How ‘Local’ are Action Plans submitted by Open Government Partnership Local Participants?

This paper asks, “Are the action plan commitments of Open Government Partnership (OGP) Local Program participants designed to address locally identified problems in a way that is responsive to their unique technological, organizational and contextual envir onments?” To answer this question, document analysis is used to critically examine the OGP Local Action Plans to assess whether unique contextual characteristics influence the application of broad open government tenets within individual commitments. In addition, a Problem Driven Iterative Adaptation (PDIA) lens is employed to determine whether action plans address locally relevant problems and construct viable and measurable solutions in collaboration with local stakeholders. This analysis allows for the identification of commonalities in the conceptualization and application of open government as well as providing a window into how local context (geographic, social, political, economic, technological) and institutional environment influences the nature, scope and characteristics of implementation.


Introduction
Over the last decade, governments around the world have adopted a variety of "open government" reforms in an attempt to increase government legitimacy, improve living standards and provide more effective service delivery (Meijer et al., 2012;Chatwin and Arku, 2017).While the early conceptualization of open government focused primarily on freedom of information as a means to accountability, the current perspective demands that freedom and access to information are facilitated by a fundamental transformation in the relationship between government and residents (Yu and Robinson, 2012;Clarke and Francoli, 2014;Clarke, 2019).Broadly, open government can be understood as a policy approach to governance reforms that aims to build a culture of transparency, civic participation and accountability to increase confidence in government and strengthen the social contract (OECD, 2016;Williamson and Eisen, 2016).
Leading the global push for openness reforms, the Open Government Partnership (OGP) positions itself as a convener of government and civil society with the purpose of securing commitments to collaborative action.In 2011, at its inception, membership in the OGP was only open to national level governments.However, in 2015, recognizing the importance of local government in improving resident-centered governance and responsive public service delivery, the OGP launched the Subnational Pioneer's Pilot Project (Robinson and Heller, 2015).In 2018, the OGP made the program permanent and increased the number of participating governments from 15 to 20 with plans for future growth (OGP Local, 2019).Following the pilot phase, the program was renamed "OGP Local", in an effort to recognize geopolitical sensitivities in relationships between some local and national level governments.Participants include a variety of entities such as states, regions, provinces, counties and cities. 1   The open government packaging of reforms follows a long history of broad sweeping prescriptions that have promised to address governance deficits and support governments to address their most complex challenges.Often, globally homogenized policies have dominated the governance narrative, presenting externally constructed solutions that minimize the value of adapting to local context (Grindle, 2017).With each new policy prescription comes a new set of implementation challenges and tensions between the approach and the social, economic and political conditions in which it has to operate.These broad governance reforms more often hinder than empower and can lead to superficial changes that are neither effective nor sustainable (Andrews et al., 2015;Grindle, 2017).Particularly at the national level, this has led to what has been termed 'openwashing' or the superficial implementation of open government reforms without the required systemic transformation (Chatwin and Arku, 2017).Similar to previous governance reforms, the empirical data on the efficacy of open government is mixed (Williamson and Eisen, 2016).
In contrast to other predominant approaches to reform, the OGP prioritizes local knowledge and leveraging effective and trusting relationships within a network of stakeholders to implement commitments aimed at immediate change to the status quo (OGP Local, 2019).Participants are encouraged to build their systemic capacity for transparency, civic participation and accountability by addressing locally relevant problems.Taking this into consideration, this paper asks: Are the action plan commitments of OGP Local Program participants designed to address locally identified problems in a way that is responsive to their unique technological, organizational and contextual environments?Document analysis is used to critically examine the OGP Local Action Plans to assesses the extent to which local context influences the application of broad open government tenets within individual commitments.In addition, a Problem Driven Iterative Adaptation (PDIA) lens is used to determine whether action plans address locally relevant problems and construct viable and measurable solutions in collaboration with local stakeholders.The PDIA approach suggests reforms build government capacity by solving locally relevant problems within an environment that encourages innovation, iteration and learning in collaboration with a broad group of local stakeholders (Andrews et al., 2015).
The research draws on 15 of the 20 action plans for the second cycle (2018)(2019)(2020) of the OGP Local participants (see Table 1 for participating jurisdictions).Of the 20, ten participants are on their second round of action plans, five have submitted their first action plan and five action plans were not included.Ontario, Canada, Bojonegoro, Indonesia and Kigoma Ujiji, Tanzania, have not submitted updated action plans in the second cycle.Further, the English version of the action plans for Jalisco, Mexico and La Libertad, El Salvador, have not been published at the time this paper was written.The action plans represent a diverse range of geographic and political-economic contexts and represent a rare opportunity to analyze coordinated efforts to apply open government tenets at the local level.This analysis allows for the identification of commonalities in the conceptualization and application of open government as well as providing a window into how local context (geographic, social, political, economic, technological) and institutional environment influences the nature, scope and characteristics of implementation.The paper starts with a brief review of literature on open government and Problem Driven Iterative Adaptation.This is followed by a detailed overview of the methodology employed.The results of the analysis are then presented and a discussion of their implications is articulated before the paper concludes.

Open Government: Transparency, Civic Participation, Accountability
The Open Government Partnership (OGP) constitutes a relatively new international initiative serving as a mechanism to foster governance reforms aimed specifically at the values inherent in the concept of open government.This includes transparency, civic participation and accountability.However, it is important to note that the notion that transparency and freedom of information are vital to accountable government well predate the OGP.In 1766, Sweden (one of OGP's founding members) became the first country to enshrine press freedom in its constitution and guaranteed freedom to publish and read public documents.The term 'open government' itself was first used in the 1950s in reference to draft freedom of information legislation in the United States and later popularized by Barack Obama in 2009 during the first term of his presidency in the form of the 'Open Government Directive.'In its more contemporary usage, the term was afforded an expanded understanding to include transparency, participation and collaboration as is articulated in the introduction.
The sort of transparency required for contemporary open government takes multiple forms.It relates to the quantity and quality of government information and data available to the public (Florini, 2007).Information and data help residents to better understand government decision making, and to determine their own preferences among government decisions, programs or policies (Bimber, 2003).Contemporary open government and discussions of open data have drawn attention to the issues of quality, format, usability and diversity of the information and data.In part, this should theoretically lead to greater accountability, but there is also an economic rationale built on the understanding of information and data as public resources.Belonging to the public, these resources should be made available, not just to enhance accountability, but to allow the public an opportunity to use them in the development of their own economic pursuits.In essence social and economic innovation is fostered through good transparency.
Furthermore, transparency of government processes helps residents to further engage with governments and to hold them to account.In this sense, it is foundational to the values of civic participation and collaboration which are at the heart of the Open Government Declaration signed by all OGP members.As Harrison and Sayogo (2014: 513) observe "Public participation supplements these processes (transparency and accountability) in that the public is viewed as the ultimate audience for and consumer of information provided through transparency processes."Information and data are needed to inform residents, but they also require access to mechanisms, both informal and formal, to participate in matters of governance.It should also be noted that not all residents will require the same information, data, or participatory mechanisms.In essence, designing open government for the 'general user' runs the risk of compromising transparency and civic participation (Meijer et al., 2012).
While open government reforms driven through OGP are supposed to be aimed at improving transparency, accountability and civic participation within OGP member jurisdictions, members must also be accountable to OGP.Most importantly, they must publish their action plans in a timely manner, ensure a minimum level of civic participation in the development and implementation of their action plans, publish information on their open government work in an online repository, and demonstrate that progress has been made toward their open government commitments (OGP Procedural Review, 2020).Should these criteria not be met, members receive a notice that they have acted 'contrary to process' and risk their membership being deactivated.
OGP's expansion to include local governments was a recognition that they tend to be much closer to the needs and desires of residents than national governments.In many cases, the need for improved transparency, civic participation and accountability is most felt at the local level, giving reforms the greatest opportunity for impactful change.However, there is a temptation simply to adopt international or national level solutions at the local scale.This approach runs the risk of being ill suited to local challenges, and stakeholders may be alienated if a solid co-creation process is not followed.
OGP Local participants are encouraged to build their systemic capacity for transparency, civic participation and accountability by addressing locally relevant problems.The OGP mandates a timeline for each commitment within action plans, establishing an action-oriented expectation.While not explicitly stated, these OGP guidelines reflect a path-dependent belief in the need for iterative change and in response they promote government commitments that align with a non-traditional approach to governance reform called Problem-Driven Iterative Adaptation.

Problem-Driven Iterative Adaptation
Problem Driven Iterative Adaptation (PDIA) was originally introduced by Pritchett et al. (2010).It challenges the dominant prescriptions of externally developed best practices and proposes they be replaced by local solutions for local problems.This does not imply that global initiatives like the OGP are not relevant to addressing local problems, but the tenets of open government must be applied in a way that builds the capacity of government to resolve such problems.Each country and their respective local governments have a unique balance of political, administrative, civil society and resident interests that creates politicization (Moon and Ingraham, 1998).By identifying and addressing local challenges in ways that adapt to local context, the capacity of governance stakeholders is built (Andrews et al., 2015).To align the PDIA approach to open government we have adapted the order and articulation of the four steps originally proposed by Andrews et al. (2013) (see Figure 1).These four steps are: problem identification; engaging a broad network of stakeholders; creating an enabling environment; and feedback through monitoring and evaluation.While these steps are discussed in a linear fashion, they are cyclical and as new information emerges, revisiting previous steps may be required.A brief overview of the four steps follows before the methodology used is elaborated in greater detail.

Problem Identification
The first step of PDIA is to define a locally relevant and tangible problem to address.This requires government entities to collaboratively conduct a contextual analysis related to the presenting problem and why they are struggling to address it.Grindle (2017: 20) states: "Thus, the particular context of a problem -its micro-level political economy of winners and losers, power, reform champions, and coalitions -is a jumping-off point to generate interventions…" Consistent with the literature on systems or complexity thinking, problems must be perceived as a part of an overall system, so that solutions are designed in anticipation that they may merely shift the problem rather than solve it.A shift is required from asking what solution should be adopted to what problem do the local stakeholders want to address (Andrews et al., 2015).A well-defined problem is one that has been constructed, and deconstructed, and defined through shared consensus by local stakeholders.Too often, problems are constructed by stakeholders outside the local network who have a vested interest in seeing a particular solution implemented.
In the context of complex social challenges, the problem is two-fold: the experienced problem in the community or the result of the problem; and the capability or resource gaps that limit the government's ability to sufficiently address it.The emphasis of this component of PDIA is on the capacity building that can develop through the process of identifying and solving problems rather than defaulting to off-the-shelf solutions.Government reforms are more successful when they are motivated by, or directed towards, problems that local stakeholders care about.This involves an active search for solutions that can be achieved within the context and with an understanding of the existing socio-political environment.The more complex the challenge the more modest the scale and scope of the first iterations will be (Andrews et al., 2015).An intentional approach to choosing these preliminary actions must consider past practice, existing stakeholder engagement, quick wins and the potential for rapid results and existing capacities and strengths.A properly framed problem with a compelling narrative is required to communicate the importance to build the engagement and motivation of local stakeholders.OGP has recognized this and has attempted to encourage governments to carefully consider the rationale guiding the commitments outlined in their action plans.This is evidenced by the template for the plans where governments are prompted to clearly outline what problem or 'status quo' their commitment attempts to address, the commitment itself, and then to outline how they see the commitment contributing to a solution to their given problem (OGP Handbook, 2020).

Engaging a broad network
The second step of PDIA focuses on building a broad network of stakeholders who co-create an understanding of the problem and collaborate to address it in a way that is sustainable within the local context (Andrews et al., 2015).The network will include those acting directly to address the identified challenge and those acting as support and accountability partners.This process does not necessarily exclude external actors, but rather it centres the contextual knowledge of local stakeholders.It means that external best practices need to be adjudicated by local stakeholders who understand the problem, context and feasibility of new solutions (Andrews, 2012;Andrews et al., 2013).
Focusing the energy of a network on a particular problem assists in highlighting the weaknesses of the existing structures in addressing it (Seo and Creed, 2002;Vis and van Kersbergen, 2007).Stakeholders on the periphery, often underrepresented groups, can provide new perspectives.A central role needs to be created for those who have lived experience in dealing with the effects of the problem (OGP Participation, 2017).Local stakeholders begin with existing practices and collaboratively define why they are not working and how new solutions could address barriers.Pritchett et al. (2010) discuss the paradox of embeddedness: the stakeholders who are within proximity to identify challenges and barriers do not have the authority to address them, but the stakeholders with the authority do not have the proximity to identify the problems or the willingness to acknowledge them.The collaboration of an engaged network of stakeholders facilitates the co-creation of a change strategy that identifies the necessary conditions for an enabling environment, entry points from within existing practice and permissions, and strategies for creating an environment conducive to realizing the desired change.
As was noted earlier, co-creation is one of the few mandatory conditions of membership in OGP.Members of the OGP Local Program detail the engagement and cocreation exercises that took place in the development of their commitments in the action plans that they submit to OGP.These are subsequently assessed by the OGP's Independent Reporting Mechanism (IRM).Co-creation varies widely across members, with some being more successful than others in engaging the broad range of stakeholders outlined within the literature on PDIA.

Enabling environment
The third step of PDIA relates to the type of environment needed to foster reform.Complex social challenges are rarely 'solved', rather the goal is for the local government entity to develop the capacity to manage and minimize their negative effects.As stated by Muamar (2016: 698): "Poor progress comes from the fact that you do not learn from your wrong decision, but it also comes from the fact that you do not learn from your good decision".The ability to confidently make decisions requires an environment that is conducive to active iteration, experimentation and learning from ongoing monitoring and evaluation (Andrews, 2018).Here, we can identify three necessary conditions: political support of change; acceptance of the need to change by those with authority; and the ability to present new and compelling ideas for implementation (Andrews, 2008;2012;2017).Resistance must be anticipated and the role of reformers, inside and outside government, is to prepare for windows of opportunity and engage the political economy of those positively or negatively impacted by changes to the status quo (Grindle, 2017).
Often for an enabling environment to exist, disruption in the context is necessary, with an event or series of events occurring that make the status quo untenable (Brady and Spence, 2010;Andrews, 2013).This is consistent with the literature on meaningful public participation (Lee, 2013).Those in power need an incentive to question the way things occur and how challenges are addressed.However, disrupting events only catalyze change if an engaged network has already been preparing the environment for the eventual challenge to the status quo.Gradual improvement, quick wins and problem clarity serve to solidify the enabling environment.Similarly, the focus on active iteration and learning to mitigate the risks involved with long-term governance reform that is based on untested assumptions (Pritchett et al., 2010).Local stakeholders, working directly to address a network capacity deficit that hinders their ability to solve a problem, need to set a direction, but also to allow flexible processes, structured adaptation and the creation of opportunities for consistent feedback and data from monitoring (Andrews, 2018).An enabling environment that supports this experimentation is sustained by tight feedback loops that facilitate learning (Muamar, 2016).
While not well documented, OGP has attempted to provide an enabling environment.It has allowed members to change their action plans after they are published in an attempt to recognize that problems, and the way they are approached, can change as learning occurs.The narrative-based assessments completed by the IRM allow researchers flexibility to comment on the local context and the extent to which it helps or hinders change.PDIA suggests that all commitments within the action plan should be ambitious, but viable within the current political, administrative and financial environment of participating jurisdictions.

Feedback through monitoring and evaluation
In the final step of PDIA we see the necessity of a specific stage for reflecting on progress and lessons learned (Andrews, 2012).There is an important distinction between the learning from ongoing monitoring and the learning from pre-identified © 2022 The Author People, Place and Policy (2022): 15/3, pp.149-168 Journal Compilation © 2022 PPP stages for formative evaluation.Evaluation literature argues that monitoring and evaluation strategies should be developed and implemented at programme ideation and be embedded into each stage of action (Kane et al., 2017;Chatwin and Mayne, 2020).Ongoing monitoring and learning are designed to facilitate real-time iteration and pivoting of strategy.They allow for new learning and ideas to be combined into hybrid approaches that are best suited to the context (Andrews, 2013).Monitoring minimizes the risk of long lag times inherent in periodic or end of programme evaluation.For stagebased evaluation, it is valuable to include stakeholders from outside the immediate working group to provide perspective and allow those actively engaged to process their collective experience (Kane et al., 2017).There can be a review of the progress made, lessons learned, and iterations of the approach.The rapid feedback component of PDIA is responsive to the literature on civic inclusion that emphasizes the need for reporting back to stakeholders, both institutional and residents, on what is happening with their input (SFU, 2018).Transparency on why decisions are made and how input from stakeholders influenced them is essential to ensure those in positions of power are being responsive to the needs of the residents.
An analysis of individual and collective contributions can highlight changes in behaviour, trust and relationships through shared impact measurement and learning (Mayne, 2019;Hemelrijck, 2013).Further benefits include contributing to the sustainability of the enabling environment and necessary authorizations to continue the work.It also helps government actors to build legitimacy within their institutions and the broader community (Andrews, 2017).
In the context of OGP, iteration might be considered to happen as members participate over time and develop successive action plans.As this occurs, problems should theoretically be re-examined, approaches evaluated, stakeholders regularly consulted, and evidence of early results presented.However, OGP is limited in the extent to which it can comment on results.Its IRM has clearly maintained that it is not an impact assessment mechanism.It is the responsibility of the participating jurisdictions to establish an ongoing monitoring and evaluation strategy.The narrative focus of the IRM is accommodating for mid-point corrections within commitments.

Methodology
A comprehensive content analysis of OGP Local participants' action plans for the 2018-2020 cycle for 15 of the 20 participating jurisdictions was undertaken (see Table 1).These action plans are publicly available via the OGP website (OGP Local Members, 2020).
The current OGP process of bi-annual action plans provides a unique opportunity to systematically analyze documents from distinct socio-political environments because participants are provided with a template to follow and the action plans are published in both the official language of government and English for each jurisdiction.As can be seen in Table 1, the OGP Local participants represent a broad range of states, regions, provinces, counties and cities from diverse geographic regions.Content analysis is an appropriate approach for the research question as broad learnings and patterns can be drawn from a limited, but representative, number of sources (Moynihan, 2006;Cleave et al., 2017).The diversity in the participating governments allows for the trends and patterns that emerge from the document analysis to be accepted as a reflection of broader government actions within the context of emerging government processes to 'open' government both within and external to the OGP movement (Stemler, 2001;Cleave et al., 2017;Chatwin et al., 2019).The methodology for the research builds on similar studies conducted by Chatwin et al. (2019) on subnational governments, De Blasio and Selva (2016) on European governments, and Clarke and Francoli (2014) on national governments.The analysis of action plans aligns with research by Chatwin et al. (2019) and Clarke and Francoli (2014) in its interest in the 'salience' of open government ideas with a specific focus on how much emphasis is placed on an idea in comparison to other ideas and the entirety of the action plan text.
The research for this current project began with a codebook adapted from Chatwin et al. (2019), with 22 codes (see Table 2).It should be noted that co-creation and progress updates were added as new codes for this review.As previously mentioned, cocreation between the participating jurisdiction, civil society and the public is a mandated component of the OGP Local process and helps researchers understand the process that governments use to identify the 'status quo' or challenge that they are trying to address (OGP Participation, 2017).Using the textual coding software NVivo, the entire corpus of action plans was analyzed using qualitative computer-assisted manual coding to calculate the occurrences of distinct codes.The information was then systematically recorded as counts for a quantitative analysis of the documents.There is a measure of subjectivity inherent in the coding process, but through an iterative process with two reviewers, the subjective decisions were standardized for each action plan (Cleave et al., 2017).For example, if the text of the commitment in an action plan reflects numerous codes, but they are in service to one particular "end", then the entirety of the commitment was coded to the code reflected by the end.In the implementation breakdown, the distinct steps taken to achieve the commitment were coded to the theme they specifically reflected.Another area vulnerable to subjectivity is in the decisions made between coding text that closely reflects multiple nodes.For example, steps taken to improve access to information are often with the intent of facilitating civic participation.Similarly, opening data is often a step towards more access to information.Coded themes that met a threshold of three per cent or greater of an action plan were analyzed for empirical contributions to the research question.Themes that did not meet the three per cent threshold within each action plan were assessed for their alignment with existing themes or removed from analysis.This was done to enhance the readers' ability to see and compare what themes were the most salient for each action plan (see Table 3).If a code did not have one or more jurisdictions with above three percent, it was removed from the final codebook.
Following the coding, an in-depth reading of each plan was conducted by the researchers to gain a better understanding of the context around each commitment, the structure of the plans, and the co-creation process undertaken in each jurisdiction.This in-depth analysis also allowed the researchers to determine whether commitments were clearly assigned to a particular authority with a specified timeframe for completion.

Results and Discussion
The results of the document analysis show how unique contextual characteristics influence the application of broad open government tenets -transparency, civic participation and accountability -within individual commitments.Table 3 provides a visual summary of the analysis, displaying the percentage of text dedicated to each theme, sorted by participating jurisdictions.Only 16 of the original 22 codes represented three per cent or more of the text in one or more the 15 action plans reviewed.Consistent with prior research, civic participation was the most prevalent theme overall, accounting for 26.6 per cent of all text and present in all 15 action plans reviewed (Chatwin et al., 2019).This is perhaps unsurprising given that the template provided to members prompts them to spend some time outlining the co-creation process undertaken in the development of their plan.

Table 2: Codebook
Access to information, foundational to the tenet of transparency, was the second most prevalent theme, present in 14 action plans and representing 14.7per cent of all coded text.A noticeable change from the 2016 action plans was the theme of accountability.It was present in all 15 of the first-wave action plans, but only met the threshold in five of the current action plans and accounted for three and a half per cent of all the text as opposed to seven per cent previously (Chatwin et al., 2019).Likewise, in the first-wave action plans, transparency represented six per cent of the text and was People, Place and Policy (2022): 15/3, pp.149-168 Journal Compilation © 2022 PPP above the threshold in 14 of the 15 plans reviewed.In this round of action plans only two action plans had over three per cent of their text dedicated to transparency and it represented 2.3 per cent of the overall text.One theme that was substantially more salient to participating jurisdictions for this round of action plans was the 'strengthen governance' code.It was present in eight of the 15 action plans and represented 4.4 per cent of all coded text.A new code, co-creation, represented 22.4 per cent of all coded text across the 15 action plans.There was substantial variability in the amount of text dedicated to co-creation in the individual action plans.For example, the Municipality of Sao Paulo (2018) dedicated nearly 80 per cent of their action plan to discussing its cocreation process and how they arrived at their commitments.In contrast, the City of Paris (2018) only dedicated 4.5 per cent and focused more on updates from its previous action plan.

Table 3: Action plan document analysis summary
Looking at these findings, we can see that there was a homogenization of the open government codes that were privileged (those related to transparency, civic participation, and accountability), but variability in how they were applied across levels of government and geographies.This is consistent with previous research findings (Clarke and Francoli, 2014;Chatwin et al., 2019).Some of these discrepancies are detailed here as they relate to the three core tenets of open government.

Transparency
A consistent pattern in the action plans was a passing reference to transparency as a principle that is achieved through activities to improve access to information and opening data.As was noted earlier in this paper, transparency involves providing the public with access to government information and data and ensuring the public understands government decision-making processes (Dawes and Helbig, 2010;Shauer, 2011).Within the action plan text coded to 'access to information', there was an appreciable focus on making information available in ways that facilitate engagement between governments and residents to collectively solve locally relevant problems.For example, Tbilisi's action plan states: "it is important to have mechanisms for processing and simplified usage of published information which can be an important prerequisite for the increase of the citizen engagement" (Tblisi Government, 2018: 6).Indeed, Tbilisi's action plan applied this understanding of residents' need for information to solving a local problem related to construction and tree cutting issues.The first commitment in Tbilisi's plan addressed a locally relevant need for residents to have updated information on government funded construction projects in order to equip them to monitor their progress and permissions for removing green cover (Tblisi, 2018).Similarly, Sekondi-Takoradi aimed to provide the public with user-friendly information on public infrastructure so that they can follow up on the implementation of public investments (Sekondi-Takoradi Metropolitan Assembly, 2018).
In the Elgeyo-Marakwet action plan (Elgeyo Marakwet, 2018), access to information was a critical means to the declared goal of enhancing government actions aimed at capacity development and job creation for youth and other special interest groups (Elgeyo-Marakwet, 2018: 22).Additionally, Elgeyo-Marakwet's text related to access to information was in service to equipping residents and CSOs to hold the government accountable and improving the acquisition and distribution of medical drugs in government operated facilities (Elgeyo-Marakwet, 2018).Iasi demonstrated an in-depth understanding of successfully applying the 'access to information' tenet to a local challenge.In their second commitment, aimed at reducing waste and increasing recycling, the action plan states: "The highest information retention rate is registered when the information has practical, contextual value.Those people who remember the information the best are those who need it the most in a specific context" (Iasi, 2019: 22).
In contrast to the salience of the informational aspects of transparency, the action plans contained limited text focusing on facilitating transparency of process and decision-making.One of the few mentions was in the action plan for Scotland which committed to increasing the knowledge of the public about "who makes decisions about how public services are designed and delivered, and how they make those decisions" (Scottish Government, 2018: 17).The limited focus on opening up existing government processes is potentially addressed by the significant focus on new avenues for civic participation, but only if the avenues become embedded within sustained government practice.

Civic Participation
Broadly, the references to civic participation were in service to a tangible end goal such as addressing environmental concerns or collectively addressing economic development.Narino, a department of Colombia, demonstrated a unique approach by using civic participation to build knowledge about local coffee production (Narino, 2019).In some cases, the action plans studied fell into the trap of designing open government for a 'general user' as was discussed earlier.The Tblisi Government plan, for example, talks about access to public services, but does not specify any particular underserved communities.However, other governments detailed consultations with specific populations, or noted the demographics they saw their commitments impacting.The plan for the Municipality of Sao Paulo, for example, referenced the involvement of Periferia invisivel, an arts and culture organization that works to democratize art in the outskirts of the city.One of the commitments outlined in Seoul's plan (City of Seoul, 2018) was to develop a subway transfer map to help vulnerable people such as those with disabilities, the elderly and pregnant women.The map was made by Muui, a civil society organization.However, it is unclear whether the vulnerable populations impacted were included in its development.
Some emphasis was put on the importance of fostering civic participation and expanding stakeholder engagement.Buenos Aires (City of Buenos Aires, 2018), for example, noted the need to build capacity for effective public participation.Similarly, the City of Iasi plan (2019) identified civil society as part of public participation, demonstrating an understanding that non-governmental organizations can actively advocate on behalf of underserved communities.Going further, the Iasi plan comments specifically on the OGP Multi-Stakeholder Forum indicating that there should be © 2022 The Author People, Place and Policy (2022): 15/3, pp.149-168 Journal Compilation © 2022 PPP transparency in how its members are selected.Other plans focused explicitly on capacity and engagement within specific sectors, or around defined themes.South Cotabato (Province of South Cotabato, 2019), for example, outlined the need to further engage residents in mining governance, while Narino focused on engagement around the defence of water.
In some cases, civic participation and accountability were intimately intertwined.Elgeyo-Marakwet County, for example, points to the need to understand who is participating in engagement and how their actions might impact vulnerable persons.

Accountability
As was noted earlier, accountability did not feature significantly in the action plans examined.Secondi-Takoradi Metropolitan Assembly's plan incorporated the value of accountability most explicitly.It included references to a civil society implemented Citizen Report Card (CRC) where residents' views on government activity are captured.It also discussed the need to protect against the perception of corruption, and the need for improved access to information regarding public infrastructure.
It should be noted that Kaduna State (2019) also mentioned a Citizen Perception Survey, similar to Secondi-Takoradi Metropolitan Assembly's Citizen Report Card.These sorts of opportunities for residents to provide feedback on government action and progress is essential to accountability, particularly as it is identified and measured within OGP where a redress mechanism is fundamental.However, these opportunities need to be institutionalized in order to be effective.Once the civil society organization stopped funding the CRC in Sekondi-Takoradi, the opportunity for resident input ended.
It is impossible to capture all nuances in a short paper, but it is clear that while the prioritization of civic participation, access to information and accountability were consistent, unique contextual characteristics were discernible in how they were applied.It could be argued that the deprioritization of transparency detected in both Chatwin et al's 2019 study and this study looking at the second action plan cycle reflects a growing nuanced understanding of applying open government to local problems, rather than aspiring to a vague notion of openness.Throughout the last decade when open government has gained prominence, the tenets of transparency, civic participation and accountability have been presented as an end in themselves (Heald, 2006;Chatwin, and Arku, 2017).Through an increase in practical outcomes within codes such as 'improve public service', 'strengthen governance', 'driving economic growth' and 'pursuing environmental outcomes' the current action plans demonstrate an emerging view of open government tenets being in service to achieving such results rather than just an end in themselves.
To better understand whether this is occurring, and whether OGP Local participants are better defining the problems they face, anticipating resistance, and learning from this to reform their activities accordingly, it is useful to return to the Problem-Driven Iterative Adaptation (PDIA) lens outlined earlier in this paper.Doing this, we see examples of global policy migration that are not adapted to local context, but also examples of participating jurisdictions committing to addressing locally relevant problems.One risk associated with the global rise in popularity of open government is the transfer of homogenized policies being prescribed by external entities without adapting to local influence (Chatwin et al., 2019).From a PDIA perspective, if open government, as promoted by the OGP, is to make meaningful change through the commitments of participating jurisdictions it must avoid superficial changes to organizational structure and surface level policy change that are neither implementable given the current context nor address the systemic nature of a problem (Andrews et al., 2013;Andrews, 2013;Muamar, 2016).What follows is a brief discussion of the four © 2022 The Author People, Place and Policy (2022): 15/3, pp.149-168 Journal Compilation © 2022 PPP steps of PDIA -problem identification, engaging a broad network, enabling environment, and feedback through monitoring and evaluation -in relation to the findings from the document analysis.

Problem Identification
The OGP Local template requires that participating jurisdictions articulate the local context and challenges that demonstrate the need for each commitment.However, the degree to which residents, particularly underrepresented and vulnerable groups, are involved in articulating these problems and devising solutions is varied.Broadly, the action plans suggested that the involvement of residents was more geared towards prioritization of issues rather than identification and definition.PDIA, in alignment with literature on liberal democracy, suggests that meaningful engagement occurs when initial decisions or definitions are being made through a process of constructing and deconstructing people's experiences of the challenge (Dahl, 1989;2006;Andrews et al., 2015).
There is little concrete evidence that the lived experiences of the people impacted by the plans have been fully captured or incorporated into commitments.Diverse groups such as children, youth, women and persons with disabilities have often been treated like a monolith.This was certainly the case in the Secondi-Takoradi Metropolitan Assembly's plan and in Elgeyo-Marakwet County Government's plan.The lack of nuance and understanding demonstrated by combining diverse demographics in this manner has resulted in weak problem identification.
In other cases, there were clearer attempts to understand and include peoples' experiences, but in limited ways.The Madrid plan (City of Madrid, 2018), for example, included an opportunity for participation in an observatory, but recruitment was done via a lottery.While this perhaps guaranteed equal opportunity to participate, it ran the risk of alienating the voices of those who might be most impacted by a decision.Similarly, the Madrid government left one of its five commitments to be identified and co-created by residents of Madrid.The other four were decided by the city council with some consultation.In Paris, interviews with residents to find out the reasons for homelessness helped to develop a more nuanced understanding of why people were sleeping rough on their streets.What this demonstrates is that OGP Local participants have room for improvement when it comes to the first step of PDIA: that is, defining the specific problems that they wish to address in their action plans.As stated in the OGP Participation and Co-Creation standards, those most intimately impacted by the commitments need to be better identified and engaged in thinking through the articulation of problems and their corresponding solutions (2017: 2).

Engaging a broad network
Part of the weak problem identification identified above is potentially related to the failure of many governments to engage the sort of broad network identified through PDIA.Action plans tended to provide little specificity around the extent of engagement.As was noted above, for example, many failed to specify whether they have included vulnerable populations in their co-creation processes.When they did, it was in a general way.In Sao Paulo, there was mention of engagement of 'different people' and 'vast territory.'The Scottish plan referenced 'equalities groups' that included young people, black and minority ethnic groups and disabled person's organizations.In many cases no clear network of actors was clearly identified in the plans -this included a lack of clearly identified civil society actors, residents, or even involved government actors.
The Austin plan (City of Austin, 2018) was somewhat more comprehensive here.In response to suggestions made by the OGP's Independent Reporting Mechanism, Austin assigned a specific department and Manager to each commitment, improving the overall accountability.Similarly, the plan for South Cotabato (Province of South Cotabato, 2019) noted specific individuals dedicated to most of their commitments.This is really an exception to the norm.
In the Paris plan (City of Paris, 2018), it would seem that engaging a broad network went beyond institutional engagement (though did not exclude them).It specified training volunteers to participate (The Solidarity Factory).It also talked about a 'conference' for residents to outline opinions and to look toward future reform.
In some cases, it is difficult to know how broad a network had been engaged and whether they were truly local stakeholders.Kaduna State's plan (Kaduna State, 2019), for example, is one of the few that explicitly talked about having local organizations (even community level) indicated in the stakeholders.In contrast, Elgeyo-Marakwet County Government's plan mentioned World Vision Kenya, and the Center for Innovation in Open Governance as their civil society partners.While such external international NGOs have benefit, the focus should be on local civil society partners with intimate understanding of the context.

Enabling Environment
The third step of PDIA is to create an enabling environment for positive change and learning.The diversity of geographic and socio-political contexts within the OGP Local participants highlight the complex nature of developing and maintaining an enabling environment.For example, the Department of Narino, Columbia struggled with the governance of territories with ongoing violence, occupation by illegal groups, illicit crops and attacks on civil society leaders (Narino, 2019).Within the Department of Narino's action plan commitments were activities designed to keep social leaders and residents safe in their civic participation.In contrast, the City of Seoul (2018) acknowledged the role its institutions play in creating disincentives to participation and that as a government they were responsible for creating an enabling environment for residents to actively participate in governance (City of Seoul, 2018).Despite this awareness, there were limited activities specifically directed at improving the enabling environment.
The Municipality of Sao Paulo (2018) offers an interesting example of the tension between avoiding the potential for isomorphic mimicry and maintaining an enabling environment.To improve their co-creation Sao Paulo created several interdepartmental meetings to ensure the action plan had more sectoral, interdisciplinary and relevant commitments.The co-creation process was extensive and led to a set of preliminary commitments that were submitted for technical and legal review.Commitments included equipping local municipal counsellors with tools and information to fulfil their duties to strengthen transparency and increase public integrity.However, every commitment submitted for review was significantly altered to fit within the current authorizing environment of the municipality.When the revised commitments were voted upon by the multi-stakeholder forum, the votes were split along the lines of government and civil society.The multi-stakeholder forum contained one more government vote than civil society and so each revised commitment was passed without any civil society votes in favour (Sao Paulo, 2018).
The document analysis broadly indicates that participating jurisdictions were working within their current authorizations, but there was no discussion about actions that could further prepare an enabling environment for ambitious commitments.For example, South Cotabato's administration decided to exclude a commitment on open planning and budgeting because the environment was not conducive, and the desired reforms  2019).This demonstrates an awareness of the need for an enabling environment, but there was a noticeable absence of forward planning on how to begin to cultivate the necessary conditions to include this type of commitment in future plans.In fact, none of the plans clearly highlighted the assumptions or necessary conditions for change that is vital to PDIA.Part of the problem here is a lack of specificity in the wording of commitments and a failure to elaborate key indicators of success or failure.In the Kaduna State (2019) plan, for example, there was an 'access to information' commitment that had a legislative component and an administrative function component.However, the evaluation section did not indicate how they would evaluate the culture changes necessary for success.It is important to note that this study was limited to the text of the action plans; it is feasible that government efforts to improve their enabling environments and expand authorizations are occurring outside the plan.

Ongoing Feedback through Monitoring and Evaluation
The final step in PDIA focuses on rapid feedback through ongoing monitoring and evaluation.It is to promote agility and allow jurisdictions to capitalize on quick wins and pivot away from obvious failures.While there was limited focus on agility and adaptation within the action plans reviewed, South Cotabato's introduction most closely aligned with the PDIA approach when it stated that it was taking an iterative approach to their action plan.It described its process as a continuous cycle of "consultation-implementationevaluation-replanning/improvement" and indicated that the plan would be continually revisited during the implementation phase until it became closer to realizing the desired development results through the OGP principles (Province of South Cotabato, 2018:10).While this aligns with the PDIA perspective, it is the exception to the norm within the reviewed documents.Some participants such as Basque Country (2019) and the City of Buenos-Aires have detailed evaluation plans attached to their commitments, but there was a consistent and discernible absence of feedback collection and/or monitoring for rapid iteration opportunities.The majority of references to evaluation were outcome indicators to demonstrate that an activity was undertaken rather than qualitative assessment of change to the status quo or the experience of residents.For example, within the limited milestones documented by the City of Madrid, the deliverable within commitment one was the creation of the digital mailbox and the deliverable for commitment two was "development of the project" (City of Madrid, 2018: 3-4).The City of Buenos-Aires, recognizing the value of monitoring and "collaborative assessment", built a milestone into its commitment on comprehensive sexual education that outlined a desire to use evaluation to strengthen its implementation (City of Buenos-Aires, 2018: 67).While potentially well-intentioned, Kaduna State (2019) established evaluation to 'curtail' the Thematic Working Groups from deviating from the established work plan.This approach to evaluation is the antithesis of PDIA, indicating the lack of a sufficiently enabling environment and stifling iteration and learning.The overall lack of focus on both early iteration and outcomes was potentially influenced by the OGP strategy to primarily monitor for task completion.

Conclusion
In closing, it is useful to return to the research question guiding this paper: Are the action plan commitments of Open Government Partnership Local Program participants designed to address locally identified problems in a way that is responsive to their unique technological, organizational and contextual environments?Ultimately, the document analysis has demonstrated that there are commonalities in the conceptualization and application of open government.It has also offered a window into how local context (geographic, social, political, economic, technological) and institutional environment has influenced the nature, scope, and characteristics of implementation.
It is recognized that the document analysis approach used here is only able to scratch at the surface of this question.A more nuanced answer to the question would necessitate a broader review of documents and in-depth stakeholder interviews to get a richer understanding of the environment in which OGP Local Participants operate.Further, the documents analysed here are limited in themselves.As has been noted throughout the paper, OGP Local Participants were provided with a broad template to follow when creating their action plans.This has influenced both content and potentially space dedicated to demonstrating co-creation, commitment design, implementation and evaluation.Stakeholder interviews would allow for a better idea of any restrictions that might have been felt as a result of the template, or other guidance from OGP.
In spite of these limitations, the document analysis and the PDIA lens have shown evidence that local governments have room for improvement when it comes to thinking through the design, implementation and evaluation of their action plans.Two particular areas for improvement stand-out: 1) problems could be better articulated so that they clearly relate to a specific need that has been identified through consultation or collaboration with those most directed impacted; and 2) commitments could be designed in a more agile way so that activities can adapt to ongoing feedback provided through monitoring and evaluation.
Additionally, there is room for improving co-creation processes in general, or at least in the way that they are articulated.When identifying stakeholders, it is important to identify their area of expertise, to explain how they are best positioned to contribute to the success of the commitment, and how they played a role in the definition of the problem.This sort of attention to detail helps to ensure that residents and civil society are not treated as a monolithic entity of 'general users'.As such it will improve problem definition and the identification of potentially more successful reforms that are generated through a clear understanding of lived experience.In essence, it will help to ensure that future local plans are in fact reflective of the local populations they impact.*Correspondence address: Merlin Chatwin, Department of Geography, The University of Western Ontario, 1151 Richmond Street, London, Ontario, N6A 5C2 Canada.Email: Mchatwin@uwo.caNotes

Table 1 :
OGP Local Participants Included in Study not be implemented within the action plan cycle (Province of South Cotabato, could