Path-dependency and path-shaping in translation of borrowed policy: outsourcing of teaching in public schools in Hong Kong and South Korea

Abstract

Purpose

The study systematically analyses the path dependency and path-shaping of borrowed education policy, tracing it from the global through the national to individual schools. It also revisits the case schools after five years to map the school level policy paths.

Design/methodology/approach

Recently, path-dependency heuristics have drawn attention in predicting educational policy trajectories. However, these studies are primarily theoretical, and those empirical studies do not capture what happens at the school level. This paper fills the research gap by presenting a model that synthesises the research from diverse fields and is informed by findings from a longitudinal case study of educational outsourcing in public schools in Hong Kong and Korea.

Findings

The findings highlight path dependency interactions across educational levels diachronically and synchronically, while aptly incorporating the creative ways school leaders exercise their agency therein. The paper concludes with new insights into policy trajectory and education outsourcing.

Originality/value

The study substantiates and extends previously suggested theoretical models on the paths of travelling educational policies and identifies the factors that shape the paths. It also sheds light on how school leaders navigate the structures that constrain their actions or create a new path and pursue their educational goals.

Citation

Choi, T.-H. (2022), "Path-dependency and path-shaping in translation of borrowed policy: outsourcing of teaching in public schools in Hong Kong and South Korea", International Journal of Comparative Education and Development, Vol. 24 No. 3/4, pp. 144-159. https://doi.org/10.1108/IJCED-01-2022-0004

Introduction

The influence of history and other contextual features on the trajectory of educational policy has continuously attracted scholarly attention. That is, researchers have identified how past decisions affect future policy outcomes, making it difficult to adjust the policy course over time (e.g. Cohen, 2017; Feeney and Hogan, 2017; Gulson et al., 2017; Hartong and Nikolai, 2017; Tan and Yang, 2021). Many of these studies concern themselves with local policies (e.g. Zhang, 2005, the privatisation of postsecondary education in China; Tan and Yang, 2021, the case of South Korea reforms) and reform initiated by the government of a political block for its member countries (e.g. Feeney and Hogan, 2017, on a policy of the EU).

What has not received due attention is the path of travelling policies outside a nation's boundary or political block. Many governments borrow or transfer policies in the expectation that their success in another context will be replicated or to legitimise an initiative that was to be implemented even without such reference (Morris, 2012). Others are initiated by international organisations such as OECD (Rizvi and Lingard, 2010). Whatever the causes are, it is common to see travelling policies, albeit with varying degrees of local adaptation and success in their implementation. Despite this policy trend, there is no empirical research that explores (1) how such travelling policy trickles through a national system and then into individual schools; and (2) once the policy reaches the school level, how the policy evolves within the school system. This research gap is particularly pertinent to educational outsourcing, where third parties (e.g. edu-businesses and charities) participate in public schooling, an area almost neglected in privatisation studies despite their prevalence (Ball and Youdell, 2008; Gulson et al., 2017). Using educational outsourcing as a case, the research addresses the intersection of these research gaps by tracing how, over the span of five years, a travelling policy of educational outsourcing is translated into local policies and then into school-level practices.

Framed by the policy enactment theory and path-dependency theory, this international comparative case study was guided by the following research questions (RQs):

RQ1.

How was the travelling policy of outsourcing translated into executive plans in the two case contexts, and what shaped their paths?

RQ2.

How was the executive plan translated into school-level practices in the two case contexts, and what shaped their paths?

RQ3.

How did school-level practices change after five years, and what shaped their paths?

Applying the policy enactment theory and path-dependency theory, this study discusses factors shaping the path of a policy, where the findings were collected from comparative case studies conducted in Hong Kong and South Korea. This study specifically contributes to the path dependency and privatisation debates, and suggests future research direction.

Policy enactment and path dependence

Policy enactment theory emphasises the evolutionary nature of the policy process and the agency of policy actors. Implementing a policy is not linear and top-down. It is a creative process as policy actors bring in their own beliefs and experiences, and the contextual features shape it, thus the term enactment (Ball et al., 2012). The policy plans are interpreted in interaction with different contexts' discourses and ideologies. They are sometimes side-lined by historical, cultural and structural conditions; they are appropriated or resisted by actors, leading the same policy ideas to develop differently in different societies (Choi, 2018b).

Once settled, policies become resilient or difficult to change. The path dependency theory focuses on this “inertial tendency for the initial policy choices to persist” (Feeney and Hogan, 2017, p. 280) and traces the trajectory of individual policies. Pierson (2000, p. 252), synthesising previous studies, explains the concept in plain terms as a “self-reinforcing process” that “preceding steps in a particular direction induce further movement in the same direction”. While initially developed in economic history, diverse fields such as business, technology, management, history, and political science noted the path-dependency in policy processes (e.g. Pierson, 2000; Sydow et al., 2012; Torfing, 1999; Wilson, 2014). They have identified the features that create inertia from initial decision-making, both tangible and intangible. For instance, Pierson (2000) from the field of political science identified initial setup costs, improvement of the chosen option through use, perception of being in the mainstream, benefit from the use by the majority such as well-developed infrastructure, high costs of reversal, and the entrenchment of specific institutional arrangements, as the main factors that shape the path. Torfing (1999) adds that discourses such as inherited value systems also affect the path.

Scholars in the field of education (e.g. Cohen, 2017; Takayama, 2012; Tan and Yang, 2021), though a handful, have also noted the potential of path dependency theory in explaining educational reform and change. They also note the factors contributing to the inertia. For instance, Takayama (2012) specifies how an institutional setting and existing policy have shaped and restricted the path of standardised testing in Australia and Japan. Cohen (2017) notes how shared educational ideology leads actors to resist a reform that does not fit the ideology using their power. The path dependency theory has provided a fruitful approach to understanding the stability of educational systems (e.g. Meyer, 2011).

Recently, the agency of policy actors has been integrated into the path-dependent theory to explain how inertia is disrupted. While there are different interpretations and conceptualisations, in this paper, the agency is understood as a negotiation process that an individual engages in between their beliefs and external structural constraints and enablements; the beliefs affect their goal setting (Joseph, 2002; Layder, 1997). Both the education field and others note the role of policy actors and how they bring in changes, creating a new path. For instance, Hartong and Nikolai (2017) note how socio-political changes move actors to consider changes. The consequent emergence of policy brokers brings in new ways of policy actions. Tan and Yang (2021) show how teachers created a hybrid pedagogy to meet both the force to keep the exam-centred pedagogy and the one to bring holistic education into schooling. More pertinent to this study, Choi (2019; see also, Poudel and Choi, 2021) shows how educational decision-makers appropriate the incoming policies to realise their visions and educational philosophy, by packaging their agenda in a palatable way to the resource distributors and proactively preparing for possible changes, so that they are ready to navigate through related challenges.

While enlightening, little empirical research has been conducted on how travelling educational policies are adopted by individual stakeholders across multiple levels, e.g. from the global through the local governments to individual schools (Gulson et al., 2017). Existing studies analysed the cases of local, nation-bound policies, e.g. high-stake exams (Tan and Yang, 2021) and charter school policies (Cohen, 2017). Others did investigate transnational policies such as the European Higher Education Area's Qualifications Framework (Feeney and Hogan, 2017) and the OECD-coordinated Programme for International Student Assessment (PISA) (Hartong and Nikolai, 2017). However, they are either theoretical papers or draw on document analysis only. This study seeks to fill the research gap.

Comparative case study of educational outsourcing: data and analysis

This paper investigates how the systemic structure and discourses at government and school levels have affected the enactment of the imported policy of educational outsourcing. Outsourcing [1] or contracting out refers to delegating activities to deliver curricula to third parties, such as educational businesses, freelancers, NGOs and charities, whether they concern academic subjects or character building. Recently, a new practice of using the school budget or government fund, in contrast to the previous practice of charging the user directly, has become popular in countries across continents, including Asia, Oceania, Europe and the US (Bates et al., 2021; Choi, 2018a). This comparative multi-case study investigates publicly funded educational outsourcing, and comprises document research (e.g. government policy and executive plans, annual school plans, contracts) and semi-structured interviews. It was conducted in Hong Kong and Korea, adhering to the ethical requirement of the author's university. As the questions concern participants' daily work practice and the sharing is within the remit of professional discussion, the participation was considered to involve no or little risk. The author's contextual knowledge gained through long-term teaching experience in the case contexts enabled an emic view of the phenomenon at the school level. However, the author did not have any particular position regarding the phenomenon before the investigation. Thus, there is no known conflict of interest or biases–though no researcher can be free from the influence of their subjectivity.

The two contexts were chosen as telling contrastive cases, from the contexts that the author has contextual knowledge of, which is indispensable in tracing the policy trajectories. Since the road to outsourcing in public schooling was paved in Korea in 1995 and Hong Kong in 2000 through neoliberal education reforms, educational outsourcing has become prevalent [2]. At the same time, they exhibit contrastive practice. In Hong Kong, outsourcing is publicly acknowledged and recorded, while it is only conducted off-the-record in Korea. The comparative study was conducted twice, in 2014/15 initially, and a follow-up in 2019/20. The 5-year gap was deemed appropriate to capture the policy trajectory. Principals are the primary change agent in schools. In Korea, their usual service term is four years; in Hong Kong, principals stay with schools till they retire after the initial contractual period of 2 years, but a development plan, which becomes a main juncture of changes, covers three years.

Case studies were conducted with schools actively engaging in educational outsourcing in Hong Kong and Seoul (Hong Kong Schools A and B; Seoul Schools A and B), all of which purchased programmes in three or more subjects. The case schools from those meeting the criteria were recruited through professional networks, and if the initially invited ones rejected the invitation, the next was invited. English is considered a marker of social status in both cities and a necessary skill to secure financial stability in a globalised society, and one of the most popular areas of outsourcing (Choi, 2018a; Guo, 2012). The said language was thus chosen as the focal subject.

A total of 12 staff have participated in the initial case studies, comprising one decision-maker from each school, e.g. the principal or English panel head [3]; one teacher in charge of an outsourced case programme; and an educational service provider of that programme [4]. Typically, the decision-makers will choose the most appropriate programmes (and sign the contract). The schoolteachers in charge will be the first contact point afterwards, sometimes co-developing, co-teaching and being present in the outsourced classes during school hours. The interviewed providers were the personnel involved in contract writing. Of all the outsourced programmes, those which the teachers recommended were chosen as case programmes.

In the follow-up study, the school reports on the outsourced programmes were accessed to check for changes. The previously interviewed school decision-makers were contacted for short follow-up interviews, two of which accepted the author's request. A new principal, who possesses English teaching background and made English teaching related decisions, was appointed to School A at the time of the follow-up study. Thus, she was briefly interviewed, which made the total number of participants 13 (see Table 1 for case schools, contracted programmes, and interviewees in the two rounds of data collection).

A typical interview lasted about an hour, and if needed, follow-up interviews were conducted on- or offline. Participants in Hong Kong were interviewed in English; those in Korea, in Korean. Data in Korean were transcribed and translated into English by two bilinguals; where there were discrepancies, the two discussed and agreed upon the translation. Data were analysed thematically and compared cross-level, cross-case (Manzon, 2014; Miles and Huberman, 1994). While open to emerging themes (Gibbs, 2018; Marshall and Rossman, 2016), it was also guided by previous research on outsourcing policy (e.g. Hartong and Nikolai, 2017; Sydow et al., 2012). The data were analysed focussing on outsourcing policy translation at the government- and school-level, the role of the actors, and interaction between the levels. Parts of data were coded by research assistants according to a coding framework drafted by the principal investigator. Through this initial coding, interrater reliability was established and the coding framework was revised, to be used for coding the remainder of the data.

Through the analysis, how outsourcing strategy evolved across levels and what path-shaping factors affected its course were identified. At the school level, how the process was affected by actors was also explored. Because of limited space, most interview data will be presented in a narrative form, with direct quotations being sparingly used. The names of participants quoted below are all pseudonyms.

Localisation of educational outsourcing policy

RQ1: translation of the outsourcing policy into executive plans at the government level

The localisation of the travelling policy of educational outsourcing at the government level was traced through the analysis of the government document. In introducing the outsourcing practice into the contexts, both governments resorted to the macro discourse and global ideology of neo-liberalism. The education authorities in Hong Kong and Korea have circulated policy texts that affirmed or reconceptualised the role of the private sector as a partner in providing public services, including education. They gave the cases of other countries to justify the introduction of this new discourse (Hong Kong Efficiency Unit (HKEU), 2008; South Korea Ministry of Education (SKMOE), 1995), reflecting the fact that educational outsourcing was a “borrowed” or “travelling” policy.

The two jurisdictions' contrasting histories, with the private sector in the provision of public services and the resultant diverging public attitudes towards the private's participation in public service provision, led to differing approaches in introducing educational outsourcing. To illustrate, in Hong Kong, public services have long involved the private sector since the colonial era. The government could thus take a more open and confident approach, as it was simply expanding the partnership within education. For instance, the efficiency unit, which was created to oversee public–private partnership claims:

Hong Kong has a long history of using the private sector to deliver public services … Contracting has been widely used by governments over many years for the purchase of goods and services from private and voluntary providers. Overwhelmingly this policy has been a success. (HKEU, 2008, p. 3, 5)

In addition, public education was initially provided by third parties such as religious bodies, charitable foundations, trade bodies (95% of the schools), while the government stepped in significantly only in the 1970s (Lee and Manzon, 2014). This “normalised” role of the private sector in education is evidenced by the fact that all imaginable manifestation of educational privatisation is observed, e.g. operation of schools (early childhood education is 100% privately provided), private tuitions, and Direct Subsidy Scheme schools (similar to charter schools), to name a few.

In Korea, the privatisation of public services only started after the financial crisis of 1998, which enabled access to private funds for education. Before that, there was no public-private partnership in the country (Song et al., 2012). The government has exhibited a somewhat defensive position when introducing the partnership in education, as can be seen in the 1995 reform documents that heralded neoliberal governance of education:

[The current educational reform] will expand the choices of the educational consumers and promote real competition among the educational providers, to maximise the benefit of education … It should be understood that the market system was adopted as an effective means to achieve these goals (SKMOE, 1995, pp. 3–4, author's translation).

The cautious approach of the Korean government is understandable, considering the “purist” sentiment that the public has toward education. People believe that education should not be treated as a commodity for market activities (Kim, 2005, p, 28). The new discourse of education privatisation has met massive resistance from the public. For instance, as Autonomous Senior High Schools (similar to chartered schools) became prevalent in Korea, the public engaged in heated debates over education becoming commodified, when such schools could determine school fees. Most people did not welcome this new type of school, which was reflected in the Seoul Metropolitan Office of Education survey. Over 60% of the citizens demanded its abolition, foreseeing that market activities through education would increase educational inequity between the affluent and financially disadvantaged (Park, 2014). Opposition to the policy was adopted in the election platform of presidential candidate Moon Jae-in. When he was elected, his cabinet shrank the sector and recently announced the abolition of the school type altogether (Chun et al., 2019).

With the policy only being a broad sketch, executive plans had to be made before the policy could be put into practice. In this translation, the existing paths exerted considerable influence. The deep relationship between the private and the public in Hong Kong education enabled the government to openly promote partnership between third parties and schools. For instance, it created over 20 funds that can be used to outsource the curriculum delivery, as well as establishing grants such as the Capacity Enhancement Grant and School-Based Support Scheme Grant for Schools with Intake of Newly Arrived Children. The executive documents explicitly use terms such as “procurement” or “purchase of services”, and official government documents reflect such use of funds. In 2013/2014, over HK$36.1 million was spent in supporting secondary schools' outsourcing of teaching (Education Bureau, 2014). The amount was equivalent to 2.48% of the annual budget for publicly funded secondary education for the academic year (HK$1,455.6 million) (Hong Kong SAR Government, 2013).

The Korean government, in contrast, did not officially promote the outsourcing of teaching. Official discussion only concerned participation by charities and public entities, and that of the business sector, only when the programme was offered for free (e.g. internship programmes). There were no specific funds allocated for the outsourcing of curriculum delivery, while schools could exercise autonomy in using their annual budgets since the introduction of school-based management system, including purchasing educational services. Also, schools designated as “centre” or “leader” schools in a particular subject area were given extra funds to use at their discretion.

The regulatory frameworks in the two contexts were also different. In Hong Kong, where the partnership was officially promoted, the government provided clear guidelines, for instance, on the remit to use the funds and documentation of the partnership process, including procurement and evaluation of the programmes (e.g. Education Bureau, 2013). In Korea, as the practice was happening with schools' discretion, there were no official governmental level guidelines. These structural differences have a bearing on school-level practice, as to be discussed later. The diverging ways of introduction of educational outsourcing and its translation into executive plans across the two contexts confirm the crucial role played by the governments, existing policy and public values in the localisation of a travelling policy (Takayama, 2012; Torfing, 1999).

RQ 2: school-level initial translation of the outsourcing policy

Translation of the outsourcing policy from the government executive plans to school-level policy was traced through document analysis and interviews. The school-level document analysis shows that the contrastively translated executive plans constitute the path for implementing educational outsourcing policy at the school level, along with other school-level contextual features. In Hong Kong, because of the requirement to document and publish the usage of specific-purpose funds and the quality management of the purchased services, the practice is comparatively well traced. A document-based study that explored the grant reporting from a third of Hong Kong secondary schools (n = 137) for the 2014/5 academic year shows that most case schools (65%) made reports on purchased programmes publicly accessible, including quality control measures (Choi, 2018a).

In contrast, the outsourcing of teaching in Korea is not captured in the public domain. To illustrate, the two Korean case schools each purchased two-year-long English programmes. However, in the official documents such as yearly school plans and reports, the only information given on these programmes were generic names of the programme, such as “extensive reading” and “speaking classes” and the name of in-school staff charged to oversee them. No information to identify the programmes as outsourced was given. However, the interviews reveal that the schools purchased a wider range of programmes and services. The principal of Seoul School A remarks:

Here we hire coaches for activities within school hours for Sports Clubs. They are also hired for Creative and Experiential Learning, in addition to Physical Education lessons, one hour per week. We hire coaches for baseball, basketball, and others … To enforce Arts and Sports education, to defocus from academic learning, we provide very many experiences and sports programs … These days the government distributes a huge amount of money to enhance cultural education, sports, and arts education.

The fact that this extensive outsourcing practice does not need to be documented leaves the quality monitoring to the discretion of schools, which may not always work out. The English teacher in charge of outsourcing partnerships at Seoul School B acknowledged that although she had to oversee these programmes, she never monitored whether and to what degree students used the purchased online and offline programmes. She felt she “wasted the public funds” after this observation. The resources and motivation for teachers to monitor the outsourced programme were limited due to heavy teacher workload and without an accountability system.

In translating the outsourcing policy at the school level, what programme to purchase and how were guided by the decision-makers' beliefs and visions (see Choi, 2019 on the relationship between actors' agency and beliefs). For instance, Faith, the Principal of Hong Kong School A remarked:

Our students, they mainly come from a low SES [socio-economic status] background. They do not have support from their family. So, as a teacher, we have to do more for them. It's just a belief. You work according to your belief, especially your belief in education, right? I can give you an example. In our school, we have four social workers. In other schools, they can have one or 1.2 only. But I will use the LSG [Learning Support Grant] to buy services to let our school have more social workers and then let our social workers help our students, especially in their growing up, in their life, in their behaviour. This kind of students, the major problem, the core problem, the behavioural problems and the problems they face during the growing stage, are not in the academic.

One of the repeated themes in her justification of her decisions was to support the social mobility of her students. She feels that in a Band 3 School [5], like her previous and current schools, students usually have low social capital in terms of English. She explains:

Because if they fail in English, they have no chance to enter the university, right? And this is the most critical factor. Why is the enrollment [of students] from band 3 schools in the university so low? Because actually, their English is extremely weak, even in the primary, their basics. Their foundation is weak. We have to do a lot of things to push their English standard. In general, students in my school can only get 2 in English. 3 in English is really really difficult for them. For band 1 students, they get 5 or 5*!

Securing grade three (out of five) in English in the secondary graduation exam is a prerequisite to be eligible to apply for a government-funded university, while English is the subject where her students are weakest. Faith “put emphasis on the English subject” and made English “the major concern”. She tried to realise her educational belief and subsequent goal of improving students' English test results, through purchasing “quite a lot and different kinds of programmes from outside”. Many funds involved submitting a proposal to an organisation such as the Education Bureau and NGOs, and competing with other schools. Thus, she tried to figure out “their minds” and package the identified needs of her students in the funder's terms, such as “settling the newly started senior secondary school curriculum”, “promotion of career education”, “supporting the students with special education needs”, or “creating an alternative career path for academically low achievers”.

Mark, the Principal of Seoul School A, shared Faith's emphasis on English education. He felt that it was unfair for students from low-income families or academically low-performing students to receive support limited to the basic level. Thus, he purchased services that can be challenging at their current level. He believed that being exposed to up-to-date development such as AI-assisted learning or obtaining an internationally recognised certificate would help students broaden their perspectives and be motivated to reach a higher goal. He happily shared some success cases of students from highly underprivileged backgrounds. They found motivation and courage from his approach and found decent jobs. He could persuade other principals serving low-income, low-performing schools to become inspired and adopt his approach.

What is notable is that these leaders do not blindly adopt the practice of outsourcing, just because the government created the room to purchase services financially or administratively. They engage in this because it will help them realise their “beliefs” or “dreams”. Obviously, the participants in the study did not engage in the critical debate on whether privatisation of education or educational outsourcing itself is proper; however, they made a conscientious judgement in their engagement, using the maximum learning benefit of students as the single most important criterion of judgement. Thus, they ensure that the neoliberal initiative does not erode equity and exercise their professionalism with self-confidence, alleviating the usual concerns around educational privatisation (e.g. Ball and Youdell, 2008), to a degree.

Realising their beliefs or educational philosophy, however, was not without difficulties. In Hong Kong, the government fund provided on a per capita principle was not sufficient, especially if the school size was small. Faith was not thwarted by that, however. She related how she resisted being framed by the structure, giving the example of an English specific fund:

This [fund] is totally related to the head-count … [My previous school] was a really really small school, only had 400 students. We got a small amount in this scheme, but for some other big schools, they have some two million or something like that. So when you get a small amount of money, you cannot do anything to promote the English environment … So I went to the Education Bureau with my colleagues to do some presentations and then do some negotiations. We asked for more, but they just never accepted the request.

After the failure of her attempt to negotiate with the Education Bureau, she started to reach out to other organisations to secure funds. On the day of the interview as well, she had just finished writing a proposal that she had been drafting for two days to submit to the Social Welfare Department. Mark, Principal of Seoul School A, also sees securing funds as a prerequisite to realise his educational philosophy. As he has to spend considerable time on securing and managing funds, he now has shed his identity as a teacher and considers himself an administrator, though not without regrets. While acknowledging the needs of funds, not all schools write proposals to secure funds. Hong Kong School B, for instance, dealt with the issue of limited funds by recruiting volunteers to expose her students to English further.

Visions and beliefs, once institutionalised, constituted the structure to which the staff had to conform. In describing the outsourcing practice, the decision-makers of the two Seoul schools quite often referred to the school visions created when applying for a designation as leader schools. Seoul School A, which was designated as a school that develops global leadership, purchased programmes that help with the branding, such as travels to schools on different continents, support for inter-country seminars, and online English proficiency certification conferred by an international company. Seoul School B packaged itself as an English leader school for the underprivileged, due to the low average income of its catchment area. It signed up for diverse online programmes, such as a level-adjusting, interactive online extensive reading programme, aiming to compensate for the home environment where English related exposure was limited. The school also purchased the individualised support programme to prepare for the English section of the college entrance exam, which was available to students for free.

RQ 3: school-level change of practices after 5 years

The document analysis and follow-up interviews confirm that the previously existing institutional context contributed to the school-level inertia, while leadership could create new paths. Over the five years between 2014 and 2019, outsourcing practice hardly changed in Seoul case schools. While such uniformity may be surprising, considering that new principals served both case schools, it is also understandable because they have received awards for their leader school practice and renewed their leader school position. They do not have reasons to change the practice that is appreciated by the funder, especially considering the inevitable workload increase and uncertainties around developing and enacting new programmes while maintaining their hard-earnt positive public image. Changing outsourcing providers requires considerable discussion across phases of collaboration (e.g. contracting, implementation and evaluation of the programme) to learn the expectations of each other and ensure the delivered service is up to standard. One teacher who oversees an outsourced programme well noted the discursive nature of learning of mutual expectations:

Because I've been working with three different people in the company, to a great extent, I've already told them what I've got to say. Not just the contact person but also the person in charge of the company remember what our college wants, what they need and have to do.

In Hong Kong, both schools maintained their focus to cater for the needs of the lowest academic band students–the school band is hard to change in the public's eye. Panel Head of School B, in explaining her decisions relating to outsourcing, still referenced the low-income status of the communities the school serves. She referred to the sponsoring body's related vision to support underprivileged students, the importance and difficulty of developing English proficiency in a Band 3 school, and how the school still invests considerably in English. Due to its low academic standing, School B had continuing concerns regarding student recruitment. Thus, it continued to purchase a summer bridging programme to attract potential students. Finally, the annual, school-sponsored English immersion trip to Singapore and elsewhere remained the same to provide an immersion opportunity for students who did not have many opportunities to experience English in their daily lives.

In the micro-level, however, the two schools made considerable adjustments: School A, with the change of its financial situation, and School B, in response to the new trend in education policy. School reports [6] of School A show that it had a considerable change in what they purchased and how. It seems that Faith's efforts paid off. The school secured a long-term fund from a large charity linked to a consortium of businesses, with an additional HK$ 3 million income (≈ USD 400,000). The subject of English well illustrates the financial changes. In the initial round of data collection, the school could purchase only basic “remedial” programmes that support students who fall behind academically, such as phonics and tutorials to help with homework. Since the school secured the big fund in 2017, the school invited external advisors from abroad to consult on English curriculum, and purchased a broad spectrum of enriching programmes including motivational courses (e.g. English drama workshop), and building high-level skills such as creativity (Creative writing). With such changes, it is expected that the students would develop portfolios that support university admission, building their proficiency to reach the threshold level for admittance to higher education. Faith finally reached her goal to provide sufficient support to offset her students' underprivileged backgrounds. The English Panel Head of Hong Kong School B noted that she now combines English programmes with science. With the new emphasis on science, technology, engineering, and mathematics (STEM) education, there are many more STEM funds.

Implication for research on policy trajectory

This study contributes to the current debate on path dependency of educational policy by providing empirical substantiation of hypotheses made about the multi-level localisation of a travelling policy from previous studies (e.g. Cohen, 2017; Hartong and Nikolai, 2017; Sydow et al., 2012) and new insights into the school level long-term policy trajectory. The government policy document analysis shows that the newly travelling discourses of neoliberal ideology led the two governments to bring educational outsourcing into their policy agenda. Notably, the governments presented the neoliberal strategy of opening education to market activities as something to improve effectiveness. Whether the government officers, who played essential roles in adopting these measures, believed in its efficiency or were simply appropriating the discourses to implement what they wanted to do (e.g. Morris, 2012), remains to be seen.

In translating the outsourcing policy from the global to the local level, in Hong Kong, the government was more confident and open about bringing educational outsourcing into public schooling; while in Korea, the government was more cautious and obscure. This diverging translation of the policy, shaped by the disparate discourses about the private in the public education across the two contexts, triggered a “self-reinforcing process” (Pierson, 2000). The initial government path affected the following phase of localisation, i.e. creating executive plans. In Hong Kong, the subsequent measures such as financial arrangements were open. Its public accountability measures, including reporting on the quality of the outsourced programmes to the government and the public, were also integrated into the system. In Korea, however, the absence of governmental guidelines on the practice and public accountability measures left the practice to the schools' discretion. The practice became hidden from the public and untraceable by the government; even staff in the same school did not know the circumstances unless directly involved. Indeed, the action at the local government level created “reactive sequences” (Mahoney, 2000).

This study's new knowledge concerns empirically capturing a travelling policy's translation from the local to the school level, and the subsequent changes at the school level. The first round of data illustrates how the existing structure (e.g. school catchment areas, teacher workload) and discourses (e.g. the vision of the sponsoring body to help underprivileged students' social mobility) affected the outsourcing practice at the school level. It also shows how principals created a new path: the principal of Seoul School A, when he joined the school, created a new vision of global leadership and a structure to sustain the vision by seeking the designation as a leading school of global leadership, and ensured that subsequent principals inherit it. These path-creators search for ways to navigate challenges. For example, the principal of Hong Kong School A, who keenly felt the need to change the school's financial structure, reached out to the private sector after her initial failed attempt to receive extra funds from the Education Bureau. She contacted and persuaded numerous potential funders, finally securing a steady funder. With that, she brought changes in outsourcing: she shifted the focus to more motivational and high-level English courses, stepping closer to her goal of creating a foundation for students' social mobility. The direction of an agency does not always go against the existing path. The new principals of Seoul schools in the second round of data collection (and two and one more principals in between, respectively) maintained the flagship curriculum and renewed contracts with current service providers after ascertaining that the original principals' visions were still appealing to the government and potential students. If the successive principals find that the existing visions do not serve their purposes and goals, they will start a new discourse and mechanisms to realise their own beliefs, like the principals we met did.

The following diagram (Figure 1) synthesises the previous literature and findings from this study to capture the path dependency and path-shaping while a travelling policy (e.g. educational outsourcing) is localised at the two levels of government and school. It illustrates the specific path-dependent and path-shaping factors influencing school-level and government-level practices. Three characteristics feature this path dependency heuristic, concerning the educational policy. First, it reflects the policy enactment perspective. The policy is an iterative process co-constructed by different stakeholders and situated in a local context (Ball et al., 2012). The outer circular arrows represent the time dimension and show that the process of path-shaping/creation and path-reinforcing perpetuates. Policy actors at different levels of the educational system figure in the process, through selective (dis)engagement with new or existing discourses. The two-directional arrows between the path-reinforcing and path shaping elements illustrate this ambivalent position of the actors. As seen from this case, principals are main change setters. Depending on the alignment between their own educational beliefs and the existing structure (discoursal or systemic), new leaders reinforce or disrupt the inertia of the policy processes. For instance, they can reinforce the school tradition, or start a new one. They can also attract new funds or seek designation as a leader school, which will form a new path structurally. Second, the model captures the tension between the old discourses/structures and the new ones by juxtaposing them. This juxtaposition also recognises the coexistence of the path dependence and path-shaping forces in a context (Tan and Yang, 2021). The division of the two opposing forces is porous, to reflect the fact that a new discourse will become normalised to form the path-reinforcing elements with time. Thus, the heuristic captures the contribution of this paper: that the path exerts an influence both diachronically and synchronically, hence, the embedding of the snapshot tables illustrate the interactions between the government and school-level practices within the time cycle. The process is not always top-down, but what happens at the grassroots level may find a way to reach the government level, and “extend its influence beyond its local or national borders” (de Sousa Santos, 2006, p. 396) to make up another travelling policy, though empirical substantiation of this part is left to future studies. The interactive arrows between the two levels of schools and the government hence exist, and the same reciprocity will apply between the government and global levels, though not captured in the model.

Conclusion

This comparative study extends previous studies on policy path dependence by tracing the localisation process of a travelling policy through the system to the school level, and how the policy practice evolves over an extended period in individual schools. I have constructed an empirically substantiated, heuristic model that reflects the iterative nature of policy processes as embodied in policy enactment theory. It shows how a travelling policy of educational outsourcing is localised, shaped by the local discourses and the structural barriers and enablers, and how the localised policy, in turn, sets the path or direction of the subsequent policy process in individual schools. The path-dependency and path-shaping dialectics are depicted in the diagram by incorporating the actors' (dis)engagement with the thrust change.

The study also responds to the call for documentation of educational outsourcing practices in terms of how pedagogy-related decisions are made (e.g. Sperka, 2020). This study shows that school vision, initial setup costs (e.g. teacher workload, stakeholder learning), and funding affect schools' decision-making at the school level, affirming findings from other fields of studies such as political science or economic history. Other related policies, such as privatisation of public services, set a scene for the outsourcing of teaching, affirming the impact of “policy ensembles” (Ball, 1993, p. 14) (Hong Kong case). School-based management that allows for autonomy in annual budget deployment created a space for the practice of outsourcing, though off-record (Bates et al., 2021) (Korean case). Governments and educational practitioners may find these timely insights valuable in making informed decisions to respond to this global phenomenon of outsourcing of teaching.

We suggest the following direction for future studies. First, to address the limitation of this study that it mainly draws on the document analysis and the views of the decision-makers on the school side, future studies can integrate the voices of teachers and the educational outsourcing providers. Such design will enable analysing how the agency of various actors across levels collectively affects the outsourcing policy trajectory. Second, the replicability of findings in a larger context can be examined through a larger scale study, for instance, whether and how individual schools' school-level policy practices interrelate with other schools' practice, thus, affecting the trajectory of the policy at the societal level. Such investigation is worth attention, as some outsourcing providers work with multiple schools, some more than a hundred in the respective case contexts, creating a trend in policy translation. Such studies will help trace whether and how the school-level practice feedbacks into the societal level practice, and in return, the global level, in a bottom-up way. Finally, some significant local, national and global stakeholders in the policy network, such as policymakers, outsourcing service providers and parents, are not included in this study but some comments from the participants suggest that they also contribute to path-dependency and path-creation. Therefore, a longitudinal empirical study that explores how they interact with each other and shape the policy path over time is needed. Such study will help draw out a comprehensive, multi-level trajectory map of borrowed educational policies, verifying and extending understandings contributed by theorists (e.g. Hartong and Nikolai, 2017; Gulson et al., 2017).

Figures

Figure 1

Factors affecting the localising trajectory of a travelling policy

Table 1

Case schools, contracted services, interviewees

Notes

1. With policy enactment theory framing this study, the grassroots practice of outsourcing is considered part of the policy process, whether the actors are aware of the policy or not.

2. For discussion of the potential harms educational outsourcing and privatisation at large can cause or reproduce, including educational inequity and deprofessionalisation of teachers, see Bates et al. (2021).

3. At this school-level translation, decision-makers differed across schools. In Hong Kong School B and Korea School A, the principal made the overall blueprints for English education, and delegated some responsibilities such as writing the actual plans to others, but made the final decision. For instance, the principal of Hong Kong School B says she “will assign a teacher to take charge of each fund, and they give [her] the proposal, and then [she] make[s] the final decision”. In Korea School B and Hong Kong School A, the panel heads had a large degree of autonomy with the matters concerning their departments. Regarding the autonomy, the panel head of Korea School B noted that due to limited understanding of English pedagogy by the Principal or Vice Principals, she could have her own ways in English subject related decisions.

4. There are two main types of education service providers. Freelancers or one-person companies will make contracts with schools directly and teach; larger companies separate managers and teachers.

5. The school band represents the prestige of a school in terms of its academic performance. The band is decided by the grade of students who are admitted by the school. If the majority of the admitted students are high performing students, it becomes a Band 1 school, with Band 3 being the lowest. The Education Bureau does not release the data, however, there are private websites which publish schools' prestige according to their academic performance in the previous year, and other factors such as its public perception. The band of a school rarely changes.

6. Hong Kong School B declined to participate in the interview in 2019/20.