5.4 Health and education workforce mandates Ngā whakature mō te kāhui kaimahi hauora, mātauranga
5.4. Health and education workforce mandates | Ngā whakature mō te kāhui kaimahi hauora, mātauranga
During the pandemic, many countries required employees in specific occupations to be vaccinated against COVID-19 to continue working in their fields. These vaccination requirements became known as 'vaccine mandates'. The reasons behind mandates varied by country and occupation. Reasons included increasing overall vaccination levels, reducing COVID-19 transmission in specific workplaces, and reducing the risk of essential service disruption by staff illness.
The New Zealand Government first introduced vaccine mandates in April 2021 to cover some employees working at the border (that is, ports, and airports). Over the course of 2021, government vaccine mandates were expanded to cover other border workers, fire and emergency staff, education and health workers, prison staff, police officers and New Zealand Defence Force staff. A Cabinet paper from October 2021 estimated that Government-issued vaccine mandates for workers in border, health, education and correctional settings would cover roughly 15% of the country's total workforce.19
The vaccination mandates for the health and education workers were announced on 11 October 2021.20 Affected workers were required to have their first dose of a COVID-19 vaccine by 15 November 2021, and their second dose by 1 January 2022.21 Overall, the mandates applied to more than 350,000 workers in the health and education sectors (Figure 54).
Figure 54: Total jobs affected by health and education mandates in New Zealand
Total jobs, by ANZSIC06 Level 3 class, March 2020 to December 2023
Source: Stats NZ, Infoshare, Linked Employer-Employee Dataset
5.4.1 Many workers were vaccinated before the mandates | He tokomaha ngā kaimahi i werohia i mua i ngā whakature
Around 70% of the New Zealand population had received at least one vaccine dose by 11 October 2021, the date on which the health and education mandates were announced.
Vaccination rates for particular sectors or occupations are not directly available from the Ministry of Health or other public sources. To investigate the employment impacts of these mandates on affected workers, we examined a longitudinal dataset of skilled workers that were employed in the health or education sectors and aged between 20 and 74 years in September 2021 (the month before mandates were announced).22
By the time the mandate for the education and health sectors was announced, 88.4% of the education workforce and 94.2% of the health workforce had received one or more doses of the vaccine (Figure 55). The rates for the health sector were high because health workers were some of the first to be eligible to receive the vaccine. For example, nurses and other health workers in frontline roles in managed isolation and quarantine, and medical staff who were frontline vaccinators, were eligible to receive vaccines from February 2021.
Figure 55: Cumulative vaccinations, health and education workers
Vaccination rate (one or more doses), skilled education and health sector workers, 2021–2022
Source: Royal Commission Phase Two analysis of Stats NZ Integrated Data Infrastructure data
Notes: Workforce determined by a combination of employer industry and worker qualification.
By 1 January 2022, the time workers were required to have received their first dose of the vaccine, these rates had risen to 95.8% of education workers and 97.2% of health workers. By mid-2022, the proportions of staff that had received one or more doses had risen to 96.7% and 97.9%, respectively.
5.4.2 Some workers left the workforce before the mandates were introduced | I wehe ētahi kaimahi i te kāhui kaimahi i mua i te whakatūnga o ngā whakature
Worker turnover is a fundamental part of employment in all sectors. People leave and join new organisations for a range of reasons – a new challenge, a promotion or pay rise, through to changes in family circumstances.
Figure 56 shows separations, that is, the number of workers who left their health-sector employer for any reason between 2015 and 2024. There was a general upwards trend in separations from around 2015, and some signs of more workers leaving health sector jobs when mandates were introduced at the end of 2021.
Figure 56: Worker separations, New Zealand health sector
Worker leaving their employer for any reason, quarterly, by employer type, Q1 2010 to Q3 2024
Source: Stats NZ, Infoshare, Linked Employer-Employee Dataset.
By contrast, no longer-term trends in the education worker separation data are evident over the same period (dotted lines in Figure 57). The peaks and troughs in the solid lines reflect the school year.
Figure 57: Worker separations, New Zealand education sector
Worker leaving their employer for any reason, quarterly, by employer type, Q1 2010 to Q3 2024
Source: Stats NZ, Infoshare, Linked Employer-Employee Dataset.
Notes: Dotted lines indicate 4-quarter rolling averages.
5.4.3 Understanding the effects of mandates on workers | Te mārama ki ngā pānga o ngā whakature ki ngā kaimahi
Using the longitudinal dataset of skilled workers, we divided workers into three groups based on if and when they received their first dose of the vaccine (Table 4).
Table 4: Vaccination groups, health and education workers
| Vaccination group | Employees in sample | |
|---|---|---|
| Health sector | Education sector | |
| Early vaccinated – those who had received their first dose before the mandate was first announced (11 October 2021) | 111,276 | 88,596 |
| Late vaccinated – those who had received their first dose between the announcement and the first dose deadline (15 November 2021) | 3,543 | 7,443 |
| Unvaccinated – those who were still unvaccinated after 15 November 2021 | 3,291 | 4,182 |
Health sector analysis
Figure 58 shows the probability that people in each of the three vaccination groups being employed in the health sector (panel (a)). The bottom panel (b) shows the probability of being in employed in any job in New Zealand.
The probability of being employed in the health sector is 100% for all three vaccination groups in September 2021 (by design), the month before the mandate announcement (panel (a) of Figure 58). Each vaccination group had quite similar labour market experiences prior to the announcement of the vaccine mandate.
In the period after the announcement, the employment probabilities of the three groups diverged. By 1 January 2022, the booster dose deadline, a large proportion of the Unvaccinated group had left the health sector, and more had left by a few months later. Some of the Unvaccinated group returned to the health sector from September 2022, after the mandates were revoked.
The introduction of the mandate for health workers encouraged 3.0% of the workforce to get vaccinated (the ‘Late vaccinated’ group in Table 4).
Figure 58: Health sector worker employment by vaccination group in New Zealand
Probability of being employed in the health sector or any sector, by vaccination group, January 2020 to July 2024
(a) Probability of being employed in the health sector
(b) Probability of being employed anywhere in New Zealand
Source: Royal Commission Phase Two analysis of Stats NZ Integrated Data Infrastructure data
Notes: Workforce determined by a combination of employer industry and worker qualification.
The difference in employment patterns for the three groups may not be solely due to the imposition of mandates. There could be other differences between the people in each vaccination group. For example, those in one group might be older, on average, and so be more likely to retire. Because of the richness of the data available in the IDI, we can control for multiple factors, such as age, gender, employment history (for example, years in sector; years outside sector; total years in work), wages, ethnicity, where workers live, and, for migrants, their country-of-birth and year of arrival in New Zealand.
In Figure 59 we control for these factors. It shows the difference in probability of being employed in the health sector or another sector for the Unvaccinated, compared with the Early vaccinated groups. Once we control for differences between groups, the employment patterns prior to the mandate announcement are very similar (January 2020 to September 2021).
There are however differences in employment patterns after the announcement of the vaccine mandates. For most of 2022, the likelihood of an Unvaccinated worker remaining in the health sector is over 40% lower than for an Early vaccinated worker. This is shown in the green line in Figure 59. Unvaccinated workers were around 10% more likely than Vaccinated workers to be working outside the health sector (orange line).
After two years, Unvaccinated workers were still 25% less likely to be employed in the health sector.
Figure 59: Difference in probability of employment by vaccination status, in health sector or non-health sector job
Difference in employment probability between Unvaccinated and Early vaccinated workers, in health sector or non-health sector jobs, January 2020 to July 2024
Source: Royal Commission Phase Two analysis of Stats NZ Integrated Data Infrastructure data
Notes:
1. Dotted lines depict 95% confidence intervals.
2. Control variables include age effects that vary by sex, employment history, wage, ethnicity, post-graduate degree, residential location, country-of-birth, and year arrived in New Zealand (for migrants).
3. Sample excludes people who died or left New Zealand.
Those who declined to get vaccinated before the mandates' implementation (2.8% of healthcare staff) faced income and employment losses. Employment rates were 70%, compared with 89% for Early vaccinated. Once other factors were controlled for, the gap in employment rates reduced from 19% to 16%. Conditional on having a job, average annual real wage growth for the Unvaccinated group was 3.9% lower, or 6.3% lower if they had left the health sector, showing real wage losses compared with pre-mandate wages.
Education sector analysis
Most education worker movements happen between the end of one school year and the beginning of the next. This seasonal pattern is reflected in a dip in the probability of employment in January and an increase in February. These can be seen in both panels of Figure 60.
The initial drop in employment for the Unvaccinated group was more than for the usual January dip. This appears to have been followed by a further outflow of workers to a low point in March 2022.
The introduction of the mandate for education workers encouraged 7.4% of the workforce to get vaccinated (the ‘Late vaccinated’ group in Table 4).
Some Unvaccinated workers returned to the sector over the remainder of the 2022 school year, and their employment rose further in 2023.
Unvaccinated workers were less likely to be employed in any sector compared with the vaccinated groups, as shown in panel (b) of Figure 60.
Figure 60: Education sector worker employment by vaccination status in New Zealand
Probability of being employed in the education sector or any sector, by vaccination group, January 2020 to July 2024
(a) Probability of being employed in the education sector
(b) Probability of being employed anywhere in New Zealand
Source: Royal Commission Phase Two analysis of Stats NZ Integrated Data Infrastructure data
Notes: Workforce determined by a combination of employer industry and worker qualification.
Figure 61 controls for age, sex, work history and other factors that could differ between the Unvaccinated and Early vaccinated groups. There is a rapid decline in the proportion of Unvaccinated workers remaining in the sector (to around 50%), but this bounces back more quickly than was the case for the health sector Unvaccinated group. The education mandates were revoked in April 2022, several months before those covering the health sector.
In March 2022, after the mandate vaccination deadline, the likelihood of an Unvaccinated worker remaining in the education sector was around 50% lower than for an Early vaccinated worker (greenline in Figure 61). However, this changed over 2022 to around 25% lower. Unvaccinated workers were around 5% more likely than Vaccinated workers to be working outside the education sector (orange line).
After two years, Unvaccinated workers were nearly 20% less likely to be employed in the education sector.
Figure 61: Difference in probability of employment by vaccination status, in education sector or non-education sector job
Difference in employment probability between Unvaccinated and Early vaccinated education workers, in education sector or non-education sector job, January 2020 to July 2024
Source: Royal Commission Phase Two analysis of Stats NZ Integrated Data Infrastructure data
Notes: Workforce determined by a combination of employer industry and worker qualification.
1. Dotted lines depict 95% confidence intervals.
2. Control variables include age x sex, employment history, wage, ethnicity, post-graduate degree, residential location, country-of-birth, year arrived in New Zealand (for migrants)
3. Sample excludes people who died or left New Zealand
Those who declined to get immunised before the mandates' implementation (4.2% of education workers) faced income and employment losses. Conditional on having a job, average annual real wage growth was 2.5% lower, or 4.7% lower if the workers had left the education sector.
19 Cabinet Paper, Supporting vaccination requirements in the workplace, 26 October 2021, para 31, https://www.mbie.govt.nz/dmsdocument/19926-supporting-covid-19-vaccination-requirements-in-the-workplaceproactiverelease-pdf
20 The cut-off dates for these mandates were updated on 23 October 2021.
21 Press release by Hon. Chris Hipkins, 'Expanded vaccination order for health and disability, education and prison workers', 23 October 2021, https://www.beehive.govt.nz/release/expanded-vaccination-order-health-and-disability-education-and-prison-workers
22 This research linked vaccination data to information on employment and qualifications using the anonymised data in Stats NZ's Integrated Data Infrastructure. It is possible to identify the sector in which a person is employed (from income tax data) with qualifications and occupation (from the Census) to identify teachers, doctors, nurses and other education and health professionals. See the inside front cover for the disclaimer that applies to this research, and the Appendix for further information on the methodology used.