Introduction

On an international scale, school systems are struggling to find sufficient staff to meet demand (Allen et al., 2019; Sullivan et al., 2019). As the largest school system in the southern hemisphere, the New South Wales public education system is facing a significant staffing crisis. Projections indicate an additional 13,000 teachers will be needed in the system over the coming decade to meet enrolment growth (Norris, 2021). In response the system is investing $125 million over four years in recruitment. Substantial attention has been directed to issues affecting the attractiveness of teaching such as escalating workloads (Stacey et al., 2022), uncompetitive salaries (Gallop, 2021), and negative public images (Mockler, 2022), however what has remained under-explored to this point is the role of outside of the system factors on the current and aspiring teaching workforce.

Among the public services, schooling has a unique geographic scale. With all school-age children and youth (6 to 17 years) legally required to attend school, wherever there are children and youth there is a need for schooling. While under pandemic conditions, digital technologies made access to schooling multi-modal, physical schooling remains the default form of provision. Constitutionally, and articulated in the 2019 Alice Springs (Mparntwe) Declaration, the State is the guarantor of education to its citizens through a social contract. However, if housing costs make living near some schools inaccessible on a teachers’ salary, the execution of the social contract is compromised.

New teachers, those fundamental to addressing workforce shortages, are particularly vulnerable to unaffordable housing. Based on currently available data and the analysis presented in this paper assuming a single income household, there are 675 schools representing 22,703 full-time equivalent (FTE) teaching positions located in Local Government Areas (LGA) where the median rent for a one-bedroom dwelling is unaffordable for graduate teachers. The issue is not limited to new graduates. For top-of-the-scale teachers (non-promotional positions), there are 70 schools representing 2059 teaching positions located in LGA where a one-bedroom place is unaffordable. While drawing on empirical data from a single Australian state, the significance of this work extends well beyond state borders. Globally school systems are struggling to attract and retain teachers, and affordable housing is a significant, albeit somewhat invisible, policy issue.

More than just an empirical paper on housing affordability, the contribution of this work is to highlight the systemic implications of housing affordability on the teaching workforce. Taking the key focus to be the overall health of the school system, the objective of this study is to develop an improved and expanded data-driven approach to the relations between housing affordability and teaching salaries in New South Wales. Despite taking a descriptive rather than causal focus, the findings of this paper point to a larger sovereignty issue within the teacher shortage and its systemic implications.

Background and context

An intensifying shortage of affordable housing represents one of the most pressing policy challenges for governments in Australia (Pawson et al., 2019). Stagnant wages, or at least those not keeping pace with inflation, are an amplifier of housing affordability issues. To maintain adequate housing arrangements, many young Australians have been forced to live further away from major economic and employment hubs, choose smaller living spaces, or spend longer periods of time living with family (Abidoye et al., 2022; Yates, 2016).

While housing affordability has worsened in recent years and gained more attention from city planners, economists, unions, and politicians, peer reviewed research in education devoted to the topic has not kept pace (Dizon-Ross et al., 2019; Holme, 2022). What has been published has focused more on the impact of community gentrification and segregation on student outcomes (e.g., Berkowitz, et al., 2017; Pearman, 2019). This has allowed the systemic implications of housing affordability on teacher supply to advance without adequate attention (Eacott et al., 2022). As with other essential workers who could once afford to purchase their own home within a reasonable distance of their place of employment, teachers are being pushed to the fringes of greater metropolitan region or satellite housing markets as increases in housing prices have out stripped wage growth (Dempsey, 2017; Morris, 2021).

In many industries, the pandemic has re-cast the relationships between home and the workplace, making proximity less of an issue (Baxter et al., 2021). School based education is different. The provision of schooling has remained a primarily in-person activity. Any opportunities for re-thinking and re-designing long-standing rules and regulations covering how teachers live and work in Australia have been limited to tinkering at the edges while the core tasks remain relatively untouched. Through the Education Act 1990 (NSW) s 22, there is legislated compulsory attendance for school-aged children and youth. In combination with Commonwealth legislation (e.g., The Australian Education Act 2013) and 2019 Alice Springs (Mparntwe) Declaration, there is the establishment of a social contract between the State and its citizens in the provision of schooling. Needing to match the geographic spread of population the interplay of school provision and housing is more significant than the paucity of education research on the topic indicates. This is particularly so given the current teacher workforce issues and escalating housing costs in population dense locations.

In non-scholarly communities there has long been recognition for new kinds of reforms and incentives to teacher housing in the context of emerging teacher shortages (e.g., Honawar, 2006). A common project is the construction of teacher (or key worker) housing, sometimes referred to as ‘teacher villages’, and frequently tied to low or no interest loans that are forgiven if a specific period is served in the district (e.g., Duvall, 2017; Galley, 2001; Housing Issue, 2006; Nott, 2018). In Australia, initiatives have targeted rental rather than purchase pathways. Aware Super (one of the largest superannuation funds) has undertaken 15 projects across Sydney, Melbourne, Perth, and Canberra to deliver more than 1800 units, most of which will be rented at a 20% discount on market value. This includes a $70 million project in Miranda (24 kms south of Sydney CBD) located close to schools, hospitals, and public transport hubs, and with more than 50% of the 115 units available for key workers who pay 70% of the market rental (Coomber, 2018).

For an ownership pathway, the NSW Government is trialling a two-year $780 million shared equity scheme to help essential workers (e.g., nurses, teachers, and police officers) enter the housing market.Footnote 1 This scheme will allow up to 6000 eligible key workers and low-income households to own their own homes with a minimum of 2% deposit, but without the burden of lenders mortgage insurance and no interest on the Government equity share (up to 40% for a new home or up to 30% for an existing home) in the property. To be eligible the maximum gross income is $90,000 for singles and $120,000 for couples and the maximum purchase price is $950,000 for Sydney and major regional centres (Central Coast, Lake Macquarie, Newcastle, and the North Coast), and $600,000 in the rest of the state. This shared equity scheme creates a purposeful link between housing, income, school provision and systemic responsibility under the social contract. While new construction and population projections have impacted systemic decisions regarding where new schools are needed, housing and education policy have rarely been co-ordinated (DeBray, 2021).

Australia is a highly urbanised nation with population centred in major cities. These cities have large, sprawling, low-density housing, and favour car-oriented transport (Paijmans & Pojani, 2021). Culturally, Australia has long been a nation of homeowners. Home ownership is widely embraced as a housing policy priority with more than $20.5B (in 2021 terms) in stamp duty concessions and cash grants from governments to first homebuyers in the last decade (Pawson et al., 2022). These measures target those already close to entering ownership rather than assisting those who would otherwise have been unable to access an ownership pathway at the time. Even with these incentives, the rate of homeownership is falling as even high-income first home buyers struggle to save a deposit for an entry-level property in major cities (Gurran et al., 2022).

The importance of context or place is a core argument in education research (e.g., Roberts et al., 2022). Visibility and commitment to the community in which you teach, primarily demonstrated through home ownership in the area, has been identified as a major attribute of high-impact educators and school leaders (Eacott et al., 2021). Essential service personnel, such as teachers, nurses, and police officers are critical features of community fabric (Fernando & Hearne, 2017). Any sense of residential segregation achieved through cost of housing (Ng et al., 2022) will have a significant impact on not just employment accessibility but also connections between educators and communities—and potentially, student outcomes. With annual movement in the Consumer Price Index (CPI) the highest since 1990, with key rises in the costs of groceries and electricity,Footnote 2 areas traditionally seen as desirable and without staffing issues are becoming, if not already, unaffordable for teachers.

Addressing this issue requires long-term strategic thinking at all levels of government and mechanisms where investment in infrastructure is shared between government, landowners, and developers (Rowley et al., 2022). The unique policy architecture of Australian federalism, where states and territories have constitutional responsibility for education, but the Commonwealth has the fiscal resources has been part of the problem. Sovereignty over the teacher housing affordability problem is further diluted across portfolios. No specific department within state or territory government has ultimate responsibility for housing the teaching workforce, even though everyone has a stake in the issue. For example, while the NSW Minister for Education and Early Learning is responsible for ensuring ‘that every child receives an education of the highest quality’ under the Education Act 1990 No. 8 (NSW) s 4(c), Housing Services, which cover teachers and the policy force but only in rural and remote communities, is under the Department of Planning and Environment. The existing fragmentation of the problem of adequately staffing schools has led to an absence of checks and balances to ensure delivery of the social contract. Addressing this requires a shift in the unit of analysis from the individual teacher or even the school to the system. For it is only at the systemic level where a comprehensive analysis of the implications can be understood.

Data

This study exploits public data sets on housing costs (rental and sales), teacher salaries, and school locations (see Table 1), to investigate the extent of housing affordability as a contributor to the current and forthcoming teacher shortage in New South Wales, Australia.

Table 1 Summary and sources of data
Full size table

The primary spatial unit for analysis is the LGA as defined by the Australian Bureau of Statistics. An LGA is a legally designated part of a state or territory for which incorporated local governing bodies have responsibility. The choice of LGA is twofold. First, there is time series data on rental costs and house sales available allowing for a greater nuance in the argument. Second, as the local governing body, LGA have a significant role in housing development policies.

Housing data

As the basis for claims on the affordability of housing, data concerning weekly rental costs, and house (strata and non-strata dwellings) sales are used. Quarterly reports for sales and rental data are available from the NSW Department of Communities and Justice as Microsoft Excel spreadsheets.

Sales data are derived from information provided on ‘Notice of Sale or Transfer of Land’ forms lodged with Land and Property Information NSW. The quarterly rent and sales reports are considered the sole authoritative source of data. Reporting data is at the LGA level, and the Greater Metropolitan Region (GMR), its constituent parts (e.g., Newcastle SSD, Wollongong SDD, outer ring, middle ring, inner ring), and the rest of the state.

Strata title properties include town houses, terraces/villas, flats/units (multi-unit dwellings) whereas non-strata title properties refer to separate houses. An exception is any multi-unit dwelling with a Torrens title being counted as a non-strata property.

Rental statistics are derived from information provided on the rental bond lodgement form that is lodged with Renting & Strata Services Branch (RSSB) of the Office of Fair Trading.

Despite the presence of extensive historical data, new reporting procedures were introduced in the 2017 September quarter (Q3) impacting the validity of time series analysis at the sub-market level. Fortunately, this aligns with the introduction of a new salary scale for teachers in NSW.

Teacher salaries

In 2017 the NSW Department of Education re-structured its salary scale to align with the Australian Professional Standards for Teachers (APST). The former 13 Step salary scale (with new university graduates entering at Step 5) was replaced with a new three band (graduate, proficient, highly accomplished / lead [HALT]) approach (see Table 2). This is described in Schedule 2 of the Crown Employees (Teachers in Schools and Related Employees) Salaries and Conditions Award (NSW). Whereas progress in the previous scheme was linked to time served (one step per year served), this new approach is based on accreditation against the APST (although progress through Band 2 is based on time served, each step reflecting one year’s service).

Table 2 New South Wales teacher salary schedule, 2012–2022
Full size table

As it is impossible, within existing privacy laws, to obtain data on the household occupancy and income conditions for individual teachers, the analysis of this paper assumes a single-income household. While a limitation, it is equally problematic to assume that teachers will have, or need to have, a secondary income in the household to live within commuting distance of their school.

School data

For school data, ACARA profiles are used. These data are available in downloadable Microsoft Excel spreadsheets. Two specific data sets were drawn from. First, school location data (e.g., latitude and longitude) and geolocation structures (e.g., LGA) were used to identify the number of schools within an LGA. Second, the school profile data was used to identify the number of reported staff (head count and full-time equivalent [FTE]) and students at schools.

Analytical methods

After initial extraction of the data from various sources, the first analytical task was data linkage. This is a technique that connects pieces of information from multiple sources to create a new dataset. Australia has a long history of data linkage (Smith & Flack, 2021), but less so in education. Previous attempts in education have targeted student-level data and stalled due to complex legal and privacy matters within and across jurisdictions. However, by shifting the unit of analysis to the systemic provision of schooling, it is possible to curate datasets capable of informing systemic policy-making, program design, and further research.

Housing affordability analysis required thresholds for affordability. To establish the affordability of rental properties, the threshold levels used by the Rental Affordability Index (RAI) published by National Shelter, Community Sector Banking, Brotherhood St Laurence, and SGS Economics & Planning (see Table 3) are used. The RAI draws on the average gross weekly household earnings in a region and relates that to the median rental price of dwellings for which bonds were lodged in the last quarter. The key threshold value for housing stress is 30% or more of income on rent.

Table 3 Rental Affordability Index (RAI) threshold levels
Full size table

The 30% threshold for affordability is also used in the Australian Bureau of Statistics’ (ABS) recently introduced new variable, the Rental Affordability Indicator (RAID), in its Census of Population and Housing. Holme (2022) offers an overview of the history and critiques of the 30% threshold, mostly consisting of the arbitrary nature of the cut-off and the additional hidden costs of housing not captured in the 30% of gross income, but it does remain as a standard threshold used by government for identifying rental stress.

For housing prices, Demographia’s rating (see Table 4) is used. Demographia International Housing Affordability uses the median multiple to rate middle-income housing affordability. The median multiple is a price-to-income ratio, which is the median house price divided by the gross median household income. The key threshold is three times the annual gross income, with increasing unaffordability as the ratio increases.

Table 4 Demographia housing affordability ratings
Full size table

Descriptive data, presented in subsequent sections, was calculated using pivot tables within Microsoft Excel using the new dataset. To visualise the data QGIS v3.28.2 was employed. It is a free and open-source software used to work with spatial data. Maps were created for rental and house sales prices at the LGA level with a graduated colour (magma) ramp based on cost thresholds. Two key levels of teacher salary, Band 1 (graduate teacher) and Band 2.3 (top-of-the-scale proficient teacher) were used for analysis. Through a focus on new graduate salaries the analysis targets those newly recruited into teaching, and the top-of-the-scale enables some comment on the general affordability of housing—both rental and ownership pathways—for teachers.

To add predictive value for this work, Autoregressive Integrated Moving Average (ARIMA) modelling, a common time-series approach in econometrics, is used to relate housing costs to teacher salaries. ARIMA is a class of linear models utilizing historical values (e.g., weekly rental costs) to forecast future values. Non-seasonal ARIMA models are generally denoted ARIMA(p, d, q) where parameters p, d and q are non-negative integers, p is the order (number of lags) of the autoregressive model (AR), d is the degree of differencing (the number of times the data have had past values subtracted), and q is the order of the moving averages (MA) model. Given there is minimal change in rental costs by the seasons (compared to, for example, the availability and sales of fresh produce), there is no requirement to factoring in seasonality.

To establish the appropriate ARIMA model was a multi-stage process undertaken in IBM SPSS v26. Initially the data were checked through a sequence chart to identify the trend in the data. There was a positive trend for rental costs, and it was sufficient to warrant ARIMA modelling. Autocorrelations were then used to make the data stationary. This is achieved by correlating observations against a time lagged version of itself, removing the trend and allowing for a constant mean based on similarity in data points. The order of differencing (d) can then be identified—in this case, 1. The Autocorrelation Function (ACF), which plots the correlation coefficient against the lag, can be used to identify the q (in this case, 3), and the Partial Autocorrelation Function (PACF), is used to identify the p (3). Resulting from this procedure, the most appropriate model is an ARIMA(3,1,3). The choice of model was supported by the SPSS expert modeller (R2 = 0.996, Normalized BIC = 4.307) and was statistically significant (p = 0.028).

Analytically, the goal across all the methods described above is to identify how many LGA and then how many schools and teaching roles, are unaffordable on a contemporary teaching salary.

Findings

As an initial snapshot of the affordability of rental properties throughout the state based on teacher salaries, Fig. 1 provides an overtime (by quarter) comparison of rents (by bedrooms of dwelling) and the 30% threshold for Band 1.0 (new graduate) and 2.3 (top-of-the-scale) teachers.

Fig. 1
figure 1

Median weekly rent (by bedrooms) in NSW compared to teacher salary (Band 1.0 and 2.3), 2017–2022

Full size image

Figure 1 demonstrates that on an aggregate state-wide level, a newly graduated teacher in a single-income household experiences rental stress with all sizes of dwellings based on median weekly costs. In other words, it is unaffordable (on average) for a single-income household to rent on a new teacher salary in most of the state. Top-of-the-scale teachers, still assuming single-income household, fared better, with only a recent uptick in three- and four-bedroom locations exceeding the 30% threshold, although two-bedroom is similar ($530 to $529) to the 30%. The trajectory means that income is not matching rent rises.

Figure 2 displays the same analysis but focused on housing (non-strata and strata) affordability. Unlike rental properties, both non-strata and strata dwellings are above the 3:1 ratio of affordability for new graduate and top-of-the-scale teachers. In short, the median house sales price in NSW is unaffordable for a single income teacher family.

Fig. 2
figure 2

Median house sales prices (NSW) and teacher salary (Band 1.0 and 2.3), 2017–2022

Full size image

To provide greater nuance, Fig. 3 displays the affordability using median non-strata house sales from the June quarter (Q2) 2022 at the LGA level. Almost the entire eastern seaboard (where most of the population lives) is severely unaffordable for a top-of-the-scale teacher. To better understand the systemic impact of this unaffordability, Table 5 displays the reported number of teachers (head count, FTE), non-teaching staff (head count and FTE), and enrolled students in LGA based on affordability ratios.

Fig. 3
figure 3

Non-strata purchase affordability, Band 2.3, June Quarter 2022

Full size image
Table 5 Number of schools (by type) located in LGA sorted by affordability of non-strata purchases for a top-of-the-scale (Band 2.3) teacher, June quarter 2022
Full size table

For the NSW public school system, using the assumption of a single income household, 87% of teaching positions (n = 50,519) are in LGA (with data for the June 2022 quarter) where non-strata housing ownership is severely unaffordable, with the impact felt slightly more for primary school (89%) than secondary school (87%) roles. In contrast, only 2% of teaching positions are in LGA where the median non-strata house sales price is affordable. Put simply, a single teacher income household—even top-of-the-scale—is unable to enter home ownership in all major markets.

Strata properties are often thought of as cheaper than non-strata properties and Fig. 2 reflects this gap. However, there is greater likelihood of strata properties being in larger housing markets where population is denser. Figure 4 displays the median LGA sales price for strata properties compared to top-of-the-scale teacher salaries.

Fig. 4
figure 4

Affordability of strata properties, by LGA, Band 2.3 Teacher, June quarter 2022

Full size image

The context-specific nature of the Australian housing market means that strata properties are less common than non-strata. As a result, far fewer LGA had data in the last available quarter (June 2022). However, as with non-strata properties, all major eastern seaboard markets are not affordable to households with a single teacher—even top-of-the-scale—salary. Consequently, many schools in what have traditionally been thought of as desirable areas are not accessible to individual teachers or those families dependent on a single teacher salary.

Table 6 breaks down the number of schools and reported teaching (FTE) positions in LGA against the level of affordability. Based on the available data in the June 2022 quarter, 1226 schools representing 64% of teaching positions (n = 37,101) in the state, are in severely unaffordable LGA (n = 44). In comparison, only 2 LGA with 21 schools and 556 teaching positions are affordable.

Table 6 Overview of LGA, Schools, and Teaching position by affordability of strata properties for a top-of-the-scale (Band 2.3) teacher, June 2022
Full size table

Addressing the teacher shortage, at least from an attraction or recruitment standpoint is dependent on new graduates. Table 7 displays a heat map of the affordability, since quarter three 2017, for a one-bedroom rental for a new graduate teacher across the major reporting areas of the Greater Metropolitan Area ([GMR], inner, middle, and outer ring, including Newcastle and Wollongong sub-markets) and the Rest of the State.

Table 7 Rental affordability, graduate teacher (Band 1) salary, for a one-bedroom place, 3rd Quarter 2017–3rd Quarter 2022
Full size table

Unlike house sales prices, there are more affordable areas for rent with a new graduate teacher salary, especially in peripheral markets of Newcastle and Wollongong. It is the unaffordability (moderate-severe), of a one-bedroom place for a new graduate in the outer, middle, and inner ring of Sydney that creates an issue for the system in making teaching in many schools accessible to all potential applicants.

To further nuance this unaffordability in the GMR, Figs. 5, 6, 7, 8 and 14 are maps visualising the affordability of a one-bedroom dwelling for a new graduate at the LGA level. Figure 5 provides the overview of the entire GMR (inner, middle, and outer ring, plus Newcastle and Wollongong SSD) by LGA. While the periphery remains affordable (e.g., Newcastle, Wollongong), the closer one gets to the centre the more unaffordable rents become. To further demonstrate this, Figs. 6, 7 and 8 displays the outer (Fig. 6), middle (Fig. 6), and inner rings (Fig. 7) of the GMR. The outer ring remains somewhat affordable, but the inner and middle rings have many LGA that are seriously and severely unaffordable. The inner ring is exclusively seriously and severely unaffordable.

Fig. 5
figure 5

Affordability of median rental costs in the greater metropolitan area for a graduate teacher (Band 1.0), April–June Quarter 2022

Full size image
Fig. 6
figure 6

Outer ring, greater metropolitan region, LGA level of affordability of a one-bedroom rental for a new graduate (Band 1.0), September quarter 2022

Full size image
Fig. 7
figure 7

Middle ring, greater metropolitan region, LGA level of affordability of a one-bedroom rental for a new graduate (Band 1), September quarter 2022

Full size image
Fig. 8
figure 8

Inner ring, greater metropolitan region, LGA level of affordability of a one-bedroom rental for a new graduate (Band 1), September quarter 2022

Full size image

The unaffordability of the inner ring, and similar issues with the middle ring, mean that schools can become difficult to access. To demonstrate the systemic scale of this issue, Table 8 provides an overview of the number of public schools, their teaching staff (both head count and FTE), non-teaching staff (head count and FTE), and the number of students enrolled, as reported to ACARA. There are over 3500 teaching positions where it is at least seriously, if not severely, unaffordable for a new graduate teacher to rent a one-bedroom place in the inner ring of Sydney.

Table 8 Overview of public schools in the Inner Ring, 2021
Full size table

Expanding the analysis across the entire state, for a graduate teacher, there are 10 LGA that are moderately unaffordable (234 schools, 7683 teaching positions),Footnote 3 17 that are seriously (371 schools, 12,961 teaching positions),Footnote 4 and 4 LGA that are severely unaffordable (70 schools, 2059 teaching positions).Footnote 5 Not limited to the GMR, the systemic implications of having 675 schools and 22,703 teaching positions in LGA where the median rental for a one-bedroom place is unaffordable for a new graduate makes it difficult to attract new teachers into those area. While it may be possible to argue that these are desirable areas, inaccessibility is still an equity problem for the system and potentially obscures a growing staffing issue within a larger teacher shortage.

To highlight the significance of this data, Fig. 9 displays the forecasting projections of one-bedroom rental costs into the immediate future (through to 2027) and overlaid with the 30% of a new graduate income from 1990. Projection assumptions are that teacher salaries will rise by 3% annually.Footnote 6 In the period 1990–2008, the new graduate salary was greater than rental costs, but the gap was 4.5% in 2012, rising to 19% in 2017 and reaching 5.8% in 2022. Projections based on current trajectories have the gap at 12.7% in 2025 and increasing to 15.6% in 2027—a gap of $79 a week. In sum, the level of unaffordability of a one-bedroom dwelling for new graduates is rising.

Fig. 9
figure 9

ARIMA(3,1,3) model of median rental costs of a one-bedroom dwelling in NSW against 30% of new graduate teacher income

Full size image

This accessibility issue is not limited to new graduate positions. There are four LGA (Bayside, Canada Bay, Sydney, and Waverley), all in the inner ring, where a one-bedroom place is moderately unaffordable on a top-of-the-scale teacher salary (70 schools, 2059 teaching positions).

To take a state-wide look at affordability and its systemic implications (e.g., the number of teaching positions affected), Table 9 displays a heatmap of the median sales price (all dwellings) ratio (June quarter, 2022) and median rental percentage (September quarter, 2022) against the top-of-the-scale teacher salary at the LGA level. Excluding 13 LGA with missing data in either sales, rent or both,Footnote 7 there is a significant housing affordability issue in NSW for households’ dependent on a single teacher income. A total of 91.2% of teaching positions (90.8% if counting the excluded LGA as affordable) are in LGA where both rental and ownership pathways are severely unaffordable.

Table 9 Heatmap of median rental (June quarter) and ownership (September quarter) affordability (all dwellings) for LGA, NSW for top-of-the-scale (Band 2.3) teacher income
Full size table

Discussion

Addressing the teacher shortage is dependent on increasing the supply of qualified teachers to at least meet, if not exceed, demand. Fundamental to achieving this goal is introducing new candidates into the profession. Assuming that demand can be met, ensuring that schools are accessible for teachers is important. The analysis presented in this paper has demonstrated that newly graduated teachers are especially vulnerable to housing affordability. Across the state some 22,703 teaching positions are in LGA where a one-bedroom dwelling is unaffordable on a new graduate teacher salary. The entire inner ring of the GMR, representing 107 schools and 3551 teaching positions is unaffordable. This introduces the very real prospect of schools in densely populated areas without teachers.

Housing affordability is a significant policy issue in Australia (Pawson et al., 2019) and a fast emerging one for education systems. Rental pressures are increasing, and home ownership is not just unlikely, but impossible for a single teacher income household. Throughout NSW, 90.8% of teaching positions, over 50,000 FTE are in LGA where the median rent and house sales price are severely unaffordable on a top-of-the-scale teacher salary. The challenges of housing affordability are complex, diverse and play out differently throughout the state. Systemically, the prospect of inaccessible schools due to housing cost stresses is a very real prospect. In the context of the social contract of education, government (at all levels) have a legitimate role and indeed responsibility to use policy interventions to improve the efficiency, efficacy, and affordability of housing for the teacher workforce.

Guaranteeing sufficient staff for schools is a matter of national importance. The cascading effects through not just school systems but communities and the economy are substantial. Investment in education is a social good. The NSW Government’s $780 million shared equity scheme reflects a focus on housing affordability for key workers. However, median non-strata house prices in the inner, middle, and outer ring, plus Wollongong SSD already exceed the maximum purchase price for the scheme. Orthodox saving for house purchase deposit assumptions (20% of gross income) are difficult to sustain if under rental stress or forced to live further away and introducing additional travel costs. Travel by car remains the dominant mode of transport to work (53% in 2021 census), with the median commute for NSW ‘education and training’ professionals 9.28 kms (M = 15.48, SD = 20.23). The latest (Q3 2022) Australian Automobile Association transport affordability indexFootnote 8 indicates that the average Sydney household spends $25,256 annually in transport costs. Combined with rising CPI, travel costs mean housing affordability issues in major centres are not going away.

To address the teacher shortage requires integration of policy and reforms across multiple portfolios (DeBray, 2021). Nationally, 98% of housing is delivered by the private sector (Rowley et al., 2022). In the absence of a major public funded essential workers housing initiative, any shift in the supply of affordable housing will be a shared responsibility of government, landowners, and developers. This highlights the sovereign complexity of the issue. No specific government department nor the private sector has ultimate responsibility for delivering teacher housing, but all have a stake. Arguably the greatest stake is held by the NSW Department of Education. However, its $125 million teacher supply strategy makes no reference to affordable housing. Existing locality allowances for eligible schools focus on climatic disability (e.g., hot and cold), isolation from socio-economic goods and services or motor vehicle.Footnote 9 There are no allowances for housing affordability. Analysis in this paper has demonstrated that timely action is required if the system is to fulfil its obligations to deliver equitable and excellent schooling to the state’s citizens.

Solutions require a mixed strategy of incentives for developers to construct affordable properties, subsidies for transportation or commuting costs for key workers, and increased salaries or allowances for staff living in unaffordable areas. Whatever the approach, there is a need to deliver new supply of affordable housing if all schools are to be staffed at capacity. Without sufficient volume of affordable housing for those currently excluded or marginalised, the affordability of housing will not alter. Cross-subsidising new construction (Benedict et al., 2022), such as the Aware Super funded projects, are a common response but their success hinges on a cultural change where renting is not seen as a short-term transition period before home ownership. Renting de-centres proximity to work as costs become prohibitive. Existing subsidies for teachers do not address journey to work, having considered place of residence an individual choice. If housing affordability is making living close to schools not just problematic but financially impossible, the viability of many schools and the system is called into question.

Limitations

The current study should be interpreted within the constraints of several limitations. First, the guiding assumption of the analysis was a single teacher income household. Although this is an appropriate approach for investigating housing affordability for teachers across the state, it will be important to conduct further research with a greater array of household income arrangements to better model housing affordability and access to schools across all LGAs. Second, the analysis conducted drew on publicly available data which while replicable (both in NSW and other jurisdictions), a more comprehensive analysis could include participant reported data to build a more nuanced account of accessibility to schools and housing affordability and its impact. Third, only touched on in this paper but open to further analysis is commuting distance and its impact (time, financial) on educators and the accessibility of schools. A practical, or engagement and impact strength of the present work is that it provides systemic authorities, government, professional associations, unions, and developers with an evidence informed basis from which to build a collective and distinctive context sensitive suite of options for engaging with the problem of housing affordability for educators.

Conclusion

When the gap between the supply and demand of teachers reaches a certain tipping point, a school system faces a major crisis. As demonstrated in this paper, factors outside of the orthodox view of the school system are creating a rapidly escalating tipping point as many schools not traditionally considered difficult to staff become inaccessible for teachers. Contextually situated solutions, however partial, require not just a whole of government response but a mixed strategy of public and private investment. This highlights the sovereignty crisis of the current and forthcoming problem of the teacher shortage—no one body owns the crisis, but everyone has a stake in it. Failure to address the issue is to abdicate responsibilities of the State to provide high quality schooling to its citizens. Asserting jurisdictional independence and sovereignty means that problems (e.g., housing affordability and access to schools) can deteriorate substantially before any remedial action is taken. Blurring sovereign boundaries can initiate important failsafe mechanisms targeting the overall health of the school system.