# Quantifying the macroeconomic impact of Covid-19 school closures on human capital

**Christine de la Maisonneuve, Economist, Balazs Egert, Senior Economist, OECD, Dave Turner, Head of Macroeconomic Analysis, OECD. Originally published on VoxEU.**

The COVID-19 pandemic has led to partial or complete school closures in almost every country in the world. On average across OECD countries, school buildings were completely closed for 13 weeks and partly closed for another 24 weeks between March 2020 and October 2021, cumulatively equivalent to about one full school year. [1]. The learning losses associated with school closures can be difficult to compensate for and therefore can have long-term economic impacts on affected students, with possible long-term macroeconomic consequences (Ilzetzki 2020, Kuhn et al. 2020, Popova et al. 2020).

**figure 1** Duration of school closures from March 2020 to October 2021

*Note*: Complete school closures refers to situations where all schools have been closed across the country due to COVID-19. Partial school closure refers to the closure of schools in some regions or for some grades, or with a reduction in in-person instruction. The total number of closes is defined as the simple unweighted sum of these two aggregates. *A source*: UNESCO.

We use a new measure of human capital derived from Égert et al. (2022), which combines mean years of schooling (MYS) and OECD data from the Program for International Student Assessment (PISA). The new indicator is a cohort-weighted average of past PISA scores (representing the quality of education) of the working-age population and the respective mean years of schooling (representing the quantity of education). Weights for PISA scores and mean years of schooling are estimated based on regressions that take into account how well the cohort-weighted variables explain scores from the Program for the International Assessment of Adult Competence (PIACC).

Based on this new metric, we can separately calculate the impact of the pandemic on PISA scores and average years of schooling (MYS) and feed these data into the overall measure of human capital. For each affected cohort, we summarize the impact of the pandemic on MYS and PISA scores to estimate the overall impact on human capital. We calculate them using the human capital elasticities of MYS and PISA estimated by Égert et al. (2022). We then calculate the population-weighted average impact of each affected cohort to represent the global impact on human capital.

The new measure of human capital shows a robust correlation with productivity for OECD countries in cross-country time series panel regressions. This helps us quantify the macroeconomic losses due to school closures reflected in the loss of PISA scores and average years of schooling.

Using these estimates, we consider three scenarios:

**Impact of school closures in spring 2020**in many OECD countries, roughly equivalent to the end of one-third of the school year. This closure period results in a 2.6% reduction in average years of study.[2] and, using the rule of thumb described above, a 0.14 standard deviation decrease in PISA scores.[3]which corresponds to a decrease in PISA scores of 1.1%.[4]**Consequences of school closure for one year**, broadly in line with the average total (total and partial) school closures seen in OECD countries since the start of the pandemic and, according to a first estimate, the learning loss of the most disadvantaged students in the US (US Department of Education, 2022). This scenario results in a -8.2% decline in MYS and a -0.37 SD decline in PISA scores, corresponding to a 2.9% decline in PISA scores.**The effect of a two-year school closure**which has been rare and broadly consistent with complete (complete and partial) school closures in Colombia, Chile, Korea and Mexico since the start of the pandemic, resulting in a -16.5% and 5.6% decrease in MYS and – A drop in PISA scores by 0.72 SD

We estimate the impact of school closures on productivity through the human capital effect for these three scenarios. Multivariate productivity regressions relate productivity to human capital in the presence of a number of control variables such as innovation intensity, product market regulation, and trade openness. The impact will gradually increase as students affected by the pandemic enter the labor market, peaking in 2067. the first, second and third scenarios respectively. The impact will then gradually dissipate until the last affected cohort retires in 2083 (Figure 2). The impact will be greatest in 2067 as that is when all affected cohorts will be in the older workforce and the impact on human capital is most important.

**figure 2** Impact of school closures on productivity

*A source*: Calculations of the authors.

**Comparison with estimates in existing literature**

Empirical literature data, standardized for one-year school closures, imply a significant crisis impact on GDP levels ranging from -1.1% to -4.7% between 2040 and 2050 (Dorn et al. 2020, Hanushek et al. etc.). Woessmann, 2020 and Viana Costa et al., 2021). The researchers used different methods. Dom et al. (2020) have developed different scenarios for precalculations. Viana Costa et al. (2021) calculate economic costs using microsimulation model calculations. Hanushek and Woessmann (2020) use a macro-regression analysis that links GDP per capita to student error correction test scores in several countries. Our results are broadly consistent with most of the literature, with the exception of Hanushek and Woessmann (2020), who found a much larger effect (-4.7%). These results will be equivalent, *ceteris paribus*to influence GDP per capita.

**Mitigation policy**

Mitigating the impact of COVID-19 on human capital is a major policy challenge as most, if not all, education policy reforms have long delays in implementation, meaning that education policies that mitigate the impact of the pandemic will fail to reach the oldest COVID-affected student cohorts. . -nineteen. An additional difficulty is that some of the rules apply to the youngest students. Measures that can be implemented to help catch up with the affected generations of students include the following (OECD 2020, OECD-Education International 2021 and Molato-Gayares et al. 2022):

- Increasing study time by temporarily reducing school holidays and/or adding hours to the school day.
- Revise curriculum to focus on key skills.
- Providing teachers with training.
- Considering the use of digital technologies to improve the diagnosis of learning gaps and promote more personalized learning methods.
- Spreading collaboration and professional ways of working to increase teacher effectiveness

For post-school cohorts, it is important to strengthen young adult education programs. However, they are notoriously not very cost-effective, and compensating for lost learning at a younger age can be very costly to the government budget.

Further action could include expanding and improving the quality of early childhood education, considered by many to be the best value for money, which would be too late for almost all student groups affected by the pandemic. Other education policy reforms found to be positively correlated with student test scores during normal times, but which may also help offset some of the post-pandemic losses for younger generations, include strengthening school accountability and school autonomy, reducing early tracing, and increasing the quality and qualifications of teachers.

[1] These average figures hide large differences between countries. While schools in Switzerland and Iceland were closed in less than ten weeks, school closures in Korea, Chile and Colombia lasted almost a year and a half (Figure 1). It is assumed that the full academic year is 38 weeks.

[2] The percentage loss of MYS is calculated as the loss in training, expressed in school years, divided by the average MYS for the entire workforce. For example, for a loss of 0.32 school years, assuming an average MYS for the entire workforce of 12 years would imply a 2.6% loss in MYS for this cohort (= 0.32/12 x 100%).

[3] At 12 weeks, the decline in PISA is equivalent to a standard deviation of 0.14 (12*0.012); for one year it is 0.37 (12 * 0.012 + (38-12) * 0.009) standard deviation, and for two years it is 0.72 (12 * 0.012 + (76-12) * 0.009) standard deviation.

[4] Loss Percentage in PISA = (Estimated Impact * PISA Standard Deviation)/PISA Baseline Score = (-0.14 * 36.1)/462.