statement regarding commitment to communicating the code. While, only 6% were interested in external monitoring or inspection.

Most, if not all, codes are voluntary and few require formal and independent verification when used by companies. Their coverage is uneven - some provide principles of overall behaviour (OECD, EU, ILO Multi-National Codes), some refer to one stakeholder such as management (corporate codes of governance such as the King Commonwealth Report, Turnbull report in UK, Bush's ten principles on corporate governance report in USA[6]), or the environment (CERES principles, UNEP Financial Statement) or labour (SA8000, ICFTU Codes of Labour Practice, ILO TNC principles, FLA principles, ETI principles). Some refer to one or more stakeholders (US Model Business Practices refer to labour and the environment), the UN Global Compact refers to labour, human rights and the environment, the Global Sullivan Principles refer to external stakeholders, while others, particularly some of the more recent, cover most if not all stakeholders in a firm (GRi, AA1000s, Social Venture Network Standards on CSR etc.). As the USCIB notes:

"The trend toward multi-issue codes has led to the development of substantially identical codes that vary only in sponsoring organizations and/or geographic representation. Recent examples indicate that geography plays a key role in such cases" (USCIB Report p6).

The table below groups some of the main ones - striking is the concentration in Brussels, Geneva, London, Paris, New York and Washington DC.

Table: Most Well-Known Business Principles

Government and Intergovernmental Initiatives


Main Issues

ILO Tripartite Declaration on TNCs

Geneva,1977 revised March 2000


OECD Guidelines

Paris, 1976, revised June 2000

Multi-stakeholder including employment & triple bottom line

UN Global Compact

New York, July 2000

Human Rights, labour and environment

UNEP Financial Statement

Paris,1992, revised May 1997


EU Principles

Brussels, 2001 revised 2002


Voluntary Principles on Security and Human Rights, US and UK Govts

Wash. DC and London.Dec 2000

Security and Human Rights for extractive and energy industries

US Model Business Practices, US Dept. of Commerce

Wash. DC, 1996

Health & safety, labour, environment, corruption, community,law

Ethical Trading Initiative, UK Govt and NGO

London, Sept 1998

Labour practices essentially on trade links

Company led Initiatives    

Caux Principles



Fair Labor Association (FLA), US

California, Nov 1998

Labour practices

Global Sullivan Principles

USA, Nov. 1999

Mainly external stakeholders

ICC Business Charter for Sustainable Development

Brussels, 1991

Environment, health & safety

World Economic Forum

Davos/Geneva, 2002

Corporate Governance, Managers

NGO led Initiatives    

AccountAbility 1000

London, 1999 (revised 2002)

Social & ethical 'assurance' not environment, stakeholder model (SKs not defined)

Amnesty International HR Guidelines

London, Sept 1998

Human rights & security

CERES Principles

USA, 1989

Environmental ethical standards

Global Reporting Initiative (GRi)

Boston, 1997 revised 2002

Multi-stakeholder, triple bottom line indicators

ICFTU Code of Labor Practice

Brussels/Geneva, 1997

Labour and Trade Union issues


London, 1998 (revised 2002)

Says CSR but is mainly labour

Interfaith Center Global Codes of Conduct

New York, 1995


Social Venture Network Standards on CSR

Brussels, 1999

CSR multi-stakeholder, corporate governance mentioned, social audits

Sigma Sustainability Principles

London, 1999

Multi-stakeholder bias toward environment, no management s/h

Q-Res Codes of Ethics

Italy, 1999


Governance Initiatives


General Motors Board Guidelines



OECD Principles/Millstein Report (International)



Bosch Report, Australia



Merged Code Recommendations, Belgium



Turnbull Report, UK



King Report, South Africa

Johannesburg, 2002

Management & Ethics

King Report, Commonwealth

London, 2001

Management & Ethics

World Bank Corporate Governance Forum

Wash., DC, 2000

Management and finance


Recent prominent codes

Four prominent codes of conduct (following Urminsky's definition above) are the GRI, AA1000, SA8000 and the UN's Global Compact[7]. The GRi intends to help companies to produce social reports and, as its most recent report states:[8]

"the Guidelines are not: a code or set of principles of conduct; a performance standard (e.g., emissions target for a specific pollutant); or a management system"

One can only wonder what they are then?

Denial is also associated with the UK NGO AccountAbility 1000s[9] who associate themselves with GRi "the Guidelines can be used in conjunction with emerging approaches to standardised reporting such as the GRI"[10] but then assure themselves that they are really process orientated (to what? Certainly a code of best practice) when they say in the same document:

"Assurance is an evaluation against a specified set of principles and standards of the quality of specified public reports and the systems, processes and competencies that deliver the associated information and underpin the reporting organisation's performance."

SA8000[11] and the UN's Global Compact[12] admit to setting standards for company behaviour - the former for the labour stakeholder group and the latter to the three stakeholder groups labour, environment and the community (human rights).

Of these four, only the first two purport to be multi-stakeholder and the latter two are limited to a selection of stakeholders. Of them all, there is no doubt that the Global Reporting Initiative is currently the most wide encompassing of them all. It has been brave enough to attempt to devise a set of indicators so that companies can report on progress to meeting 'triple bottom line' objectives. It is not without its critics however[13]. Deloitte Touche have stated[14]

"We do believe that the core indicators required by the 2002 Exposure Draft are too voluminous and will discourage too many organizations from even attempting to report under the GRI guidelines. Further, we believe that the required boundaries of a sustainability report should not exceed the reporting entity's circle of control because it is unlikely that the reporting entity would have the ability to obtain the requisite information or determine its accuracy. Management may present supplementary, or additional, information on such matters, where relevant. We are concerned that the GRI is proceeding down a path of attempting to make a sustainability report be everything to everyone rather than focusing on how the reporting entity's sustainability performance can be measured overall. While we recognise that the latter form of a sustainability report will not suit each and every stakeholder's perceived needs, we believe it will result in a far more meaningful presentation."

Nike, too, were critical but appreciated GRi's flexibility:

"The depth of the questions is overwhelming at times, however the flexibility allowed by the structure makes the GRI more digestible and tenable than most surveys. It can also serve as a useful catalyst in engaging internal leaders in substantive discussion around governance and triple bottom line accountabilities."[15]

Conceptual difficulties

One of the biggest problems devisers of codes of conduct or principles of behaviour face is what conceptual basis to use. There is a proliferation of terms - CSR, corporate citizenship, corporate sustainability, business ethics etc[16]. I, obviously, prefer the definition of CSR above. Others tend to include confusing terms like 'sustainability' or 'triple bottom line' to weave their codes. The conceptual basis is often poor - I shall briefly show this with the TBL approach below.

A useful framework by which to think about codes of conduct is one developed by Donna Wood and is one on which I have developed a number of indicators. It is also multi-stakeholder in concept[17]. In this conceptual model, the question is asked does the company have a clear statement of principles, is this followed by a number of processes to implement these principles and what outputs can be measured. Thus CSR is measured following a business organization's configuration into three levels - The Triple-P Approach to CSR:

Principles of social responsibility

  • Processes of social responsiveness
  • Products (or Outcomes) as they relate to the firm's societal relationships

These can then be further divided, naturally, into the principles, processes and outcomes for each stakeholder group.

The GRi, in fact, uses implicitly this approach but gets a little confused by jumping in the triple bottom line bed. That latter notion was developed by John Elkington in his book Cannibals with Forks to describe his own process from the environmental field into the wider fields of social and economic considerations. Given that GRi took its first steps in the environmental field since it grew out of the environmental CERES principles, it is not surprising that it is a little hamstrung by environmental considerations.

The triple bottom line approach (TBL) is simply too confusing, and intellectually suspect, as a basis for a code of conduct. The initial attraction, of course, is that TBL appears to bring in concerns of the environment and society neatly alongside the usual business notions of profitability (the economic bottom line). However the TBL concept suffers from, at least, four main difficulties:

1. Company's cannot simply put profitability on the same level as social and environmental, a company cannot survive by either behaving socially or environmentally responsibly while making losses

  • 2. Social and environmental benefits tend to be long-term before impacting on stakeholder value
  • 3. TBL equates social with environmental, social clearly encompasses environmental as one among many other concerns
  • 4. The notion of stakeholders are not necessarily defined in the TBL approach

Concluding Remarks

Therefore, let corporations focus on creating stakeholder value as measured by profits, in a socially responsible manner. Don't let us add on a surplus less deficit approach on the environment or social considerations. A company that does poorly on one line - that of profits - but wonderfully on the environment or social component of TBL is not going to last long in a competitive world! So let's abandon that approach and concentrate on defining exactly what we mean by a corporations responsibility to its stakeholders, who they are and how to measure progress through precise, and a limited number, of key indicators.

Perhaps the key problem to date is that designers of codes suffer from a lack of context and history. None of them seem to survey the literature and then suggest either improvements or radically different approaches. If we in the CSR movement wish to bring more and more companies on board then please no more codes that don't situate themselves on what has gone on before!

Further References

1. Michael Kane: "Resources for Promoting Global Business Principles and Best Practices: A Directory of People, Organizations and Web Sites", United States Government, Environmental Protection Agency, Washington DC, updated regularly, August 2002, contact: This e-mail address is being protected from spambots. You need JavaScript enabled to view it.
  1. Rania Jamal: "USCIB Compendium of Corporate Responsibility Initiatives", US Council for International Business (USCIB), CR Working Group, 2001.
  2. Holly J. Gregory: "International comparison of corporate governance guidelines and codes of best practice in developed markets", 2001, Weil, Gotshal & Manges LLP, New York
  3. Michael Urminsky (ed.): "Self-regulation in the workplace: Codes of conduct, social labeling and socially responsible investment", Draft, ILO, Geneva, April 2002
  1. Business for Social Responsibility: "Comparison of Selected Social Responsibility related standards", Nov., 2000

[1] Michael Hopkins: CSR Matters, Earthscan, London, forthcoming
[2] See (
[3] Michael Urminsky (ed.): "Self-regulation in the workplace: Codes of conduct, social labeling and socially responsible investment", Draft, ILO, Geneva, April 2002
[4] Based upon the US Conference Board's 2001 survey of around 100 large companies.
[5] Rania Jamal: "USCIB Compendium of Corporate Responsibility Initiatives", US Council for International Business (USCIB), CR Working Group, 2001.
[7] However, each prefers a different label to code of conduct - GRi (reporting guidelines), AA1000 (Assurance), SA8000 (Standard), UN (principles)
[10] Claire Nacamuli and Tracey Swift "Corporate Responsibility: The Business Case for Assurance", Institute of Social and Ethical AccountAbility, London, UK, 2002
[11] See
[12] See

[13] See for instance "The Global Reporting Initiative - Raising the Bar too high?" by Mallen Baker, August, 2002,


[17] For a fuller description see