Digital governance: A conceptual framework and research agenda
数字治理：概念框架和研究议程

https://doi.org/10.1016/j.jbusres.2023.113777Get rights and content 获取权限和内容
Under a Creative Commons license
在知识共享许可协议下
open access 开放获取

Highlights 亮点

  • •
    We introduce digital governance as a distinct form of governance.
    我们引入了数字治理作为一种独特的治理形式。
  • •
    We shed light on mechanisms of digital governance in terms of control, coordination, incentives, and trust.
    我们阐明了数字治理中的控制、协调、激励和信任机制。
  • •
    We distinguish between analog, augmented, and automated forms of governance.
    我们区分了模拟、增强和自动化的治理形式。
  • •
    We predict that high transactivity favors augmented and automated forms of governance.
    我们预测高交易性有利于增强和自动化形式的治理。
  • •
    We present opportunities for future research related to digital governance.
    我们提出了与数字治理相关的未来研究机会。

Abstract 摘要

The rapid expansion of digital technologies has paved the way for new forms of organizing, facilitated by increased data and knowledge exchange between individuals and organizations. However, this poses major new challenges for designing effective governance mechanisms. This paper highlights the critical role of digital governance in facilitating digitally enabled exchange relationships. To this end, we propose a typology of analog, augmented, and automated governance modes, each associated with specific control, coordination, incentive, and trust mechanisms. Additionally, we provide a heuristic for determining the optimal governance choice via the interplay of transactivity (i.e., the contributors, connections, and consistency in an exchange network) and corresponding governance costs. Our study advances the governance literature by defining digital governance as a distinct form and outlining key governance mechanisms and choices in the digital era. Finally, we identify avenues for future research in this field.
数字技术的快速扩展为新的组织形式铺平了道路，这些新形式得益于个人与组织之间数据和知识交换的增加。然而，这也为设计有效的治理机制带来了重大挑战。本文强调了数字治理在促进数字赋能的交换关系中的关键作用。 为此，我们提出了一种治理模式的分类法，包括类比、增强和自动治理模式，每种模式都与特定的控制、协调、激励和信任机制相关联。此外，我们提供了一种通过交互活动（即交换网络中的贡献者、连接和一致性）与相应的治理成本之间的相互作用来确定最优治理选择的启发式方法。 我们的研究通过将数字治理定义为一种独特的形式，并概述数字时代的关键治理机制和选择，推进了治理文献。最后，我们确定了该领域未来研究的方向。

Keywords 关键词

Algorithmic management
Artificial intelligence
Blockchain
Corporate governance
Digital governance
Digital transformation
Interorganizational governance

算法管理
人工智能
区块链
公司治理
数字治理
数字化转型
组织间治理

1. Introducing digital governance
1. 引入数字治理

The proliferation of digital technologies has expanded the opportunities for data and knowledge exchange (Hanelt et al., 2021, Verhoef et al., 2021, Vial, 2019), yet it also presents new challenges for governance. Digital exchanges, such as platform-based transactions and online communities, frequently occur in large networks with numerous simultaneous interactions, pushing analog governance mechanisms such as contracts and relational norms to their limits. For instance, it would seem almost impossible to negotiate a contract for every job performed via Amazon Mechanical Turk or to establish a trust relationship for every Airbnb home stay. Thus, it is crucial to better understand the governance mechanisms and choices that meet the demands of the digital age.
数字技术的广泛应用扩展了数据与知识交流的机会（Hanelt 等, 2021, Verhoef 等, 2021, Vial, 2019），但同时也为治理带来了新的挑战。 数字交换，如基于平台的交易和在线社区，往往发生在具有大量同时互动的大型网络中，这使得合同和关系规范等模拟治理机制面临极限挑战。例如，为通过亚马逊的 Mechanical Turk 完成的每一份工作商定合同，或是为每一次 Airbnb 的住宿建立信任关系，似乎几乎是不可能的。 因此，更好地理解符合数字时代需求的治理机制和选择至关重要。
To enable large-scale digital exchanges, there is increasing reliance on digital governance, which leverages algorithmic protocols to automate control, coordination, incentives, and trust (Hanisch et al., 2022, Vaia et al., 2022). Digital governance touches on fundamental issues of organizing, e.g., enhancing task programmability to improve process control, automating task division and allocation to facilitate coordination, conditioning incentives through dynamic inputs, and creating the transactional transparency required for trust. For example, digital governance can create verification mechanisms for transactions, e.g., oracles and consensus protocols, which are used in blockchain networks (Al-Breiki et al., 2020, Zheng et al., 2017). Similarly, artificial intelligence (AI)-supported analyses can enable automatic checks on accounting data and raise red flags early, thus enhancing firm monitoring (Commerford et al., 2022, Möhlmann et al., 2021). These technological solutions are important precursors of new, data-driven business models that require the regulation of data ownership, storage, transfer, access, and use across individual, functional, and organizational boundaries.
为了实现大规模数字交流，越来越多地依赖于数字治理，它利用算法协议来自动化控制、协调、激励和信任（Hanisch et al., 2022, Vaia et al., 2022）。数字治理涉及组织结构的根本问题，例如：,增强任务的编程性以改进流程控制，自动化任务分工与分配以促进协调，通过动态输入调整激励，以及建立交易透明度所需的信任。例如，数字治理可以为交易创建验证机制，例如。,预言机与共识协议，这些在区块链网络中得到应用（Al-Breiki 等人，2020, Zheng 等人，2017). 同样地，人工智能（AI）支持的分析能够自动检查会计数据，并及早发出警示，从而加强企业监控（Commerford 等，2022，Möhlmann 等，2021）。 这些技术解决方案是新型的、数据驱动的商业模式的重要前驱，这些模式要求在个人、职能和组织边界之间对数据所有权、存储、传输、访问和使用进行规范。
We advance the debate on digital governance by developing a conceptual framework that distinguishes between analog, augmented, and automated forms of control, coordination, incentives, and trust. In particular, we show how 1) automated control no longer relies on hierarchical control but on decentralized checks-and-balances protocols; 2) automated coordination transforms bilateral agreements into omnilateral arrangements; 3) incentives transition from bureaucratic rules to cybernetic protocols that update autonomously via dynamic inputs; and 4) trust can be algorithmically enhanced by shifting from individual actors to a complete system. In addition to the poles of analog and automated governance, our model highlights augmented governance modes, which blend elements from both domains. Importantly, our framework does not suggest that analog governance is completely displaced by digital governance; rather, they complement and constrain each other.
我们推进了关于数字治理的讨论，通过构建一个概念框架，区分模拟、增强和自动化的控制、协调、激励和信任模式。 特别是，我们展示了 1)自动化控制不再依赖于层级控制，而是依赖于分散的制衡协议；2)自动化协调将双边协议转变为多边安排；3)激励机制从官僚规则转变为自适应协议，这些协议通过动态输入进行自主更新；以及 4)通过从个体行为者转向整个系统，信任可以通过算法得到增强。 除了模拟治理和自动化治理的极点，我们的模型还突出了增强治理模式，它是两种领域元素的融合。重要的是，我们的框架并不认为模拟治理完全被数字治理所取代；相反，它们相互补充并制约。
Our distinction between analog, augmented, and automated governance also informs a governance choice model that predicts the optimal governance mode as a function of transactivity and the resulting governance costs. Specifically, we predict that automated governance becomes more cost-efficient than augmented governance and, ultimately, analog governance as transactivity—defined by the number of contributors, connections, and level of exchange consistency—increases. More broadly, we facilitate a deeper understanding of the mechanisms and strategic tradeoffs governance designers and exchange participants face when establishing effective governance solutions for the digital age. We conclude with an extensive research agenda and identify opportunities to study the governance by and of algorithms.
我们对模拟、增强和自动化治理的区分也提供了治理选择模型，该模型根据交互动态和相应的治理成本来预测最优治理模式。具体而言，我们预测随着交互动态的增加——由贡献者数量、连接强度和交换一致性水平定义——自动化治理将比增强治理乃至最终的模拟治理更加成本高效。 更广泛地说，我们促进了更深入的理解治理设计者和交易所参与者在建立有效的数字时代治理解决方案时面临的机制和战略权衡。最后，我们提出了一项广泛的研究议程，并确定了研究算法治理及其应用的机会。
This paper makes three important theoretical contributions. First, we provide definitional clarity regarding the concept of digital governance, a distinct form of governance that has spawned a new field of research requiring a conceptual foundation (e.g., Hanisch et al., 2022, Lumineau et al., 2021, Möhlmann et al., 2021, Vaia et al., 2022). Hence, we answer calls for an enhanced conceptual distinction of emerging digital phenomena and clarification of the extant knowledge of corresponding analog phenomena (Adner, Puranam, & Zhu, 2019). Second, we extend previous governance research by unraveling the strategic decision-making parameters and tradeoffs associated with digital governance, and we define relevant governance mechanisms associated with digital exchange that are critical in discussions of advanced system designs such as AI and blockchains (e.g., Chhillar and Aguilera, 2022, Goldsby and Hanisch, 2022). Third, our work contributes to the wider discussion on digital transformation that has gained prominence in management and organizational research (Hanelt et al., 2021, Verhoef et al., 2021, Vial, 2019) by shifting the focus from organizational processes and business models to how digital technology impacts governance.
本文作出了三项重要的理论贡献。首先，我们对数字治理的概念提供了明确的定义，这是一种独特的治理形式，催生了需要概念基础的新研究领域（例如，Hanisch 等人, 2022, Lumineau 等人, 2021, Möhlmann 等人, 2021, Vaia 等人, 2022)。因此，我们响应了对新兴数字现象概念上更明确区分的呼声，并阐明了现有对应模拟现象的知识（Adner, Puranam & Zhu, 2019）。 其次，我们通过剖析数字化治理相关的战略决策参数及取舍，扩展了原有的治理研究，并定义了与数字交换相关的治理机制，这些机制在讨论诸如人工智能和区块链等高级系统设计时至关重要（例如，Chhillar 和 Aguilera, 2022, Goldsby 和 Hanisch, 2022）。 第三，我们的工作对数字转型在管理和组织研究中日益重要的广泛讨论作出了贡献（Hanelt 等，2021，Verhoef 等，2021，Vial，2019），将焦点从组织流程和业务模型转向数字技术如何影响治理。

2. Taking stock of the governance literature
2. 治理文献的盘点

The need for governance arises from the division of labor and the associated dissipation of information and control of inputs and outputs (Gulati and Singh, 1998, Puranam et al., 2014, Sundaramurthy and Lewis, 2003). Ultimately, it entails creating and capturing value through exchange amid competition and asymmetric information (Gnyawali & Ryan Charleton, 2018). The governance challenge involves creating mechanisms that help integrate, direct, and monitor the distributed efforts in productive exchange relationships (Dekker, 2004). To meet this challenge, exchange partners must find ways to control relevant exchange processes (e.g., allocation of resources and tasks), outcomes (e.g., generation and distribution of financial, environmental, and social value), and relationships (e.g., opportunistic behaviors) (Goold and Quinn, 1990, Sundaramurthy and Lewis, 2003). The design of control mechanisms can be complemented and substituted by appropriate coordination (Bechky and Chung, 2018, Gulati et al., 2012), incentives (Makadok and Coff, 2009, Rutherford et al., 2007), and trust mechanisms (Cao and Lumineau, 2015, Westphal, 1999) to achieve desired governance benefits. Hence, governance broadly concerns the establishment of rules that help verify inputs and outputs (i.e., control mechanisms), divide and allocate tasks (i.e., coordination mechanisms), align competing interests (i.e., incentive mechanisms), and attenuate relational vulnerabilities (i.e., trust mechanisms). Fig. 1 summarizes these complementary and substitutive mechanisms that are central to the governance literature.
治理的需求源于劳动分工及其伴随的信息分散和输入输出控制（Gulati 和 Singh, 1998, Puranam 等, 2014, Sundaramurthy 和 Lewis, 2003）。 最终，它涉及在竞争与信息不对称的环境中通过交换创造并获取价值（Gnyawali & Ryan Charleton, 2018）。治理挑战则在于建立机制，以帮助整合、指导并监督分布式努力在生产性交换关系中的表现（Dekker, 2004）。 为了应对这一挑战，交换合作方必须找到方法来控制相关的交换过程（例如，资源和任务的分配）、结果（例如，财务、环境和社会价值的产生与分配）以及关系（例如，机会主义行为）（Goold 和 Quinn，1990，Sundaramurthy 和 Lewis，2003）。 控制机制的设计可以通过适当的协调（Bechky 和 Chung, 2018, Gulati 等, 2012）和激励措施（Makadok 和 Coff, 2009, Rutherford 等, 2007），以及信任机制（曹与卢米诺，2015，韦斯特法尔，1999），以实现所需的治理效益。因此，治理广泛涉及建立规则，以验证输入和输出（即控制机制），分配和分配任务（即协调机制），并对齐竞争利益（即，激励机制），并缓解关系脆弱性（如信任机制）。图 1总结了治理文献中这些互补和替代机制的核心内容。
  1. Download: Download high-res image (347KB)
    下载: 下载高清图像 (347KB)
  2. Download: Download full-size image
    下载: 下载全尺寸图像

Fig. 1. Governance as a mediator between value creation and value capture.
图 1. 治理作为价值创造和价值获取之间的中介。

Governance research has identified various responses to the control, coordination, incentive, and trust challenges in transactions (Brown et al., 2011, Furlotti, 2007, Hambrick et al., 2008). A first response lies in the creation of bureaucratic organizations that define and delimit transactions and provide mechanisms of hierarchical fiat and authority to improve control and coordination, set incentives, and ensure cooperation (Powell, 1990, Williamson, 1991). A second generic response involves using contracts as a means of codifying agreed-upon control, coordination, and incentive mechanisms, backed by institutional support, to enforce legal rights (Grossman and Hart, 1986, Hart and Moore, 1990, Tirole, 1999). Finally, a third response entails fostering relational bonds that emerge organically through social interdependencies and positive exchange experiences to provide a bedrock for the emergence of trust (Gulati, 1995, Uzzi, 1997). While these analog governance mechanisms are vital to organizational life and are likely to persist, emerging digital technologies present new challenges and opportunities for the design of governance mechanisms.
治理研究识别了对交易中控制、协调、激励及信任挑战的多种应对策略(Brown 等, 2011, Furlotti, 2007, Hambrick 等, 2008)。 第一个解决方案在于创设官僚机构，这些机构定义并界定交易范围，提供层级命令与权威机制以增强控制与协调，设立激励机制，并确保合作（鲍威尔，1990，威廉姆森，1991）。 第二种一般性的回应是通过合同来编纂各方同意的控制、协调和激励机制，并辅以制度支持，以强化法律权利的执行（Grossman 和 Hart, 1986，Hart 和 Moore, 1990，Tirole, 1999）。 最后，第三个回应涉及培养通过社交相互依赖和积极交流经历自然产生的联系纽带，为信任的产生提供坚实基础（Gulati, 1995, Uzzi, 1997）。 尽管这些模拟治理机制对组织生活至关重要，并可能持续存在，但新兴的数字技术为治理机制的设计带来了新的挑战和机遇。

3. Facing governance challenges in the digital age
3. 应对数字时代的治理挑战

The need to revisit and advance governance theories in the digital age is closely linked to the novel challenges arising in this context. We examine these new challenges in terms of establishing/building, maintaining/adapting, and restoring/terminating exchange relationships. For each stage, we highlight exemplary new forms of competition and cooperation in the digital age (Fig. 2) because a critical function of governance is to enable cooperation amid competing interests.
数字时代下对治理理论进行重新审视和推进的必要性，与这一背景下涌现的新挑战密切相关。我们考察了这些新挑战在建立/构建、维持/适应以及恢复/终止交换关系中的表现。在每个阶段，我们重点介绍了数字时代下竞争与合作的新典范形式（图表 2) 因为治理的一项关键职能是在利益冲突中促进合作。
  1. Download: Download high-res image (554KB)
    下载: 下载高清图片（554KB）
  2. Download: Download full-size image
    下载: 下载完整尺寸的图片

Fig. 2. Exemplary governance challenges in digital exchange relationships.
图 2. 数字交换关系中的代表性治理挑战。

3.1. Establishing and building digital relationships
3.1. 建立和构建数字关系

The strategic relevance of governance results from its ability to ensure and enhance performance in exchange relationships (Jones, 1983, Leiblein, 2003). This logic is subject to a new dynamic in the digital age, where governance serves as not only a performance enabler but also a strategic differentiator. Since the value of digital networks is heavily reliant on the realization of network effects (Shapiro & Varian, 2008), companies are increasingly recognizing that their governance decisions strongly influence the overall attractiveness of their network (Chen, Tong, Tang, & Han, 2022). In light of this, Apple markets its privacy policy across the ecosystem as a strategic differentiator from its competitor Google.
治理的战略重要性源于其能够确保并提升交换关系中的绩效（琼斯, 1983, 莱伯恩, 2003）。这一逻辑在数字时代面临新的动态变化，治理不仅作为性能促进者，更成为战略性的差异化工具。 由于数字网络的价值在很大程度上依赖于网络效应的实现（Shapiro & Varian, 2008），企业越来越意识到其治理决策对整体网络的吸引力（Chen, Tong, Tang, & Han, 2022）有重要影响。 鉴于此，苹果将其生态系统中的隐私政策作为与竞争对手谷歌的战略差异化点进行营销。
As organizations have begun to compete on governance, there is also growing evidence of novel collaborative dynamics, e.g., the transparency movement concerning governance decisions via the publication and development of algorithmic protocols using open-source software code. Organizations such as OpenAI develop open-source AI to promote transparency, while decentralized autonomous organizations (DAOs) discuss algorithmic governance choices publicly in Discord forums (Taulli, 2022). These collective efforts represent new forms of collaboration largely absent from the analog world, where governance choices (e.g., contract design) have typically been negotiated behind closed doors.
随着组织在治理方面展开竞争，有关新型协作动态的证据也日益增多，例如，通过发布和开发使用开源软件代码的算法协议，推动治理决策透明度的运动。 诸如 OpenAI 这样的组织开发开源人工智能以促进透明度，而分布式自治组织（DAOs）则在 Discord 论坛上公开讨论算法治理选择（Taulli, 2022）。这些集体努力代表了在模拟世界中基本上缺失的新型协作形式，模拟世界中的治理选择（如合同设计）通常在幕后进行谈判。

3.2. Maintaining and adapting digital relationships
3.2. 维护和适应数字关系

The transition from analog to digital governance shifts the locus of decision-making from the actors involved in the exchange to those who develop digital governance tools (Benlian et al., 2022). In the analog world, the governance mode is primarily negotiated and agreed upon by the exchange participants. However, in the digital world, decisions are often made by those who develop digital tools, disconnecting them from the parties that execute the transaction (Renwick & Gleasure, 2021). This warrants the need for stronger consideration of the governance setters and principal actors and the processes necessary for alignment. Failing to align exchange participants and governance setters can lead to serious tensions (Huber, Kude, & Dibbern, 2017), as evidenced by Epic’s lawsuit against Apple over its pricing policies on the iOS platform. These tensions can also bring new forms of collaboration, such as solidarity among content creators expressing their dissatisfaction with governance decisions (Ricart, Snihur, Carrasco-Farré, & Berrone, 2020).
从模拟到数字治理的转变将决策中心从参与交换的参与者转移到了开发数字治理工具的人身上（Benlian 等，2022）。在模拟世界中，治理模式主要由交换参与者协商和达成一致。 然而，在数字世界中，决策往往由开发数字工具的人做出，使他们与执行交易的主体脱离关联（Renwick & Gleasure, 2021）。这迫切需要加强对治理制定者、主要参与者的重视，以及确保各方协调一致的必要流程。 未能协调交易所参与者与治理制定者可能会导致严重的紧张关系（Huber, Kude, & Dibbern, 2017），正如 Epic 对 Apple 因其 iOS 平台定价政策发起的诉讼所证明的那样。 这些紧张关系也可以带来新的合作形式，例如内容创作者之间表达对治理决策不满的团结（Ricart, Snihur, Carrasco-Farré, & Berrone, 2020）。

3.3. Restoring and terminating digital relationships
3.3. 恢复和终止数字关系

The termination of interorganizational and interpersonal relationships has been extensively studied in the context of strategic alliances (e.g., Asgari et al., 2017, Bakker, 2016) and CEO dismissals (e.g., Marcel et al., 2017, Oehmichen et al., 2017). However, comparatively little is known about participants’ departure in predominantly digital exchange relationships (Shah, 2006, Tiwana, 2015). This termination can be attributed to various factors, one of which is the design of governance mechanisms. Well-designed governance and increased trust between parties in platform markets can paradoxically lead to a platform’s disintermediation as users seek to bypass platform fees and continue their transactions (Gu & Zhu, 2021). Governance choices regarding the interoperability of digital platforms can also raise coordination costs for developers and increase the likelihood of platform abandonment (Tiwana, 2015). Finally, the addition of non-human agents through governance choices can alienate human participants (Newlands, 2021). For instance, the use of algorithms to control information presentation on social media platforms presents challenges in designing governance mechanisms that avoid negative consequences and potential legislative interventions (Riemer & Peter, 2021). Hence, preventing activities that drive human participants away is a crucial challenge in digital governance.
组织间和个人间关系的终止已在战略联盟（例如，战略联盟的背景下被广泛研究（例如，Asgari 等人, 2017, Bakker, 2016) 和 CEO 解职（例如，Marcel 等人, 2017, Oehmichen 等人, 2017). 然而，对于以数字交流为主的关系中参与者的离去，我们所知甚少（Shah, 2006, Tiwana, 2015）。这种终止可归因于多种因素，其中包括治理机制的设计。 良好设计的治理与平台市场中各方之间信任的增加，虽然看似矛盾，却可能导致平台去中介化，因为用户寻求绕过平台费用以继续其交易（顾 & 朱, 2021）。 关于数字平台互操作性的治理选择，也会增加开发者的协调成本，并增加平台被废弃的可能性（Tiwana, 2015）。最后，通过治理选择引入的非人类代理可能会疏远人类参与者（Newlands, 2021）。 例如，算法用于控制信息在社交媒体平台上的呈现方式，这在设计治理机制时带来了挑战，需要避免负面后果和可能的立法干预（Riemer & Peter, 2021）。因此，防止活动导致人类参与者流失是数字治理中的一个重要挑战。
In the most extreme cases when governance decisions are met with widespread resistance, a coordinated campaign can lead to mass platform exit, as Facebook experienced after the Cambridge Analytica scandal (Zhang, Wang, Karahanna, & Xu, 2022), or the concerted effort by major consumer products companies to stop advertising on Twitter over controversial policies set by new CEO Elon Musk. Thus, exchange participants can pressure governance designers by employing exits as effective strategies to undermine the size and value of a network. Recognizing these novel exigencies, scholars have begun to identify and explore the mechanisms and tradeoffs of governance in digital contexts across disciplines.
在最极端的情况下，当治理决策遭遇广泛抵制时，协调一致的行动可能导致大规模平台退出，正如 Facebook 在剑桥分析丑闻后所经历的那样（张、王、卡拉汉纳、徐，2022），或者是大型消费品公司因新任 CEO 埃隆·马斯克制定的争议性政策而集体停止在 Twitter 上投放广告。 因此，交换参与者可以通过采用退出策略，有效地对治理设计者施加压力，从而削弱网络的规模和价值。认识到这些新的紧急情况，学者们已经开始跨学科地识别和探索数字环境中治理的机制和权衡。

4. Recognizing the potential of digital governance
4. 认识数字治理的潜力

Digital governance is facilitated by digital technologies, i.e., different combinations of information, communication, and other connectivity technologies (Bharadwaj, El Sawy, Pavlou, & Venkatraman, 2013). In digital governance, there is a strong focus on digital technologies that can process data relevant to value-added exchanges (e.g., advanced databases such as blockchains; Lumineau et al., 2021) and the heuristics that can make autonomous decisions to support exchange continuity (e.g., complex algorithms such as matching algorithms and AI; Malgonde, Zhang, Padmanabhan, & Limayem, 2020). Such digital technologies permit a shift toward automated modes of governance, which differ radically from their analog counterparts (Strich et al., 2021, Tarafdar, Page, & Marabelli, 2023). Hence, we view digital governance as a distinct governance category that sustains novel forms of organizing, value creation, and value capture, and thus goes beyond the digitization of existing analog governance mechanisms.
数字治理借助数字技术得以实现，即信息、通信和其他连接技术的不同组合（Bharadwaj, El Sawy, Pavlou, & Venkatraman, 2013）。在数字治理中，重点关注那些能够处理与增值交换相关的数据的数字技术（例如，区块链等先进数据库；Lumineau et al., 2021)以及那些能够自主决策以支持交易连续性的启发式方法（例如，复杂算法如匹配算法和人工智能；Malgonde, Zhang, Padmanabhan, & Limayem, 2020）。这类数字技术促使治理模式向自动化转型，与传统模拟模式大相径庭（Strich 等人。, 2021, Tarafdar, Page, & Marabelli, 2023). 因此，我们将数字治理视为一个独特的治理类别，它支撑着组织、价值创造和价值获取的新形式，从而超越了现有模拟治理机制的数字化。
The extent to which digital technologies are used to govern exchanges can range from augmenting to fully automating governance (cf. Raisch & Krakowski, 2021). On the one hand, digital governance can augment aspects of governance, reducing reliance on human intervention. For example, classic exchange relationships between buyers and sellers can be augmented by digital governance in the form of digital platforms that act as intermediaries connecting buyers and sellers (Constantinides et al., 2018, de Reuver et al., 2018). On the other hand, technological solutions can help organizations automate governance. For example, algorithmic surveillance in the gig economy allows platform firms to automate control of their workforce and complementors (Bellesia et al., 2023, Möhlmann et al., 2021, Newlands, 2021). The advantage of augmenting and/or automating governance through digital technologies is that these digital governance structures can drive efficiency (e.g., through repeat, rule-based transactions) and transparency between exchange participants (e.g., recommender systems and digital identities). Additionally, digital governance structures can have formally superimposed controls (e.g., role management and access rights) that increase certainty and reduce tolerance for erroneous transactions (e.g., through approval systems, voting rights, and rigorously programmed workflows).
数字技术在管理交易中的应用程度可以从增强到完全自动化治理（参见Raisch & Krakowski, 2021）。一方面，数字治理可以增强治理的各个方面，减少对人力的依赖。 例如，传统的买卖双方交易关系可以通过数字治理得到增强，这种治理以数字平台的形式存在，这些平台作为中介连接买家和卖家（Constantinides 等人，2018，de Reuver 等人，2018）。另一方面，技术解决方案可以帮助组织自动化治理。 例如，零工经济中的算法监控使得平台公司能够自动化管理其劳动力和补充者（Bellesia 等，2023，Möhlmann 等，2021，Newlands，2021）。 通过数字技术增强和/或自动化治理的优势在于，这些数字治理结构能够推动效率（例如，通过重复、基于规则的交易）以及交换参与者之间的透明度（例如，推荐系统和数字身份）。此外，数字治理结构还可以正式叠加控制（例如，（例如，通过审批系统、投票权和严格编程的工作流程）增加了确定性，并减少了错误交易的容忍度（角色管理和访问权限）。
Nevertheless, augmented and automated governance are often complemented and constrained by analog governance, which balances technological affordances with its strong focus on interpersonal and contractual mechanisms. For example, machine learning algorithms, while able to automate workflows such as content review processes, are prone to biases and therefore require human oversight in certain circumstances (Akter et al., 2022, Kordzadeh and Ghasemaghaei, 2022). In other cases, algorithmic governance is simply undesirable, especially in regard to assuming significant responsibilities. For instance, corporate legislation generally prohibits non-human representatives from assuming a supervisory role on the board of directors. Consequently, digital governance relies heavily on the complementary and constraining influence of analog governance to mitigate the pitfalls of fully automated governance.
尽管如此，增强型和自动化治理往往得到并受限于传统治理模式，它通过人际与合同机制平衡了技术能力。例如，机器学习算法虽能自动执行诸如内容审核流程，但由于易受偏见影响，在某些情况下仍需人工监督（Akter 等人。, 2022, Kordzadeh 和 Ghasemaghaei, 2022）。在其他情况下，算法治理是不可取的，尤其是在承担重大责任方面。例如，企业法规通常禁止非人类代表担任董事会中的监督角色。 因此，数字治理严重依赖于模拟治理的互补和约束作用，以缓解完全自动化治理的弊端。
One limitation of automated governance can arise when its highly programmatic nature promotes rigidity and thereby harms adaptability (Zhu, Kraemer, Gurbaxani, & Xu, 2006). A related downside is that automated governance requires explicit information, while governance may involve tacit information that are difficult to codify. Furthermore, automated governance can quickly succumb to compliance and regulation, as seen with the European Union’s General Data Protection Regulation (GDPR), which strictly regulates personal data storage and may conflict with the append-only nature of blockchains (Rieger, Guggenmos, Lockl, Fridgen, & Urbach, 2019). Thus, automated governance must be carefully constrained and complemented with analog governance to overcome its shortcomings. Automated governance alone cannot be considered a cure-all solution.
自动化治理的一个局限性在于其高度程序化的特性可能导致僵化，从而损害适应性（朱、克莱默、Gurbaxani 和 Xu，2006）。另一个相关的不利因素是，自动化治理需要明确的信息，而治理可能涉及难以编码的隐性信息。 此外，自动化治理可能会迅速陷入合规和监管的困境，正如欧盟的《通用数据保护条例》（GDPR）所示，该条例严格规范了个人数据的存储，并可能与区块链仅追加的特性产生冲突（Rieger, Guggenmos, Lockl, Fridgen, & Urbach, 2019）。 因此，自动化治理必须受到严格限制，并辅以传统治理，以克服其不足。仅凭自动化治理不能被视为万能解决方案。
To theorize about the role of analog, augmented, and automated governance mechanisms, we consider them to be interrelated and complementary. Our goal is to comprehend how digital governance can augment and automate analog governance, such as administrative procedures, contracts, and relational norms, thereby establishing digital mechanisms of control, coordination, incentives, and trust. It is crucial to emphasize that analog governance plays a significant role in complementing and constraining digital governance mechanisms where they fall short.
要探讨模拟、增强和自动化治理机制的作用，我们将其视为相互关联且互补的。我们的目标是理解数字治理如何增强和自动化模拟治理，例如行政程序、合同和关系规范，从而建立控制、协调、激励和信任的数字机制。 强调模拟治理在弥补和约束数字治理机制不足方面的重要作用至关重要。

5. Understanding the shift from analog to automated governance
5. 理解从模拟到自动化治理的转变

Ongoing digital transformation is leading to profound changes in the governance of exchange relationships, with analog forms of governance being supplemented and, in some cases, replaced by automated forms. Here, analog governance refers to instances when governance is predominantly based on centralized control structures, bilateral task coordination, bureaucratic incentives, and actor-based relational trust. In contrast, automated governance entails that governance is based largely on decentralized control, omnilateral coordination, automatic (“cybernetic”) incentives, and algorithmic system trust. Given these two extremes, an intermediate mode is augmented governance, where actors and algorithms intertwine. Augmented governance involves distributed control, multilateral coordination (assisted by digital channels), programmatic incentive structures, and actorithmic trust. Below, we elaborate on these generic governance modes (analog, augmented, and automated) and provide a comprehensive definition of the underlying four governance mechanisms—(1) control, (2) coordination, (3) incentives, and (4) trust.
持续的数字化转型正引领着交换关系治理的深刻变革，模拟形式的治理正在得到补充，并在某些情况下被自动化形式所取代。这里，模拟治理指的是治理主要基于集中控制结构、双边任务协调、官僚主义激励和基于行为者的信任关系。 相比之下，自动化的治理意味着治理在很大程度上基于去中心化的控制，全方位协调，自动化的（“控制论”）激励，以及算法系统信任。在这两种极端之间，存在一种中间模式，即增强型治理，其中行为者和算法交织在一起。 增强型治理涉及分布式控制、多边协调（借助数字渠道）、程序化激励结构以及算法化信任。下文将详细阐述这些通用治理模式（模拟、增强和自动化），并全面定义其背后的四种治理机制——（1）控制，（2）协调，（3）激励，以及（4）信任。
Fig. 3 illustrates how each of the three governance modes manifests across the four mechanisms. When using the terms “analog,” “augmented,” or “automated” governance, we refer to discrete, pure sets that exist within the broad spectrum of the combinations of these endpoints of the multidimensional space we depict. In general, we view the transition between analog and automated governance as a fluid, multilevel continuum that allows decision-makers to automate certain governance mechanisms individually or in conjunction with others. Thus, it is possible to “mix and match” analog, augmented, and automated governance mechanisms in various hybrid forms of governance. For parsimony, we focus our theorizing on discrete points in the option space; however, we fully recognize the configurational logic underlying this framework (Furnari et al., 2021). Thus, in practice, we expect to observe configurations that blend analog governance (e.g., for control) forms with elements of augmented governance (e.g., for coordination) and automated governance (e.g., for incentives).
  1. Download: Download high-res image (511KB)
    下载：下载高清图片（511KB）
  2. Download: Download full-size image
    下载：下载完整尺寸图片

Fig. 3. Configurational building blocks of governance mechanisms and modes.
图 3. 治理机制和模式的构建模块。

5.1. Control: from centralized to decentralized
5.1. 控制：从集中到分散

In analog governance, control is centralized through contracts and fiat (Williamson, 1991). Centralized control relies on mechanisms of vertical authority and primarily focuses on behavior and outcome enforcement (Eisenhardt, 1985, Ouchi and Maguire, 1975). Outcome-based contracts formalize agreed-upon metrics, such as product sales, and behavior-based contracts improve adherence to performance-related metrics, such as task completion time. Both types of contracts are typically supported by vertical authority mechanisms, such as hierarchical fiat and institutional power, to enforce outcomes or behaviors.
在模拟治理中，通过合同和法令实现集中式控制（Williamson, 1991）。集中式控制依赖于垂直权威机制，主要关注行为和结果的执行（Eisenhardt, 1985, Ouchi 和 Maguire, 1975）。 基于结果的合同明确了约定的指标，如产品销售，而基于行为的合同则提高了对与绩效相关的指标的遵守程度，例如任务完成时间。这两种类型的合同通常由垂直权威机制支持，如层级命令和制度权力，以强制执行结果或行为。
Automated governance in the form of decentralized controls replaces vertical authority with autonomous algorithms that improve outcome certainty and enforce rigid behavioral control. These algorithms can be used to monitor a workforce (e.g., Amazon’s fully automated warehouses; Baraniuk, 2015) or user-generated content in social networks (e.g., Gilbert, 2021). Decentralized control systems have several advantages, such as following preprogrammed rules that are automatically executed (e.g., through smart contracts) and not requiring enforcement by vertical authority (Murray, Kuban, Josefy, & Anderson, 2021). Additionally, they structure a checks-and-balances system that increases certainty through transaction and information validation by each participant in the control structure. Furthermore, decentralized information systems, such as blockchains, can digitize routinized workflows, providing stricter process control and increasing behavioral certainty.
自动化治理以去中心化控制的形式，用自主算法替代垂直权威，提升结果的确定性并实施严格的行为管控。这些算法可用于监控工作团队（例如，亚马逊的全自动化仓库；Baraniuk, 2015）或在社交网络上管理用户生成的内容（例如，Gilbert, 2021）。 去中心化控制系统具有多项优势，例如遵循预编程规则并自动执行（例如，通过智能合约），且无需垂直权威的强制执行（Murray, Kuban, Josefy, & Anderson, 2021）。 此外，他们建立了一个制衡系统，通过控制结构中每个参与者对交易和信息的验证来增加确定性。此外，去中心化的信息系统，如区块链，可以数字化常规工作流程，提供更严格的过程控制并增加行为确定性。
In addition to centralized and decentralized control, digital tools can enable the emergence of distributed control mechanisms. Distributed control is a form of control that operates through lateral authority and is enabled by digital tools. For example, open-source software development relies on distributed controls among developers that regulate access to read, write, or contribute code (Dahlander & O’Mahony, 2011). This type of control is anchored in the developer community and digital tools such as GitHub or Bitbucket. The advantage of distributed control is that it gives partial control over bureaucratic tasks, such as version control, to an automated and decentralized system while accounting for contingencies beyond the program code.
除了集中式和分散式控制外，数字工具还能促进分布式控制机制的涌现。分布式控制是一种通过横向权限运作的控制形式，由数字工具所驱动。例如，开源软件开发依赖于开发者之间的分布式控制，这些控制规范了代码的阅读、编写或贡献权限（Dahlander & O'Mahony, 2011）。 这种类型的控制在开发者社区和 GitHub 或 Bitbucket 等数字工具中得以体现。分布式控制的优势在于，它将诸如版本控制等行政任务的部分控制权交给自动化和去中心化的系统，同时考虑程序代码之外的意外情况。

5.2. Coordination: from bilateral to omnilateral
5.2. 协调：从双边到多边

In the analog context, coordination occurs bilaterally, which means that an actor divides labor into tasks that can be assigned to and performed by another party, as would be the case in buyer–supplier relationships (Jones, 1984). One benefit of this type of coordination is that task assignments are usually routinized, leading to increased reliability and efficiency in organizational performance (Cohen & Bacdayan, 1994). However, this type of procedural memory can also be inarticulate and challenging to transfer between actors.
在模拟环境下，协调以双边方式进行，这意味着主体将劳动分解为任务，这些任务可以分配给另一方执行，正如在买方与供应商关系中那样（琼斯，1984）。 这种协调的一个好处是任务分配通常是例行公事，从而提高了组织绩效的可靠性和效率（Cohen & Bacdayan, 1994）。然而，这种程序记忆也可能是不明确的，且难以在不同参与者之间转移。
In contrast, omnilateral coordination relies on mechanisms where a system divides labor into tasks that can be assigned to any party automatically. Importantly, omnilateral mechanisms do not rely on implicit procedural memory but on rigid task codification, division, and assignment, which are fully autonomous. For instance, platform firms such as Uber, Lyft, Deliveroo, and GrubHub extensively adopt algorithmic coordination to manage task allocation, goal setting, and scheduling for their workforce (Tarafdar, Page, & Marabelli, 2023). Another example of omnilateral coordination concerns DAOs, whose first use cases (e.g., MakerDAO, a stablecoin issuance platform) illustrate how fully autonomous task division and task allocation occur through on-chain voting (Zhao, Ai, Lai, Luo, & Benitez, 2022). As routines are considered dynamic (Feldman, 2000), a risk of any omnilateral coordination is that its assisting algorithm does not persist over time and cannot account for possible changes in the organization’s task environment, which can impede overall task division and assignment.
相比之下，全方位协调依赖于一种机制，即系统将劳动分解为可自动分配给任何方的任务。重要的是，全方位机制不依赖于隐性的程序记忆，而是依赖于刚性的任务编码、划分和分配，这些过程是完全自主的。 例如，像 Uber、Lyft、Deliveroo 和 GrubHub 这样的平台公司广泛采用算法协调来管理任务分配、目标设定和工作安排（Tarafdar, Page, & Marabelli, 2023）。另一种全方协作的例子涉及 DAO，其最初的应用案例（例如……,MakerDAO，一家稳定币发行平台) 展示了如何通过链上投票实现完全自主的任务分工和分配（赵，艾，赖，罗，& 贝尼特斯，2022）。 由于惯例被视为动态的（Feldman, 2000），任何多方协调的风险在于其辅助算法无法随时间持续，且无法考虑组织任务环境中可能的变化，这可能阻碍整体任务的划分与分配。
Between the extremes of bi- and omnilateral coordination mechanisms, coordination can take an augmented form of multilateral coordination where tasks are divided and assigned by physical actors through digital channels. The advantage of multilateral coordination is that digital channels partly codify routines that would otherwise be stored in procedural memory. For example, the Catena-X network, which fosters cross-company data exchange in the automotive industry, relies on an open data ecosystem where parties use network services to share data and on an actor-based consortium to coordinate its development. Simultaneously, with multilateral coordination, direct actor involvement helps account for salient routine instability (Pentland, Hærem, & Hillison, 2011). Using multilateral coordination through digital channels, parties can better anticipate when a change in routine is required and whether a bypass of automatic task division and allocation is warranted.
在双边和多边协调机制的极端之间，协调可以采用增强的多边协调形式，其中任务通过数字渠道由物理实体划分和分配。多边协调的优势在于，数字渠道部分地编码了原本存储在程序记忆中的例程。 例如，促进汽车行业跨公司数据交换的 Catena-X 网络，依赖于一个开放的数据生态系统，各方通过网络服务共享数据，并依托基于角色的联盟来协调其发展。同时，通过多方协调，直接参与者的投入有助于应对显著的日常不稳定性（Pentland, Hærem, & Hillison, 2011）。 通过数字渠道的多边协调，各方可以更好地预测何时需要调整常规，并判断是否需要绕过自动任务划分与分配。

5.3. Incentives: from bureaucratic to cybernetic
5.3. 激励机制：从官僚化到网络化

In the analog world, incentives are bureaucratic in the sense that they are set in contracts that align the objectives of the partners, such as explicit compensation arrangements between owners and managers (Oehmichen, Jacobey, & Wolff, 2020). Incentives are important complements to controls; they can help reduce conflicts between parties, e.g., agency problems due to the separation of decision- and risk-bearing functions (Fama & Jensen, 1983). In their bureaucratic form, incentives are agreed upon by parties and subject to potential renegotiations. For example, when a bonus is paid to an employee, this is an incentive that rewards the employee with variable compensation that is typically agreed upon contractually (Shaw, Gupta, & Delery, 2000).
在模拟世界中，激励机制具有官僚主义色彩，因为它们是在合同中设定的，旨在对齐各方目标，如所有者与管理者之间的明确补偿安排（Oehmichen, Jacobey, & Wolff, 2020）。激励机制是控制的重要补充，它们能帮助减少各方之间的冲突，例如。,由于决策与风险承担功能的分离而产生的代理问题（Fama & Jensen, 1983）。在官僚形式中，激励措施由各方商定，并可能受到重新谈判的影响。 例如，当向员工支付奖金时，这是一种激励措施，奖励员工以可变薪酬，这种薪酬通常是在合同中约定的（肖、古普塔和德勒里，2000）。
Bureaucratic mechanisms differ sharply from cybernetic incentives, which are set and re-evaluated by a self-adapting algorithm in a feedback loop where outputs continuously serve as inputs (Green and Welsh, 1988, Vergne, 2020). For example, an increasingly popular cybernetic incentive mechanism is “proof-of-stake” rewards in cryptocurrencies. Here, network participants serve as “validators” by staking their cryptocurrencies or tokens for a set period, which is documented in a smart contract on the blockchain (Edelman, 2022). In return, when participants validate new block transactions, they are rewarded for their validation efforts with cryptocurrency. Another example of a cybernetic incentive is Google Maps Local Guides; contributors to Google Maps who share reviews, photos, and knowledge are rewarded with points, which can be exchanged for rewards or used in exclusive community events (Tajedin, Madhok, & Keyhani, 2019). Underlying these cybernetic incentive systems are algorithms that automatically distribute and adjust rewards when certain input conditions (e.g., price levels, demand) are achieved.
官僚机制与控制系统的激励措施截然不同，后者通过自适应算法在反馈环路中被设定和重新评估，其中输出持续作为输入（格林和韦尔士，1988，韦尔热，2020）。 例如，一个日益流行的网络激励机制是“权益证明”奖励在加密货币中。在此，网络参与者通过在一定时间内质押他们的加密货币或代币来充当“验证者”，这一过程记录在区块链上的智能合约中（Edelman, 2022）。 作为回报，当参与者验证新区块交易时，他们以加密货币作为其验证工作的奖励。另一个网络激励的例子是 Google Maps Local Guides；为 Google Maps 分享评论、照片和知识的贡献者会获得积分，这些积分可以兑换奖励或用于专属社区活动（Tajedin, Madhok, & Keyhani, 2019）。 这些网络激励系统的基础是算法，它们会根据某些输入条件（例如，价格水平、需求）的达成，自动分配和调整奖励。
In addition to the range of bureaucratic and cybernetic incentives, programmatic incentive structures that are recorded in code but subject to manual review and adjustment could become the norm. In contrast to cybernetic incentives, programmatic incentives are predefined, rigid, and automatized rules that do not work in a constant feedback loop. An example of programmatic incentives is preferred supplier programs: Sourceability, a global distributor of electronic components, has developed an automatized supplier rating system that scores each supplier and prefers high-quality, timely suppliers, incentivizing suppliers through the communication of rank order and related order placement.
除了官僚主义和控制论的一系列激励措施外，记录在代码中但可以进行人工审查和调整的程序性激励结构可能会成为常态。与控制论激励不同，程序性激励是预定义的、刚性的、自动化的规则，不会在持续的反馈循环中工作。 程序化激励的一个例子是首选供应商计划：Sourceability，一家全球电子元件分销商，开发了一套自动化供应商评级系统，该系统对每个供应商进行评分，并偏好高质量、准时的供应商，通过沟通排名顺序及相关订单安排来激励供应商。

5.4. Trust: from actor-based to algorithmic
5.4. 信任：从基于个体的到算法的

In the analog world, trust is actor-based and describes the expectation that an exchange partner will not behave opportunistically, even when the affected party has limited abilities to detect such behavior (Mayer et al., 1995, Puranam and Vanneste, 2009). In interorganizational collaboration, trust can be built between partners that engage with each other repeatedly, where each delivers the outcomes expected of the other, and where each entity behaves responsibly toward the other, which can ultimately build trust in competence and goodwill (Das and Teng, 2001, Gulati, 1995).
在模拟世界中，信任是基于角色的，描述了对交换伙伴不会在受影响方能力有限的情况下采取机会主义行为的预期（Mayer 等人，1995，Puranam 和 Vanneste，2009）。 在跨组织合作中，通过反复互动的伙伴间可建立信任，其中每一方都履行对对方的期望成果，且每一实体都以负责任的态度对待另一方，这最终能建立对能力与善意的信任（Das 和 Teng，2001，Gulati，1995）。
In automated contexts, trust resides in the algorithmic system itself and is not dependent on personal relationships (Lumineau, Schilke, & Wang, 2023). Rather than relying on the (unobservable) actions of partially known exchange partners, trust can be placed in a system that automatically validates each party without having to reveal its identity. For example, cryptocurrencies, such as Bitcoin and Ether, rely fully on transactions based on automated consensus mechanisms, eliminating the need for trusted intermediaries to help safeguard against opportunistic behavior (Seidel, 2018, Werbach, 2018). In enterprise applications, blockchain technologies such as IBM Food Trust enable information in the food sector to be stored on an immutable blockchain, allowing customers to track their entire value chain while increasing confidence in the product’s provenance and quality (IBM, 2022). Trust in the system, rather than in the actors, is particularly crucial in digital exchange (Ba & Pavlou, 2002).
在自动化环境中，信任建立在算法系统本身，并不依赖于个人关系（Lumineau, Schilke, & Wang, 2023）。相较于依赖于部分已知交易伙伴（无法观察到的）的行为，信任可以寄托于一个自动验证各方身份而不必泄露其真实身份的系统。 例如，像比特币和以太币这样的加密货币，完全依赖基于自动化共识机制的交易，消除了对可信中介的需求，以防范机会主义行为（Seidel, 2018, Werbach, 2018）。 在企业应用中，诸如 IBM Food Trust 这样的区块链技术使得食品行业的信息能够存储在不可篡改的区块链上，让客户可以追踪整个价值链，同时增强对产品来源和质量的信心（IBM, 2022）。 信任系统，而不是信任角色，在数字交易中尤为关键（Ba & Pavlou, 2002）。
Along the spectrum of actor- and system-based trust, an augmented mode of trust can emerge as a principal form that is actorithmic in nature, where actor-based trust is algorithmically enhanced. In other words, actorithmic governance emerges when trust depends partly on the actions of the parties involved in an exchange and partly on digital technologies. Actorithmic governance is evident in digital marketplaces and platforms such as eBay and Airbnb, where trust between unacquainted transaction parties is partially established through the parties fulfilling their agreements and partially through digital technologies. The necessary trust building is facilitated by the recommender systems of these platforms, which assess parties based on their transaction performance, combining human evaluation with automated ranking systems (Malgonde et al., 2020).
在基于参与者和系统的信任谱系中，一种增强型的信任模式可以成为主要的形态，其本质是算法化的，即基于参与者的信任通过算法得到增强。换言之，当信任部分依赖于交易中各方的行为，部分依赖于数字技术时，算法化治理便应运而生。 算法治理在数字市场和平台中显而易见，例如 eBay 和 Airbnb，陌生交易方之间的信任部分通过各方履行协议建立，部分通过数字技术实现。 这些平台的推荐系统促进了必要的信任建立，它们根据各方的交易表现进行评估，结合了人工评价与自动化排名系统（Malgonde 等，2020）。

5.5. Combining analog, augmented, and automated governance forms
5.5. 结合模拟、增强和自动化治理形式

Governance designers can blend control, coordination, incentives, and trust mechanisms from analog, augmented, and automated governance to create new governance configurations that can be gradually implemented. In many organizational settings, some governance mechanisms are analog, while others are augmented or automated. As an illustration, the Linux kernel development community utilizes a combination of analog and augmented governance. Control is augmented through write access granted solely to maintainers (authorized developers) who formally approve patches reviewed by the community. Coordination is facilitated through digital mailing lists. Incentives are augmented by a repository that stores version copies and developer names, allowing for credit attribution. Finally, trust is placed in “trusted lieutenants,” who work closely with founder Linus Torvalds (Lee & Cole, 2003). Another example that leans more strongly toward automation are DAOs, where control is automated through voting rights and code-embedded rules; coordination is augmented with digital tools such as Discord; incentives are automatically distributed through consensus mechanisms and smart contracts; and trust is fully automated using blockchain technology (Hsieh and Vergne, 2023, Kaal, 2021). This combinatory power of digital governance enables designers to experiment with various configurations before making a commitment or opting for partial automation of governance.
治理设计者可以将来自模拟、增强和自动化治理的控制、协调、激励和信任机制融合，以创建可逐步实施的新治理配置。在许多组织环境中，一些治理机制是模拟的，而其他可能是增强型或自动化的。以 Linux 内核开发社区为例，它运用了模拟与增强治理的结合方式。 控制通过仅授予维护者（授权开发者）的写访问权限得以增强，他们正式批准社区审查的补丁。协调通过数字邮件列表进行。通过存储版本副本和开发者姓名的存储库，激励机制得以加强，便于归属认可。 最后，信任被寄托在“值得信赖的副手”身上，他们与创始人林纳斯·托瓦兹（李&科尔，2003）密切合作。另一个更倾向于自动化的例子是去中心化自治组织（DAOs），其控制权通过投票权和嵌入代码中的规则实现自动化；协作通过 Discord 等数字工具得到增强；激励通过共识机制和智能合约自动分配；而信任则完全通过区块链技术实现自动化（Hsieh 和 Vergne, 2023, Kaal, 2021）>）。这种数字治理的组合能力使得设计者在做出承诺或选择部分自动化治理之前可以进行各种配置的实验。
Notably, governance choices are dynamic and adaptable. In fact, each of the four governance mechanisms may experience a dynamic development; a shift from analog to automated ways of governing exchanges, or vice versa. However, it is unlikely that automated forms of governance will emerge as the only dominant form. Rather, the dominant governance mode may gradually stabilize in augmented forms of governance, where analog and digital forms overlap and fill each other’s voids. In the next section, we describe a heuristic for determining when analog, augmented, or automated forms of governance are best suited for exchange.
值得注意的是，治理选择是动态且可适应的。实际上，这四种治理机制中的每一个都可能经历动态发展，从模拟方式转变为自动化方式，或是反向转变。然而，自动化治理形式不太可能成为唯一主导形式。相反，主导的治理模式可能会逐渐稳定于增强形式的治理中，在那里，模拟与数字形式相互交叠，互补不足。 在下一节中，我们将描述一种启发式方法，用于确定何时最适合使用模拟、增强或自动化的治理形式进行交换。

6. Making the right governance choice
6. 做出正确的治理选择

An important consideration when evaluating governance choices concerns the associated governance costs of designing, implementing, and adapting the necessary control, coordination, incentive, and trust mechanisms. Each of the three focal governance modes (analog, augmented, and automated) generates specific costs. From an efficiency perspective, the governance choice should provide the desired benefits at the lowest possible cost.
在评估治理选择时，一个重要的考虑因素是设计、实施和调整必要的控制、协调、激励和信任机制所涉及的治理成本。每种三种主要治理模式（模拟、增强和自动化）都会产生特定的成本。从效率的角度来看，治理选择应提供所需的好处，同时尽可能降低成本。

6.1. A governance choice framework for the digital age
6.1. 数字时代的治理选择框架

A core tenet of transaction cost economics is that transactional attributes, particularly the “bilateral dependency [that] builds up as asset specificity deepens” (Williamson, 1991, p. 282), determine governance choices. However, this argument requires serious reconsideration in the digital age, where exchanges often occur between multiple parties simultaneously, the primacy of assets gives way to digital data, and reliance on institutional enforcement is supplanted by algorithmic rules. Moreover, digital exchanges often occur outside the spectrum of markets and hierarchies, generating new forms of organizing as in blockchains, digital platforms, and online communities (e.g., Benkler, 2002, Puranam et al., 2014). We are interested in an extension of the classical governance choice model, taking into account augmented and automated forms of governance in addition to the analog form. Our objective is to incorporate a discriminant logic that explains when and why each form of governance is selected.
交易成本经济学的核心信条之一是，交易属性，尤其是随着资产专用性加深而建立的“双边依赖”（Williamson, 1991，第282页），决定了治理选择。 然而，在数字时代，这一论点需要严肃地重新考虑，因为在这一时代，交换往往在多方之间同时进行，资产的主导地位让位于数字数据，而对机构执行的依赖则被算法规则所取代。 此外，数字交易常常发生在市场与层级结构之外，催生出区块链、数字平台和在线社区等新的组织形式（例如，本克勒，2002，普拉纳姆等人，2014）。 我们对经典治理选择模型的扩展感兴趣，考虑了除传统模拟形式外，增强型和自动化形式的治理。我们的目标是纳入辨析逻辑，解释何时以及为何选择每种形式的治理。
To better explain digital governance choices, we introduce the notion of transactivity—a composite construct that encompasses the overall extent of contributors (i.e., participants), connections (i.e., relationships), and consistency (i.e., flows) in an exchange network. In network terminology, the first element concerns network size (i.e., number of nodes) and the second concerns network density (i.e., realized connections between nodes), two important network governance determinants (Provan & Kenis, 2007). The third element, consistency, indicates whether exchanges occur in a standardizable or homogeneous manner rather than a customized or idiosyncratic fashion. Notably, a linear increase in the number of contributors (i.e., network size) can induce exponential growth in the number of connections (i.e., network density), which puts particular strain on any analog governance design. In contrast, consistency acts as a critical boundary condition for the scalability of any automated governance solution; algorithmic solutions require predictability and reliability for seamless execution.
为了更好地解释数字治理的选择，我们引入了互动性的概念——这是一个复合构念，涵盖了交换网络中贡献者（即参与者）、连接（即关系）和一致性（即流程）的总体范围。在网络术语中，第一个元素涉及网络规模（即节点数量），第二个元素涉及网络密度（即（节点间实现的连接），两个重要的网络治理决定因素（Provan & Kenis, 2007）。第三个要素，一致性，表示交换是以可标准化或同质化的方式进行，而非定制化或特异性的方式。值得注意的是，贡献者数量的线性增加（即网络规模）可诱导连接数量的指数级增长（即，网络密度），这对任何模拟治理设计都特别有压力。相比之下，一致性作为自动化治理解决方案可扩展性的关键边界条件；算法解决方案需要可预测性和可靠性才能无缝执行。
Since the three elements that constitute transactivity strongly interact with each other, very high transactivity values occur when all three elements take on high values. However, low transactivity values can occur if only one element has low values while the others have high values. This would apply to a large network (e.g., hundreds of members) with low connectivity (e.g., a density of ten percent) and high consistency (e.g., all exchanges are similar in nature). In this setting, most interactions occur on a bilateral basis with little external exchange interdependency, fostering a dispersed network structure that reduces the need for a unified and integrated governance solution. In formal terms, the multiplicative nature of the interrelationship between these three elements can be expressed as follows:transactivity=contributors×connections×consistency
We expect the costs of each governance alternative (analog, augmented, and automated) to increase exponentially as a function of transactivity but at different rates. Analog governance mechanisms are particularly cost-effective for low transactivity as the necessary arrangements can or must be negotiated bilaterally, taking into account specific transactional attributes. Here, the parties select a combination of transaction-specific relational and contractual agreements, as is the case with  strategic alliances or mergers and acquisitions. However, the costs of analog governance become particularly punitive as transactivity increases because the number of agreements to be negotiated increases at an exponential rate.
In contrast, for automated governance, the setup costs tend to be much higher than analog solutions because the required algorithms are expensive to design, implement, and adjust (Rimba et al., 2020, Zhu et al., 2006). However, they become comparatively low when they can be spread across a large number and volume of similar exchanges because digital solutions can be scaled at low marginal costs. Therefore, automated governance solutions have a relative cost advantage over analog governance options in settings with high transactivity, as is the case with many financial transactions, standardized service contracts, and small purchase agreements involving many parties.
Finally, the augmented governance solution falls between the two extremes; it incurs costs on both the analog and algorithmic sides. Augmented governance involves the combination of analog and augmented governance solutions. The associated costs tend to be comparatively low at medium levels of transactivity, which might be the case in small, high-density networks where some customization is required but many exchange attributes are standardized, e.g., a supply chain network where certain facets require specific relational and contractual governance (such as discussing product specifications and agreeing on relationship scope), but many aspects can be transferred to fully automated solutions (e.g., delivery, pricing, and orders). In Fig. 4, we illustrate the general logic underlying the relationship between transactivity (horizontal axis) and governance costs (vertical axis) as well as the resultant governance choice.1
  1. Download: Download high-res image (139KB)
  2. Download: Download full-size image

Fig. 4. Governance choice framework.

6.2. A contingency perspective on the benefits and costs of digital governance

Our analysis has focused primarily on governance costs, but in practice, it is important to weigh a complex set of costs and benefits for different stakeholders, particularly governance designers and exchange participants. For governance designers, the toolbox of automated governance offers many benefits, such as better insight into user behavior, improved unwanted action monitoring, and greater efficiency, especially in scaling operations (Benlian et al., 2022). Such benefits are offset by significant upfront costs when designing, implementing, and adapting automated governance solutions, which can hinder their deployment for small-scale transactions. Hence, organizational characteristics may play an important contingency role; smaller organizations may lack requisite funds for designing and implementing automated governance modes. Moreover, lawmakers worldwide are increasing pressure, demanding greater accountability for online activities (e.g., fake news and illegal activity) and putting bounds on the use of governing algorithms (e.g., the AI Act in the European Union).
While focusing on the governance designer's perspective is reasonable for analytical purposes, the exchange participants may have a different perception of the costs and benefits of digital governance. For participants, automated governance offers many promises, such as increased predictability through technologies such as smart contracts, improved inclusivity due to the low barriers to participation in the digital economy, and a high level of reliability in conducting transactions based on transparent rules (Santana & Albareda, 2022). Nevertheless, automated governance solutions can also be abused, leading to a loss of autonomy through surveillance, a sense of voicelessness amid quasi-monopolistic digital incumbents, and dependencies on specific services with high switching costs (e.g., Möhlmann, Alves de Lima Salge, & Marabelli, 2023). Network effects introduce another contingency for participants, wherein they may be coerced into accepting certain governance modes due to peer pressure, while the simultaneous increase in switching costs and pull effect exerted by the network further amplifies this pressure. In Table 1, we provide an overview of some of the key benefits and limitations of digital governance from the perspectives of governance designers and exchange participants. We also elaborate on these issues below in our future research agenda.

Table 1. Benefits and costs of digital governance for designers and participants.

Empty CellEmpty CellHow digital governance impacts
Empty CellEmpty CellBenefitsCosts
Whom digital governance impactsGovernance designers
  • •
    Insight:
    e.g., behavioral data analytics in real time, latent pattern analysis
  • •
    Oversight:
    e.g., automated fraud detection, transparent operational data
  • •
    Efficiency:
    e.g., low costs per transaction, easy scalability
  • •
    Design:
    e.g., programming efforts, testing algorithms
  • •
    Implementation:
    e.g., infrastructure set-up, conversion of IT systems
  • •
    Adaptation:
    e.g., bug corrections, feature updates
Exchange participants
  • •
    Predictability:
    e.g., clarity of incentive mechanisms
  • •
    Inclusivity:
    e.g., opportunities for individuals to contribute economically
  • •
    Reliability:
    e.g., high standardization, clear if-then conditions
  • •
    Policing:
    e.g., loss of autonomy, surveillance of behavior
  • •
    Impotence:
    e.g., lack of influence on governance decisions
  • •
    Dependence:
    e.g., lock-in effects due to peer pressure and network effects

7. Developing a research agenda on digital governance

We propose a research agenda with two avenues through which scholars can deepen and broaden their understanding of digital governance. The first avenue adds to our focal discussion and highlights the governance challenges posed by digital technologies (governance by algorithms). The second avenue extends our discussion and addresses the accountability of digital governance (governance of algorithms). Table 2 provides a summary of these future research opportunities and key research questions.

Table 2. Summary of future research opportunities in digital governance.

Governance by AlgorithmsCognition and emotions
  • ▪
    Does digital governance lead to a “switching off” mentality (similar to a loss of orientation due to using Google Maps), and if so, how can it be prevented?
  • ▪
    When do digital governance mechanisms harm vigilance?
  • ▪
    When and why does digital governance affect human emotions, either positively (e.g., empowerment, trust, and confidence) or negatively (e.g., fear, frustration, and helplessness)?
  • ▪
    How can digital connections and identities strengthen or weaken relational ties between network participants?
Standardization and biases
  • ▪
    How can organizations reap the benefits of digital governance (e.g., efficiency and standardization) while minimizing its costs (e.g., rigidity and technological dependencies)?
  • ▪
    Are organizations losing their “human touch” as a result of increasing digital governance?
  • ▪
    What are the limits of digital control and programmability of (inter)organizational processes?
  • ▪
    How does digital control influence social and creative tasks in organizations, e.g., the formation of friendships and innovation activities?
Contingency and

boundary factors
  • ▪
    When is augmented and automated governance superior to analog governance and vice versa?
  • ▪
    How can digital technologies enable forms of process controls that relieve the oversight role of managers?
  • ▪
    What is the optimal balance between digital and analog governance?
  • ▪
    When do the costs of digital governance outweigh its benefits, and when is “no governance” a better solution?
Governance of AlgorithmsDesign and responsibility
  • ▪
    Where do organizations locate the responsibility for designing (e.g., setting the parameters) digital governance mechanisms (e.g., internal development vs. outsourcing)? What are the decision parameters?
  • ▪
    Who takes responsibility and checks for biases and technical errors in technical governance solutions?
  • ▪
    How can consensus mechanisms be designed to provide security and effective dispute resolution?
  • ▪
    When should analog governance complement and constrain automated governance?
Accountability and regulation
  • ▪
    Should policymakers actively regulate digital governance?Should digital board members and auditing firms (e.g., AI-based accounting controls and blockchain-based transparency mechanisms)
  • ▪
    be allowed, and if so, when?
  • ▪
    With whom and under what circumstances should the algorithms and data underlying governance decisions be shared or even made public?
  • ▪
    How can digital governance mechanisms contribute to or prevent antitrust problems?
  • ▪
    In large digital networks where pseudonymity prevails, who is responsible for illegal activities, biased decision-making, and technical errors?
Cybersecurity and risk
  • ▪
    How can companies increase their resilience against malicious attacks and hacking?
  • ▪
    How can cybersecurity become a strategic issue?
  • ▪
    How can companies maintain their strategic autonomy as digital technologies require ever more expertise, which often resides outside the organization?
  • ▪
    How can companies secure their critical infrastructures amid increasing data integration and processes?

7.1. Avenue 1: governance by algorithms

Our study underscores the nature of digital governance, a shift toward using digital technologies to provide automated mechanisms of control, coordination, incentives, and trust. Consistent with findings on other digital technology-driven phenomena such as digital transformation (Hanelt et al., 2021, Verhoef et al., 2021, Vial, 2019), important questions arise regarding the broader implications of such technological changes. While algorithms enable the control of numerous participants and foster perceived fairness and impartiality within organizations (Dolata et al., 2022, Fu et al., 2022), their use also promotes rigid standardization and the risk of losing sight of the social side of organizations, e.g., human cognition and emotion (e.g., Massey, 2002). Thus, insights into the cognitive, emotional, and organizational processes that accompany the introduction of digital governance are needed.
Cognition and emotions. The use of digital governance raises concerns about the role of “soft factors” such as emotions (e.g., enthusiasm or frustration) and perceptions (e.g., valuation or sensemaking), and whether people are comfortable being monitored, controlled, and potentially challenged by algorithms. Previous research highlights the impact of technology on customer emotions and discomfort, suggesting the importance of emotions in digital interactions (Holthöwer & van Doorn, 2022). The use of algorithms can also result in frustration when they challenge human intuition and compromise accountability in the workplace (Allen and Choudhury, 2022, Lebovitz et al., 2022). These examples illustrate the need to consider the impact of digital governance on human emotions and perceptions to ensure its effective and responsible use.
The role of fairness in determining people’s compliance with algorithmic solutions highlights the importance of a sociotechnical perspective in evaluating the interplay between technology and society (Dolata et al., 2022, Lee, 2018). Research has shown that individuals engage in complex sensemaking processes to interpret algorithms and their perceived fairness, and optimize their behavior based on incentives set by the algorithms, leading to widespread behavioral adaptation (Bellesia et al., 2023, Cameron and Rahman, 2022, Möhlmann et al., 2023). In light of the behavioral changes that may accompany the deployment of digital governance, it is crucial for further research to investigate the broader societal consequences of digital governance from both a research and a policy perspective (Lamy et al., 2022).
Standardization and biases. While some arguments suggest that more technology and governance produce linear benefits, marginal returns could decrease over time due to increasing technological saturation (Karr-Wisniewski & Lu, 2010). Therefore, digital governance implementation evokes fundamental discussions in the management literature when increasing formalization is beneficial or harmful to organizations (Walsh & Dewar, 1987). Further research is needed to determine whether digital governance promotes increasing “hyperbureaucratic” organizations where every step is mapped digitally. Consequently, hyperbureaucracy may hinder creativity and innovation in such ventures, negatively impacting organizational performance (Adler and Borys, 1996, Pesch et al., 2021). Furthermore, digital technologies can create barriers between an organization and its customers. For example, research on chatbots (which are based on algorithms and AI) as mediators in customer interactions shows that technology can effectively serve as a barrier, impacting customer satisfaction (Crolic, Thomaz, Hadi, & Stephen, 2022).
A particularly critical issue is built-in biases in the algorithms (Akter et al., 2022). Research has provided disturbing evidence of how machine learning algorithms, if not carefully trained, can reinforce and proliferate malicious gender and racial stereotypes (Hundt, Agnew, Zeng, Kacianka, & Gombolay, 2022). Future research should therefore deepen our knowledge of how organizations can use digital governance to counter biases and ensure safe and inclusive digital environments (see Bolukbasi, Chang, Zou, Saligrama, & Kalai, 2016 for an example of a gender debiasing algorithm). This also applies to the risk of misinformation triggered by algorithmic systems: The OpenAI-developed chatbot, ChatGPT, was banned from Stack Overflow and faced restrictions on being listed as an author by academic publishers due to concerns about its potential for producing incorrect answers (Sample, 2023, Vincent, 2022).
Contingency and boundary factors. Finally, the contingency and boundary factors of digital governance require elaboration (see our discussion of Fig. 4). Research suggests that organizational characteristics, such as the size of the company (i.e., large incumbent vs. small to medium enterprise), can significantly impact the adoption of digital technologies (Fabian, Dong, Broekhuizen, & Verhoef, 2023). Furthermore, contingency aspects concern the normative question of whether companies should embrace the further digitization of their governance. For instance, Griesbach, Reich, Elliott-Negri, and Milkman (2019) uncover irritating forms of “algorithmic despotism” in food delivery platforms where algorithms essentially dictate coworkers’ schedules and activities. Worth keeping in mind is that the motivation for adopting digital governance is not only a firm-internal consideration but may also be influenced by interorganizational ties (e.g., supply chain interdependencies) and the institutional environment (Aguilera and Jackson, 2003, Oehmichen et al., 2017). Understanding organizational embeddedness, such as through regulations, cooperative, and competitive dynamics, can help explain variance in digital governance adoption. For instance, digital governance may be adopted due to mimetic pressures and because it has already been adopted by other firms (Mithas et al., 2013, Wang, 2010).

7.2. Avenue 2: governance of algorithms

In addition to recognizing the potential of automation, digital governance necessitates discussion of responsibility and accountability, affirming our second research avenue concerning the governance of algorithms (Haenlein et al., 2022, Haenlein and Kaplan, 2021, Kaplan and Haenlein, 2020, Loebbecke and Picot, 2015, Martin, 2019). Insights into who designs algorithms and oversees the parameters of digital governance are important (see Chhillar & Aguilera, 2022 for a review in the AI context). While this issue is important from a societal perspective, it also involves legal considerations as long as the algorithms are not considered legal entities that can be held accountable for their actions (Drummer & Neumann, 2020). Moreover, from a legal perspective, it is critical for policymakers to understand the implications of digital governance and its potentially detrimental effects on society. Finally, the shift toward algorithmic and internet-based technologies also has implications for cybersecurity and how organizations can protect themselves against cyberthreats.
Design and responsibility. The definition of responsibilities for the design and consequences of digital governance remains a largely unexplored area. The impact of ambiguous responsibilities is exemplified in the Bitcoin protocol, kickstarted by Satoshi Nakamoto, who remains anonymous and cannot be held accountable for the abundance of illegal activities that this cryptocurrency enables (Foley, Karlsen, & Putniņš, 2019). From a corporate point of view, the question arises whether digital governance should be viewed as a technical matter for IT departments or whether it requires the involvement of top management, such as the Chief Digital Officer (Firk, Hanelt, Oehmichen, & Wolff, 2021). In addition to new roles such as the CDO, specific expertise can play a role, such as how the digital expertise of managers and board members influences the development and use of algorithms (Fabian et al., 2022). The shift toward algorithmic modes of management also presents opportunities for exploring the cognitive capabilities of managers (Helfat & Peteraf, 2015) because using automated governance technologies requires a new set of skills to effectively utilize them. From an alliance perspective, further research is needed on how to govern collaboration with algorithm providers such as AI startups (Oehmichen, Schult, & Dong, 2023) and AI-as-a-service providers (Zapadka, Hanelt, Firk, & Oehmichen, 2020).
Accountability and regulation. With the advent of digital governance, the question of control is critical for organizations (Kellogg, Valentine, & Christin, 2020). If algorithms lead to adverse consequences in established organizations (e.g., discrimination), who ultimately bears responsibility for them? If the shift toward decentralized and transparent blockchain-based systems continues, how can actors within such systems be held accountable, especially if online and legal entities remain separate (Sun Yin et al., 2019, Tumasjan, 2021)? These questions suggest that digital governance presents macrolevel challenges that are relevant for policymakers (Nambisan, Wright, & Feldman, 2019). If governance is increasingly becoming more automated, what policies should be designed to protect network participants and how? How can algorithms be designed to comply with national and supranational regulations such as those enacted by the European Union? Interestingly, many large technology firms are calling for regulation at the national and supranational levels (Bajarin, 2020, Feloni, 2018, Knight, 2019), raising the intriguing question of whether these digital incumbents are truly interested in ethical digital governance or whether they are primarily trying to devolve responsibility.
Cybersecurity and risk. As organizational processes and decision-making are increasingly reliant on algorithmic protocols, the supporting infrastructure has become an attractive target for malicious attacks that compromise the operations of an organization and sensitive data (Angst, Block, D’Arcy, & Kelley, 2017). For example, the cloud company Akamai reported the largest DDoS attack ever launched against a European customer—an aggressive attempt to cripple the operations of the business (Sparling, 2022). Given such vulnerabilities, companies are called upon to develop appropriate security measures to protect their IT infrastructure from them, entailing critical tradeoffs between a higher level of automation, which enables greater efficiency, and vulnerability to cyberattacks, which can cause severe reputational damage (Triche & Walden, 2018). With the rise in digital governance and the growing importance of data, cybersecurity is becoming a strategic issue for organizations, requiring careful consideration not only to prevent malicious attacks, but also to avoid reputational damage.

8. Conclusions

In this paper, we provide a new perspective on governance in the digital age. As organizations are coalescing into ever-larger value networks, we argue that governance mechanisms for mitigating the tension between cooperation and competition between different exchange participants in digital environments are crucial. Our research emphasizes the critical importance of regulating control, coordination, incentives, and trust in ways that enable new forms of organizing, value creation, and value capture. Hence, we define digital governance as one of the long-term cornerstones of management in the digital age.

CRediT authorship contribution statement

Marvin Hanisch: Writing – review & editing, Writing – original draft, Conceptualization. Curtis M. Goldsby: Writing – review & editing, Writing – original draft, Conceptualization. Nicolai E. Fabian: Writing – review & editing, Writing – original draft, Conceptualization. Jana Oehmichen: Writing – review & editing, Writing – original draft, Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors would like to express their sincere gratitude to the editor, Arnd Vomberg, for his valuable insights and constructive feedback throughout the review process. We would also like to extend our thanks to the two anonymous reviewers for their thoughtful comments, which helped improve the quality of the paper. Their contributions are greatly appreciated.

Appendix A. Summary table of key concepts with examples

The following online appendix provides an overview of key concepts and examples related to digital governance:Download: Download Word document (137KB)

Summary table.

References

Cited by (27)

  • Blockchain technology and privacy regulation: Reviewing frictions and synthesizing opportunities

    2024, International Journal of Information Management

  • Breaking the resource curse: Heterogeneous effects of digital government

    2024, Resources Policy

  • Agency in the algorithmic age: The mechanisms and structures of blockchain-based organizing

    2023, Journal of Business Research

  • Blockchain for compliance: an information processing case study of mandatory supply chain transparency in conflict minerals sourcing

    2024, Supply Chain Management

  • Prescriptive Theorizing in Management Research: A New Impetus for Addressing Grand Challenges

    2024, Journal of Management Studies

  • The impact of digital transformation on corporate innovation: Roles of analyst coverage and internal control

    2024, Managerial and Decision Economics
View all citing articles on Scopus
Marvin Hanisch is an Assistant Professor in the Innovation Management & Strategy Department at the University of Groningen. His research focuses on governance mechanisms in strategic alliances, open-source software communities, digital platforms, and blockchain networks. His work has been published in leading management journals, such as the Journal of Management Studies, and in practice-oriented outlets, such as California Management Review, and has received numerous international awards. Marvin Hanisch is co-affiliated with the University of Passau and is a visiting lecturer at Maynooth University.
Curtis Goldsby is a research scholar in management in the Technology and Operations Management Department at the Rotterdam School of Management, Erasmus University. As a Managing Consultant at IBM Consulting in Düsseldorf, Germany, he combines his research interests in platform and blockchain governance with practical engagements in the automotive, retail and industrial sectors for DAX 30 clients. In his PhD research, Curtis Goldsby studies how digital technologies can be deployed in intra- and interorganizational governance contexts. His work has been featured in California Management Review and conference outlets, such as the Strategic Management Society. Curtis Goldsby is an alumnus of the London School of Economics with an M.Sc. in Management.
Nicolai Fabian is an Assistant Professor in the Innovation Management & Strategy Department at the University of Groningen. His research focuses on the digital transformation of incumbent companies. He is also interested in digital business models and digital innovations and their impact on companies and society. His research has been published in the Journal of Business Research, the European Journal of Information Systems, and in a peer-reviewed academic book by Routledge.
Jana Oehmichen is a Full Professor of Organization, Human Resources, and Management Studies at the University of Mainz and Honorary Professor of Leadership & Governance at the University of Groningen. Her work has been published in journals such as Strategic Management Journal, Organizational Behavior and Human Decision Processes, Leadership Quarterly, Global Strategy Journal, and the Journal of Management Studies. Her research is focused on digital leadership and strategy as well as corporate governance. She serves as an associate editor for the Schmalenbach Journal of Business Research and as an editorial board member for Corporate Governance: An International Review.
1
Augmented governance often permits a transition from analog to automated governance and allows certain governance voids within each pure governance mode to be filled. For example, when programmatic errors or cyberattacks undermine algorithmic protocols, analog governance mechanisms can provide valuable contingency plans and help restore the system. Thus, regarding these potential cost inefficiencies entailed by a combination of governance modes, the benefits of redundancy and associated system stability may outweigh them.
© 2023 The Author(s). Published by Elsevier Inc.