Regulation, the hybrid market, and species conservation: The case of conservation banking in California

Conservation Banking in California is a long-established offset program. Banks are hybrid instruments that hover between market autonomy and regulatory oversight. Challenges that may affect outcomes of the program include aligning regulation with the scales and objectives of the hybrid market and conservation and interaction with other compensation instruments. I use an analytical framework combining social-ecological fit (does the regulation fit the spatial, functional, and temporal scales of the market or conservation?) and instrument interaction (are compensation instruments redundant, synergetic, etc.?) to analyze the institutional framework of the conservation banking program. Results show that the program fails to reflect the hybrid market or species conservation objectives, creating a social-ecological mismatch. The institutional framework disincentivizes banking, while its contribution in conserving species cannot be measured. Competing and redundant instruments can lead to weaker compensation. The program needs equal standards that reflect conservation objectives for all compensation instruments. Findings on fit can be useful for other banking programs, and considerations on instrument interaction could improve offsets anywhere.


INTRODUCTION
For decades, governments have made an effort to reduce global biodiversity loss. These efforts include the prohibition of specific impacts and the development of offset programs (Hrabanski 2015). Offsets are 'measurable conservation outcomes' implemented to compensate for unavoidable adverse impacts of development projects to at least maintain the current status of biodiversity (BBOP 2012). They exist in over 100 countries (Maron et al. 2016(Maron et al. , 2018Bull et al. 2020). However, questions concerning the proper implementation of offsets and their success in conserving nature remain (Yu et al. 2022).
The California conservation banking program was one of the first offset programs for impacts on species and habitats. A conservation bank conserves and manages protected species habitat for which they receive credits from the permitting agencies. A project developer who needs to compensate for unavoidable impacts to such species can then purchase these credits. Banks are strictly regulated, and many stakeholders agree that the program holds ecological and economic benefits (Grimm 2020). Researchers and agencies have studied program design or the ecological contribution of individual banks (Bunn et al. 2014;Carreras Gamarra and Toombs 2017;Sonter et al. 2019;Grimm 2021), but due to limited tracking and data on species ecology, none have evaluated its actual success in conserving species. However, scholars agree that the institutional setting, including legislation and market arrangements, affects offsets' effectiveness (Boisvert 2015;Gelcich et al. 2016;Sonter et al. 2020b;White et al. 2021).
Some major challenges of conservation banking include a lack of standards and competition with other marketbased compensation instruments (White et al. 2021). This raises the ''problem of governance' ' (p. 362), which is Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s13280-022-01803-2. rooted in the difficulty of fitting scales of ecological processes with scales of market activity and regulation (Robertson 2004). Van Teeffelen et al. (2014) analyze economic and ecological prerequisites for conservation banking and show that the two differ in terms of objectives, requiring trade-offs. The challenge lies in bridging these spheres to allow for trading ''environmental intangibles'' (Barral 2020;Chiapello and Engels 2021) in a market restricted by regulation. Previous research has discussed this hybridity of banking instruments (Vaissière and Levrel 2015;Rea 2017) and the mismatch of economic and ecological objectives (Calvet et al. 2015b). I analyze the interaction of the regulatory, hybrid market and ecological spheres from a social-ecological systems perspective, including the relation of banks to other compensation instruments.
Conservation banking takes place in a social-ecological system (SES) (Mann and Absher 2014;White et al. 2021). An SES perspective allows for a systemic view of the regulatory, ecological, and market sphere. Previous research has also focused on matching institutional settings to the systems they organize (Folke et al. 2007;Mann and Absher 2014). Because most research combining SES and offsets has focused on stakeholders and their values, Gelcich et al. (2016) advocate analyzing the institutional settings for offset programs to better understand institutional differences and enabling conditions. Banking takes place in a hybrid regulatory market, where credit demand is created by the developer's responsibility to compensate for unavoidable impacts. This article aims to fill the research gap on such hybrid governance instruments identified by Baker et al. (2020) and adds to a better understanding of institutional settings in SES (Guerrero and Wilson 2017).

State of the art and research gap
This research contributes to existing research by analyzing the institutional framework from an SES perspective (the regulatory framework, species conservation and hybrid market objectives) to provide a systemic view of the conservation program in California and draw conclusions about its future. Conservation banking in California is a long-established program, and an analysis of its institutional framework can provide insight into the challenges and practicalities for banking approaches developed elsewhere (e.g., the Netherlands or Spain, cf. Burgin 2008; Gorissen et al. 2020;Maestre-Andrés et al. 2020).
The performance of conservation banks can be discussed in economic and ecological terms (Bonnie 1999;Cline 2013), which is reflected in previous literature: Research on conservation banking has often applied either an economic (Poudel 2017;Higashida et al. 2019) or ecological perspective (cf. Bunn et al. 2014). However, both must be connected because the regulatory framework should reflect conservation and market objectives. Although many publication on conservation banking exist (cf. Ruhl et al. 2005;Fleischer and Fox 2009;Bunn et al. 2013Bunn et al. , 2014Boisvert 2015;Grimm 2020Grimm , 2021, few analyze the ecological outcomes (Bunn et al. 2014;Sonter et al. 2019). Considering the regulatory framework and banking guidance, Carreras Gamarra and Toombs (2017) or McKenney and Kiesecker (2010) provide overviews, analyzing whether the program meets key principles of offsetting (e.g., providing additional outcomes, advance mitigation, etc.). Although the importance of the institutional framework for the outcomes of offsets is prevalent in literature (Santos et al. 2015;Gelcich et al. 2016), many researchers focus on specific aspects of the program, such as crediting metrics (Bezombes et al. 2017) or ecological equivalence of losses and gains (Boisvert 2015).
However, very few scholars have analyzed where identified challenges in the institutional framework stem from while including ecological and economic perspectives. Some have discussed trade-offs between the two spheres (Wissel and Wätzold 2010;Mann and Absher 2014;van Teeffelen et al. 2014;White et al. 2021), but have not analyzed the origin of these trade-offs or explored which spatial, functional or temporal scales are reflected in the current institutional setting.
This research contributes to closing this gap by analyzing the current regulatory framework in California to the objectives of the conservation of species and the hybrid credit market, addressing specific functional, spatial and temporal considerations. Findings from a mature and strictly regulated offset program like California conservation banking can be relevant for the design and implementation of conservation banking across the USA and banking programs elsewhere (see Sect. 5.2). Agencies and scholars have voiced the need for more research on market dynamics, factors affecting the demand and supply of conservation banking credits (Cline 2013;DOI 2013), institutional challenges, and bureaucratic constraints (Fox and Nino-Murcia 2005;Cline 2013). There is no previous research that explores how different compensation instruments interact withing this institutional setting. This research thus explores the institutional challenges of bureaucratic requirements for banking and California as well as interactions with other compensation instruments under ESA.

California conservation banking
Conservation banking originated in California in the early 1990s, and most banks (126 out of 194) are located in the state (RIBITS 2021). Conservation banks were modeled after wetland mitigation banks used to implement compensation requirements for impacts on wetland under the US Clean Water Act (CWA). Conservation banking is rooted in state and federal legislation: The California Endangered Species Act (CESA) and the US Endangered Species Act (ESA). The ESA prohibits the ''take 1 '' of threatened and endangered species. When a project impacts such species or their habitat, the developer may need to compensate for unavoidable impacts by conserving, enhancing or restoring habitat elsewhere. Developers may use different instruments to fulfill this compensation requirement: The developer can implement compensation themselves (permittee-responsible mitigation) or go through a third party. Third-party compensation includes conservation banks, in-lieu fees, habitat credit exchanges, turnkey mitigation projects, paying a fee to a regional habitat conservation plan/ natural community conservation plan administration, or going through a mitigation credit agreement. Table 1 lists these instruments, explains how they work, whether they are permitted under federal or state jurisdiction and whether compensation projects are implemented in advance of impacts, consolidate multiple compensation requirements and whether they can be strategically sited. Different compensation options may be available to the developer depending on the location of the impact and the impacted species. For example, regional habitat conservation plans do not exist for every county, banks tend to be located in more densely developed areas, and permittee-responsible or turnkey mitigation are implemented on a project-by-project basis. Also, some instruments can only be used to compensate for impacts on state listed species and not federally listed species. Still, some regions offer multiple compensation options for developers and the selected choice must be approved by the permitting agency.
A conservation bank is a site where a third party conserves and manages protected species habitat for which they receive credits from the permitting agencies. By providing compensation measures, banking is intended to contribute to species conservation. By definition, a bank is established before the impact occurs, and the site consolidates multiple compensation requirements. Credits can only be sold in-kind (for the same species or habitat) and for impacts taking place within a designated service area (impacts for which mitigation credits are purchased must be located in this geographic area). This limits the distance from the impact to the compensation site and provides a functional relation. When bank sites include wetlands, they can also produce wetland credits to compensate for impacts under the Federal Clean Water Act. In that case, the US Army Corps of Engineers is involved in the approval process. To streamline the approval process for such multicredit banks, the agencies signed a Memorandum of Understanding (MOU) in 2011, which provides the basis of bank approval and oversight in California (CNRA, CDFW, et al. 2011).
Policy instruments are often categorized as either legal, voluntary or economic (Blackstock et al. 2021). However, instruments can also be hybrids, combining state and nonstate actors (Vatn 2015(Vatn , 2018Ghosh and Wolf 2021). Banking approaches are not market-based instruments, but they provide compensation in a regulation-based hybrid regulatory markets (Boisvert 2015;Vaissière and Levrel 2015;Koh et al. 2019) or ''command-and-commodify'' instruments (Rea 2017): Demand and supply only exist because legislation requires developers to compensate and regulatory oversee both supply and demand. The bank implements conservation measures, which are commodified and traded within this market. Thus, the institutional framework has developed from regulating impacts to also regulating a hybrid market with different compensation instruments and suppliers, requiring a balancing act between ecological and market spheres. Thus, hybrid market is created by the regulatory framework, and banks within this market must be feasible for the hybrid market to function. The rules by which banks can be established and trade credits make or break this functionality, which is why this research explores the institutional framework in more detail.

Aim and methods
This article analyzes the institutional setting of conservation banking in California and draws conclusions regarding the future of the program (i.e., conservation banking in California) and associated outcomes for conservation and the hybrid market. I examine a regulation torn between serving a hybrid market and species conservation as well as interactions between different compensation instruments. The questions to be answered are: • Do conservation banking regulations and standards reflect the scales (spatial, temporal, functional) and objectives of the hybrid market or species conservation? • How do different compensation instruments interact?
What is the current level of standardization? • What does that mean for the outcomes and future of the program?
I show which functional, spatial and temporal aspects of the regulatory framework reflect the objectives and scales of the hybrid market and which aspects reflect the objectives of species conservation. I also explore how the institutional framework manages a multitude of compensation instruments. In the end, I discuss the likely effects of the current regulatory setting in California and contextualize the results for conservation banking on a federal level and offsets beyond the US.
This research draws on environmental governance literature in the context of SES to provide a conceptual understanding of institutional settings in SES and establish a framework for the analysis of the conservation banking program. I conducted a configurative review (Gough et al. 2012) of journal articles published 2000-2021 in Scopus (searching for ''social-ecological systems'' in combination with ''offset'', ''conservation banking'', ''banking'', ''environmental governance'', ''market-based''), followed by a review of titles and abstracts and through snowball sampling. I then selected the theoretical framework outlined in Sect. 3 (instrument interaction, social-ecological fit) as a framework for the analysis of the conservation banking program.
To apply this framework, I use information on conservation banking gathered through 30 semi-structured qualitative expert interviews conducted in California in 2019. These interviews enabled collecting empirical data from the field that is not included in publicly available documents, ensure a discussion of predefined topics and allow new content to emerge (Helfferich 2014;Brinkmann 2018). Stakeholders can also help understand policy design and outcomes (Mann and Absher 2014). The interviewed experts were agency staff (n = 12), bank sponsors (n = 11), consultants (n = 10), and others (attorneys, NGOs, permittees, researchers, n = 4). The interviews were coded deductively and iteratively (Mayring 2014). Using codes and subcodes reflecting the analytical framework outlined in Sect. 3 (e.g., spatial fit -service area; interaction with Regional Habitat Conservation Plans; etc.), I analyzed interview files, highlighting statements pertaining to these specific aspects. Some information was gathered through targeted questions (i.e., ''How are bank sites selected?''), whereas other aspects came up during more open questions In-Lieu fee A program (often run by public actors) collecting funds from developers to conserve and manage habitat sites for mitigation. Credits are approved by agencies and sold to the developer, but conservation is implemented incrementally based on sales Federal -x (x) Regional habitat conservation plan Regional conservation strategy planning for development and compensation implemented by one or multiple agencies. Some implement compensation by collecting a fee (in-lieu fee program) Natural community conservation plan As above, but under state jurisdiction State (x) x x

Regional conservation investment strategy
Similar to a regional habitat conservation plan or natural community conservation plan, but non-binding. Conservation actions may provide advance mitigation through the implementation of a mitigation credit agreement (see below) State Not applicable, see below.

Mitigation credit agreement
Conservation or enhancement actions identified in a regional conservation investment strategy implemented to generate credits for mitigation State x ? x (''Where do you see issues with the banking program?'' or ''Where do you see room for improvement?''). The information collected in these interviews is indicated throughout this article by naming the stakeholder type and interview number (e.g., ''Bank 17'' or ''CDFW 11''; see Appendix SI, stakeholder names and affiliations were excluded to maintain anonymity of interviewees). Indicating stakeholder types and number allows the reader to identify what stakeholder type the information was obtained from. Each interviewee reflected their own personal point of view. Information collected in interviews may be subject to social desirability (Sovacool et al. 2018) and affected by the informants' experience (Roulston and Choi 2018). However, according to Helfferich (2014), information collected in expert interviews can be removed from personal experience. The number of informants, the different stakeholder groups and high level of agreement among them suggest internal validity of the collected data (Krippendorff 2004;Bengtsson 2016). In addition, supplementary data was gathered from literature on conservation banking and from policy and guidance documents to confirm the empirical data collected through interviews and to include additional information. This combination of naturally occurring data (i.e. policy documents, cf. Salheiser 2014; Peräkylä and Ruusuvuori 2018) and interviews allows for analyzing legal requirements and implementation in practice (Paavola et al. 2009). Existing research on offsets and banking was also used to contextualize the results in the discussion section and to make inferences about the outcomes for species conservation and market feasibility, and to outline relevance for offsets elsewhere.

THEORETICAL FRAMEWORK FOR ANALYSIS
SES are integrated systems of humans and the environment with interdependent linkages (Berkes 2017;Partelow 2018). This section provides a conceptual understanding of institutional settings in SES and a framework for the analysis of conservation banking. The scope of the analysis, i.e. the SES, here refers to the ecological sphere (the objective of species recovery and the ecological processes affecting recovery) and the hybrid compensation market sphere (a functioning credit market reliant on bank feasibility and possibly affected other compensation instruments) as well as the institutional framework governing these spheres.

Social-ecological fit
Social-ecological fit refers to the idea that effective institutions 'match' the characteristics of the SES (Young 2002;Folke et al. 2007). Institutions are ,,sets of rules, decisionmaking procedures, and programs that define social practices, assign roles to the participants in these practices, and guide interactions among the occupants of individual roles'' (Young 2002, p. 5). Problems of fit occur when institutions fail to adequately address components or links of the ecological systems they govern (e.g., stressors affecting species survival) (Ekstrom and Young 2009). Fit can refer to political, economic and ecological dimensions (Guerrero et al. 2021). Fit is discussed often in the context of social-ecological systems (Folke et al. 2007;Ekstrom and Young 2009;Epstein et al. 2015) and conservation approaches (Guerrero and Wilson 2017;Guerrero et al. 2021).
The fit of hybrid instruments with regulations, ecological systems and relevant objectives (such as species conservation) is crucial for designing effective institutions in SES (Cash et al. 2006). Mann and Absher (2014) discuss fit and interplay when analyzing stakeholder negotiations during the establishment of the conservation banking program. I build on their work, using three dimensions of fit for a qualitative analysis of how the regulatory sphere of conservation banking aligns with the scales and objectives of the hybrid market and species conservation:

Instrument interaction
Institutions function as rules of interactions in SES (Werdiningtyas et al. 2020), and the interactions of institutions with other institutions can affect the effectiveness of environmental governance of SES (Young 2002). Policy instruments (such as banks) are a form of institutions (Corbera and Brown 2008;Mann and Absher 2014). Instruments interact with other instruments and abide by the context conditions of the institutional framework (Werdiningtyas et al. 2020). Understanding how instruments interact can help align these instruments with their objectives and may lead to effective policy solutions in SES (IPBES 2019). To analyze the relationship between conservation banks and other compensatory mitigation instruments, I use the five categories provided by Santos et al. (2015) ( Table 2).

RESULTS
This section explores compromises and mismatches between the regulatory framework and the scales and objectives of the market and species conservation. It discusses the standardization and interaction of different compensation instruments. Figure 1 shows the structure of the program, including federal and state legislation, actors, supply and demand.

Regulation, the hybrid market, and species conservation
Conservation banking regulation must combine scales and objectives of the hybrid market and species conservation: ''Banking provides an opportunity to soften impact of command and control while still protecting listed species'' (FWS 5). As hybrid instruments, banks hover between a certain degree of market autonomy (e.g., market-influences credit pricing) and regulatory oversight (e.g., requirements for bank establishment). They compensate for impacts on species and habitat and thus contribute to conservation. The bank approval process in California is a template-based negotiation (Grimm 2020), where stakeholders must balance the implementation of regulatory requirements with ecological and market objectives. This section analyzes temporal, spatial and functional fit of the hybrid market, species conservation and the regulatory framework. I argue that the regulatory framework and existing standards of the  1 The California conservation banking program: Institutional framework. The left shows the legislation and regulatory agencies, the top right shows the impact permitting process, and the bottom right the bank approval process and other compensation instruments conservation banking program in California fail to fit the hybrid market or species conservation, creating a socialecological mismatch.

Spatial considerations
Conservation bankers in the USA rely on sufficient credit demand (unavoidable impacts caused by development) for profit (van Teeffelen et al. 2014). Demand is determined by species occurrence and impacts located within the service area of the bank. This potential demand is considered by bankers when selecting a site, which in turn affects how well bank site and service area fit the spatial dynamics of species conservation. The bank site is selected by bankers and approved by the agency (FWS 5, Cons. 21, CDFW 10). According to different stakeholders, they should consider area size, connectivity (Cons. 23,CDFW 12,Bank 15,17,18) and existing plans, such as species recovery plans (describe the status, threats and objectives for protected species) (FWS 3,4,CDFW 10,12,Bank 17,18). In practice, site selection is based on expected credit demand, suitability for species habitat or occupancy (Bank 17, FWS 4, Cons. 21) and adjacent land use (CDFW 12, Bank 17). Although guidance includes other habitat and landscape values (CDFW 2019a), agencies stated that they often accept any suitable habitat (FWS 5, CDFW 11). According to agency and consulting stakeholders, bankers select regions where they expect credit demand (FWS 3, Cons. 20), which across the USA occurs in large urban areas (Poudel et al. 2019). Therefore, some suitable sites may not become banks, because of low feasibility (Cons. 20, CDFW 11). Selected regions are often densely developed, making land availability and cost critical for site selection for banks in and outside California (Poudel et al. 2018;Poudel and Pokharel 2021).
At the other extreme are areas where no banks exist due to little demand (e.g., Northern California) (CDFW 11, Cons. 21). In such regions, it may be more difficult for developers to mitigate (CDFW 9): They must use other compensation instruments that may not be implemented in advance or may not consolidate multiple compensation requirements in a large area, providing potentially weaker compensation. Also, many species aren't covered by banks (e.g., San Joaquin Antelope Squirrel, ammospermophilus nelsoni), and permittees cannot find suitable sites for permittee-responsible mitigation (CDFW 11). This creates a potential functional mismatch between the ecological objectives (by providing the best compensation options) and hybrid market dynamics.
Concerning the service area, ecological and economic interests differ. Agency representatives state that bankers prefer a large service area, and agencies prefer a smaller one to limit the geographic distance between impact and compensation site (Attorney 14, CDFW 10, FWS 4, 5). Still, the negotiations between bankers and agencies lead to a compromise between incentivizing markets and reflecting species conservation (leaning more towards the latter): The service area is usually based on species recovery plans (CDFW 11,12,Attorney 14,Bank 17,Cons. 21). They reflect the objective of species recovery (CDFW 12, Cons. 28) and allow bank feasibility (CDFW 11, 12, Attorney 14, Bank 17, FWS 5). However, if there is no other option, a developer outside of a service area may also purchase bank credits (CDFW 12, Bank 17, Cons. 28, FWS 3, County 2). Such cases do not occur often and they increase demand for bankers but may fail to spatially and functionally fit the ecological objectives. Bank feasibility and availability are also affected by bank approval and crediting requirements, which I outline in the following section.

Temporal and functional considerations
Many stakeholders suggested that the requirements for bank approval need improvement, because it has become expensive and difficult (Bank 17,18,Attorney 14,Cons. 20,24,FWS 5). Consultants and one agency stakeholder observe that agencies increased procedural requirements to reduce the risk of bank failure by holding banks to higher standards, disincentivizing habitat protection through banks (FWS 3,Cons. 20,22,24,26). The approval process takes 1.5-2 years on average (CDFW 11, Bank 16, Attorney 14) and can be affected by jurisdictional conflicts (Cons. 20, 25, County 2). Progress was made with the MOU between agencies (CDFW 10, 11, Cons. 21), but different stakeholder agree that preferences concerning details of bank approval vary between agencies, offices and individuals (Cons. 20,24,27,Bank 15,16,18,USACE 6,CDFW 12). Bankers provide an endowment fund (700 000-1.5 million USD, funded incrementally; Bank 16,17, Cons. 21, 28), pay fees, purchase property, put an easement in place (CDFW 11, 12) and standards require long-term management and monitoring (CDFW 2019b; CNRA, CDFW, USACE, FWS 2021). Bankers may also need to make yearly payments to investors (Bank 16) as they can be funded by capital investments (Cons. 28). Two agency and bank stakeholders estimate an initial investment of at least 500 000 USD (CDFW 11, Bank 15). Thus, requirements focus mostly on financial security, not on ensuring functional fit with conservation objectives.
Agencies and consultants state that the program thus moves away from the premise of banking to target ,,land rich, cash poor '' individuals to conserve species and moves towards banking businesses (CDFW 11,FWS 5,Cons. 21,22). However, cash-poor landowners can partner with companies to establish a bank, providing the land while the company absorbs most of the financial risk and profits are split accordingly (Cons. 28, Bank 17). Still, bank feasibility is tied to how fast and at what price credits are sold. Sales and prices are based on demand and initial costs (Bank 17,Cons. 21,25). Banks try to produce wetland and species credits to increase potential credit sales, but involving more agencies may prolong the approval process and thus increase initial costs (FWS 4). Return percentages across the USA can vary wildly (1.5-10%, Bank 16; 4% according to Poudel and Pokharel 2021). Poudel and Pokharel (2021) found that although the annual average operating cost for a bank in the USA ($42.78/acre) is much lower than the revenue from credit sales ($6014.72/acre), only 14 out of 26 banks across the USA have produced a positive return. According to two banking companies in California (bank 16, 18), it takes about three years to make a profit. A bank site must cover at least 50 acres to make the approval process worth it, disincentivizing the protection of smaller areas (Cons. 27). Here, synergies between the market and the ecological objectives can be observed in terms of area size: According to van Teeffelen et al. (2014), both bankers and species in the USA benefit from larger sites.
The credits and methods that are used to define the units bought and sold for banks across the USA ''allow the assessment of habitat recovery [and] they are the cornerstone of conservation banking profitability'' (Barral 2019). Crediting should therefore fit different temporal scales and functional linkages. However, the conservation banking program is struggling with some mismatches between crediting and conservation outcomes: • Whereas the land is protected and the easement is signed in advance of the impacts, credits can be sold without ensuring an additional conservation outcome besides habitat protection (Cons. 21, 28, Bank 17, Bank/Perm. 19). Generally, banks still provide advance mitigation (Attorney 14, CDFW 11, Cons. 20) because there is no time lag between the impact and the site protection. However, credit release is usually based on financial securities rather than outcome-based criteria (factors improving species conservation status). Therefore, a credit does not reflect a particular conservation benefit. Still, in-kind credits and additional approval of the purchase by the agencies (CDFW 11, Cons. 21, CDFW 10, 11, FWS 5) ensure a limited functional link. • Crediting in California is based on habitat acreage, not ecological quality (Grimm 2021). An explicit goal of no net loss or a net gain referring to species recovery and reflecting that goal in the crediting could improve functional fit because a threat to recovery might be something other than habitat loss (FWS 5, Cons. 28). Although interviewees agree that banks provide higher quality habitat than the impacted area (Grimm 2020), the program cannot ensure or measure additional conservation outcomes from banking. Consultants and state agencies criticize poor tracking and transparency and no functional methods for assessing losses and gains (CDFW 10, 12, Cons. 20, 28), which is why crediting does not match the ecological objectives. Banks provide a baseline and conduct annual reports and long-term monitoring (Bank 17,18, Bank/Perm. 19, Cons. 21, 28, FWS 3,), which could be used for crediting purposes to align measurement of losses and gains with regulatory conservation objectives. Figure 2 summarizes the regulatory framework's temporal, spatial, and functional fit with the hybrid market and species conservation discussed above. The regulatory framework aims to achieve the ecological goal of species recovery and a hybrid banking market has been introduced to help reach this goal. This market depends on credit demand, which is created and affected by regulation and guidance. The figure shows that functional, spatial and temporal consideration within regulation and guidance (middle column) is located in between the ideal scales and objectives of species recovery (left column) and the ideal preferences of the banking market (right column). It also shows, that compromises have been reached. For example, the service area allows bank feasibility but is mostly based on ecological considerations. Other aspects of the regulatory framework, such as long approval processes, do not benefit the market or species recovery.
A consultant states that regulating the approval process is easier than focusing on conservation outcomes (Cons. 27), but the requirements disincentivize bank establishment without enforcing ecological objectives and create a socialecological mismatch. They also have led to mitigation companies providing potentially weaker compensation projects. This interplay between banks and other compensation instruments is explored in the next section.

Instrument interaction and standardization
The relevant regulations do not only rely on banks to compensate for impacts but also on other types of compensation instruments (in-lieu fees, etc.). These instruments function on different jurisdictional scales (e.g., federal and state regulations, county plans and project-by-project compensation). This section outlines differences and similarities among compensation instruments and discusses their relation to banks (Table 3). Due to redundant and competing instruments, there is a problem with interplay between different compensation instruments applied under ESA/CESA.

Banks and regional conservation plans
ESA/CESA allow the establishment of regional habitat conservation plans (federal jurisdiction) and natural community conservation plans (state jurisdiction). These plans provide a regional strategic framework for impacts and the protection of species and habitat through mitigation: They anticipate certain impacts for multiple decades and receive an umbrella impact permit for such, which is tied to the implementation of a conservation strategy as mitigation (FWS 2014;CDFW and FWS 2015). Banks and regional habitat conservation plans/ natural community conservation plans were intended to be complementary (CDFW 2014): Banking was created in California as a tool for implementing the associated compensation measures and the plan as a framework for directing mitigation efforts (CDFW 10, Attorney 14). Stakeholders interviewed by Mann and Absher (2014) state that the plans would ensure that banks are located in ecologically valuable areas and ensure the achievement of ecological objectives through automatic social-ecological fit). For example, Sacramento and Placer County habitat conservation plans have www.kva.se/en arrangements with banking companies to implement their mitigation strategy, creating credit demand (Attorney 14, FWS 5). Banks can also work within/around regional habitat conservation plans by providing mitigation for actions not covered by the plan (Cons. 21, 24). However, in other regions, spatial, temporal, and functional misfit hinders a synergetic relationship between banks and regional habitat conservation plans/ natural community conservation plans. Here I explore issues of spatial, temporal and functional fit that hinder a synergetic or complementary relationship between banks and strategic plans: • Strategic plans don't exist in all regions, and bank location depends on demand and ecological suitability. Second, spatial boundaries do not align: The service area does not 100% overlap with the area of an regional habitat conservation plan, which is limited by administrative boundaries. Bankers and consultants state that the entire county is often smaller than the service area (Bank 17, Cons. 21, 28). It may not align with recovery areas outlined in species recovery plans, disturbing spatial/functional fit. • Including banks outside the administrative region could interfere with the conservation strategy of a strategic plan (Cons. 24) because they include impact permits and mitigation for certain actions. • The temporal scales may not align: Strategic plans take years to approve and anticipate impacts and conservation for decades, and (according to a consultant and an agency representative) banks work with shorter time frames for quicker returns (Cons. 24, CDFW 11).
Therefore, many existing plans now implement their own compensation measures for a fee from the developer (HCP 8, Attorney 14). Some regional habitat conservation plans/ natural community conservation plans have absorbed banks in their plan area and do not allow new banks (Cons. 28, HCP 8), because ''there is potential competition between habitat conservation plans and banks'' (Academia 1). Often, when a regional habitat conservation plan exists, developers must go through it (HCP 7) to get a permit for certain actions. Plan representatives and consultants highlight that a regional habitat conservation plan makes mitigation easier for the developer compared with getting impact permits on their own and then using a bank because the plan fee includes the impact permit and mitigation for multiple legislations (Cons. 20,24,HCP 7,8). Because the plans are run by government agencies (providing an indirect form of subsidy), some stakeholders noted that they can provide mitigation at a lower cost (Attorney 14, Cons. 20, CDFW 13). This aspect of competition is further explored in the following sub-section.

Interaction with other compensation instruments
Outside of such regional plans, banks potentially compete with permittee-responsible mitigation and in-lieu fees (mainly applied for impacts under the Clean Water Act) and habitat credit exchanges (mostly pilot projects). Although all instruments provide compensation, no equal standards exist.
There are multiple reasons for the potentially unfair competition that leads to decreased bank credit demand. One is the differences in approval processes. For example, doing advance permittee-responsible mitigation is a ''less intense process'' than bank approval according to some stakeholders (Perm./Agency 30, Attorney 14, Cons. 20), and thus companies rather implement larger permittee-responsible mitigation projects for developers (turnkey mitigation) (Cons. 28, Bank 18). Turnkey mitigation does not consolidate multiple compensation requirements and requires less initial investment (Bank 15, CDFW 12, Attorney 14). It can also provide advance mitigation for large-scale impacts (Cons. 24) but is often developed parallel to the impact permit (Cons. 28). The cost depends on the project's risks and whether the turnkey provider takes over long-term responsibility (Cons. 24,28). Other mitigation options can also be cheaper than bank credits: in-lieu fee credit prices can be 1/2 or even 1/3 of bank credits, according to a bank sponsor (Bank 16).
Rare competitors are habitat credit exchanges under ESA. These hybrid markets also exist outside of California and multiple landowners generate credits and a central clearinghouse organizes this diffuse (sometimes temporary) credit production and the sales (Galik et al. 2017;Davis et al. 2021). Habitat credit exchanges are mainly in pilot stages at this point and may exist in areas with no mandatory mitigation policy (ibid.). In recent years, CDFW has also established a new program under CESA: Regional conservation investment strategies provide a non-binding version of a regional habitat conservation plans, and six have been approved (CDFW 2018(CDFW , 2019c(CDFW , 2021b. The plans are mostly proposed by public agencies (Attorney 14) and don't include a take permit (CDFW 13). Mitigation Credit Agreements implement compensatory mitigation within these plans (Attorney 14, CDFW 13). ''Mitigation credit agreements, in essence, compete with banks'' (Bank 15) but follow different standards than conservation banks: According to an agency representative, ''[it] is easier and cheaper than banking'' (CDFW 13). Both bank and mitigation credit agreement approval require the payment of a fee with CDFW (CDFW 2021a), but mitigation credit agreements can be established on private or public lands (CDFW 13) and banks only on private land. Mitigation credit agreement requirements should be similar to bank requirements to level the playing field (Cons. 23).
Many compensation instruments are redundant, but lower standards lead to unequal competition and potentially weaker compensation (Table 3). This causes them to conflict with banks and potentially lowers program performance due to weaker compensation. This competition combined with high investment costs can lead to fewer banks being established, further reducing compensation performance. Stakeholders find that permittee-responsible mitigation is the least efficient, followed by in-lieu fee programs (Attorney 14, Cons. 20, FWS 5). Therefore, banking (FWS 3, 4, CDFW 10, 12, Cons 28) or regional habitat conservation plans / natural community conservation plans (Bank/Perm. 19, Cons. 24) should be the preferred options. However, some voiced concern that such a preference may also blind regulators for great permitteeresponsible mitigation options (USACE 6,Cons. 25). Uniform standards for all compensation instruments would create a level playing field (Cons. 20) with fairly competing instruments.

DISCUSSION
This section discusses the potential outcomes of the current institutional framework in California for conservation and the hybrid market based on the results of this analysis. I use findings from other research to contextualize this research and discuss implications for conservation banking in California and for banking approaches and offsets elsewhere.
Implications for the outcome and future of conservation banking

Social-ecological (mis)fit?
Results show that the regulatory framework and standards for conservation banking in California hover between the objectives of the market and species conservation. Banking is a hybrid instrument because ecological processes cannot be fully commodified for the economic sphere, and the market is restricted by regulation. Credit demand strongly affects bank feasibility, and the implementation of the banking program affects species recovery. Market and ecological objectives are affected by • spatial (service area, market size, bank site and size), • temporal (advance mitigation, duration of approval processes, perpetual management), and • functional (crediting, equivalence) considerations.
Results show the complexity of the institutional arrangements of conservation banking, and highlight that there is not a single correct characterization of scale to be targeted by governance instruments (Cash et al. 2006;Folke et al. 2007). For some aspects, market and ecological objectives favor different levels on a temporal and spatial scale and will always lead to trade-offs. For example, the more ecologically complex crediting methods are, the lower the tradability in a market (Calvet et al. 2015). Such trade-offs can also affect public accountability (Barral 2019) and program oversight for banking in California and in other states.
The results show that regulators aim to solve such tradeoffs by hovering between command and control nature conservation (strictly prohibiting certain actions) and a market. In some cases, they manage a compromise (e.g., service area based on recovery plans), while other requirements are closer to reflecting ecological objectives (in-kind credits) or market objectives (credits based on acreage) and some benefit neither (e.g., lengthy approval processes). The technical challenges of offsets (e.g., crediting and equivalence) (Calvet et al. 2015;Chiapello and Engels 2021) reflect the challenge of aligning economic and ecological spheres and are symptoms of problems of fit. For example, the only factor affecting species recovery considered in the program is habitat loss. However, an institutional framework in an SES should encompass all essential ecosystem functions and properties (Ekstrom and Young 2009;Paavola et al. 2009), and other factors do affect species conservation. Offsets should apply at least a functional (accounting for functions of ecosystems and species) or even a systemic accounting approach (accounting for biodiversity dynamics, processes and interactions) (Calvet et al. 2015). Current standards focus on procedural and financial aspects, which does not benefit the banking market nor measure banks' contribution to species conservation. Therefore, the banking program should reduce procedural/financial safeguards (FWS 3,Cons. 26) and move towards more functional accounting to align implementation with ecological objectives.

Instrument interaction and program performance
Federal and state regulations pursue different goals and create different instruments that overlap. Some of these instruments function at a regional scale, while others function at a more local scale and the spatial boundaries of each are not often congruent. The conservation banking program overlaps with other compensation instruments, which can increase cost-effectiveness and flexibility (Santos et al. 2015), assuming that all redundant instruments lead to the same conservation outcomes. A strong policy mix ''builds on the strengths of individual instruments, while compensating for their weaknesses with additional or complementary instruments, thereby maximizing overall policy performance'' (ibid., p. 295).
Currently, instrument interaction is mainly redundant, with non-uniform standards potentially causing conflicts. Although regional habitat conservation plans/ natural community conservation plans in synergy with banks could create larger connected areas over time (Bank/Perm. 19,Attorney 14,Cons. 24,28), they often compete due to different temporal and spatial scales. The current multitude of compensation instruments without equal requirements for all compensation projects, might disincentivize conservation banking in California and beyond and incentivize the use of other instruments. Some instruments are available across the US, whereas California state regulation offers additional instruments for compensation under CESA. Banks, in-lieu fee program and habitat conservation plans often cover species that a impacted frequently. Not all regions may offer all compensation options for a developer, but permittee-responsible mitigation and turnkey mitigation can always be an option, because they are implemented on a project-by-project basis and compensate for a specific impact.
Other compensation instruments may offer lower prices but don't necessarily provide compensation that is consolidated, can be strategically sited and implemented in advance of impacts. Regional habitat conservation plans can offer such compensation, but they only exist in some areas and only cover specific impact project types. Also, private-sector compensation reflects the true environmental cost of an impact (Cons. 20). In-lieu fee programs and habitat credit exchanges are both rare, and the first consolidate mitigation but can collect payments before implementing measures, whereas the latter can be implemented in advance but is not strategically sited and can consist of smaller and scattered sites. Thus, a lack of banks may lead to more project-by-project permittee-responsible or turnkey mitigation, which can be implemented after the impact occurs and does not offer consolidated and strategically sited mitigation (see Table 1 for characteristics of compensation instruments). Hence, a decreased use of banks or a disappearance of banks in favor of other instruments is likely problematic due to non-uniform standards and thus weaker compensation. Such a development may have negative repercussions for species conservation (as less regulated instruments may provide lesser conservation outcomes), agencies (as approving and monitoring project-by-project mitigation requires more resources), and developers (as fewer options may be available and may not provide a one-stop shop).
A shift towards weaker compensation instruments can reduce the relatively small conservation banking market and adversely affect compensation performance under ESA. Banking incentivizes the private protection and maintenance of often crucial habitat (Grimm 2021). As long as requirements are not uniform, a preference for banks and regional habitat conservation plans/ natural community conservation plans could improve conservation outcomes (Bank/Perm. 19,Cons. 24). Policies that would have established consistent standards and a net gain goal under ESA, requiring compensation across the USA (FWS 2016), were rescinded before they were even fully implemented. Uniform requirements could allow for fair competition between overlapping and complementary instruments: Holding all instruments to the same high standards (such as advance mitigation or other quality assurances) could thus improve conservation outcomes.

Beyond California: Offsets and banking elsewhere
This analysis focuses on conservation banking in California, which can differ from the application of compensation under ESA in other states. Standardization in California has been refined over decades and the framework from California was used as a template for the 2016 FWS compensatory mitigation policies (FWS 2016, rescinded by the Trump administration). A discussion thereof can inform other states or future federal standards and policies. For example, issues with proper metrics exist across the USA (Barral 2019). The rescinded ESA compensation policies also included a net gain goal but without it, the ESA is interpreted differently in different states (i.e., some do not require compensation). This explains why banking only exists in some states (White et al. 2021a) and shows how the institutional framework can (dis-)incentivize banking outside of California. Strategic planning will likely be needed to identify bank sites, anticipate development, and ensure that banks meet credit demand across the USA (Theis and Poesch 2022). This need could be met with a combined use of habitat conservation plans and banks.
Nationally added banks and acreage have increased yearly (Theis and Poesch 2022). Although a decrease in annual bank establishment is expected, conservation banking should meet increasing credit demand as more credits are released (ibid.). For wetland mitigation banks, a credit release bottleneck is likely and Theis and Poesch (2022) suggest advance credits to avoid that. One could thus argue that outcome-based credit releases for conservation banks could also lead to such bottlenecks. The expected increase in credit demand can also increase competition between credit providers and calls for equal compensation standards.
Uniform standards for compensation mechanisms and a preference for banks was established for wetland mitigation banking under the Clean Water Act with the 2008 rule to ''improve the quality and success of compensatory mitigation projects'' (EPA 2008, p. 19594). Wetland mitigation banks follow different standards than conservation banks (e.g., no averted loss offsets, clear rationale for ratios), but the programs are closely linked: Banks can produce conservation and wetland mitigation credits, and thus changes in one program may affect the other. Mitigation banks are preferred because they provide advance mitigation, consolidated mitigation (easier monitoring), and protect larger areas and thus offer better ecological outcomes (Vaissière et al. 2017). Following banks, the hierarchy lists in-lieu fee programs and lastly permitteeresponsible mitigation. In addition, the wetland compensation rule reduces bank approval times (Robertson 2008). Thus, the rule aims to reduce uncertainty for both agencies (in terms of ecological restoration by preferring banks, which is perceived less risky by regulators) and bankers (in terms of expectations and timing) (BenDor and Riggsbee 2011). Vaissière et al. (2017) show how negotiations between wetland mitigation bankers and regulators keep the program in a ''zone of ecological-economic viability'' (p.512) and I argue that more outcome-focused requirements could also place negotiations for conservation bank approval in such a zone. My results above show that a level playing field or a preference for banks (as provided for wetland mitigation banks) could improve the institutional framework for conservation banking.
Although this research has a regional focus, findings may be relevant for banking and offsets elsewhere. Banking approaches are applied in many countries outside the US, such as in Canada (Doka et al. 2022), Australia (Santos et al. 2015), the UK (Ermgassen et al. 2021) and many countries across Europe (Wende 2018), including France (Vaissière et al. 2021), Germany (Darbi and Tausch 2014) and Spain (Maestre-Andrés et al. 2020). Regardless of location, offset protocols must be improved to align with conservation objectives Watson et al. 2021). With the new Convention on Biological Diversity's post-2020 global biodiversity framework, this will be an international Net Gain objective (CBD 2021), that should be reflected in national policies across the globe (Simmonds et al. 2022). Thus, improving social-ecological fit of an institutional framework to reflect such goals and measuring long-term conservation outcomes is relevant beyond the scope of conservation banking. For example, conservation banking practice shows a way to incorporate the cost of long-term management into offset prices, which is necessary to build offsetting around the ''notion of limits'' (Damiens et al. 2020, p.2) and long-term net gain objectives (Damiens et al. 2021).
Offset programs are shaped by different stakeholders and each institutional setting is unique (Boisvert 2015), but many face similar challenges, such as currency, equivalence and dealing with trade-offs between accuracy and tradeability (Bull et al. 2013;Chiapello and Engels 2021). This is also discussed by Ermgassen et al. (2020) in terms of geographical and ecological flexibility, which can expand the market, but can have negative implications for conservation. Thus, aforementioned findings on potential barriers to market participation as well as ecological and economic outcomes of trade-offs can be relevant for the design and implementation of other offsets. For example, Ermgassen et al. (2019) identify the failure of offset implementation and failure of offsetting intervention as the main risks to reaching no net loss. As outlined in the results section of this paper, conservation banking in California requires a conservation easement for bank approval and crediting and strict monitoring schedules. This practice and the recommendation of more outcome-based crediting could potentially serve as a template for offsets elsewhere. However, local conditions (e.g., land availability, ratios) (Sonter et al. 2020a) can always affect the outcomes of offsets.

CONCLUSION
Analyzing social-ecological fit in terms of temporal, spatial and functional scales has shown that the conservation banking program in California has found some compromises between market and ecological spheres. Results also include mismatches that could be targeted for program improvement. The analysis of instrument interaction has identified differences in requirements that cause uneven competition and weaker conservation outcomes. Moving forward, the US conservation banking needs clear federal policies with equal standards for all compensation tools to stabilize the hybrid market and create synergies among compensation instruments. Such standards must be adjusted to the ecological objectives of the program to measure its contribution to species conservation. The findings on spatial and temporal scales and functional linkages could be useful for other species and habitat banking programs, and considerations on instrument interaction could help improve regulation-based offset programs anywhere.