PAPUA NEW GUINEA HIGH FREQUENCY PHONE SURVEY ON COVID-19 Data Collection: May – July 2021 Acknowledgement: This report was prepared by Darcey Johnson, Donny Pasaribu, Christopher Hoy, Shohei Nakamura, Darian Naidoo and James Carroll Waldersee of the World Bank and Stephanie Laryea of UNICEF-PNG using the third round of the World Bank’s High Frequency Mobile Phone Survey to Assess the Socio-Economic Impacts of COVID-19 in Papua New Guinea and the first and second round of UNICEF-PNG’s Social Challenges and Hardship Survey in Papua New Guinea. The team gratefully acknowledges the Korea Trust Fund for Economic and Peace-Building Transitions for financing the data collection and analysis for the World Bank survey and USAID for financing the analysis of the UNICEF-PNG survey. The team also greatly appreciates the contributions of Kristen Himelein and Shuwen Zheng in constructing the survey weights and in- forming the analysis, and Karin Hosking for her support in editing. The authors thank World Bank colleagues Da Woon Chung, Rinku Murgai, Taufik Indrakesuma, Ruth Nikijuluw, Jane Sprouster and Hassan Zaman as well as UNICEF colleagues Judith Bruno, Chiharu Kondo, Jayachandran Vasudevan and Ali Safarnejad for their comments on earlier drafts. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development / World Bank, UNICEF and its affiliated or- ganizations, nor those of the Executive Directors of the World Bank nor the governments they represent. Cover photo: Daru, Western Province (Tom Perry/World Bank) PAPUA NEW GUINEA HIGH FREQUENCY PHONE SURVEY ON COVID-19 Data Collection: May – July 2021 World Bank TABLE OF CONTENTS Executive Summary............................................................................................3 1. Background...............................................................................................5 1.1. Data.............................................................................................................5 1.2. Update on COVID-19 outbreak in PNG.............................................. 6 2. Impact of COVID-19 on Economic Activity..................................... 8 2.1. Employment............................................................................................. 8 2.2. Farm and Non-Farm Business............................................................. 10 2.3. Remittances.............................................................................................12 2.4 Financial Anxiety.....................................................................................13 2.5 Income sources......................................................................................13 2.6 Coping Strategies...................................................................................15 3. Impact of COVID-19 on Human Development............................. 16 3.1. Education and Schooling..................................................................... 16 3.2 Access to Health Care.......................................................................... 18 3.4. Water and Sanitation............................................................................. 20 3.5 Access to Staple Starch, Proteins, Fruits and Vegetables.............. 20 3.6 Food Insecurity....................................................................................... 22 3.7 Public Trust, Community Security, and Intra-household Conflict............................................................. 25 3.7 Child Discipline...................................................................................... 27 Conclusions & Policy Implications.............................................................. 28 2 Abbreviations COVID-19 Coronavirus Disease 2019 CSSE Center for Systems Science and Engineering at Johns Hopkins University DHS Demographic and Health Survey ECDC European Centre for Disease Prevention and Control FCS Fragile and Conflict Afflicted Situations GDP Gross Domestic Product HFPS High Frequency Phone Survey HIES Household Income and Expenditure Survey NCD National Capital District NDoH National Department of Health (Papua New Guinea) NGO Non-Governmental Organization PNG Papua New Guinea UNICEF United Nations Children’s Fund WHO World Health Organization 3 Collecting water near Bialla, West New Britain (Ruth Moiam/World Bank) Executive Summary survey will shed light on the impact of the Delta variant of COVID-19 in PNG. The surveys in this report show that in general there were similar levels of economic activity This joint report by the World Bank and and human development in mid-2021 as UNICEF-PNG presents the findings from two the start of 2021, however the surveys do not mobile phone surveys conducted in May, demonstrate a recovery to pre-pandemic June and July 2021 in Papua New Guinea. levels. There was stabilization across a range The first survey, conducted in May and June, of indicators, including levels of employment, was the third in a series, with the first round access to education and healthcare as well as being completed in June and July 2020 public trust and security. However, there was and the second round being conducted in some variation across regions and wealth December 2020. This survey interviewed a quintiles. The lack of deterioration of conditions total of 2,533 respondents about the impact since the start of 2021 is notable given that of COVID-19 on their economic activity and over this period there was a rapid escalation the results were weighted using information of the COVID-19 crisis in March followed by a from the 2016–2018 Demographic and Health fall in COVID-19 cases throughout May. Survey (DHS) to reflect nationally representative estimates of the socioeconomic impacts of Employment and income levels in mid-2021 COVID-19. The second survey, conducted in were largely unchanged compared to those at June and July 2021, interviewed a total of the start of 2021, however poorer households 2,541 respondents (1880 of whom were from were worse off. Just over half of household the first survey) about the impact of COVID-19 heads reported working at the start of 2021 on a range of areas that broadly can be and in mid-2021, split quite evenly between considered human development. These results agriculture and service sectors (only a very were also weighted using information from the small share reported working in the industrial DHS to develop representative estimates. sector). Despite a slight increase in employment levels (potentially due to seasonality in the The report uses data that predates the arrival agricultural sector), household heads in the of the Delta variant of COVID-19 in PNG and bottom 40 percent were more likely to report therefore should be considered a snapshot of a decline in income, particularly among those changes in economic activity and human that worked part-time. Poorer households development from January to June 2021. The were also much more likely to report that the next round of data collection is currently most common household coping strategy planned to start in November 2021 and that they used was to engage in further income 4 Executive Summary generating activities. Collectively, these findings Water, sanitation, and hygiene services were are consistent with a pattern of poorer people also insufficient to meet household needs. At having to work more to generate a similar a time when handwashing is an important or lower level of income to support their preventive measure to slow the spread of the household. virus, around 40 percent of households did not have sufficient water for handwashing in Access to education and healthcare in the week prior to the survey, and the majority mid-2021 appear to have largely returned to did not have sufficient soap. In addition, over levels similar to those prior to the wave of one-third of households did not have sufficient COVID-19 infections that peaked in March. drinking water, and this was most acute in the During the height of the COVID-19 crisis Islands region and least acute in the Highlands schools were closed and the survey shows that region. most children did not continue their formal education from home. This was primarily be- The surveys did not show large declines in cause schools did not offer distance education public trust and security issues since the options, with 94% of respondents reporting start of the year, however experiences varied this as the reason their children stopped their across regions. Compared to the rest of the learning. Almost all children promptly resumed country, higher shares of respondents in the their studies once schools reopened. Concerns Southern region noted deterioration since about exposure to COVID-19 at health clinics January 2021 in situations related to theft, only deterred a relatively small number of alcohol and drug abuse, and intimidation by respondents from accessing urgent or preven- police, as well as higher declines in overall tive healthcare. Most respondents who want- community trust, which could be an indicator ed to access some form of healthcare were of rising tensions. In addition, there were able to, however the survey did not capture potential warning signs of the impacts of the information about the quality of healthcare prolonged crisis on children, with more than received. one-third of adults in the 15 days prior to the survey having spanked, hit or slapped their The incidence and severity of food insecurity, child with their bare hand or foot – though as well as access to stable starches, proteins, again a lack of baseline data limits the ability fruits and vegetables, varied across regions. to establish a causal link specifically with The severity of food insecurity was lowest COVID-19. in the Highlands region and highest in the Islands region. Coupled with the results on Despite the limited change in conditions declining food affordability (particularly in since the start of 2021, the survey shows regard to rice, which is a key stable starch in there are significant medium-term challenges the Southern and Islands regions), this finding facing PNG in terms of economic activity is further evidence of households having and human development. More than half of difficulty in purchasing and growing sufficient households reported relying on subsistence food. In the absence of a comparable baseline, agriculture as a main source of employment it remains difficult to attribute these issues to and income and there is widespread food the COVID-19 pandemic. insecurity. There are low levels of participation in pre-primary education and some households 5 report being unable to access health care due The objective of the surveys was to measure to distance to facilities. Levels of crime, domestic the continued impacts of COVID-19 on violence, theft and police intimidation are economic activity and human development alarmingly high. Collectively, the survey findings in PNG. The length of each survey was limited highlight many areas for policy makers to to 15 minutes. The survey instruments consisted prioritize, even though they suggest that of the following modules: Basic Information, the impact of COVID-19 on economic activity Employment and Income Loss, Food Access and human development may not have led to and Food Security, Public Trust and Security, a substantially greater deterioration in conditions Assets and Wellbeing, Service Delivery, Roster throughout 2021. However, it is important to of Children Living in the Household (including keep in mind that the surveys do not demon- schooling status), Access to Health, Education, strate a recovery to pre-pandemic levels and and Child Discipline. The recall period for the findings pre-date the Delta variant of CO- current employment was in the previous VID-19 being detected in PNG. week. In addition, retrospective questions were asked about “the start of 2020” as well as 1. Background the situation at “the start of this year”. The dates of implementation for the World Bank survey were May 26 through June 6, 1.1. Data 2021, and the UNICEF survey was collected The data collection for the analysis underlying between June 11 and July 13, 2021. Data this report was from two surveys: the third were collected through a call center set up by round of the high frequency mobile phone Digicel – Papua New Guinea and managed survey (HFMPS) conducted by the World using the Survey Solutions software package. Bank (the World Bank survey) and the second As the original objective of the survey was to round of the Social Challenges and Hardship measure changes as the pandemic progresses, in PNG survey conducted by UNICEF-PNG the third round of data collection sought to (the UNICEF survey). As the surveys used recontact all 2,820 households contacted in overlapping samples, they were analyzed round 2, however only 387 households were together, and the results are presented in this able to be recontacted. This is partly because joint report. For the World Bank survey, as with there is high turnover of phone numbers in the first and second rounds of data collection, PNG as most SIM cards are unregistered, and all survey instruments and procedures were they expire within 6 months. As such, to reach designed in accordance with the best practices the target number of respondents, 2,146 laid out by the World Bank’s COVID-19 replacement households were added to the metho­ dology and measurement task force.i World Bank survey using Random Digit Dialing The UNICEF survey was based on UNICEF’s targeted based on geography and economic COVID-19 Monitoring and Analysis Framework, status (further details about the sampling Multiple Indicator Cluster Surveys (MICS) methodology is in round 1 and 2 reports), for a and MICS Plus modules. Further details about total sample size of 2,533 households. Due to the data collection principles, approach to the very high rate of attrition between survey weighting, and data analysis methodology are rounds, this report focuses on changes in provided in the round 1 and 2 reports.ii conditions during the first half of 2021 as opposed to examining trends since the start of the pandemic. 6 I. Background There was substantial overlap between the To assist in addressing the rising number of World Bank and UNICEF surveys. In total, 1,880 cases, PNG received its first shipment of respondents completed both surveys (all 2,533 COVID-19 vaccines through the COVAX respondents from the World Bank survey were Facility in April. Additional doses of the attempted to be recontacted, however this AstraZeneca vaccine were provided by the was only successful 74.2% of the time). To Australian Government and in May doses of reach the target number of respondents, 661 the Sinopharmiv vaccine that were provided respondents were added to the UNICEF survey by the Chinese government began to be using Random Digit Dialing targeted based on administered (both vaccines require two geography and economic status for a total doses). At the start of August, the Government sample size of 2,541 respondents. of PNG received additional doses of the Johnson & Johnson vaccine via the COVAX 1.2. Update on COVID-19 facility. It is expected that this type of COVID-19 outbreak in PNG vaccine will be easier to distribute as only one Since the start of 2021, the COVID-19 situation dose is required. in PNG has rapidly deteriorated. The cumula- The overall vaccination rate remains very tive total of confirmed cases grew from 834 low. Less than one percent of the population and 9 related deaths on 1st January to 17,775 had received at least one dose of a COVID-19 confirmed cases and 192 COVID-19 related vaccine as of August 2, 2021. Around 86,000 deaths by 2nd August 2021. Most of these people had received at least one dose and cases were reported between the end of 18,000 had been fully vaccinated. Priority was February and the start of June (see Figure 1). given to health and other essential workers, The actual case numbers are likely to have which is why around 61,000 of the people been much higher due to low overall testing, who had been vaccinated fell into one of underreporting and limited contact tracing these categories. capacity.iii Figure 1 – COVID-19 cases and Vaccinations in PNG 250 350 300 2000 Daily vaccines administered 250 daily vaccines smoothed Daily confirmed cases 1500 200 new cases smoothed 150 1000 100 500 50 0 0 01-Jan-21 01-Feb-21 01-Mar-21 01-Apr-21 01-May-21 01-Jun-21 01-Jul-21 01-Aug-21 7 The National Pandemic Act 2020, which Brisbane, Australia were limited. Domestic air was passed on June 16, 2020 to provide a travel was restricted to medical evacuations, legislative mechanism to coordinate the students travelling to and from their usual public health response, remains in effect. place of residence and educational institution, Under a declaration of a pandemic, the legis- and any persons returning to their usual lation establishes the Office of the Controller place of residence. To boost COVID-testing, as the lead agency to combat COVID-19 and positive results from antigen rapid diagnostic manager of the Joint Agency Task Force of tests (Ag-RDT) no longer had to be confirmed the National Control Centre for COVID-19, by PCR testing. With the number of cases granting the Controller the regulatory authority decreasing through May the National Control to impose health response measures and Centre again eased a number of the restrictions oversee their implementation. The health that were imposed through March-April. response is guided by the PNG National Government authorities have responded to Department of Health (NDoH) COVID-19 risks of the pandemic by approving a fiscal Emergency Response Plan, which provides a package of emergency health and economic framework for emergency health response in- relief measures. In early April 2020, the terventions to be implemented from March government mobilized its resources and 20, 2020 to end of February 2022 and will appealed to development partners and the feed into the development of the forthcoming private sector for additional support to protect health sector plan, the National Health the economy and livelihoods of vulnerable Plan 2021–2030. The COVID-19 Emergency households and businesses. The ensuing Response Plan includes COVID-19 response package of health and economic support interventions around clinical management measures totaled roughly US$525 million (2.2 ces, servi­ communications, surveillance, percent of GDP) in 2020. Reflecting the laboratory and operations, procurement and government’s limited fiscal space and logistics, operational research, and develop- anticipated revenue shortfalls, the health and ment partner coordination. economic support package was financed by The Government responded to the recent external low-cost loans, foreign grants, and outbreaks by enacting several Pandemic the use of employees’ pension savings in Measures and Directives in the first half of superannuation funds in job-loss cases. 2021. In March, the Government introduced Commercial banks agreed to provide loan strict control measures including the closure repayment holidays to affected households of schools and only essential travel permitted. and businesses for three months. The tax The Government also requested - after several authority provided deferrals for tax filing and rounds of development partner meetings - payment for two months and prioritized that international Emergency Medical Teams processing of goods and services tax refunds deploy in PNG. All land and sea borders for medical supplies. with Australia, the Torres Strait, Indonesia, International development partners have and Solomon Islands would remain closed, also increased their support to the health including any traditional cross-border travel sector response and to bolster the economy. to West Papua. Flights between PNG and Several support packages have been provided Cairns, Australia were halted and flights into 8 2. Impact of COVID-19 on Economic Activity 2. Impact of COVID-19 In November 2020, Australia approved a AUD$140 million loan to provide additional budget support, which was used to repay Government debt owed to PNG businesses on Economic Activity and inject cash into local markets. The Asian Development Bank, on November 26, This section draws on the World Bank survey approved a US$250 million loan to support to report on levels of employment and income. PNG’s health services, budget and economy, For employment, it focuses on household the agriculture sector, food rations for heads’ employment between the start of vulnerable groups, and loans and subsidies for 2021 (based on retrospective questions) small businesses, in addition to increasing its and mid-2021. Any kinds of work, such as budget support to the health sector from wage, self-employment, and farm activities, is US$100 million to US$150 million in 2020. counted as employment in the questionnaire. In June 2021, the World Bank approved For household-level income, including a US$100 million Crisis Response and remittances, the comparison is made for the Sustainable Recovery Development Policy period from the start of 2020 to mid-2021. Operation to support the Government of Papua New Guinea in: protecting lives and 2.1. Individual employment livelihoods of the population during the (household heads) COVID-19 crisis; and laying foundations for a more sustainable recovery over the Employment levels of household heads were medium-term.” PNG has also benefited from similar in mid-2021 as to what they were at a time-bound suspension of its debt servicing the start of 2021.v The proportion of working due to Australia, Japan, and Germany under household heads remained at almost the the Debt Service Sus­ pension Initiative, same level: 53.7 percent at the start of 2021 committing the Government to devote its and 54.4 percent in mid-2021 (Figure 2). There newly freed resources toward the health, was a slightly faster increase in employment economic, and social impacts of COVID-19. among female household heads compared to male household heads (Panel A). There was a As the situation in PNG continues to evolve, small convergence in employment levels it is critical to continue monitoring the between urban and rural areas. The share of impacts of the COVID-19 pandemic on working household heads in rural areas economic activity and human development. was similar at the start of 2021 to mid-2021 A prolonged health crisis coupled with a whereas the share of working household struggling economy and limited fiscal space heads in urban areas fell marginally during for Government intervention could exacerbate this period (Panel B). Similar to the case in existing development challenges, particularly previous survey rounds, the main reason for the poor and near-poor, and frequent and people cited for moving in and out of the reliable data are needed to inform Government workforce in rural areas was seasonality of and development partner responses going the agricultural calendar. forward. 9 Figure 2 – Proportion of household heads that are working by sex and location (%) (A) By sex (B) By location 70 70 66.2 64.6 % of household heads % of household heads 65 65 60 60 54.1 54.8 53.7 54.4 55 54.4 55 53.7 50 50 53.4 52.4 51.2 49.3 45 45 Start of 2021 Mid 2021 Start of 2021 Mid 2021 All Male Female All Urban Rural Source: Round three of the high frequency mobile phone survey. Employment of those in the bottom 40 percent points) and in contrast to the decline in employ- wealth group slightly increased since the ment for the middle quintile (2.2 percentage beginning of 2021, though the proportion points). This is probably due to agricultural of working household heads in this group seasonality, as the majority of poor households remains much lower than wealthier house- work in the agriculture sector. Nevertheless, holds. From the start of 2021 to mid-2021, the only 45.3 percent of household heads were proportion of working household heads in the working in the bottom 40 group, compared to bottom 40 percent increased by around 2.3 66.2 percent among the top 40 households percentage points (Figure 3). This is a faster and 57.7 percent among the middle wealth pace than the top 40 percent (0.8 percentage quintile group. Figure 3 – Proportion of household heads that are working by wealth group (%) 70 65.4 66.2 % of household heads 65 60 55 59.9 57.7 50 45 40 45.3 43.0 35 Start of 2021 Mid 2021 Bottom 40% Middle quintile Top 40% Source: Round three of the high frequency mobile phone survey. Groups carry their goods up the beach at the East Cape Point, Milne Bay, East New Britain. (Conor Ashleigh/World Bank) 10 2. Impact of COVID-19 on Economic Activity Figure 4 – Proportion of working household heads by economic sector (%) (A) Sectoral composition (B) Shift in economic sectors % of household heads (total=100) 45.5 46.6 Start of 2021 Mid-2021 Agriculture Industry Services Not working 2.4 2.3 Agriculture 27.8 83.7 0 0.9 15.4 Industry 1.3 0 97.4 0.1 0 52.1 51.1 Services 25.4 0.9 0.6 89.4 2.5 Not working 45.6 15.4 0 7.3 89.7 Total 100 Start of 2021 Mid 2021 Agriculture Indu stry Services Source: Round three of the high frequency mobile phone survey. The sectoral composition of employment has Employment income remained stable for been stable, almost equally split between those who were employed in mid-2021 the agriculture and service sectors. As of across all groups and categories. Nationally, mid-2021, 51.1 percent of working household the majority of household heads who were heads engaged with agricultural jobs, while still employed reported receiving the same 46.6 percent of working household heads income from any kind of work as usual – were employed in the service sector (Panel compared to the level at the beginning of A in Figure 4). The share of the industry 2021 – in the past week (Figure 5). The shares sector was only 2.3 percent. A similar sectoral receiving decreased income and no income composition existed at the start of 2021. were only 11 percent and 1 percent, respec- Among the household heads not currently tively. There were no substantive differences working, 89.7 percent did not work at the start between men and women, between the of 2021 (Panel B). agriculture, industry, or service sectors, or between wealth groups. Figure 5 – Changes in income from household heads’ employment compared to start of 2021 11 13 9 11 11 9 5 10 14 81 81 81 84 84 79 83 82 79 3 3 3 2 1 5 2 3 5 All Men Women Agriculture Industry Services Bottom 40% Middle Top 40% qui ntile Sex Sector Wealth Increased Remained the same Decreased No payment Source: Round three of the high frequency mobile phone survey. 11 Part-time informal sector workers, which full time in the informal sector (accounting for mainly consisted of workers in the bottom 54 percent of working heads). In contrast, 40 percent and the middle quintile, were most among those working part-time in the informal likely to see reductions in pay if working. sector (accounting for 13 percent of the working Among the household heads in full-time heads) about half of them were working for formal employment (accounting for a third of the same wages, while 38 percent had seen a working heads), 86 percent were working for decrease and 11 percent received no payment. Figure 6 – Changes in household heads’ 2.2. Household-level farm employment income by full time and and Non-Farm Business formality status While income from non-farm enterprises 4 4 11 7 6 remained the same as usual for most households for the first six months of 2021, a 38 larger proportion of households in bottom 40 percent still report lower than usual or no 86 86 income. In mid-2021, 75 percent of those 50 operating a non-farm enterprise received the same level of income as compared to the start 3 4 of 2021, while 5 percent reported higher 1 Formal, full time Informal, full time Informal, part time income, 17 percent reported lower income, and 3 percent reported no income (Figure 7). Increased Remained Decreased No payment the same There was no significant difference across Source: Round three of the high frequency mobile phone survey. household demographics and regions. It is the same payment and 3 percent had seen an worth highlighting that households in the increase in payments, compared with 7 percent bottom 40 percent wealth group are more that had seen a decrease and 4 percent that likely to report lower than usual or no income received no payment (Figure 6). A similar pattern (26 percent) than those in the top 40 group is observed for household heads that worked (18 percent). Figure 7 – Change in income from non-farm enterprises in past month compared to start of 2021 3 3 2 2 3 6 0 4 1 3 5 2 17 17 15 20 16 22 12 22 12 16 16 23 78 70 72 78 77 75 75 75 76 77 69 73 ! ! ! $ % ) 6 6 8 6 6 1 All Male Female Urban Rural Highland Islands Momase Southern Bottom 40 Middle Top 40 Sex Location Region Wealth Higher than usual Same as usual Less than usual No income Source: Round three of the high frequency mobile phone survey. 12 2. Impact of COVID-19 on Economic Activity Most households expected the same income (Figure 8). Households with female heads from agricultural activities in the current were much more likely to expect no income growing season compared to previous years, compared to those with male heads, at 13 but female-headed households dispropor- percent and 2 percent, respectively. Households tionately expected lower or no income from in urban areas were more likely to expect agriculture. Overall, 85 percent of households higher income from agricultural activities, at expected to see the same income from their 10 percent, as compared to rural households, agricultural production in the current growing at 4 percent. There was limited variation season compared to the last growing season, between different wealth groups and regions. while 4 percent expected higher income Figure 8 – Expected Agricultural Earnings 3 2 2 2 2 1 2 2 4 8 13 7 8 8 6 9 8 9 9 8 6 9 9 6 85 85 74 86 87 89 84 87 83 85 80 78 4 4 10 4 5 4 6 5 1 4 4 2 All Male Female Urban Rural Highland Islands Momase Southern Bottom 40 Middle Top 40 Sex Location Region Wealth Higher than usual Same as usual Less than usual No income Source: Round three of the high frequency mobile phone survey. Handwashing hygiene stations in Kokopo, East New Britain (World Bank/Kalo Fainu) 13 2.3. Remittances Three-quarters of households that usually receive remittances received the same Roughly one-quarter of households usually amount as usual, but certain groups were receive remittances, primarily domestic more likely to have seen payments decrease remittances. Nationally, 23 percent of house- or stop. Households in urban areas were more holds reported usually receiving remittances likely to indicate that they received lower or in round 3 (Panel A in Figure 9). Virtually all the no payment, at 31 percent as compared to remittances were domestic in origin, with 23 percent among households in rural areas small numbers also coming from Australia (Panel B in Figure 9). Female-headed house- and New Zealand. Households in the highest holds were more likely to report receiving 40 percent of the wealth distribution continued the usual amount compared to male-headed to be the most likely to usually receive remit- households, at 82 percent and 73 percent, tances (30 percent) compared with the respectively. middle quintile (24 percent) and the bottom 40 percent (19 percent). There were no substantial differences between households with female and male heads and between regions. Figure 9 – Change in remittances compared to start of 2020 (A) Receipient status of remittances 77 77 76 75 77 76 76 72 76 70 82 81 23 23 24 25 23 24 24 28 24 30 19 19 All Male Female Urban Rural Highland Islands Momase Southern Bottom 40 Middle Top 40 Sex Location Region Wealth Received Not received (B) Amount of received remittances 11 12 5 11 11 5 13 12 8 11 15 7 14 13 13 13 12 12 12 17 13 12 13 16 82 84 76 73 73 63 75 74 65 72 73 70 3 3 3 6 3 3 3 4 2 1 5 3 All Male Female Urban Rural Highland Islands Momase Southern Bottom 40 Middle Top 40 Sex Location Region Wealth Higher than usual Same as usual Less than usual Stopped Source: Round three of the high frequency mobile phone survey. 14 2. Impact of COVID-19 on Economic Activity The vast majority of respondents were some- express financial anxiety than those in urban what or very worried about their households’ areas, at 89 percent and 79 percent, finances in the next month, with some respectively. Significant differences were groups expressing more financial anxiety also evident between wealth groups, with than others. A total of 88 percent of respon- 92 percent of households in the bottom dents reported that they were very worried or 40 percent expressing financial anxiety, somewhat worried about household finances compared to those in the middle quintile (86 in the next month (Figure 10). Households in percent) and the top 40 percent (83 percent). rural areas were substantially more likely to Figure 10 – Financial anxiety 2 1 2 4 1 3 1 2 2 2 10 10 10 10 8 14 11 7 12 17 11 15 42 41 45 46 43 45 50 51 46 48 46 49 45 44 49 51 43 42 36 39 38 38 33 34 All Male Female Urban Rural Highland Islands Momase Southern Bottom Middle Top 40 40 Sex Location Region Wealth Very worried Somewhat worried Not too worried Not worried at all Source: Round three of the high frequency mobile phone survey. 2.5 Income sources (26 percent), and assistance from within the Since the start of 2020, more than half of country (17 percent). There was no significant households had received income from agri- difference between male-headed households cultural activities, including farming, raising or female-headed households in terms of their livestock or fishing, and more than a quarter income sources. Households in the top 40 had income from either non-farm business percent were more likely than other groups to or wage employment. In total, 61 percent of have wage employment (45 percent) and non- households reported that farming had been a farm businesses (34 percent) as their income source of income for their household since sources, while households in the bottom 40 2020 (Figure 11). This is followed by wage percent and the middle quintile were more likely employment (30 percent), non-farm business to have income from farm related activities. Figure 11 – Top 5 sources of income since the start of 2020 (% of all households) % of households 0 10 20 30 40 50 60 70 Family farming, livestock or fishing 61.1 Wage employment of household members 29.6 Non-farm family business 25.6 Assistance from family within the country 17.3 Unemployment benefits 1.8 Source: Round three of the high frequency mobile phone survey. 15 Most households reported that their income region, at 43 percent. The proportion was from all sources had stayed the same since significantly lower in the Islands region, the the start of 2020, but one-third reported Momase region, and the Southern region (less their income had either reduced or stopped than 30 percent). Slightly more than half of entirely, with uneven distribution across households in the bottom 40 percent reported regions and wealth groups. Nationally, 62 the same level of income, while 45 percent percent of households reported that their reported that their income had either reduced income had stayed the same and 35 percent or stopped entirely. This contrasts with of households reported reduced or no income households in the middle quintile and top since the start of 2020 (Figure 12). Only 3 percent 40 percent, where only 30 percent and 22 households reported an increase in income. percent of households, respectively, reported Households in rural areas (37 percent) were lower or no income. These results paint a significantly more likely than households in worrying picture of the uneven impacts of urban areas (23 percent) to experience a decline COVID-19 since the start of 2020. Households or a stop in their income. Geographically, with more vulnerable situations were more households that experienced lower or no likely to report a reduction or a cessation in income were concentrated in the Highlands income from all sources. Figure 12 – Household income compared to the start of 2020 (% of all households) 11 11 11 3 6 4 8 7 4 12 17 17 19 24 28 21 23 20 23 24 24 25 26 28 63 73 67 69 66 73 62 62 61 66 54 53 3 3 5 4 3 3 4 2 6 3 4 5 All Male Female Urban Rural Highland Islands Momase Southern Bottom 40 Middle Top 40 Sex Location Region Wealth Increased Stayed the same Reduced Stopped Source: Round three of the high frequency mobile phone survey. 2.6 Coping Strategies region reported employing at least five strategies, compared with 43 percent of The most common strategies were finding households in the Highlands region. However, ways to earn extra money (71 percent), these strategies are difficult to attribute to spending from personal savings (52 percent), COVID-19 as there are no pre-pandemic and receiving non-cash assistance from baseline measures to compare against. friends or family (47 percent) (see Figure 13). Approximately 93 percent of households in Most households that employed coping PNG had used at least one coping strategy strategies were poor. Nearly 70 percent of since the start of 2021. Nearly 40 percent households who reported they found ways reported that their household had employed to earn extra money were in the bottom five or more coping strategies. While there 40 percent of the wealth distribution. In was little difference geographically for contrast, just 9 percent were in the middle households using at least one coping strategy, quintile, and 21 percent were in the top 40 just 31 percent of households in the Southern percent. Similarly, for the second most common 16 2. Impact of COVID-19 on Economic Activity strategy, 70 percent of survey respondents spending from personal savings, which was who spent from their savings were from the reported by more than half of households bottom 40 percent of households compared (52 percent), households also sold livestock with 23 percent who were from the top 40 (33 percent), sold harvests in advance percent of households. There was not a single (28 percent), and purchased items on credit strategy that, when employed, was employed (19 percent), all of which will impede future by wealthier households more than poorer recovery efforts and increase the risk of food households. insecurity within the household. Of the top ten coping strategies, two led directly to Households with young children were more increased household debt: purchasing items likely to change spending patterns or seek on credit and borrowing from friends or additional sources of income than to decrease family. Delaying making repayments – though consumption. While 8 percent of households not in the top ten – was also commonly with children under 5 years of age indicated employed, with 12 percent of households they had reduced food consumption, 17 percent reporting they had done this. of households without children under 5 had done the same. Likewise, just 7 percent of Access to safety nets was primarily through households with young children reduced informal channels. Approximately half of non-food consumption while 25 percent of households received assistance from friends households without young children did the or family, while one-quarter received cash same. In contrast, households with children from friends or family. It was also somewhat under 5 years were more likely to look for ways common for households to receive assistance to earn extra money, to spend their savings, from churches or religious bodies (12 percent), and to receive either cash or assistance from from the government (7 percent), or – less friends and family. commonly – from an NGO (3 percent) or community-based organization (2 percent). Households continued to employ coping Less than 5 percent of households took loans strategies that both depleted assets and from either informal moneylenders or financial savings as well as increased debt. Outside of institutions. Figure 13 – Coping strategies employed by households since January 2021 Find ways to earn extra money Spend from savings Receive other assistance from friends or family Sell livestock Sell harvest in advance Receive cash or borrow from friends or family Reduce non-food consumption Purchase items on credit Reduce the number of children attending school Reduce food consumption Delay making re-payments Receive assistance from church / religious body Sell assets Receive government assistance Take a loan from an informal moneylender Take a loan from a financial institution Receive assistance from NGO Receive assistance from a community based org Take an advance from an employer Receive payout from superannuation/provident fund 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Source: Round three of the high frequency mobile phone survey. 17 3. Impact of COVID-19 Figure 14 – Attendance rates at the start on Human Development of the 2020 and 2021 school years 100% 98% 94% 98% 92% 90% 85% 80% This section draws on the UNICEF survey to 70% 70% report on levels of education, healthcare, water 60% and sanitation, food security and public trust, 50% community security, and intra-household 40% conflict (including child discipline). 30% 3.1.Education and Schooling 20% 12% 10% 2% There was a modest increase in attendance 0% rates between the start of the 2020 school 3-5 years 6-8 years 9-11 years 12-14 years year and the start of the 2021 school year. At 2020 2021 the start of the 2020 school year (before Source: Round three of the high frequency mobile phone survey. schools closed as part of the first COVID-19 pating in distance learning during this time State of Emergency containment measures), (Figure 15). There was a considerable degree of 60 percent of children aged 3–14 were geographical influence on distance learning enrolled in school, compared with 69 percent at the elementary schoolvi level, as less than of children at the start of the 2021 school year. 12 percent of students in the Southern region Attendance rates for children 6–8 years of participated in distance learning, compared age increased from 70 percent to 85 percent with 36 percent in the Islands region. (partly because many children in this age Participation in distance learning at the cohort met the official age that children are primary school level was somewhat more supposed to enter school in 2021) and from uniform across regions. A smaller proportion 92 percent to 98 percent for children 9–11 of elementary students in the top three years of age. At the time of the survey, 68 quintiles participated in distance learning than percent of children aged 3–14 and 94 percent those in the bottom two quintiles. In contrast, of children 6–14 were currently attending primary school students in the middle 20 and school (Figure 14). This includes 85 percent of top 40 were more likely to participate in those aged 6–8, 97 percent of those aged distance learning (25 percent and 24 percent, 9–11, and almost every student aged 12–14 respectively) than the lowest 40 (20 percent). (98 percent). Only 10 percent of children Of those who participated in distance learning, aged 3–5 were attending school at the time 90 percent used printed study materials. Less of the survey, considerably lowering the overall than 2 percent reported using softcopy study average. materials (to be completed online or on any Participation in distance learning while device) and less than one percent reported schools were closed in March and April 2021 using two-way video communications (such was quite low. At the national level, 24 percent as Zoom, Skype, or WhatsApp). of elementary school students and 21 percent of primary school students reported partici- 18 3. Impact of COVID-19 on Human Development Figure 15 – Participation of elementary and primary students in distance learning 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% National Male Female Highlands Islands Momase Southern Bottom 40 Middle 20 Top 40 Sex Region Wealth Elementary Primary Source: Round three of the high frequency mobile phone survey. The vast majority of households did not school had reopened, 93 percent of students participate in distance learning simply because returned immediately, and 6 percent returned it was not provided by the school. Of the six but with some delay. Only one percent of options respondents were given when indicat- students did not return. The most common ing that children in their household had not reasons for not returning were a lack of money participated in distance learning, 94 percent for school fees (26 percent), no interest in of respondents indicated this as a reason. returning (25 percent), the child needs to help Following this, the need for a computer/ support the family financially (21 percent), or smartphone (28 percent), the need for a TV/ that the child had missed out on learning radio (23 percent), the inability of a parent/ during the closure (16 percent). caregiver to support children in accessing distance learning (19 percent), and a lack of internet connection (11 percent) were the most common reasons. Nearly all children resumed in-person classes once schools reopened following the March– April 2021 closure and only a small portion delayed their return. Almost every school that had closed was open again at the time of the survey, with just one percent opening but then closing again, and one percent remain- ing closed. Of those respondents whose Kokopo Market, East New Britain (World Bank/Kalo Fainu) 19 3.2 Access to Health Care Similarly, it was rare for COVID-19 to inhibit access to routine or preventive medical care. Access to urgent medical care, when needed, Overall, less than five percent of households was generally unobstructed, but occasionally reported requiring routine or preventive impacted by COVID-19. Nationally, 17 percent medical care (such as access to family of respondents reported that a household planning services, childhood vaccines or member had required urgent medical care tuberculosis care). Unlike urgent medical care, (that is, for an acute illness or broken bone) there were considerable regional differences in the previous month (Figure 16). Of those for this indicator, as less than 3 percent of requiring care, 96 percent had been able to households in the Islands and Momase regions access it. There was little variation by gender required routine or preventive medical care of respondent or geographical area in the compared with 10 percent of households in the need for such care or in the ability to access it. Southern region. However, of those requiring Slightly fewer households with children under routine or preventive care, almost all were Figure 16 – Rate of access to urgent medical care and the rate of those needing but unable to access urgent medical care Unable to accesss Able to access 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% National Male Female Highlands Islands Momase Southern Children No under 5 children under 5 Sex Region Children in HH Note: This is based on the total percentage of households. Source: Round 3 of the high frequency mobile phone survey. 5 reported requiring urgent care than those able to access the care they needed (98 per- without (17 percent and 19 percent, respec- cent nationally). Every household with chil- tively); however, those with young children dren under 5 was able to access routine or who required urgent care were almost always preventive medical care if they needed it. able to access it (98 percent), while those While more than 98 percent of both the bot- without had slightly more trouble (95 per- tom 40 percent and top 40 percent of house- cent). Households across the wealth index holds were able to access routine or preven- (poor, middle income, and wealthy) each tive care when needed, just 88 percent of the needed and accessed urgent medical care at middle quintile was able to do so. The most remarkably similar rates. Amongst all respon- common deterrent to access was again the dents, the main reasons for not being able to inability to travel to the health facility (39 per- obtain urgent medical care were the inability cent), though lack of medical personnel (34 to travel and fear of visiting the facility due to percent) and fear of visiting facilities due to COVID-19. COVID-19 (25 percent) were also common. 20 3. Impact of COVID-19 on Human Development 3.3 Water and Sanitation the highest proportion of sufficient access to soap reported by households in the Islands Nearly 40 percent of households reported region (58 percent) and the lowest in the insufficient access to both drinking water Southern region (54 percent). More than 60 and water for handwashing in the 7 days prior percent of the top 40 percent of households to the survey (Figure 17). Barriers to access had sufficient access to soap for handwashing, for both types of water were most acute in the compared to just 51 percent of the bottom Islands region and least acute in the Highlands 40 percent of households. region. The largest geographical discrepancy was observed for drinking water, as nearly 70 3.4 Access to Staple Starch, Proteins, percent of households in the Highlands had Fruits and Vegetables sufficient access to drinking water compared to just 54 percent of households in the Islands Markets and supply chains appear to be region. Very minor differences were present functioning and accessible, though when regarding drinking water in households with obstructed, affordability remains the most young children (under 5 years of age) and common barrier. Nationally, two-thirds of those without young children. In the absence households attempted to buy their preferred, of baseline data, it is not possible to know main staple starch in the week prior to the what proportion of these difficulties was related survey and the majority were able to access it to COVID-19 or deteriorating economic (77 percent). Nearly 90 percent of households conditions as opposed to systemic or seasonal attempted to purchase their staple protein constraints. and 85 percent were successful in doing so. Purchases of fruits and vegetables were less Just over half of households with children common, with less than 60 percent attempting (under 14 years) had sufficient access to soap to purchase fruits and vegetables (83 percent to wash hands with when needed. At the na- of whom were successful). Of those households tional level, 56 percent of survey respondents who attempted to purchase a staple food but in households with children reported that were unable to do so, cost was the greatest their household had sufficient access to soap constraint to purchasing each of the three (Figure 17). Geographic disparities were less staple food types. apparent for soap than for water sources, with Figure 17 – Proportion of households unable to access sufficient water and soap 50.0% 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% Drink ing water Handwashing water Soap National Highlands Islands Momase Southern Source: Round three of the high frequency mobile phone survey. 21 Four out of every five households purchasing Access to staple foods – particularly starches, rice in the past week had noticed a small to fruits and vegetables – was lowest in the moderate price increase since the same time Highlands region and highest in the Southern last year. Rice is a popular starch in PNG, with region. Though the proportion of households approximately one-quarter of households – attempting to purchase both staple starches especially in the Southern and Islands regions and fruits and vegetables in the past week was – consuming it as their main starch, and 86 also lowest in the Highlands region to begin percent of all other households purchasing at with (Figure 18), the proportion who tried and least some quantity in the past week. Among successfully accessed their staple starch or households who had purchased rice in the fruits and vegetables in the past week was also past week, 78 percent reported an increase the lowest. Just over half (53 percent) of in price since the same time last year. This households in the Highlands region attempted Figure 18 – Proportion of households attempting and able to purchase essential food items 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Highlands Highlands Highlands Southern Southern Southern National National National Momase Momase Momase Islands Islands Islands Staple starches Staple protein Fruits & vegetables Yes No Did not try Source: Round three of the high frequency mobile phone survey. increase was noted as “small” by the majority of to purchase their main staple starch, compared respondents (70 percent), though 24 percent with 68 percent of households in the Momase reported it as a “moderate increase”, and 6 region, 74 percent of households in the Islands percent as a “large increase”. region, and more than 80 percent of house- holds in the Southern region. Of those who Poor households had the most difficulty attempted to purchase their staple starch, just accessing all essential food groups. While 62 percent of households in the Highlands 90 percent of the top 40 percent of households region were able to access it, compared with were able to access their main staple starch in more than 85 percent of households in each the past week, just 71 percent of the bottom of the other three regions. Similarly, less than 40 percent were able to do so. Similarly, while half of households in the Highlands region 91 percent of the top 40 percent of households attempted to purchase fruits and vegetables, were able to access their main staple protein, compared with approximately 70 percent of only 82 percent of the bottom 40 percent of households in the Islands and Southern regions. households were able to do the same. Finally, Just 67 percent of those in the Highlands only 80 percent of the bottom 40 percent region who attempted to purchase fruits and of households were able to access fruits vegetables were able to do so, compared with and vegetables compared with more than 91 more than 91 percent of households in the percent of wealthy households. Islands, Momase, and Southern regions. 22 3. Impact of COVID-19 on Human Development More than two-thirds of households surveyed reported at least one episode of food insecurity (Conor Ashleigh/World Bank) 3.5 Food Insecurity region, 70 percent of households in the Momase region and 73 percent of households More than two-thirds of households reported in the Islands region (Figure 19). Sex of the re- at least one episode of food insecurity in the spondent had very little impact on whether previous 30 days. This indicator is defined as the household had experienced at least one experiencing any of the full set of eight indi- episode (68 percent for male respondents cators derived from questions recommended versus 65 percent for female respondents). by the Food and Agriculture Organization In contrast, there were vast differences (FAO) to measure food insecurity.vii Some when considering the education level of the regional variation was apparent, with just 63 respondent. More than 80 percent of respon- percent of households in the Highlands re- dents with no education reported experien­ gion reporting at least one episode of food cing at least one episode of food insecurity in insecurity in the past 30 days, compared with the past 30 days, compared to just 65 percent 67 percent of households in the Southern of those with tertiary or higher education. Figure 19 – Households experiencing at least one episode of food insecurity 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% National Male Female Highlands Islands Momase Southern No education Primary or below Secondary/ vocational Tertiary or above Bottom 40 Middle 20 Top 40 Sex of respondent Region Education of respondent Wealth Source: Round three of the high frequency mobile phone survey. 23 Food insecurity in Papua New Guinea Households with children under five expe­ remains pervasive though fairly intermittent. rienced similar levels of food security to Nationally, 31 percent of households reported other households. A larger proportion of that there had been a time in the past 30 days households with children under 5 reported at they had been hungry but had not eaten least one episode of food insecurity. However, because there was not enough money or other 23 percent of households with children under Figure 20 – Proportion of households experiencing food insecurity, by region 80% 70% 60% 50% 40% 30% 20% 10% 0% At least one episode Hungry but did not eat Household ran out of food Went without eating for a whole day National Highlands Islands Momase Southern Source: Round three of the high frequency mobile phone survey. resources for food (Figure 20). Just over 42 5 had an adult or other family member report percent of households reported that there being hungry but not eating, while 32 percent had been a time in the last 30 days that their of households without young children reported household had run out of food because of a the same (Figure 21). lack of money or other resources. The most Figure 21 – Proportion of households severe level of food insecurity – going with- experiencing food insecurity by presence of out eating for a whole day because of a lack of children in in household money or other resources – was reported by 21 percent of all households. Wealth had a 100.0% small but notable impact on the proportion of 90.0% households experiencing this indicator, with 80.0% 25 percent of the top 40 reporting going a 70.0% whole day without eating compared to 20 60.0% percent of both the bottom 40 and middle 50.0% quintile. The lowest proportion of households 40.0% experiencing each of these three food 30.0% security issues was consistently reported by 20.0% the Highlands region. However, the frequency 10.0% of each episode is remarkably low; less than 3 0.0% At least one Hungry but Household ran Went without percent of households reported that the event episode did not eat out of food eating for a occurred often, while the vast majority (97 whole day National Children under 5 No children under 5 percent or higher) reported that it occurred only sometimes. Source: Round three of the high frequency mobile phone survey. 24 3. Impact of COVID-19 on Human Development Perceptions of community relations and More respondents believed that both property trust appear stable and largely unchanged crimes and theft had become worse since since the beginning of 2021, with moderate the start of the year than believed they had improvements reported in the Islands region. improved. Though the majority indicated the Most respondents (66 percent) indicated that situations in regard to both property crimes trust and community relations had remained and theft had remained unchanged (61 percent unchanged since January 2021, while an and 60 percent, respectively), twice as many additional 15 percent believed they had respondents indicated a deterioration in the improved (Figure 22). Geographically, situation regarding theft in their community Figure 22 – Change in community trust and relations since January 2021 (% of households) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% National Men Women Highlands Islands Momase Southern Bottom 40 Middle 20 Top 40 Sex Region Wealth Improved Stayed the same Worsened Source: Round three of the high frequency mobile phone survey. improvements appeared most noticeable (27 percent) rather than an improvement (13 in the Islands region, where 25 percent of percent), while 23 percent of respondents respondents indicated that community indicated a deterioration in property crimes relations had improved since the beginning of and only 16 percent indicated an improvement. the year, compared to 20 percent of respon- There was little variation when considering dents in the Southern region, 19 percent of the gender of the respondent. The greatest respondents in the Highlands region, and proportions of displeased respondents were 13 percent of respondents in the Islands in the Southern region, where 33 percent of region. Household wealth had little impact on respondents indicated the theft situation had perceptions of community relations, with the been deteriorating since the start of the year largest difference observed in the proportion and 31 percent indicated the property of wealthy believing there had been some damage situation had been deteriorating. deterioration (18 percent) compared to the Notably, wealthy households were more likely proportion of poor that believed there had to believe the situation regarding property been deterioration (14 percent). crimes had deteriorated (29 percent) than poorer households (20 percent), and poor households were more likely to believe it had improved (19 percent) than wealthy households (12 percent). 25 Overall, 15 percent of respondents indicated theft, and police intimidation, deterioration in there had been an increase in police intimi- the alcohol and drug abuse situation again dation since January 2021. Just 7.5 percent appears most evident in the Southern region. believed the situation had improved with In this region, 55 percent of respondents regard to police intimidation, with similar indicated the situation regarding alcohol and sentiments from both men and women. drug abuse had worsened, compared to 50 There were considerable geographic variations percent of respondents in the Islands region, regarding police intimidation, as increases 45 percent of respondents in the Momase were most apparent in the Southern region region and 43 percent of respondents in the (25 percent of respondents cited increases in Highlands region. A much larger proportion of intimidation), and least apparent in the Islands wealthy households believed alcohol and region, where less than 12 percent of respon- drug abuse had deteriorated since the start dents cited increases. However the source of of the year than poorer households. While this reported increase in police intimidation 43 percent of the bottom 40 believed the requires further research. Household wealth situation had deteriorated, 56 percent of the had very minimal impact on respondent top 40 believed the same. perceptions of police intimidation in commu- Despite nearly three-quarters of respondents nities. indicating that the situation related to domes- Alcohol and drug abuse in communities has tic violence remained unchanged, a substan- considerably worsened since the start of the tial number also indicated it had worsened. year. Nearly half of all respondents (46 percent) Nationally, 74 percent of respondents indicated reported that the situation regarding alcohol that the domestic violence situation in their and drug abuse in their community had community had remained the same since the deteriorated since January 2021, with an start of the year, while 9 percent indicated it additional 46 percent reporting that it had had improved and 17 percent indicated it had remained unchanged and only 7 percent of deteriorated. Both men and women shared respondents – driven by men and those in the remarkably similar impressions of the situation. Highlands region – reporting an improvement Unlike other aspects of public trust and safety, (Figure 23). As was seen in property damage, there was minimal variation across regions or Figure 23 – Change in situation with alcohol and drug abuse since January 2021 (% of respondents) Improved Stayed the same Worsened 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% National Men Women Highlands Islands Momase Southern Bottom 40 Middle 20 Top 40 Sex Region Wealth Source: Round three of the high frequency mobile phone survey. 26 3. Impact of COVID-19 on Human Development household wealth levels, with respondents in was necessary, while the lowest was in the each of the four regions and three wealth Southern region (44 percent). The wealthiest brackets reporting similar concerns over the and poorest households were equally least increase in domestic violence. likely to believe physical punishment was necessary (55 percent), while 63 percent of 3.7 Child Discipline the middle wealth quintile believed physical Most adults and caregivers believed that punishment was necessary. discipline is necessary in order to raise Physical discipline and suspension of or educate children properly. More than privileges were the most commonly cited 86 percent of adults and caregivers surveyed methods used by adults to teach children believed that to raise or educate a child what they consider to be proper behavior. pro­perly, a child sometimes needed to be The survey randomly selected one child aged disciplined. This was a consistent belief across 3 or older in the household and covered five all four regions of the country. Though not waysviii that children could be disciplined to substantial, a larger proportion of poorer teach what adults considered to be proper households believed discipline was necessary behavior or to address a child’s behavioral than wealthier households (88 percent of the problem. More than one-third (37 percent) of bottom 40 believed discipline was necessary adults in the 15 days prior to the survey had compared with 83 percent of the top 40). spanked, hit or slapped their child with their Physical discipline, though less common, was bare hand or foot, while 34 percent had still believed by 56 percent of respondents to revoked the child’s privileges or forbade be necessary for raising or educating a child something the child liked. Also common were properly. Unlike other measures of discipline, physical discipline with an object (27 percent) there were considerable disparities in support and calling the child dumb, lazy, or another for physical punishment by both geographical name (21 percent). Less than 10 percent area and education level of respondent of adults reported that the child had been (Figure 24). The most widespread support shaken. for physical discipline of children was in the Highlands region, where 62 percent of respondents believed physical punishment Figure 24 – Respondents believing physical punishment is necessary, by region and education 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Highlands Islands Momase Southern Never attended Pre-preparatory Elementary Primary Secondary Vocational Tertiary or higher Bottom 40 Middle 20 Top 40 Region Education level of respondent Wealth Source: Round three of the high frequency mobile phone survey. 27 World Bank Conclusions & Policy most in need as there are worrying signs they may have exhausted all, or at least Implications most, of the coping strategies available to them. • Continuing to prepare distance learning The surveys show that in general there were options– More than three-quarters of students similar levels of economic activity and human did not participate in any distance learning development in mid-2021 as was the case at activities during the school closures in March the start of 2021. This stabilization is notable and April, primarily because this was not given that over this period there was a rapid offered by schools Given future COVID-19 escalation of the COVID-19 crisis in March outbreaks may result in school closures, followed by a fall in cases throughout the continued preparation to ensure more May. However, also does not demonstrate a schools can offer distance learning options recovery to pre-pandemic levels. It is important is critical for learning continuity. to note that all data in this report was collected • Increasing access to soap and water for prior to the arrival of the Delta variant of handwashing – Around half of households COVID-19 in PNG and future survey rounds reported having inadequate access to will analyze its impact on economic activity soap and water for handwashing. Full and human development. implementation of the National WASH The surveys in this report identify several Strategy is a critical policy lever to increase areas worthy of immediate attention from WASH access across the country, especially policy makers. These are listed below, with as PNG is likely to have COVID-19 and other suggestions for actions that could be taken: communicable diseases circulating in the • Expanding the coverage of formal safety community well into the future. nets – Poorer households were more likely • Improving the affordability of staple starches to report having to implement coping – Rice is a key staple starch for households, strategies that may increase their economic particularly in the Southern and Islands vulnerability, including depleting assets and regions, and over three-quarters of house- savings as well as increasing household holds reported price rises over the past year. debt. Only 7 percent of households reported This warrants investigation into whether receiving any assistance from governments. supply chains are causing this and if there Expanding the coverage of formal safety are actions governments can take to ensure nets would greatly assist the households rice is affordable. 28 Conclusion & Policy Implications • Addressing rising alcohol and drug use – report being unable to access health care due Almost half of households indicated a rise to distance to facilities. Reports of crime, do- in alcohol and drug abuse in their commu- mestic violence, theft and police intimidation nities since the start of 2021. Such an are alarmingly high. Collectively, the findings alarming increase is worthy of immediate illustrate many areas for policy makers to focus action by governments and non-govern- on in the medium term, even though thus far ment community groups as this may be the impact of COVID-19 does not appear symptomatic of deeper social unrest that to have resulted in a substantially greater may come to the surface in the future. deterioration in conditions throughout 2021. The surveys also reveal there are significant medium-term challenges facing PNG. More than half of households reported relying on subsistence agriculture as their main source of employment. There are high levels of food insecurity, low levels of participation in pre-primary education, and many households 29 References i The five-volume best practice methodological guidelines are available at https://documents.worldbank.org/en/ publication/documents-reports/documentlist?repnb=148213. ii The round 1 and 2 reports are available through the World Bank’s Open Knowledge Repository at http://hdl. handle.net/10986/34907 and https://openknowledge.worldbank.org/handle/10986/35585. iii PNG National Department of Health, World Health Organization. Papua New Guinea: Coronavirus Disease 2019 (COVID-19) Health Situation Report #84. 02 August 2021. Period of Report: 26 July – 01 August 2021. iv Of which 21,000 would be taken up by the Chinese citizens while the rest would be available for PNG citizens and other residents on a voluntary basis. v Employment data for the start of 2021 is recall data collected at the time of the Round three survey vi Elementary school covers the first three grades of primary education. vii http://www.fao.org/fileadmin/templates/ess/voh/FIES_Technical_Paper_v1.1.pdf viii These include: taking away privileges; forbade something the child liked, or did not allow them to leave the house for a period of time; called the child dumb, lazy or another similar term; shook the child; spanked, hit or slapped the child with bare hand (or foot); or spanked or hit the child with an object. 30