Primary school reopenings and parental work

In this paper, we exploit the geographical pattern of primary school reopenings during COVID-19 pandemic in Canada to estimate impact on parental employment and work hours. We use a triple-difference approach, which first compare parents primary-school children regions where schools reopened similar remained closed add older, secondary-school as an additional control group. separately for mothers fathers, single living dual-parent households. find positive actual hours worked. The effects tend be stronger mothers, but are also present fathers households spring 2020. Overall, experienced 18 percentage point increase their at rate following reopenings. split our sample according whether job can done from home, impacts those whose jobs cannot easily home. La réouverture des écoles primaires et le travail parents. Dans cet article, nous exploitons la dispersion géographique réouvertures d’écoles durant pandémie de au afin d'estimer l'impact sur l'emploi leurs heures travail. Nous utilisons une approche triple différence, dans laquelle comparons tout d'abord les d'enfants à l’école primaire régions où ont réouvert aux similaires sont restées fermées, puis ajoutons plus âgés, secondaire, comme groupe contrôle additionnel. estimons séparément pour mères pères, monoparentaux ceux ménages deux trouvons un positif effectivement travaillées. Les effets tendent être marqués monoparentales, mais aussi présents pères printemps En tout, monoparentales connu hausse points pourcentage leur taux d'emploi (au travail) suite d’écoles. séparons notre échantillon selon que puisse fait maison forts dont ne peut facilement maison. response order limit infections, childcare services, secondary down across March Most did not reopen through end year. Children were home schooled, with varying levels support teachers online schooling, access materials (Frenette al. 2020). Home schooling caring young put pressure parents’ capacity work, it or not. While lockdown measures have affected everyone's ability expected that school-aged above beyond childless workers (Alon 2020a). Our contribution is specific status Because 30.2% Canadian (aged 12 below),11 Source: Authors’ calculations Statistics Canada's Labour Force Survey, average year 2019, January December. K-12 education has vital role economy (Green 2021), understanding measure essential given closures likely affects could deliver services pandemic. Recent documented labour market situation pandemic, essentially showing decrease (Qian Fuller go further by exploiting differences reopening patterns identify causal effect work. Schools facilities within few days between 13 20, School combination host other measures, triggered important force country (Lemieux It therefore possible variation time space capture initial closures. Instead, investigate reopenings—the flip side closures—because these varied space.22 microempirical approach line Alon (2020b), who gender inequality United States using quantitative macroeconomic model. Primary indeed uniform country. May 2020 Quebec outside Montreal, then opened fall finally, 2021, they open everywhere except Southern Ontario. These variations used quasi-experimental design gain insights into decision may been combined changes unit, outcomes. Specifically, Related literature suggests prolonged slow reverse gains towards equality market. Qian (2020) gap probability widened 6.5 February youngest child was age. Adams-Prassl show April 2020, women spent significantly more than men Germany Kingdom—on top one-hour differential care tasks “typical” day. separately, document parents, well Heggeness immediate timing US finds initially took week leave reduced indicators fathers. More specifically, results suggest increased employed 9.0 (pp) 9.8 pp, respectively. For large 21.8 highlighting clear Additionally, number worked 8.4 per 3.2 households, limited Finally, had larger negative mothers’ relative activity recovered only partially once reopened, whereas fathers’ almost completely. closures, directly contributed exacerbation gap. Reopening should seen opportunity mitigate socioeconomic inequalities. outline paper follows. Section 2 briefly reviews related literature. Sections 4 5 data empirical strategy. presented section 6. 7 concludes. outcomes extremely does current situation. Jaume Willén (2018) teachers’ strikes participation. They strong evidence temporary Argentina maternal participation, translated loss earnings mothers. On average, when main breadwinner. (2018), companies laying off thousands same closing, so context, while informative preferences, very different generated A though se, looks schedules pay Duchini Van Effenterre introducing instruction Wednesdays France, thus going four-day five-day week. shifted part-time full-time employment. Fathers respond new schedules, result closing gap, both terms pay. topic actually comes research. Studies reform provided low-fee contributes rise participation short long term (see example Lefebvre Merrigan 2008, Baker Haeck 2015). reveals tightly linked availability affordability childcare, seldom depends it, provides context Lemieux (2020a) documents authors weekly decreased 32% among aged 20 64 Canada. At time, rates 15%, decline concentrated bottom distribution (in quartile). declines women. study own, wake segment age strongest six years old. This especially true low-educated Thus, everyone various degrees, expect faced constraint because (all else being equal). Multiple studies raise that, contrary some affirmations coronavirus great leveller, inequalities fact (Blundell Wright Gender particularly interesting case. Large shocks, such birth child, disproportionate compared men, contributing substantial part (Kleven 2019a;b, Connolly Kingdom lose men. Collins (2021) 20% 50%, argue direct consequence Yet, rapid working flexible arrangements, involvement responsibilities, may, longer term, benefit 2020b). epicentre crisis shows get infected SARS-CoV-2 suffer emotional challenges 2020b, Springmann There children's learning representative reliable sources. Maldonado De Witte Engzell administrative Belgium Netherlands, respectively, academic achievement children. vulnerable interruption lasted eight weeks, Canada, over 16 weeks most comparable provinces much 30% small interruptions exploits identification strategies, summer interruptions, strikes, previous pandemics entry regulations (e.g. Davies Aurini 2013, Meyers Thomasson Frenette 2008). unequal long-lasting consequences. losses summer, appears overall masks heterogeneous (Cooper 1996). favourable family environments seem strike, least lose, less experience net abilities (Atteberry McEachin Belot Webbink 2010). polio beginning 20th century long-term completing high (Meyers (2019) suffered due lower later life (by 3.2% 1.9% males females, respectively). perseverance yet measured hard measure, no periodically collecting youth. know extent affect children, bound factor how best allocate housework, leisure schooling. families negatively afffected. spend begin (Guryan Ramey 2010), external resources help them effectively, Bacher-Hicks areas higher income better internet searches school- parent-centred summary, will reflect adjustment adjustments occupation sectors others economic slowdown come back sections 6 describe strategy results. shut effective 13, Ontario 14, days, all closed. year, before province its schools, starting 11 Greater Montreal Area (GMA) (Quebec Government 2020a, b). full until third June surroundings GMA), kids special needs. Quebec, started basis June. example, Manitoba, Prince Edward Island British Columbia, 1, mainly basis. Unfortunately, documentation rather incomplete. Also, breaks vary typically start fourth As result, exclude months (June, July August) analysis, unless stated otherwise. fall, person country, September. High students went often Ontario, choice return continue virtual Furthermore, second wave (Ontario 2021). Elsewhere January, continued mix in-person (for Alberta 2021. Table 1 summary closure opening each Clearly, conjunction easing measures. instance, sectors, businesses, mandated provincial government applied entire (albeit slightly schedule). reopenings, however, implemented GMA, leading Quebec. Other eased many place periods, openings supply stems one other. preschoolers spaces. Between close Columbia. Some kept open. gradual uneven relatively low take-up rates: 5% reported attending (Statistics source information spaces one-time survey taken last (Friendly available. discuss implications having household below. confidential microdata Survey (LFS) 2017 2021 LFS monthly captures state official unemployment 58 Employment Insurance Regions (EIRs) conducted month, personal interviews, telephone electronic questionnaires. Participation mandatory regulated under Act. designed provide conditions EIRs. target population includes non-institutionalized individuals 15 older. study, household-level framework, unit observation household, rely obtained available public-use (the files contain household). allows us presence thereby improving analysis (2020). sample, keep either 12, high-school 17; including primary- Households types differently older siblings take younger ones. 6, test robustness inclusion. without 17 adults preschool 40% exclusion Reassuringly, excluding preschoolers. restrict attention 55 include prime-aged avoid grandparent. All estimations weighted sampling weights Individuals assigned 94 non-overlapping r defined variables LFS. 83 Census Metropolitan Areas (CMAs) rural CMAs, province, two (Northern Ontario). CMAs well-defined units included must tight links CMA's core city. Out regions, located GMA. generally throughout winter. Classes individually closed, periods cases class, completely LFS, (22 regions), effort wave. January. presents descriptive statistics (inclusively). vast majority (around 90% 80%, respectively) live 5.1% 16.4% percentages 7.9% 21.5%, table postsecondary education; educated about seven consistent younger. With respect location residence, around 11% lives 38% Similar observed children.33 appendix, tables A1 A2 industries occupations construction manufacturing (13.8% 13.3%, health social assistance sector educational (25.2% 13.9%, focus three indicates if parent absent look reference 91% 81% mothers). note necessarily mean work: reasons one's own illness, relatives, maternity leave, vacation. Therefore (either person). (excluding months), four percent During (March May), spiked 15% 10% fathers.44 (January May). distinction (including not) maintained bring insight mechanisms shape Actual (those employed, zeroes) 35 25 Conditional 40 41 week, 33 34 Figures fraction rest Canada.55 appendix. request. seasonally adjusted Here, included. what figures months. markers hollow (during pandemic). marked arrows red Fraction region NOTES: Y-axis (at work). included, just active Summer exceptionally Parents set 0. Vertical bars denote 95% confidence intervals. Arrows dots indicate observations ROC = Rest SOURCE: Survey. [Color figure viewed wileyonlinelibrary.com] Res