The Big Short: Early Childhood Education Expansion in Post-Pandemic Canada

by
Brad Seward – University of Toronto
Elizabeth Dhuey – University of Toronto
Annie Pan – University of Toronto

Abstract:

The demand for early childhood educators is expected to grow as the Canadian economy
recovers from the COVID-19 pandemic and Canadians continue to return to work. To support
this transition, the federal government has signed bilateral agreements with Canada’s territories
and provinces to invest more than 30 billion dollars to create a “10-dollar-a-day” universal child
care system. This ambitious movement is a landmark in Canadian child care, yet it is unclear
whether Canada’s current pipeline of early childhood education (ECE) graduates is sufficient to
meet this increased demand. Using data from the Education and Labour Market Longitudinal
Linkage Platform (ELMLP), we find that early childhood education program graduates tend to
be concentrated in relatively few provinces, come from primarily college backgrounds, and
acquire considerably modest labour market outcomes. We investigate the career transition of
ECE graduates and argue that low pay and its signal of the devaluation of the child care
workforce are contributing factors for field exit among ECE professionals. Policy implications of
this potential shortfall in the supply of ECE professionals are discussed.

1
1. Introduction

The past three years of the pandemic have underscored the necessity and importance of publicly
available child care for Canada’s productivity and efficiency (Friendly et al., 2020). The
government of Canada has acknowledged the need to provide universal, affordable, and high-
quality child care to Canadian families. Starting in 2021, the federal government committed a
staggering $30 billion to bolster the early child care sector over the next decade (Department of
Finance Canada, n.d.; Child care Resource and Research Unit, 2021). 1 The most prominent of
0F

this investment is a promise to enact a national $10-dollar-a-day child care effort in collaboration
with each province and territory. Following this ambitious announcement, however, Canada
experienced a third, and then the fourth wave of COVID-19, followed by a federal election call
in late 2021. The election threatened to halt the inertia Ottawa had generated to sign bilateral
agreements with the provinces (Pacini-Ketchabaw & Prentice, 2021a). However, as of March
2022, the federal government has signed agreements with all territories and provinces.

The demand for early childhood educators is again growing as the economy recovers from the
pandemic and most Canadians return to in-person work. Enrollments in child care centres have
recovered considerably from the earliest stages of the pandemic—an average of 43 children per
centre in 2021 compared to eight children in 2020—signalling a gradual return to the pre-covid
mean enrollment of 57 (Vickerson et al., 2022a). While persisting fears of the virus and the
wider acceptance of remote work may cause some parents to keep their children out of
traditional early learning and child care (ELCC) in the short term, there is little doubt that there
will be increasing demand for early childhood educators (ECE) due to the large federal
expansion.

And yet the turbulence of the past few years has weakened Canada’s supply of early childhood
educators. A reported 58 percent of centres laid off either all of their staff or all but the director
in the earliest stages of the pandemic (Friendly et al., 2021). 2 More troubling, however, is that
1F

more than a quarter of these centres had not hired back the staff lost during that first wave in the
following months, which may have forced these ECEs into other career streams (Vickerson et
al., 2022a). Subsequent waves of the pandemic have been slightly more promising for ECE
workers with just under a third of centres experiencing layoffs, particularly as ECE centres have
been deemed essential and permitted to stay operational. 3 However, it remains to be seen
2F

whether Canada is heading towards a shortage of ECE professionals, particularly in light of the

1
To put this figure into perspective, prior to the massive disruptions caused by COVID-19, early childhood care and
education programs in Canada were undergoing what was previously considered a significant development. In 2017
the federal and provincial governments adopted a multi-lateral framework to support Early Learning and Child Care
(ELCC), focusing on improving the sector across the areas of accessibility, affordability, quality, flexibility, and
inclusivity (see Child care Resources and Research Unit, 2021). The first phase of this three-year bilateral agreement
concluded just prior to the onset of the pandemic, in March of 2020, with an estimated $1.195 billion having been
dedicated to the program.
2
Between the period of March to May 2020, when most centres were closed after Canada experienced the first wave
of the pandemic, 71 percent of ECE providers reported laying off some, or all their staff (Vickerson et al., 2022a).
3
With a national average of centres reporting layoffs at 29 percent, New Brunswick was the least likely to have laid
off staff (14 percent of centres) and Alberta at the other end of the spectrum (40 percent of centres experiencing
layoffs) (Friendly et al., 2021).

2
federal and provincial/territorial governments’ commitments to the sector. All the provinces and
territories have committed to creating new ECE spaces in early learning and child care centres as
part of their agreement with the federal government. Alberta, for example, is planning to add
42,500 new full-time spaces by 2025-26 (Government of Alberta, n.d.-a), and British Columbia
has committed to 30,000 within five years and 40,000 within seven years (Department of
Finance Canada, n.d.). Yet to recruit enough educators for these new spaces and ensure the
provision of high-quality child care, training enough graduates and journeypersons in early
childhood education becomes more than essential. 4 3F

Staffing shortages are already a known issue in the child care workforce. Concerns for staffing
shortages were the single most commonly communicated worry among ECE operators in 2021—
even greater than concerns for financial stability and operational capacities (Vickerson et al.,
2022a). At least part of the issue can be attributed to the returns to careers in the ECE field,
issues which the influx of support from the federal and provincial governments may be able to
address. To that point, financial concerns ranked high among the issues reported among ECE
employers, with over 40 percent of providers reporting difficulties meeting wage and salary
costs, and another 21 percent experiencing difficulties affording benefits (Vickerson et al. 2022).

Yet an issue that has received virtually no acknowledgement among the public policy and
academic communities is whether the pipeline of graduates from ECE higher education programs
will be enough to meet the increased demand proposed by governmental investments in early
child care. To our knowledge, no empirical assessments exist on whether the graduation rate will
be too low to significantly undermine this sector revitalization.

Part of the problem is a result of empirical limitations, where data linking the supply of new ECE
entrants to employment trends over time has been difficult to access (Friendly et al., 2021;
Beauregard et al., 2020). Instead, researchers have had to accept somewhat of a trade-off in what
could be captured. For example, aggregate analysis of ECEs has relied upon the information
provided in the Canadian Long-Form Census (i.e., Beach 2013). However, the discontinuation of
the data in 2010 means that these analyses have largely focused on the early-to-mid 2000’s
(Halfon 2021).

Statistics Canada employs several other surveys that capture ECE graduates’ labour market
outcomes (the National Graduate Survey) and the child care workforce more broadly (the
Canadian National Child Care Study, and the National Longitudinal Survey of Children and
Youth), though these traditional sources of data do not provide adequate detail on the
proportionately small population of graduates in ECE-related programs specifically, or otherwise
are becoming outdated. Other attempts have tried to collect data on ECEs directly (Flanagan et
al. 2013; Beach and Flanagan 2007; Doherty, Lero, Goalman, LaGrange & Tougas 2000;

4
We define journeypersons according to the Statistics Canada designation: Registered Early Childhood Education
(RECE) journeypersons are apprentices who have successfully completed their entire apprenticeship programs and
have received a certificate of qualification.” (see Statistics Canada, 2020a)

3
Cleveland & Hyatt 2000; Schom-Moffatt 1984), though the samples collected similarly fall short
of what would be required to carry out the analysis presented in this paper. 5 4F

Using Statistics Canada’s recently realized Education and Labour Market Longitudinal Linkage
Platform (ELMLP) researchers now have access to unprecedented precision regarding the school
to work characteristics of Canadian graduate cohorts. With this data, we can utilize
administrative data linkages to examine graduates of early childhood education programs in
novel ways that were previously out of reach for the research community. 6 5F

We therefore use the ELMLP data to focus on the following three research questions aimed at
understanding the near future of early childhood education:

1. In Canada, given historical levels of ECE training, how much will the supply of ECE
graduates need to grow to meet demand due to the ELCC agreements?
2. What are the demographic characteristics, or profiles, of these ECE graduates and what
are their earnings after entering the labour market?
3. What is the career and earnings trajectories among graduates of ECE programs? How
quickly do these graduates enter into an ECE-related career after graduation? Do they
remain in or exit from the ECE-related career over years? Is there an association between
retention and exit behaviours and earnings?
2. Canadian Context

In Canada, many different child care options exist, ranging from receiving care from an
individual such as a nanny or home child care provider to receiving care in a regulated centre-
based program. Canadian ECEC services – similar to health, social services, and education - are
primarily under the jurisdiction of the provinces and territories and each has developed their own
regulated child care system. Therefore, there is much variation across the country. For instance,
nine out of 13 provinces/territories require ECE professional certification for some fraction of
their staff in regulated care. Nine out of 13 require kindergarten educators to have ECE
qualifications. (see McCuaig, Akbari, and Correia, 2022 for provincial and territorial
differences). No jurisdiction requires that all staff hold a post-secondary level credentials. Two-
thirds of staff working directly with children holding qualifying credentials is considered the
international minimum and only six jurisdictions have two-thirds or more requirement in their
regulations. For instance, in Saskatchewan, New Brunswick, Northwest Territories, and Yukon
only half need to be qualified, whereas Alberta only requires one-third, both British Columbia
and Ontario require three fifths and Newfoundland and Labrador, Prince Edward Island, Nova

5
The largest and most comprehensive survey project in this area was conducted in 1998 as a part of a series of
reports under the banner “You Bet I Care!” originally carried out by the Child Care Advocacy Association of
Canada, before it was transferred to an academic team at the Centre for Families Work and Well-Being at the
University of Guelph in 2000 (Doherty, Lero, Goalman, LaGrange & Tougas 2000). The data contains information
on 4,154 staff across 848 centres representing all 10 provinces, the Northwest Territories, and the Yukon (Halfon
2021).
6
We recognize the very valuable work carried out by the Child care Resource and Research Unit, which has
collected administrative data on the child care workforce since 1992. However, this data has primarily been used to
generate summary reports on wages and changes in regulatory requirements in the field, and has not been used to
track the transition into and out of the field among postsecondary graduates in the ways presented in this paper.

4
Scotia, Manitoba, and Quebec all require two-thirds (the international standard for the lower
bound acceptable rate) of the staff to be ECEs or RECEs. Even prior to the expansion of the
ELCC system, educator shortages have led to legislation in a number of provinces to reduce
qualification requirements including in Ontario and Alberta. All jurisdictions allow for allowance
for a child care to operate without the required qualified staff on a temporary basis (McCuaig,
Akbari, and Correia, 2022). Additionally, child care providers can operate in the non-profit or the
for-profit sectors.

Training to become a qualified staff in the ELCC system in Canada also varies. The length of
program can differ by learning institution attended and by jurisdictional regulations. There are a
variety of college programs in Early Childhood Education that can range in length from
accelerated programs that take only eight months to college programs that take between 2-4
years. In addition, traditional four-year university programs exist along with a few Master’s and
Ph.D. programs in early childhood education. Some provinces, such as Ontario, require
individuals to become members of the College of ECE to use the designated titles of RECE or
ECE. The pathway to become a RECE is under the jurisdiction of the Ontario College of Trades
and is consists of on-the-job and in-school training. It requires 5280 hours of on-the-job work
experience and 730 hours of in-school training, which generally takes about three years. No other
province or territory has RECE certification.

Child care is a service that is bought and sold in a market that contains a number of market
failures (see Tekin, 2021 for a nice review). First, the benefits of high-quality child care affect
not only the children and families that are receiving the care but also reduces a number of
societal costs such as the effects of poor labour market outcomes, increased crime and teenage
motherhood (e.g. Duncan and Magnuson, 2013; Magnuson and Duncan, 2016; Currie 2001,
Carneiro and Ginja, 2014). Like education, individuals will under-invest if they do not consider
the societal costs of under-investment. In addition, parents are very bad at recognizing high-
quality child care (Blau, 2001; Gordon et al., 2021; Herbst et al., 2020) and child care is often a
large component of a family’s budget (Cleveland, Krashinsky, Colley, & Avery-Nunez, 2016;
OECD, 2020). Most parents do not have the ability to borrow against their own future labour
income to pay a higher fraction of their income on child care (Caucutt and Lochner, 2004; Cunha
and Heckman, 2004). Therefore, with all these market frictions, it is likely that the usual natural
adjustments to ECE wages that would happen in a traditional market may not occur and that
government intervention for wage support may be necessary. In fact, many OECD countries
view child care as a responsibility of the government and not necessarily individual families
(Tekin, 2021).

3. Data

We utilize confidential administrative information from Statistics Canada’s Education and

Labour Market Longitudinal Linkage Platform (ELMLP). The ELMLP is a platform that allows
researchers to access rich information about past cohorts of postsecondary students and
registered apprentices linked to their education and training records and their Canadian tax files. 7 6F

7
While the ELMLP is currently Canada’s premier data source on education-to-work pathways, it unfortunately
suffers from similar coverage issues as other Statistics Canada surveys that exclude for-profit institutions.

5
For this analysis, we will use three administrative datasets contained within the platform: (1) the
Post-Secondary Information System (PSIS), 2009-2018, 8 (2) the Registered Apprenticeship
7F

Information System (RAIS), 2008-2018, and (3) the tax records -T1 Family File (T1FF), 2004-
2018. 9
8F

When drawing from the PSIS 2009-2018 files, we target ECE graduates who have completed
their program from 2010 to 2017 10, 11 and document their sociodemographic profiles as well as
9F 10F

the institutions they graduated from. Then we link these graduates to their longitudinal tax files
to allow us to follow their career and earnings trajectories for a period of five years after
graduation. Similarly, using records contained in RAIS, 12 we also explore Registered Early
11F

Childhood Education (RECE) journeypersons who have completed apprenticeship programs and
have received a certificate of qualification from 2009-2017 in this analysis. To ensure a proper
comparison group, we included only those graduates that had employment income records for
the full observation period in our analyses—therefore graduates that experienced bouts of
unemployment, or failed to file their taxes were excluded from analysis. 13 This provides us with
12F

all programs in Canadian public postsecondary institutions and certified trade or apprenticeship
programs that may lead to careers as an ECE or RECE.

Accordingly, our analyses are unable to tap into the share of graduates who acquired their education from private
colleges and universities.
8
Information of 2009 to 2018 are available for all provinces in PSIS. However, 2005-2008 are only available for
maritime provinces.
9
The PSIS is a national survey providing detailed information on graduates of all Canadian public postsecondary
institutions including enrolments and programs. And the RAIS collects information on individuals who receive
vocational training and those who obtain certification via a trade/apprenticeship. So, the ELMLP covers all
programs in Canadian public postsecondary institutions and certified trade or apprenticeship programs that may lead
to careers as a ECE or RECE. Although private postsecondary institutions are not covered in ELMLP, this is the
only platform which contains this level of information of Canadian graduates with linkages to their administrative
tax data after graduation.
10
Due to the design of the PSIS, the reporting year cycle is from May to April rather than a calendar year period
from January to December. Therefore, two consecutive reporting years in PSIS are required to cover a completed
graduate cohort who graduate from January to December in a given year. For example, graduate cohort 2010 is
derived from both PSIS 2009 (May 2009 - April 2010) and PSIS 2010 (May 2010 - April 2011). In this case,
graduate cohorts 2009 and 2018 are incomplete and excluded from this analysis.
11
A graduate might obtain multiple credentials in the same year and these credentials could be either ECE-related or
not. In our analytical sample, we focus on graduates who have obtained ECE credentials only because people who
have also obtained credentials in other fields might have different career goals compared to those with only ECE
credentials. Additionally, among graduates with multiple ECE credentials in the same year, we follow four steps
suggested by Statistics Canada to break the tie in order to include only one record per graduate for a given cohort.
“Step 1: Keep the record with the highest PSIS program type, e.g. graduate program level is retained over
undergraduate level. Step 2: If program types are the same, then keep the record with the highest PSIS credential
type, e.g. degree is retained over diploma. Step 3: If credential types are also the same, but one educational
qualification is in the field of study of ‘Education’ then it is retained over the other fields of study. Step 4: If
credential types are also the same, but neither educational qualification is in the field of study of ‘Education’ then
one of the fields of study is chosen at random.” For more details, see (Statistics Canada, 2020b)
12
The reporting year cycle in RAIS is based on a calendar year period. Therefore, the RECE journeypersons cohorts
2009-2017 are derived through the same years in RAIS (2009-2017).
13
According to CANSIM, the tax filer rate was approximately 94% in 2018 (the latest year in our data). This means
that approximately 6% of our target demographic (new graduates largely between the ages of 25 to 29 years old) are
not represented in the T1FF data.

6
4. Descriptive analysis

4.1 How much will the supply of ECE graduates need to grow to meet demand due to the
ELCC agreements?

It is assumed that most of the new supply of ECE workers to staff the child care expansion in
Canada will have to come from new graduates. Therefore, it’s imperative to examine the supply
of ECE graduates in Canada. Appendix Table 1 lists the institutions in Canada that have trained
ECE graduates in 2017 (the most recent cohort available in our data). We report the institution
names and how many ECE students have graduated from each institution by province and by
program type: college program, above college and below bachelor, and bachelor program.

Understanding the geographical distribution of graduates within a province is important as the

role of geographic distance to school has a strong effect on the probability of attending in Canada
(Frenette, 2002). Luckily the pandemic has increased the ability of post-secondary universities in
providing online education which will be paramount to make sure that the supply of ECE
graduates is located in the communities in which they are demanded. Table 1 shows the average
number of ECE graduates per year by province. As shown, Ontario and Quebec are the two
provinces that have trained the majority of ECE graduates every year, 2,880 in Ontario and 2,170
in Quebec, which counts for 78.7 percent of the ECE graduates in Canada per year. Interestingly,
those two provinces only have 60.9 percent of children under six in Canada, so they are
graduating a higher percentage of ECE’s compared to their level of children. This could to due to
a higher percentage of parents opting to send their young children into child care and the demand
for ECEs in the education system within the provinces or it could be due to the capacity to train
the number of educators that is demanded.

Table 1: Average number of ECE graduates per year

Average Percentage Percentage
number of of of
graduates Graduates Children Difference
Panel A: ECE college/university programs
Alberta 390 6.07% 13.74% 7.67%
British Columbia 450 7.01% 12.33% 5.32%
Manitoba 140 2.18% 3.94% 1.76%
New Brunswick 150 2.34% 1.80% -0.53%
Newfoundland and Labrador <30 < 0.47% 1.13% -
Nova Scotia 40 0.62% 2.28% 1.66%
Ontario 2880 44.86% 38.02% -6.84%
Prince Edward Island <30 < 0.47% 0.36% -
Quebec 2170 33.80% 22.86% -10.94%
Saskatchewan 130 2.02% 3.52% 1.50%

Panel B: Registered ECE

Ontario 370 100%

7
 Sources: Post-Secondary Information System (PSIS) 2009-2018; Registered Apprenticeship
Information System (RAIS) 2010-2017: Statistics Canada Table 39-10-0041-01
Notes: For Newfoundland and PEI, the average number of graduates per year in each province cannot
be released due to confidentiality rules.

Given these average number of ECE graduates, it is important to examine how likely the historical
levels of ECE graduates will be able to fill the newly promised child care spaces contained in the
ELCC agreements with each province. The number of promised new full-time spots for each
province is listed in Table 2 14. We estimate the number of new staff needed for those jobs by making
13F

a few assumptions. First, the agreement in each province generally only provides the net new spaces
that are being promised and does not disaggregate the spaces by age group. This is important as
staffing ratios are based on age groups. For this calculation, we assume that the spaces will be
allocated evenly across age groups and use the staff/child ratios for each province to calculate the
total number of staff needed for these additional new spaces. We next assume that child care
providers will only staff their child cares with ECEs only to the legislated minimum which is listed in
Table 2 as the ratio of qualified staff required. Using this ratio and the number of total staff needed
we can estimate the number of new ECE graduates needed assuming that all new spaces will be staff
with new graduates instead of ECEs that have previously left the sector. Finally, using the average
number of new ECE graduates per year we can estimate how many years needed to train enough new
ECE graduates to fill the promised spaces. We find that this varies widely from only 1-3 years in NB
PE, and ON to 69 years in AB (despite requiring the lowest ratio of qualified staff in the country).
Given the number of assumptions, the estimates in Table 2 need to be interpreted with care but
highlight the large differences in training capabilities across provinces and promised new child care
spaces.

4.2 Who are ECE graduates likely to be?

Table 3 presents a demographic snapshot of ECE graduates and RECE journeypersons who
completed the degree or apprenticeship. On average, college ECE graduates complete their
certificates at an older age than those who complete a university ECE degree. Additionally, RECE
journeypersons typically complete their apprenticeships at an older age than both college and
university ECE graduates. We see that women are significantly dominant in this field (under five
percent are men), which is consistent with other studies in this area (Halfon 2021). This disproportion
among women and men is true for both ECEs and RECEs. Immigrants and international students are
underrepresented, especially at the university level among ECEs. 15 About 50 percent of college ECE
14F

graduates are married/common-law (including whether they were ever married or common-law) by
one year after graduation. This proportion is even higher among college ECE graduates in Quebec
(62 percent) and RECE journeypersons in Ontario (74 percent). Also, about two-thirds of graduates
are parents with at least one child by one year after graduation. High proportions of graduates in
parenthood and being married/common-law is likely because ECE graduates are typically older than
graduates from other fields (an overall average of roughly 27 years old). Over half of ECE graduates
hold multiple jobs one year after graduation, and this high proportion might suggest that newly ECE

14
Data from Quebec is missing from Table 2 because they will not adhere to the new Federal framework and will
continue to receive federal funding and continue to provide programs and services for families and children.
15
Immigration status is not collected among RECE journeypersons.

8
 Table 2: Estimates of New Supply of ECE Needed by 2025/26 due to New ELCC Agreements

Average Number Years Needed to

Number of Number of New of New ECE Train Enough ECE
Promised New Full- Number of Total Ratio of Qualified ECE Graduates Graduates per Graduates to Fill
Time Spaces Staff Needed Staff Required Needed Year Promised Spaces
Alberta 42500 81800 33% 26994 390 69
British Columbia 30000 6250 60% 3750 450 8
Manitoba 23000 3833 66% 2530 140 18
New Brunswick 3400 730 50% 365 150 2
Newfoundland and Labrador 2790 612 66% 404 <30 20
Nova Scotia 11900 2149 66% 1418 40 35
Ontario 76700 15980 60% 9588 3250 3
Prince Edward Island 578 122 33% 40 <30 2
Quebec - - 66% 2170
Saskatchewan 28000 5911 50% 2956 130 23
Sources: Number of Promised New Full-Time Spaces: Government of Canada (2022); Number of Total Staff Needed: staff ratios obtained from Friendly et al.
2020; Ratio of Qualified Staff: McCuaig, Akbari, & Correia (2022); Number of New ECE Graduates: Post-Secondary Information System (PSIS) 2009-2018;
Registered Apprenticeship Information System (RAIS) 2010-2017
Notes: For Newfoundland and PEI, the average number of graduates per year in each province cannot be released due to confidentiality rules because they are
less than 30 graduates a year. For this table, we make the assumption that there are roughly 20 graduates per year. The agreement in each province generally only
provides the net new spaces to add in total rather than by each age range. We calculated the new spaces by age range based on an equal allocation assumption.
Years needed to train enough ECE graduates to fill promised spaces is calculated by number of new ECE graduated needed divided by the average number of new
ECE graduates per year.

9
graduates typically start with multiple part-time jobs when entering the labour market after graduating. This
level of multiple jobs is much higher than the Canadian average. In comparison, based on the Labour Force
Survey, in 2018, 6.8 percent of employed women held more than one job and this proportion is a bit lower
among employed men (4.7 percent) (Fulford & Patterson, 2019). Part-time workers are more likely to hold
multiple jobs and the rate is more than twice as high as that of full-timers. Regardless, the large number of
multiple job holders in this occupation is staggering.

Table 3: Snapshot of ECE graduates and RECE Journeypersons in 2017

University ECE College ECE RECE
Rest of
Canada Canada Quebec Ontario Ontario
Canada
Age at graduation (Mean/S.D.) 27.3/0.3 30.1/0.1 31.8/0.2 29.0/0.2 29.3/0.2 38.8/0.6
Female (%) 95.7 97.2 98.2 96.5 97.9 94.3
Immigration status (%)
Canadian citizen 92.8 82.8 82.2 83.5 82.5 -
Permanent resident or 5.8 10.6 16.9 7.4 7.0 -
International students 1.4 6.6 0.9 9.1 10.5 -
Marital status at 1 year after
graduation (%)
Married or Common-law 27.0 44.5 56.0 36.2 40.6 64.7
Divorced/separated/widowed 4.8 5.5 4.5 4.3 8.8
73.0
Single 50.7 38.5 59.3 55.1 26.5
Child status at 1 year after
graduation (%)
No child 31.3 33.9 33.2 32.1 39.1 26.5
1 child 21.9 22.9 21.7 23.5 23.9 26.5
2+ children 46.9 43.1 45.2 44.4 37.0 47.1
Program level (%)
College or below - 91.2 -
Above college below bachelor - 8.8 -
Bachelor degree 75.0 - -
Above bachelor 25.0 - -
Fields of study (%)
ECE for special needs - 0.6 -
ECE regular 44.9 1.8 -
Kindergarten/preschool 4.3 1.6 -
Child care provider 50.7 95.9 -
Multiple job status at 1 year
after graduation (%)
No job or one job 39.7 47.3 40.5 48.0 56.6 60.0
Multiple jobs 60.3 52.7 59.5 52.0 43.4 40.0
N 690 6170 2200 2540 1430 350
Sources: Post-Secondary Information System (PSIS) 2017-2018
Registered Apprenticeship Information System (RAIS) 2017
Note: Program and fields of study information among colleges ECE graduates is not releasable by province due to the small
sample size. Also, this information is not applicable among RECE people. Immigration status is available in PSIS but not in
RAIS.

10
4.3 Earnings Profiles of ECE and RECE professionals

Previous literature has shown that ECE salaries are lower than other female-dominated
professions requiring comparable education and training and are far below the average Canadian
salary (McCuaig et al., 2022). Our estimates in Figure 1 presents the average annual earnings of
ECE graduates one year after graduation for each graduate cohort 2010-2017 by program type
and find similarly low levels of earnings. The earnings are deflated with a base year in 2016. The
first year after graduation for different cohorts refers to different calendar years. For instance,
2011 is the first year after graduation for cohort 2010, while 2012 is the first year for cohort
2011. The earnings of college ECE graduates one year after graduation are relatively stable
across all cohorts (about $26,000) and lower than university ECEs and RECEs. The earnings of
university ECEs are the highest (about $36,000) and stable across cohorts 2010-2013. Then the
earnings increase for cohort 2014 and decrease significantly for cohort 2016. This may be due to
the last of the full day kindergarten rollout in Ontario schools in which they were hiring new
ECEs. The earnings of RECEs are stable (about $31,300) across most cohorts and increase for
cohort 2017.

Figure 1: Average earnings at 1-year after graduation, by program type

40000

35000
Earnings ($ 2016)

30000

25000

2010 2012 2014 2016 2018

Graduation Cohort

College ECE
University ECE
RECE

Figure 2 presents the average earnings and standard deviation bands at one year after graduation
of ECE graduates by province. 16 This figure does not include RECE wages in Ontario to make
15F

16
T-tests were used to observe whether the mean differences are significant across provinces. Using Ontario as the
reference group, we compared the mean of each province to the mean of Ontario to see whether or not the
differences are significant. The results of these t-tests range from 1.67 to 14.32, indicating that the difference in
means between each province and Ontario are statistically significant.

11
the data more comparable across provinces. ECE graduates in PEI have the highest average
earnings (about $36,100), followed by Manitoba and Alberta (about $32,400 and $32,100,
respectively). The average earnings of ECE graduates in Ontario and Quebec are $27,400 and
$25,800, respectively. ECE graduates in New Brunswick and Nova Scotia have the lowest
earnings (about $23,500 and $23,900, respectively). All provinces have large variations across
the sector with average standard deviation of $13,240 a year.

Figure 2: Average earnings at 1-year after graduation, by province

60000

50000
Average Earnings, $ 2016

40000

30000

20000

Mean
Standard Deviation
10000

AL BC MB NB NL NS ON PE QC SK
Province

4.4 Where do ECE professionals find careers after completing their studies?

We next investigate the career trajectories of ECE graduates over a period of five years to gather
insights on the recruitment and retention of the child care workforce. Figure 3 presents the career
trajectories of ECE graduates (university ECEs and college ECEs, separately) after graduation.
Specifically, we track the likelihood of graduates working in the education service industry (this
would include ECEs in Ontario’s kindergarten system), social assistance industry, or moving to a
career outside of these two areas over a five-year timeframe. 17 For clarification, child care
16F

centers are included in the social assistance industry category.

17
Our data, the ELMLP, only provides 3-digits of the North American Industry Classifications Systems (NAICS)
code which does not allow us to further investigate these industries in more detail. Unfortunately, a linkage to the
more-informative National Occupational Classification (NOC) does not yet exist in the ELMLP. However, Statistics
Canada does provide some examples of the types of jobs one might work in social assistance and education streams.
Social assistance codes include those working in adoption services, child support services, and young person
counselling. Education services includes work in both public and private schools across kindergarten to high school
(excluding pre-school and pre-kindergarten). Education services also refer to work in specialty schools like boarding

12
 Figure 3: Career trajectories of ECE graduates after graduation
College ECE Graduates University ECE Graduates
.6 .6

.5 .5

.4 .4
Proportion

.3 .3

.2 .2

.1 .1

1 2 3 4 5 1 2 3 4 5
Year after graduation Year after graduation

Work in Education Service

Work in Social Assistance
Work in Other Industries

We see that more than 60 percent of college ECE graduates transition into the social assistance
industry after graduation and the rest enter the education service or other industries. However,
the likelihood of working in the social assistance industry decreases over time while the
likelihood of working in the education service increases. This suggests that college ECEs might
initially find work in a child care center right after graduation but may later exit the sector in
favour of the education service industry. The story is quite different among university ECE
graduates. For instance, more than half of the university ECEs enter the education service
industry right after the graduation, with the remaining proportions divided among social
assistance or other industries. University educated ECEs are also more likely to move into
education service roles over time, with the probability of working in social assistance decreasing
over a five-year period.

Since the vast majority of ECE graduates tend to come from college programs, we next explore
the career pathways of college ECE graduates by province. In Figure 4 we see that the previously
identified trend of graduates moving from social assistance to education services industries is
mainly driven by Ontario, and to a lesser extent the remaining provinces aside from Quebec.

schools and schools specifically meant for person with physical disabilities. “Other” occupations refers to a move
out of either education or social assistance ECE roles, which may or may not utilize their education and training
(i.e., arts, entertainment and recreation jobs, public administration, or jobs in manufacturing or trade).

13
Alternatively, career trajectories in Quebec appear to be quite stable over time for ECE
graduates. 18 17F

Figure 4: Career trajectories of college ECE graduates after graduation, by Province

Quebec Ontario Rest of Canada
.8 .8 .8

.6 .6 .6
Proportion

.4 .4 .4

.2 .2 .2

0 0 0

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
Year after graduation Year after graduation Year after graduation

Work in Education Service

Work in Social Assistance
Work in Other Industries

One of the primary reasons that college ECE graduates might exit the social assistance industry
in favour of employment in education services could be attributable to earnings growth between
sectors. We therefore explore this possibility in Figure 5, where we present the average earnings
trajectories of college ECEs based on their career pathways for a period of five years after
completing their studies. 19 18F

18
The findings in Quebec are interesting, given the province’s own iteration of an affordable child care system, as
the stability across sectors is perhaps encouraging, at least from the perspective of those who might be concerned
that increases in Ontario’s demand for ECE’s in education services could create a vacuum and exacerbate gaps in
other areas (like social assistance, for example). The trends we see in Quebec gives some confidence that such
concerns regarding demand may be unfounded or otherwise that any early changes in demand may stabilize over
time.
19
To aid in the interpretation of the figure, “stay in social assistance” “and “stay in education service” are
categorized using the 3-digit North American Industry Classifications Systems (NAICS) code and correspond to
those ECE workers who reported consistent NAICS codes over the 5 year period of observation. The “movers” (that
is, those that left either sector) categories refer to a move from either of the aforementioned 2 digit codes to the
other, or to any other NAICS code not captured by social assistance or education services (in the case of the “other
industries” category).

14
 Figure 5: Earnings trajectories of ECE graduates after graduation, by career path
32000

30000

28000 Stay in Social Assistance

Earnings

Move to Education Service

Move to Other Industries
Stay in Education Service
26000 Move to Social Assistance

24000

22000

1 2 3 4 5
Year after graduation

Note: The categories Move to Education Service and Move to Other Industries correspond to a movement from Social Assistance fields to these
new career paths.

We include in our sample only those respondents that were employed in all five years of our
period of observation, and then separated them into the following five groups depending on their
path/trajectory during those five years: 20 19F

1. Stay in Social Assistance: those that had the same 3-digit NAICS code for social
assistance (624) from time 1 to time 5
2. Move to Education Service: those that started in the 3-digit NAICS code for social
assistance (624) at time 1 and moved to 3-digit NAICS code for education services (611)
at any point in time 2- time 5.
3. Move to Other Industries: those that started in the 3-digit NAICS code for social
assistance (624) at time 1 and moved to a 3-digit NAICS code that was NOT education
services (611) at any point in time 2- time 5.
4. Stay in Education Service: those that had the same 3-digit NAICS code for education
services (611) from time 1 to time 5

20
We draw our pathways from the T1FF data available in the linkage platform—specifically the 3-digit industry
code which allows us to identify ECE graduates who worked in ELCC careers after graduation. We recognize that
the 3-digit NAICS code has limited detail for this application, but we argue that education services and social
assistance are distinct codes and are the two industries which are most likely to cover ELCC careers based on the
wider literature. Aside from these two categories, we have included a “catch-all” category for those graduates who
ended up in industries that do not fall under the aforementioned categories. This is largely due to sample
limitations—our sample size is not large enough to unpack these other industries.

15
 5. Move to Social Assistance: those that started in the 3-digit NAICS code that was NOT
social assistance (624) at time 1 and moved to a 3-digit NAICS code that was social
assistant (624) at any point in time 2- time 5. 21 20F

Social assistance careers appear to be the most lucrative directly after completing studies,
surpassing educational services and “other” industry roles. However, earnings trajectories appear
to change this advantage over time. Graduates that remain in the social assistance industry for the
entire five-year period see earnings growth for the first three years, which levels off before
dropping between years three and five. While initially lower paying, roles in education services
surpass the social assistance category after roughly three years and demonstrate a more
promising earnings trajectory.

This trend may suggest that the promise of better career perspectives could be a factor in the
decision to exit. At an aggregate level, it appears that leavers (regardless of the destination) have
lower earnings than those who stay in their initial choice which may take a period of time to
recover from.

We encourage caution when interpreting these findings as we do not disentangle the length of
time graduates stayed in their career prior to moving to a new path, which could potentially bias
our earnings assumptions. We expect that the earnings penalty experienced by graduates that
move careers is the result of time lost in a given field—which we presume will diminish or
disappear as graduates spend more time in their new careers. In most cases these earnings
differences are relatively minor; however, leaving the field to pursue “other” industries appears
to be a particularly low-paying trajectory with earnings that are markedly lower than any of the
other pathways identified in Figure 5. In this case, the knock-on effect of this earnings penalty
may be more detrimental the longer it takes a graduate to recover from a career change. 22 21F

For those ECE professionals who stayed in the social assistance industry up to five-year after
graduation, their earnings were the highest one year after graduation and kept increasing over
time, only dropping at the four-year mark. However, for those who were working in the social
assistance field but moved to education services, their earnings kept increasing through the five-

21
It is possible that graduates could switch more than once across the “mover” categories specified above. However,
as long as they moved out from the 3-digit NAICS code for social assistance (624) at least once, we counted them as
“movers”. Also, because T1FF is a yearly data, it is also possible that people switched industries between t=1 and
t=2 (i.e., within a year) but this trend cannot be observed. Due to limitations in our sample, we are unable to further
decompose our 5 groups into more detailed categories to isolate how many times they switched (i.e., worked in
social assistance at t=1 and moved to education service twice during t=2-5 etc ).

22
At the same time, an alternative explanation could be that graduates who enter either the social
assistance or education services sectors with lower-than-average earnings may be less likely to stay in the
sector and are therefore more likely to move to other pathways. This would appear to be at least partially
supported by the greater proportional year-over-year growth of those graduates who move into social
assistance careers which parallels those that stay in social assistance (albeit at a lower total salary). In this
way the lower earnings could be attributable to individual-side characteristics such as performance, or the
negotiation of salary expectations upon changing career streams, though this is difficult to predict with
EMLP data.

16
year time period. This might suggest the behaviours of exiting from social assistance and moving
to education service are driven by the fact that the latter industry could have better career
perspectives. College ECEs staying in education service or other industries experience
significant earnings growth as well and their earnings even surpass those who stayed in social
assistance after 3-year after graduation.

It is surprising that ECE graduates in social assistance roles are among the highest earners, given
the attrition from the field we observed previously. There are two possible explanations for this
trend. On the one hand, social assistance roles appear to offer the greatest salaries right out of
college, which could be attractive to new graduates. This attraction could wane as salary
increases stagnate over time, and the absence of attractive alternative opportunities in other
social assistance roles. Education service roles may therefore offer more attractive long-term
gains that cause professionals to leave social assistance careers to seek longer-term opportunities.

On the other hand, the decision to leave social assistance careers may be driven by factors other
than earnings. For example, Halfon (2021) has identified an overall devaluation of the child care
workforce. Therefore, it may be that education service roles, and those careers outside of ECE,
offer greater intangible benefits to quality of life that surpass the modest earnings advantages
available in social assistance pathways.

Given that the vast majority of ECE graduates come from Ontario and Quebec, we separate these
provinces to examine whether there are different characteristics in the pathways of their ECEs. In
Figure 6 we see considerable variation in trajectory by province. In Ontario earnings trajectories
for ECEs who find permanent careers in the social assistance and education service categories
are roughly similar, with a slightly greater earnings trajectory in the social assistance category
that levels off towards the end of the five-year period. This is contrasted by Quebec, and the rest
of Canada, where we see stark differences in earnings trajectories depending on the career stream
chosen by the graduate. For Quebec, work in education services appear to be the more lucrative
pathway than social assistance careers, while the opposite is true for the rest of Canada.

17
 Figure 6: Earnings trajectories of college ECE graduates after graduation, by Province
Quebec Ontario Rest of Canada
35000 35000 35000

30000 30000 30000

Earnings

Earnings

Earnings
25000 25000 25000

20000 20000 20000

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5
Year after graduation Year after graduation Year after graduation

Stay in Social Assistance

Move to Education Service
Move to Other Industries
Stay in Education Service
Move to Social Assistance

In Quebec, where both social assistance and education service categories earn the same at year
one, this quickly changes with significantly greater earnings growth in education services. We
suspect this trend is largely driven by the province’s control over the social assistance sector. The
earnings for social assistance careers remain relatively stable in the province—leading to a
sizable gap by year five. The inverse is true for the rest of Canada, however, where ECEs
working in social assistance have markedly higher earnings than those in education service or
“other” industry roles. While these earnings do appear to decay after year four, the returns are
still markedly higher than any of the other career pathways.

Figure 7 shows the career and earning trajectories of RECEs in Ontario. The career trajectory of
RECEs is similar to those college ECEs in Ontario. Greater shares of graduates enter social
assistance jobs right after graduation but start to exit and move to education-related careers over
a five-year period. The earnings trajectory of RECEs differs from the college ECEs in Ontario,
however, particularly as RECEs in social assistance do not experience the drop in earnings is
observed for ECEs after year four.

18
 Figure 7: Career and earnings trajectories of RECEs after graduation
Career trajectory Earnings trajectory
40000
.6

.5
35000

Earnings
Proportion

.4

30000
.3

.2
25000

1 2 3 4 5
1 2 3 4 5
Year after graduation Stay in Social Assistance

Move to Education Service

Work in Education Service
Move to Other Industries
Work in Social Assistance
Stay in Education Service
Work in Other Industries Move to Social Assistance

Note for Figure 7b: The categories Move to Education Service and Move to Other Industries correspond to a movement from Social Assistance
fields to these new career paths. Likewise, the category Move to Social Assistance implies a movement from Education Services.

5. Regression analysis

In this section, we employ regression modelling to further investigate the recruitment and
retention of the child care workforce as well as what factors contribute to these behaviours. 23 22F

First, we use equation 1 to analyze the likelihood of entering the social assistance industry right
after graduation after controlling related factors.

𝑦𝑦𝑖𝑖 = 𝛼𝛼0 + 𝛼𝛼1 𝑋𝑋𝑖𝑖 + 𝛾𝛾𝑃𝑃𝑖𝑖 + 𝛿𝛿𝐶𝐶𝑖𝑖 + 𝜖𝜖𝑖𝑖 (1)

Where 𝑦𝑦𝑖𝑖 is a dummy variable (either 1 or 0) indicating the likelihood of working in the social
assistance industry one year after graduation. 𝑋𝑋𝑖𝑖 is a set of related variables which includes ECE
graduates’ age, marital status, and number of children one year after graduation, as well as
gender, immigrant status, program type, and fields of study. 𝑃𝑃𝑖𝑖 are province indicators and 𝐶𝐶𝑖𝑖 are
cohort indicators. Table 4 presents the regression results based on the linear probability model
(LPM) and Logit model.

23
We only focus on ECE graduates rather than RECEs in the regression analysis because many of the variables
included in the model, such as immigration status and program type, are either not available or not applicable in the
RAIS data.

19
Table 4: Likelihood of entering the social assistance industry at 1-year after
graduation
LPM Logit
Age -0.011*** -0.048***
(0.004) (0.018)
Age squared 0.000* 0.000*
(0.000) (0.000)
Male (omitted category)

Female 0.076*** 0.366***

(0.027) (0.127)
College or below (omitted category)

Above college but below bachelor -0.163*** -0.728***

(0.021) (0.095)
Bachelor degree -0.331*** -1.525***
(0.021) (0.119)
Above bachelor -0.411*** -2.523***
(0.029) (0.266)
ECE regular (omitted category)

ECE for special need 0.032 0.232

(0.069) (0.317)
Kindergarten/preschool 0.117*** 0.639***
(0.039) (0.175)
Child care provider 0.070*** 0.351***
(0.025) (0.128)
Canadian citizen (omitted category)

Permanent resident 0.074*** 0.325***

(0.016) (0.073)
International students -0.028 -0.111
(0.074) (0.315)
Married (omitted category)

Widowed -0.007 -0.041

(0.080) (0.360)
Divorced/separated -0.024 -0.107
(0.021) (0.092)
Single 0.005 0.022
(0.011) (0.052)
No children (omitted category)

20
1 child -0.009 -0.048
(0.012) (0.055)
2+ children -0.026** -0.123**
(0.011) (0.049)
ON (omitted category)

NL -0.12 -0.509
(0.169) (0.692)
PE -0.022 -0.274
(0.037) (0.257)
NS 0.025 0.098
(0.058) (0.240)
NB 0.05 0.213
(0.035) (0.144)
QC 0.220*** 0.993***
(0.011) (0.050)
MB 0.228*** 1.021***
(0.026) (0.133)
SK 0.138*** 0.587***
(0.031) (0.139)
AB 0.009 0.04
(0.023) (0.094)
BC 0.032* 0.133*
(0.018) (0.081)
Cohort 2010 (omitted category)

Cohort 2011 -0.019 -0.088

(0.013) (0.061)
Cohort 2012 -0.028** -0.135**
(0.013) (0.060)
Cohort 2013 -0.032** -0.152***
(0.012) (0.059)
Constant 0.641*** 0.561
(0.078) (0.361)
N 11480 11480
R-squared 0.119
Adjusted R-squared 0.117
Pseudo R-squared 0.093
Chi-squared 1043.200
Sources: Post-Secondary Information System (PSIS) 2009-2014 linked to T1FF 2011-2018
Note: * for p<0.10, ** for p<0.05, and *** for p<0.01
Robust standard errors in parentheses

21
The regression results are consistent with those findings from the descriptive analysis. For
example, ECE graduates with a degree/certificate higher than a college program are less likely to
work in the social assistance industry right after graduation. ECE graduates with a field of
kindergarten/preschool or child care provider are more likely to enter the social assistance
industry. The likelihood decreases as age increases and females are more likely to pursue this
pathway compared to males. Immigrants are more likely to enter compared to Canadians.
Current marital status has no significant impact on the likelihood but having multiple children
decreases the likelihood compared to those without children. 24 23F

Then, we focus on ECE graduates who were working in the social assistance industry one year
after graduation and follow them over a period of five years. We use a survival analysis, which
estimates a hazard function, to examine the effects of related factors on the behaviours of exiting
the social assistance industry. A survival analysis estimates the likelihood that ECE graduates
may exit from the social assistance industry during a stretch of time (similar methods are used in
Bridges et al, 2011; Singer & Willett, 2003). In our case, the model is specified as follows:

ℎ𝑖𝑖 (𝑡𝑡) = ℎ0 (𝑡𝑡) exp�𝑏𝑏1 𝑥𝑥1 + 𝑏𝑏2 𝑥𝑥2 + ⋯ + 𝑏𝑏𝑝𝑝 𝑥𝑥𝑝𝑝 � (2)

Where ℎ0 (𝑡𝑡) is the baseline hazard at time t and t represents the survival time. ℎ𝑖𝑖 (𝑡𝑡) is the hazard
function for ECE graduate i determined by a set of covariates. 𝑥𝑥1 , 𝑥𝑥2 , … , 𝑥𝑥𝑝𝑝 are a set of
covariates such as gender, age, and marital status described above. The baseline hazard indicates
the value of the hazard if all the covariates are equal to zero. The coefficients 𝑏𝑏1 , 𝑏𝑏2 , … , 𝑏𝑏𝑝𝑝
measures the effects of these covariates on the hazard rate of exiting from the social assistance at
time t.

Table 5 presents the impacts of related factors on the hazard rate of exiting from the social
assistance within five years after graduation. 25 ECE graduates with higher initial earnings
24F

(earnings one year after graduation) are significantly less likely to leave this industry within five
years. This is consistent with the previous studies (Beach & Flanagan 2007; Flanagan et al.,
2013; Halfon 2021) which indicates that low pay is the main reason of workers leaving this field.
The exit hazard increases as ages increases (but at a diminishing rate). ECE graduates with a
degree higher than a college program are significantly more likely to leave compared to
graduates with a college degree. Compared to the reference group (graduates with no child), ECE
graduates with one child are significantly less likely to leave. 26 These findings are consistent
25F

24
We also analyze the likelihood of entering the education service category after graduation as well as what factors
have contributed to this likelihood. The results are consistent with our descriptive analysis as well. For instance,
ECE graduates with a bachelor or higher degree are more likely to enter the education service industry. ECE
graduates with an “ECE regular” specialty are more likely to enter this industry. The likelihood increases as age
increases (but at a diminishing rate). Females are slightly less likely to enter compared to males and immigrants are
less likely to enter compared to Canadians. Singles are less likely to enter compared to married graduates while
having multiple children increases the likelihood.
25
To ease the interpretation of results, we report coefficients rather than hazard ratios (exponentiated coefficients).
26
We conduct a couple of robustness checks regarding including different forms of time-varying variables such as
marital status and number of children in the hazard rate models. For instance, following Allison (2010), we specify
marital status in different formats and include them in the model: (1) whether the ECE graduate is currently married
in year t, (2) whether the ECE graduate is married in year t-1, and (3) whether the ECE graduate has changed the

22
across provinces. 27 The p-value of the test of the proportional-hazards assumption are larger than
26F

0.10 which suggests that there is no violation of the proportionality assumption. 28 27F

Table 5: Hazard of exiting from social assistance industry within 5 years after
graduation by province
Canada QC ON ROC
Initial Earnings ($1000) 0.020*** -0.031*** 0.012*** -0.022***
(0.002) (0.004) (0.003) (0.004)
Age 0.069*** 0.088* 0.037 0.064
(0.022) (0.048) (0.030) (0.043)
Age squared 0.001*** -0.001** -0.001 -0.001*
(0.000) (0.001) (0.000) (0.001)
Male (omitted category) - - - -
- - - -
Female 0.068 0.011 0.172 -0.163
(0.155) (0.348) (0.212) (0.271)
College or below (omitted category) - - - -
- - - -
Above college but below bachelor 0.278** 0.405** 0.418* 0.453
(0.128) (0.159) (0.230) (0.339)
Bachelor degree 0.469*** 2.185*** 0.360*** 0.785
(0.110) (0.485) (0.124) (0.614)
-
Above bachelor 0.696** 2.008** 0.858*** 39.657***
(0.282) (0.916) (0.255) (0.937)
ECE regular (omitted category) - - - -
- - - -
ECE for special need 0.482 - 0.283 -
(0.314) - (0.383) -
Kindergarten/preschool -0.12 0.113 - -
(0.312) (0.394) - -
Childcare provider 0.28 0.645** 0.245 -0.748
(0.180) (0.318) (0.244) (0.750)
Canadian citizen (omitted category) - - - -
- - - -
Permanent resident 0.053 -0.009 0.133 -0.118
(0.082) (0.135) (0.114) (0.246)

marital status within t years. The results are consistent across different specifications. Same thing for number of
children.
27
Provinces and graduate cohorts are stratified in all specifications because the baseline hazard could be different
across provinces and cohorts. Marital status is further stratified in ON to meet the proportionality assumption test.
28
The tests of the proportional-hazards assumption assume homogeneity of variance across each level of the
covariate of interest. Models do not violate this assumption can estimate overall (pooled) variance–covariance
matrix. See (Kalbfleisch & Prentice 2002).

23
 - -
International students -0.157 32.198*** 0.669** 39.257***
(0.342) (0.515) (0.284) (0.384)
Married/Common Law (omitted category) - - - -
- - - -
Widowed 0.138 1.233** - 0.331
(0.422) (0.547) - (1.046)
Divorced/separated 0.092 0.263 - 0.18
(0.115) (0.195) - (0.247)
Single 0.091 0.107 - 0.291**
(0.059) (0.106) - (0.123)
No children (omitted category) - - - -
- - - -
1 child -0.133** -0.176 -0.058 -0.247*
(0.061) (0.119) (0.086) (0.130)
2+ children -0.007 -0.097 0.096 -0.12
(0.053) (0.111) (0.071) (0.115)
Test of proportional-hazards assumption
P-value 0.132 0.793 0.625 0.909

N 6570 3200 2130 1240

Pseudo R-squared 0.006 0.015 0.003 0.019
Chi-squared 154.344 4215.699 48.113 15981.436
Sources: Post-Secondary Information System (PSIS) 2009-2014 linked to T1FF 2011-2018
Note: * for p<0.10, ** for p<0.05, and *** for p<0.01
Robust standard errors in parentheses
Provinces and graduate cohorts are stratified in all specifications, while marital status is further
stratified in ON.

Table 6 presents the impacts of related factors on the hazard rate of existing from the social
assistance within five years after graduation for college ECE graduates and university graduates,
separately. The results are consistent with those in Table 6. For instance, the initial earnings
significantly decrease the exit hazard while age increases the hazard. ECE graduates with a
higher degree are significantly more likely to leave the social assistance field compared to
graduates with a college degree. ECE graduates with one child are significantly less likely to
leave.

24
Table 6: Hazard of exiting from social assistance industry within 5 years after graduation by
program
College ECE University ECE
Initial Earnings ($1000) -0.019*** -0.016*
(0.002) (0.009)
Age 0.057** 0.107
(0.022) (0.173)
Age squared -0.001*** -0.002
(0.000) (0.003)
Male (omitted category)

Female 0.039 1.143

(0.161) (0.814)
College or below (omitted category)

Above college but below bachelor 0.431*** -

(0.123) -
Above bachelor - 0.396
- (0.334)
ECE regular (omitted category)

ECE for special need 0.966** -

(0.388) -
Kindergarten/preschool 0.319 -1.152*
(0.397) (0.645)
Child care provider 0.902*** 0.059
(0.308) (0.286)
Canadian citizen (omitted category)

Permanent resident 0.074 -0.889*

(0.083) (0.532)
International students -0.186
(0.348)
Married (omitted category)

Widowed - 1.920
- (1.227)
Divorced/separated - -0.624
- (1.216)
Single - 0.165
- (0.281)
No children (omitted category)

25
 1 child -0.121* -0.105
(0.062) (0.338)
2+ children -0.004 0.175
(0.055) (0.225)
Test of proportional-hazards assumption
P-value 0.311 0.338

N 6390 170
Pseudo R-squared 0.007 0.045
Chi-squared 139.790 22.227
Sources: Post-Secondary Information System (PSIS) 2009-2014 linked to T1FF 2011-2018
Note: * for p<0.10, ** for p<0.05, and *** for p<0.01
Robust standard errors in parentheses
Provinces and graduate cohorts are stratified in all specifications, while marital status is further stratified in
ON.

6. Discussion

The purpose of this paper has been to profile ECE graduates in the wake of the federal and
provincial commitment of more than $30 billion to support early learning and child care. Using
administrative data, we have provided a snapshot of the yearly supply of entrants to the ECE
field, as well as their sociodemographic characteristics and expected career pathways. This
evidence is helpful when charting a course forward as early learning and child care expands to
meet the targets of 2025 and beyond.

We find that the overwhelming proportion of ECE program graduates come from Ontario and
Quebec (roughly 45 and 34 percent respectively) with a significantly smaller proportion coming
from the remaining provinces. These are predominantly Canadian citizen women, the majority of
which are over the age of 25 and are parents to at least one child. This is a potentially
problematic trend for the labour supply of early childhood educators for two reasons. First,
graduates tend to gravitate towards urban centres when they complete their education (Corbett,
2005a; Tremblay, 2001; Zarifa et al., 2019a). Given that two of Canada’s largest economic
centres are producing the largest shares of graduates, there are concerns that these graduates may
not disperse to the other provinces that need them—especially if they have ample opportunities
to find employment locally while finishing their education (Finnie, 2004a). Workers also tend to
become less portable once they get married and have children (Finnie, 2004a). These
characteristics signal a potential problem for ensuring ample supplies of ECE graduates fulfill
the staffing needs of other provinces, particularly if remuneration for those jobs is not strong
enough to incentivize graduates to relocate.

This potential staffing issue is more apparent when we consider the future demands for ECE staff
in light of the federal and provincial governments’ recent investments on the sector. If the federal
government’s commitment to a $10-dollar-a-day child care system meets its five-year target, we
will need far more graduates than we are currently producing. We can expect a deficit in the
number of qualified candidates to fill the spaces intended across the remaining provinces.

26
Currently, one of the few incentives to relocate outside of Ontario or Quebec is due to the fact
that provinces like PEI, Alberta, and Manitoba pay staff in ECE centres measurably more than in
Ontario and Quebec (a real dollar difference of between $6,000 to $10,000 on average depending
on the province). Though this premium does not exist in provinces like New Brunswick and
Nova Scotia (areas where ECEs make less than Ontario and Quebec), and it is difficult to say
whether the modest increase in earnings over Quebec and Ontario would outweigh the costs of
relocating, especially if this means uprooting ones’ family.

When we talk of needing staff to fill child care spaces, we are really talking about college
graduates to fill these roles. That is, unless we incentivize university graduates to shore up this
inevitable gap, however our regression modelling indicates that the majority of university
graduates are unlikely to end up in ECE centres. Policy in this area should therefore focus on a
two-pronged recruitment strategy. On the one hand, efforts should be taken to significantly
bolster college-level enrollments in ECE programs. Encouraging out-of-work workers, or those
displaced by the pandemic to consider careers in the ECE field might serve as one avenue to
improving the numbers needed to meet the 2025-26 targets. On the other hand, communicating
ECE spaces as a viable pathway for university graduates (with appropriate incentive to balance
the higher cost of a university education) could be another avenue to ensuring that the supply of
ECE professionals is able to meet labour demands.

Yet in either case, we have identified that, outside of Quebec, employment in social assistance
industries (including ECE centres) tends to decay over time, with participation in education
services and other industries serving as attractive alternative pathways. Moreover, given that the
majority of ECE workers are parents, and with earnings that barely outweigh the average cost of
child care in many provinces, ECE professionals with children may be forced to weigh the costs
of working to child care costs—which could cause some professionals to leave the labour market
to raise children. These sector exits are a potentially serious problem considering the future
demands expected in the social assistance sector. Policies therefore need to address retention as
much as recruitment if we are to ensure a sustainable workforce of ECE professionals.

Our earnings models indicate that, at their current levels, the earnings of the child care workforce
are unlikely to encourage decisions to remain in the sector. Although our descriptive modelling
indicated that wages do grow somewhat over time, in real dollars these figures are woefully
low—earnings that are well below the national average (Beach 2013; Statistics Canada,
2020b). 29 Our work, therefore, is another contribution to an already established literature
28F

indicating the negative effect low wages have on recruitment and retention in the child care
workforce (Beach & Flanagan 2007; Flanagan et al., 2013; Halfon 2021).

Although, our analyses in Quebec add another interesting perspective to this story. In a province
with its own system of affordable child care, it is encouraging that employment across sectors
appears much more stable than the rest of Canada. This is interesting for two reasons. First, it
presents the possibility that employment levels within and across sectors may stabilize (if not
immediately, then over time) as similar adoptions of universal affordable child care are

29
For more details, see (Akbari et al., 2020).

27
implemented in the rest of Canada. A second (and perhaps even more likely) interpretation of our
findings is that earnings are only one layer of a complicated issue.

The low pay of ECE professionals can also be interpreted as a signal for a wider issue: the
perceived devaluation of the child care workforce (Halfon 2021). 30 If earnings are a signal for
29F

the recognition and value we place on a vocation, then these low earnings signal a lack of
recognition for child care workers that can influence job satisfaction and performance. ECEs
have been shown to be largely invisible when developing child care policy responses, especially
in the context of COVID-19 ( Richardson et al., 2021; Goelman et al., 2000). These feelings in
turn can devalue the quality of care, and have been found to create work environments and staff
perceptions that have a predictable and statistically significant impact on program quality and
staff-child interactions (Goelman et al., 2000).

The choice to publicly fund early learning and child care is based upon a sound rationale and is
expected to generate an estimated $17 to $29 billion per year in productivity as both parents can
participate in the labour market (Stanford, 2020). This productivity is sorely needed as Canada
recovers from the financial costs of the pandemic and fights to maintain its international position.
Yet we have identified a potentially difficult point of tension for policy to address: the success of
a universal child care system requires a sufficient and continued supply of qualified and
passionate educators to meet these rising targets; however, the current state of the field does not
appear to offer incentive enough to meet that demand.

Normally, employer demand for incumbents in an occupation would lead to labour market
adjustments (by way of increases to compensation) that would bolster the labour supply as new
entrants pursue these occupations. In this case, due to the market failures, we are concerned that
there will be tension between the costs of labour, the quantity of educators required, government
budgetary allocations, and a commitment to the $10 dollar a day cost to Canadians that may limit
the operating budgets of child care centres. In this case large labour market adjustments will be
needed (by way of large compensation changes), leading us to question the feasibility of this
plan.

For this policy initiative to be successful, we argue that the significant financial investments
being made in the ELCC sector not only consider increasing the number of child care workers,
but also prioritize improving both the quality and social standing of those jobs. At the very least,
this would mean formally enforcing higher starting salaries as we have identified from our
hazard models that doing so could markedly improve retention. It might be helpful to consider
salaries that are more competitive with the provincial or national average, and other indirect
rewards (i.e., benefits) that are competitive with the total rewards offered in comparative fields.
Another worthwhile endeavour would be to offer formal incentives for cross-provincial mobility

30
The authors acknowledge the work of Hanushek et al. (2004), among others (i.e., Stinebrickner 2002), that have
found limitations in the effect that higher-salaries can have on attracting teachers. One observation that is relevant to
our discussion is that relying upon compensating differentials to outweigh undesirable work characteristics has
diminishing returns among teaching professionals—and is largely dependent upon the overall quality of the school
(and the employment conditions it provides). We agree with the authors that raising salaries alone is unlikely to
resolve issues of attraction and retention to the field. Rather, policy efforts also need to consider improving the
factors that influence overall perceptions of the field and its working conditions to encourage young Canadians to
pursue and remain in ECE careers.

28
to areas which are likely to be in the greatest need for ECE staff by 2025 as a means to prevent
gaps in ECE support. In either case, successful policy efforts will need to ensure that the jobs
that are created in the child care sector are perceived to be “good” jobs that attract qualified and
passionate educators while at the same time discouraging industry exits.

7. Conclusion

This report serves as an important first step to identifying possible labour shortages as Canada
enters a new era of federally subsidized child care. We find that the existing supply of new early
childhood educators is insufficient to meet the demand that is expected to follow these policy
commitments. Our results indicate that even in the case that the federal and provincial
governments stymie labour market exits, the total number of graduates produced each year is
insufficient to meet the new targets. We therefore argue for policy efforts to invest in both the
pay and perceived value of these occupations as a means to bolster the pipeline of graduates at
the college and university levels.

As the country continues to finalize agreements and works to put in place the infrastructure of
this change, further study will be required to continue to monitor the supply of ECEs and the
effectiveness of this new system. In the immediate short term, further study could continue to
analyze subsequent waves of ELMLP data as they become available to see whether the supply of
ECE graduates improves to the levels the country is expected to require. At the time of writing,
the Canadian Research Data Centre Network is expecting the release of the next wave of data
that will offer greater insights on the supply and labour market outcomes of ECE graduates up to
the year 2020. Certainly, it will be especially important to monitor this demographic throughout
the waves of Canada’s response to the coronavirus to observe how, and in what ways, the
pandemic has influenced graduate pathways.

Yet there are limitations to this data that are necessary to acknowledge. First, sample restrictions
among university ECEs and registered early childhood educators (RECEs) mean that our
analyses are primarily pertinent for college-level ECE graduates. Future study should attempt to
pursue the school to work transitions of university educated ECEs. Second, data collection
within the ELMLP is still relatively recent, although the earliest education records are from
2009, systematized reporting across the provinces is unavailable until the following years.
Therefore, with each release of subsequent ELMLP waves we expect the precision of results to
improve.

8. References

Allison, P. D. (2010). Survival analysis using SAS: a practical guide. SAS Publishing.
Beach, J., & Flanagan, K. (2007). People, programs and practices: A training strategy for the
early childhood education and care sector in Canada. Child Care Human Resources
Sector Council.
Beach, Jane. (2013b). Overview of child care wages: 2000-2010. Child Care Human Resources
Sector Council.
Beauregard, P. L., Connoly, M., & Molnar, T. L. (2020). Primary School Reopenings and
Parental Work. Research Group on Human Capital Working Paper, 20(6).

29
 https://grch.esg.uqam.ca/wpcontent/uploads/sites/82/Beauregard_Connolly_Haeck_
Molnar_GRCH_WP20-06.pdf
Blau, D. M. (2001). Child care problem: An economic analysis. Russell Sage Foundation.
Bridges, M., Fuller, B., Huang, D. S., & Hamre, B. K. (2011). Strengthening the Early
Childhood Workforce: How Wage Incentives May Boost Training and Job Stability.
Early Education and Development, 22(6), 1009–1029.
https://doi.org/10.1080/10409289.2010.514537
Carneiro, P., & Ginja, R. 2014. Long-Term impacts of compensatory preschool on health and
behavior: Evidence from Head Start. American Economic Journal: Economic Policy,
6(4): 135–173.
Caucutt, E., & Lochner, L. 2004. Early and late human capital investments, credit constraints,
and the family. In 2004 Meeting Papers (No. 129). Society for Economic Dynamics.
Child Care Canada. (n.d.). https://child carecanada.org/resources/issue-files/building-canada-
wide-early-learning-and-child-care-system
Child care Resource and Research Unit. (2021). ECEC in Canada 2019: Summary and analysis
of key findings. Child care Canada. https://child carecanada.org/publications/ecec-
canada/21/02/ecec-canada-2019-summary-and-analysis-key-findings
Cleveland, G. & Hyatt, D. (2000). Child care worker’s wages: New evidence on returns to
education, experience, job tenure and auspice. Toronto, ON: University of Toronto.
Cleveland, Krashinsky, Colley and Avery-Nunez (2016) City of Toronto Licensed Child Care
Demand and Affordability Study
Corbett, M. (2005b). Rural Education and Out-Migration: The Case of a Coastal Community.
Canadian Journal of Education, 28(1/2), 52–72. https://doi.org/10.2307/1602153
Cunha, F., & Heckman, J. 2007. The technology of skill formation. American Economic Review,
97(2), 31-47.
Currie, Janet. 2001. Early childhood education programs. Journal of Economic Perspectives,
15(2): 213-238
Doherty, G., Lero, D.S., Goelman, H., LaGrange, A., and Tougas, J. (2000). You bet I care! A
Canada wide study on wages, working conditions, and practices in child care centres.
Guelph, Ontario: Centre for Families, Work and Well -Being, University of Guelph.
Duncan, G. J., & Magnuson, K. 2013. Investing in preschool programs. Journal of Economic
Perspectives, 27(2), 109-32.
Finnie, R. (2004b). Who moves? A logit model analysis of inter-provincial migration in Canada.
Applied Economics, 36(16), 1759–1779. https://doi.org/10.1080/0003684042000191147
Flanagan, K., Beach, J., & Varmuza, P. (2013). You bet we still care! A survey of centre-based
early childhood education and care in Canada: Highlights report. Child Care Human
Resources Sector Council.
Friendly, M., Forer, B., Vickerson, R., & Mohamed, S. (2020). Canadian Child Care:
Preliminary Results from a National Survey During the COVID19 Pandemic (p. 270).
Child care Resource and Research Unit. https://child
carecanada.org/sites/default/files/Canadian-Child-Care-COVID19-Survey-Data-
Report.pdf
Friendly, M., Forer, B., Vickerson, R., & Mohamed, S. (2021). COVID-19 and Child care in
Canada: A Tale of Ten Provinces and Three Territories. Journal of Childhood Studies,
46(3), 42–52.

30
Gordon, J. A., Herbst, C. M., & Tekin, E. 2021. Who's minding the kids? Experimental evidence
on the demand for child care quality. Economics of Education Review, 80, 102076.
Government of Alberta. (n.d.-b). https://www.alberta.ca/federal-provincial-child-care-
agreement.aspx
Government of Canada. (2022). Early Learning and Child Care Agreements.
https://www.canada.ca/en/early-learning-child-care-agreement/agreements-provinces-
territories.html
Friendly, M., Feltham, l.e., Mohamed, S., nguyen, n. t., Vickerson, r., & Forer, B. (2020). Early
childhood education and care in Canada 2019. toronto: child care resource and research
unit.
Halfon, S. (2021). Canada’s child care workforce (Occasional Paper 35). Child care Resource
and Research Unit.
Hanushek, E. A., Kain, J. F., & Rivkin, S. G. (2004). Why public schools lose teachers. Journal
of human resources, 39(2), 326-354.
Herbst, C. M., Desouza, K. C., Al-Ashri, S., Kandala, S. S., Khullar, M., & Bajaj, V. 2020. What
do parents value in a child care provider? Evidence from Yelp consumer reviews. Early
Childhood Research Quarterly, 51: 288–306
Kalbfleisch, J. D., & Prentice, R. L. (2002). The Statistical Analysis of Failure Time Data
(Second Edition). Wiley.
Magnuson, K., & Duncan, G. J. 2016. Can early childhood interventions decrease inequality of
economic opportunity? RSF: The Russell Sage Foundation Journal of the Social
Sciences, 2(2), 123–141.
McCuaig, K., Akbari, E., & Correia, A. (2022). Canada’s children need a professional early
childhood education workforce. Atkinson Centre for Society and Child Development,
Ontario Institute for Studies in Education, University of Toronto.
OECD (2020) Is child care affordable? Policy brief on employment, labour and social affairs.
https://www.oecd.org/els/family/OECD-Is-Child care-Affordable.pdf
Pacini-Ketchabaw, V., & Prentice, S. (2021b). Early Childhood Education in Canada During a
Pandemic. Journal of Childhood Studies, 1–2. https://doi.org/10.18357/jcs463202120385
Rosenberg, N. E., & Sirkus, L. (2011). Survival Analysis Using SAS: A Practical Guide. Second
Edition By Paul D. Allison. American Journal of Epidemiology, 174(4), 503–504.
https://doi.org/10.1093/aje/kwr202
Schom-Moffatt, P. (1984). The bottom line: Wages and working conditions of workers in the
formal day care market. Ottawa, ON: Status of Women Canada.
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and
event occurrence. Oxford University Press.
Statistics Canada (2020a). Labour market outcomes for college and university graduates, 2010 to
2016. Technical Reference Guides for the Education and Labour Market Longitudinal
Platform (ELMLP). Catalogue no. 37200001 ISBN 978-0-660-36297-7.
https://www150.statcan.gc.ca/n1/en/pub/37-20-0001/372000012020004-
eng.pdf?st=D0DNT4VG
Statistics Canada (2020b). Interprovincial and territorial mobility indicators of newly certified
journeypersons, 2008 to 2017. Technical Reference Guides for the Education and Labour
Market Longitudinal Platform (ELMLP). Catalogue no. 37200001 ISBN 978-0-660-
36297-7. https://www150.statcan.gc.ca/n1/pub/37-20-0001/372000012020002-eng.htm

31
Stinebrickner, T. R. (2001). Compensation policies and teacher decisions. International
Economic Review, 42(3), 751-780.
Tekin, E. (2021). An Overview of the Child Care Market in the United States.
https://www.nber.org/sites/default/files/2021-10/White%20Paper-Tekin-
101.06.21_revised.pdf
Vickerson, R., Friendly, M., Forer, B., Mohamed, S., & Nguyen, N. T. (2022b). One year later:
Follow up results from a survey on COVID-19 and child care in Canada.
Zarifa, D., Seward, B., & Milian, R. P. (2019b). Location, location, location: Examining the
rural-urban skills gap in Canada. Journal of Rural Studies, 72, 252–263.
https://doi.org/10.1016/j.jrurstud.2019.10.032

32
Appendix Table 1: List of institutions providing ECE programs in 2017 (ranked on number of
graduates)
Number of
Program ECE Percentage of
Institution Name Province Type graduates ECE graduates
Ryerson University - Parent Institution ON Bachelor 350 5.07%
George Brown College of Applied Arts and Technology ON College 340 4.93%
Above
Mohawk College of Applied Arts and Technology ON College 230 3.33%
Above
Humber College of Applied Arts and Technology North Campus ON College 230 3.33%
Algonquin College of Applied Arts and Technology ON College 220 3.19%
Above
Université Montréal - Parent Institution QC College 200 2.90%
Conestoga College of Applied Arts and Technology ON Bachelor 180 2.61%
Bow Valley College AB College 180 2.61%
Fanshawe College of Applied Arts and Technology ON College 170 2.46%
Centennial College of Applied Arts and Technology ON College 170 2.46%
Above
Université du Québec Montréal QC College 170 2.46%
Saskatchewan Polytechnic - Parent Institution SK College 160 2.32%
CÉGEP Marie-Victorin QC College 140 2.03%
Seneca College of Applied Arts and Technology ON College 140 2.03%
Niagara College Canada ON College 130 1.88%
St. Clair College of Applied Arts and Technology ON College 130 1.88%
Collège Boréal d'art appliqués et de technologie - Parent Institution QC College 120 1.74%
CÉGEP de Sainte-Foy QC College 120 1.74%
New Brunswick Community College - Parent Institution NB College 110 1.59%
Collège TAV QC College 110 1.59%
Red River College - Parent Institution MB College 110 1.59%
Cité Collégiale - Parent Institution ON College 110 1.59%
Loyalist College of Applied Arts and Technology ON College 100 1.45%
CÉGEP du Vieux Montréal QC College 100 1.45%

33
University of Guelph - Parent Institution ON Bachelor 100 1.45%
CÉGEP Édouard-Montpetit QC College 100 1.45%
Georgian College of Applied Arts and Technology ON College 90 1.30%
Above
Université du Québec Trois-Rivières QC College 90 1.30%
Collège Vanier / Vanier College QC College 90 1.30%
CÉGEP de Saint-Hyacinthe QC College 80 1.16%
Capilano University BC Bachelor 80 1.16%
University of Toronto - Parent Institution ON Bachelor 70 1.01%
Collège Communautaire du Nouveau Brunswick NB College 70 1.01%
Sheridan College of Applied Arts and Technology ON College 60 0.87%
Collège régional de Lanaudière à L'Assomption QC College 60 0.87%
CÉGEP de l'Outaouais QC College 60 0.87%
Red Deer College AB College 60 0.87%
CÉGEP de Jonquière QC College 60 0.87%
CÉGEP de Saint-Jérôme QC College 50 0.72%
CÉGEP Montmorency QC College 50 0.72%
Collège Shawinigan QC College 50 0.72%
Lethbridge College AB College 50 0.72%
Cambrian College ON College 50 0.72%
Campus Notre-Dame-de-Foy QC College 50 0.72%
CÉGEP de Rivière-du-Loup QC College 50 0.72%
Nova Scotia Community Colleges (NSCC) - Parent Institution NS College 50 0.72%
Grant McEwan University AB College 50 0.72%
CÉGEP de Sherbrooke QC College 50 0.72%
Douglas College BC College 40 0.58%
CÉGEP Beauce-Appalaches QC College 40 0.58%
CÉGEP d'Abitibi-Témiscamingue QC College 40 0.58%
CÉGEP Lionel-Groulx QC College 40 0.58%
Canadore College of Applied Arts and Technology ON College 40 0.58%
Portage College AB College 40 0.58%

34
 Above
Université Sherbrooke QC College 30 0.43%
CÉGEP de Valleyfield QC College 30 0.43%
CÉGEP de Sorel-Tracy QC College 30 0.43%
CÉGEP de Granby-Haute-Yamaska QC College 30 0.43%
Collège La Salle QC College 30 0.43%
Above
Langara College BC College 30 0.43%
Brock University - Parent Institution ON Bachelor 30 0.43%
Assiniboine Community College MB College 30 0.43%
North Island College BC College 30 0.43%
Lakeland College AB College 30 0.43%
College of the Rockies BC College
College of New Caledonia BC College
CÉGEP Gérald-Godin QC College
Northern Lakes College AB College
University of British Columbia - Parent Institution BC Bachelor
CÉGEP de St-Félicien QC College
L'École Technique et Professionnelle MB College
Northern Lights College BC College
CÉGEP de Drummondville QC College
770 11.16%
Collège Laflèche QC College (combined) (combined)
College of the North Atlantic - Parent Institution NL College
Camosun College BC College
Lambton College of Applied Arts and Technology ON College
Mount Royal University AB Bachelor
NorQuest College AB College
University of the Fraser Valley BC College
Institut Teccart inc. QC College
CÉGEP de Rosemont QC College
Collège Heritage / Heritage College QC College

35
Holland College - Parent Institution PE College
Vancouver Island University BC College
Medicine Hat College AB College
Concordia University ON Bachelor
University of Victoria BC Bachelor
Vancouver Community College BC College
Okanagan College BC College
Thompson Rivers University - Parent Institution BC College
CÉGEP de Rimouski QC College
Champlain Regional College - Parent Institution QC College
CÉGEP de la Gaspésie et des Îles QC College
CÉGEP de Sept-Îles QC College
Sault College of Applied Arts and Technology - Parent Institution ON College
CÉGEP de Thetford QC College
CÉGEP Saint-Jean-sur-Richelieu QC College
Durham College of Applied Arts and Technology ON College
Grande Prairie Regional College AB College
University of Winnipeg - Parent Institution MB Bachelor
CÉGEP de La Pocatière QC College
Nicola Valley Institute of Technology BC College
St. Lawrence College of Applied Arts and Technology - Parent Institution ON College
Selkirk College BC College
Simon Fraser University BC College
CÉGEP André-Laurendeau QC Bachelor
York University - Parent Institution ON Bachelor
Keyano College AB College
Université Sainte-Anne - Parent Institution NS College
Collège Ellis, campus de Drummondville QC College
CÉGEP de Lévis-Lauzon QC College
Northwest Community College BC College

36
Northern College of Applied Arts and Technology - Parent Institution ON College
Kwantlen Polytechnic University BC College
CÉGEP de Victoriaville QC College
Confederation College of Applied Arts and Technology ON College
Sources: Post-Secondary Information System (PSIS) 2016-2017
Notes: For institutions in italic, the number of graduates in each institution cannot be released due to confidentiality rules.
The figure reported in the table is a grouped number.

37