A "She-cession"? The Impact of COVID-19 on Thailand's Labour Market.

• Author: Paweenawat, Sasiwimon Warunsiri

1. Introduction

   The COVID-19 pandemic has caused an unprecedented worldwide economic recession. With lockdowns and social distancing, labour markets have been hit severely and many people are suffering from the dramatic loss of income and jobs. Several recent studies have argued that the effect of the pandemic is highly unequal across countries and depends on the characteristics of the majority of the workers and their occupations (Alon et al. 2020a; Adoms-Prassl et al. 2020). Concerns that the pandemic will increase gender inequality as women are more affected than men (World Bank 2020a).

   In Thailand, despite a relatively low infection rate of COVID-19, the impact has been dramatic due to the country's openness to trade and exposure as a global tourism hub (World Bank 2020b). Additionally, the impact is unprecedented and unequal across demographic, industrial and occupational groups. People working in certain industries have been hit harder than others. For instance, tourism and manufacturing sectors that depend on intermediate inputs from abroad have sustained significant losses, whereas finance and insurance, manufacturing of pharmaceutical and chemical products, and machinery have not suffered as much (ILO 2020a). It is also important to note that government lockdown measures have also exposed workers differently based on their occupations. In particular, whether employees are able to work from home, adapt to the "new normal", and make use of Information and Communication Technology (ICT) determine their level of government assistance.

   This study analyses the impact of the COVID-19 crisis on different demographic groups in the Thai labour market using the Labour Force Surveys conducted in 2018 and 2019, specifically those related to human capital and gender inequality. We construct a new set of indicators by using the International Labour Organization (ILO 2020a) assessment of industries on COVID-19 disruption risk for Thailand. The study utilizes the pandemic-related occupational requirements from the Occupational Information Network (O*NET) to capture the degree of risk in industries and in occupational flexibility in the Thai context.

   Several key findings emerge from this study. First, there are key differences in the impact of COVID-19 across different industrial sectors and occupations in Thailand. Workers in high-risk sectors with low probability of adjustment to social distancing measures are likely to face significant negative labour demand shocks. Second, the impact of the pandemic is not the same across demographic groups. Although women are not objectively worse off than men, COVID-19 potentially exacerbates the pre-existing inequality in the Thai labour market. We find that workers with low education, those in informal sectors, private employees and the young are more likely to be negatively affected by the crisis. Their vulnerability necessitates greater government assistance. From a gender perspective, our findings indicate that marriage shows different effects on men and women in the COVID-19 crisis; preliminary results show that the institution ends up protecting men but hurts women. Moreover, women in informal sectors and those over sixty years of age are more vulnerable than men. Consistent with the World Bank's (2021) findings on Thailand's labour market, workers' hours dropped sharply during the early stages of the pandemic, and the decrease was sharper for women than for men (7.2 per cent versus 5.7 per cent). Meanwhile, the unemployment rate is particularly high for young people; and those with low skills and less education have had their working hours reduced drastically (World Bank 2021).

   Third, the pandemic recession displays some irregularities compared to previous economic downturns in Thailand, specifically the 1997 Asian Financial Crisis and the 2008 Global Financial Crisis. This time, the health crisis has not only had a general detrimental impact on workers in the affected sectors but also adversely affected their work flexibility.

   Our study contributes to the literature by evaluating the impact of COVID-19 on the labour market in terms of both the predicted economic outcome of industries as well as the work flexibility of the affected employees. Our industry-occupation pairing can provide additional insights into risk exposure to COVID-19 by shedding light on the vulnerable groups, which can help the government develop adequate economic recovery policies.

   The paper is organized as follows. The next section provides insights into the pandemic-induced disruption in Thailand. The third section describes the data and methodology employed in the study. The subsequent section elaborates on the full impact of COVID-19 with respect to prevalent gender differences in the labour market, while the fifth section analyses the effect using individual characteristics. The sixth section compares "regular" crises to the COVID-19 crisis, and the seventh section discusses relevant policy implications. The final section concludes.

2. Recent Economic Disruptions in Thailand

   In responding to COVID-19, Thailand has implemented a host of social distancing restrictions and travel bans. While the country has achieved some level of success in flattening the infection curve, the economy has been severely hit. The National Economic and Social Development Council (NESDC) indicated a 1.8 per cent contraction of the Thai economy in the first quarter of 2020, which is expected to worsen with time. In fact, the World Bank (2020c) projected an 8.3 to 10.4 per cent contraction of the Thai economy, which is amongst the highest declines in the region. This is mainly because Thailand's exposure to trade and remittance, dependence on service exports, and the risk of travel disruption are large compared to other Southeast Asian countries.

   Different industrial sectors have been affected differently by COVID-19. According to NESDC (2020), in the first quarter of 2020, accommodation and food services, manufacturing, agriculture, transportation and storage and construction declined, while wholesale and retail trade, electricity and gas, ICT and financial and insurance expanded. The unemployment rate, however, remained low at 1 per cent (NESDC 2020). This was consistent with pre-pandemic records that showed that Thailand had the ninth-lowest unemployment rate (of around 1 per cent) among 233 countries (World Bank 2018). It must be noted that, in the Thai context, a low unemployment rate is indicative of several structural problems in the labour market, including a large proportion of informal employment and underemployment, skill-mismatch employment, and discouraged workers (Bank of Thailand 2019). In fact, over half of all Thai workers are in the informal sector, and they are amongst the most vulnerable labour groups in the country (ILO 2020a).

   Thai workers are affected differently by occupation, where female-dominated occupations in retail and related service sectors are more affected by the pandemic (World Bank 2020a; ILO 2020a). Female workers in Thailand have more flexible jobs than male workers (Lekfuangfu et al. 2020).

   In May 2020, Thailand lifted some lockdown measures (Corona Virus Epidemic Management Center 2019), which benefited employees in high physical proximity jobs. As a matter of fact, the government reopened all businesses and activities in July 2020 as infection rates decreased to single-digit figures. However, in December 2020, as the number of COVID-19 patients rose sharply--partly because of migrant workers and Thai residents returning from other countries--mild lockdown measures, including closure of high-risk places, restrictions on mobility, and bans on mass gatherings, were reimposed (World Bank 2021).

3. Data and Methodology

   This study makes use of Thailand's Labour Force Surveys of 2018 and 2019 conducted by the National Statistical Office to analyse the impact of COVID-19 on the country's workforce. The sample includes 235,958 observations (121,226 in 2018 and 114,732 in 2019) and is restricted to workers age over the age of fifteen. For the subgroup analysis, we have restricted the sample to different age groups. The composition of education (in Appendix Table A1) by gender shows that a larger share of women obtains lower and higher education than men (27.50 per cent to 24.86 per cent; 19.08 per cent to 11.91 per cent), while a larger share of men has primary, secondary and high school education. Overall, Thai women, on average, have more years of schooling than men (8.85 versus 8.58 years).

   We use the normalized factors and ordinal ranking of industry-occupation pairs by merging the information on industrial risk and occupational flexibility to assess the impact of the health crisis in the Thai context. The occupational flexibility index is based on COVID-19-related job task requirements from O*NET using factor analysis (Gorsuch 1983; Hamilton 2013), considering occupational characteristics related to adaption of work from home due to social distancing (Mongey, Pilossoph, and Weinberg 2020). Dingel and Neiman (2020) evaluated the impact of social distancing on US jobs and found that 37 per cent of all jobs can be performed entirely from home. They applied this measure to eight-five other countries by mapping the Standard Occupational Classification based on O*NET to the International Standard Classification of Occupations, and showed that less developed countries have lower share of home-based occupations, with Thailand at around 17 per cent, Myanmar at around 10 per cent and Lao PDR at 22 per cent.

   Using these matchings, we adapt the approach to assess the occupational flexibility in Thailand and add additional survey questions to account for the impact. This occupational flexibility contains information from O*NET database on the Work Context Questionnaire and Generalized Work Activities Questionnaire. The complete set of selected variables for factor analysis is presented...