The distinction matters. Pension funds, insurers and endowments hold a substantial share of listed equity, which gives them both the legal standing and the economic weight to influence how companies are governed. Whether that influence is exercised deliberately, or delegated to a default, determines whether stewardship is a discipline or merely a disclosure.
From agency to stewardship
For much of the late twentieth century, corporate governance was framed through agency theory: managers, left unchecked, would pursue their own interests at the expense of owners. Stewardship theory offered a counterpoint. Davis et al. (1997) argued that managers can act as stewards whose motivations align with those of the principals they serve, provided governance structures support that alignment rather than merely police it.
The concept migrated from the boardroom to the investment chain as ownership itself changed. Hawley and Williams (2000) described the rise of fiduciary capitalism, in which large institutions holding broadly diversified portfolios effectively own a cross-section of the economy. Such owners cannot easily exit underperforming companies, and they bear the system-wide costs that individual firms impose. Engagement, rather than divestment, becomes the rational instrument.
The Kay Review brought this argument into public policy in 2012, concluding that short-termism in equity markets had eroded the relationships of trust on which long-term value creation depends.
Codification without consensus
Regulators have since translated the principle into frameworks. The UK Stewardship Code, a reference point for many other jurisdictions, now defines stewardship as «the responsible allocation, management and oversight of capital to create long-term sustainable value for clients and beneficiaries». The 2026 edition, published by the Financial Reporting Council in June 2025 and effective from 1 January 2026, narrowed the 2020 definition, which had referred to sustainable benefits for the economy, the environment and society. It also reduced the number of principles from twelve to six.
In the European Union, the revised Shareholder Rights Directive requires asset managers and institutional investors to publish an engagement policy, or to explain publicly why they have not. Switzerland has gone further on one dimension: under articles 71a and 71b of the Occupational Pensions Act (BVG), pension funds must exercise their voting rights in Swiss companies whose shares are listed in Switzerland or abroad – on board elections, compensation and amendments to the articles of association – and report to their members at least once a year on how they have voted.
Codification has made stewardship visible. It has not, by itself, made it effective.
What engagement achieves
The empirical record suggests that engagement can work, but unevenly. Dimson et al. (2015), studying the environmental, social and governance engagements of a large institutional investor with US companies, found that successful engagements were followed by positive abnormal returns, while unsuccessful ones produced no adverse market reaction. Success, however, was the exception rather than the rule: fewer than one engagement in five achieved its objective.
Barko et al. (2022) reached a complementary conclusion in an international sample: engagement tended to target companies with weaker ESG ratings, and among companies that started from a low rating, ratings subsequently improved. Azar et al. (2021) found that companies held more heavily by the three largest index managers reduced their carbon emissions, an effect the authors attribute in part to engagement by those owners.
The pattern across these studies is instructive. Outcomes depend less on the number of meetings held than on the clarity of the objective, the credibility of escalation and the persistence of the owner. Stewardship measured by activity rather than by result tells beneficiaries very little.
The incentive problem
The difficulty is structural. Most institutional capital is managed not by the beneficial owner but by intermediaries, each with commercial incentives of its own. Gilson and Gordon (2013) described this layering as agency capitalism: the institutions that hold shares are themselves agents, and they are often «rationally reticent» about exercising the governance rights their holdings confer.
Bebchuk et al. (2017) set out why. An asset manager bears the full cost of stewardship but captures only a small fraction of any resulting increase in value, which accrues to its clients and, through overlapping portfolios, to its competitors' clients as well. Index managers face this problem most acutely, since their fee model rewards scale and low cost rather than intensity of oversight. Business relationships with portfolio companies may add a further reason for restraint.
None of this implies bad faith. It implies that stewardship delegated without specification will tend towards the minimum that clients and regulators will accept.
What credible stewardship requires in Switzerland
In Switzerland, the remedy lies with the boards of trustees of pension funds, which sit at the top of the investment chain and set the mandate. Because those boards comprise equal numbers of employer and employee representatives (BVG, art. 51), they are well placed to articulate what beneficiaries actually want. Four practices distinguish stewardship that can be evidenced from stewardship that is merely asserted.
Define the objective. A pension fund should state which issues matter most to its insured members over the relevant horizon, and why. Hart and Zingales (2017) argue that companies should maximise shareholder welfare rather than market value alone, which places members' interests at the centre of the mandate.
Look beyond the statutory minimum. The voting duty applies to directly held shares of Swiss companies (BVG, art. 71a). It does not reach foreign equities or holdings in collective investment vehicles, where much Swiss pension capital is invested. Voting and engagement expectations for those assets belong in investment management agreements and fund selection criteria.
Set escalation in advance. Binding shareholder votes on board and executive compensation give owners of Swiss listed companies an unusually direct instrument of escalation (Swiss Code of Obligations, art. 735). Collective platforms such as the Ethos Engagement Pool Switzerland and the Swiss Association for Responsible Investments (SVVK-ASIR) allow smaller funds to engage with a weight they could not command alone.
Report outcomes, not activity. The annual voting report required by law records how votes were cast (BVG, art. 71b). Members are better served by an additional account of what changed as a result, and what did not.
Lessons from abroad
Practice elsewhere shows each of these principles at work.
- Norway. Norway's Government Pension Fund Global has published its voting intentions five days before each shareholder meeting, where practicable, since 2021 – turning the vote into a public signal rather than a private record (NBIM, 2025).
- Japan. The Government Pension Investment Fund, which invests largely through external managers, requires them to comply with its stewardship and proxy voting principles and weighs the quality of their stewardship in its evaluation of them (GPIF, 2025).
- Netherlands. ABP illustrates escalation carried to its conclusion: in October 2021 it decided to divest more than EUR 15 billion of holdings in fossil fuel producers – almost 3% of its assets – having concluded that it had insufficient opportunity as a shareholder to push for a significant acceleration of their transition, while continuing to engage with large users of fossil energy (ABP, 2021).
- United States. CalPERS pioneered the public targeting of underperforming companies, and early evidence showed that many targeted firms adopted the governance changes it sought (Smith, 1996).
The test ahead
Stewardship is, at its core, a claim about responsibility: that those entrusted with other people's capital will act as owners rather than as passengers. Codes and directives have established the vocabulary. The test for the coming decade is whether asset owners can show, with evidence, that their ownership has made a difference.
References
ABP (2021), «ABP stops investing in fossil fuel producers», press release, 26 October 2021 · Azar, J., Duro, M., Kadach, I. & Ormazabal, G. (2021), «The Big Three and corporate carbon emissions around the world», Journal of Financial Economics 142(2), 674–696 · Barko, T., Cremers, M. & Renneboog, L. (2022), «Shareholder engagement on environmental, social, and governance performance», Journal of Business Ethics 180(2), 777–812 · Bebchuk, L. A., Cohen, A. & Hirst, S. (2017), «The agency problems of institutional investors», Journal of Economic Perspectives 31(3), 89–112 · Davis, J. H., Schoorman, F. D. & Donaldson, L. (1997), «Toward a stewardship theory of management», Academy of Management Review 22(1), 20–47 · Dimson, E., Karakaş, O. & Li, X. (2015), «Active ownership», Review of Financial Studies 28(12), 3225–3268 · Directive (EU) 2017/828 of 17 May 2017 (Shareholder Rights Directive II), OJ L 132, 1–25 · Federal Act on Occupational Retirement, Survivors' and Disability Pension Plans (BVG) of 25 June 1982, SR 831.40 · Financial Reporting Council (2020), The UK Stewardship Code 2020 · Financial Reporting Council (2025), The UK Stewardship Code 2026 · Gilson, R. J. & Gordon, J. N. (2013), «The agency costs of agency capitalism», Columbia Law Review 113(4), 863–927 · Government Pension Investment Fund (2025), Policy to fulfill stewardship responsibilities, 8 December 2025 · Hart, O. & Zingales, L. (2017), «Companies should maximize shareholder welfare not market value», Journal of Law, Finance, and Accounting 2(2), 247–274 · Hawley, J. P. & Williams, A. T. (2000), The Rise of Fiduciary Capitalism, University of Pennsylvania Press · Kay, J. (2012), The Kay Review of UK Equity Markets and Long-Term Decision Making: Final Report, Department for Business, Innovation and Skills · Norges Bank Investment Management (2025), Voting: First half 2025 · Smith, M. P. (1996), «Shareholder activism by institutional investors: Evidence from CalPERS», Journal of Finance 51(1), 227–252 · Swiss Code of Obligations of 30 March 1911, SR 220.