Toward a World-Class Labor Market Information System for Indonesia An Assessment of the System Managed by the Indonesian Ministry of Manpower Toward a World-Class Labor Market Information System for Indonesia An Assessment of the System Managed by the Indonesian Ministry of Manpower Cover photo © Thichaa Contents Acknowledgments VII Executive Summary VIII Abbreviations XVIII 1. Introduction: Improving Labor Market 4. Toward a World-Class Labor Market Efficiency in Indonesia 1 Information System 53 Labor Market Information Systems 4 Going Forward 53 Framework for an Advanced LMIS 5 Vision and Action Plan for Reform 54 2. The Indonesian Labor Market Appendix 61 Information System 9 AyoKitaKerja: Overview of Its Functions and Users 10 Additional Initiatives That Are Part of Indonesia’s LMIS 12 References 66 LMIS Classification 15 International Examples of Advanced LMISs 17 Boxes 2.1 Data Sources for an Advanced LMIS 16 3.1 Indonesia’s Skills Monitoring System 49 3. Benchmarking AyoKitaKerja against Work-net: What Can Indonesia Learn from the Republic of Korea? 25 Figures ES.1 Vision of a World-Class Labor Market Information System XV Data Collection 27 1.1 Potential Costs of Labor Market Failures 3 Data Validation 30 1.2 Framework of an Advanced LMIS 7 Accessibility and User Experience 32 2.1 AyoKitaKerja Job Placements, 2016–19 11 Targeted Service Delivery 36 3.1 Work-net Information Sharing 28 Strategic Marketing to Users 38 3.2 Responsive Web Design of AyoKitaKerja 33 Interoperability with Other Systems 39 3.3 Interoperability of Work-net with Other Systems 40 Human Resources Serving the System and Its Users 41 3.4 Software Architecture of AyoKitaKerja 43 Technical Infrastructure 42 3.5 Hardware Architecture of MoM’s Data Center Kemnaker 44 Data Management Systems 43 3.6 Information Security Means 46 Information Security 45 4.1 Vision of a World-Class Labor Market Information System 55 Additional Functions beyond Job Matching 46 A.1 Timeline for Work-net, 1980–2020 60 A.2 KEIS Corporate Organization and Divisions Related to Work-net 63 IV Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice V Tables ES.1 Summary of Gaps between AyoKitaKerja and Work-net XIII Acknowledgments 1.1 LMIS Classification 6 2.1 AyoKitaKerja Job Vacancies and Job Seekers, 2016–19 10 2.2 Indonesia’s LMIS Initiatives, by Function 13 2.3 Subsidiary LMIS Systems in Australia 18 This report was prepared by a team led by Mauro Testaverde (senior economist) and 2.4 Subsidiary LMIS Systems in Denmark 19 Josefina Posadas (senior economist) and including (in alphabetical order) Insung Cho 2.5 Subsidiary LMIS Systems in France 20 (manager of the Job Matching Platform Development Team in the Korean Employ- 2.6 Subsidiary LMIS Systems in the Republic of Korea 21 ment and Information Services, consultant), Julia Granata (labor market specialist, 2.7 Subsidiary LMIS Systems in the United Kingdom 22 consultant), Sara Johansson de Silva (senior labor market specialist, consultant), Gyun 3.1 Select Development and Labor Market Indicators Kim (director of Employment Service User Experience Team in the Korean Employ- Comparing Indonesia and the Republic of Korea 26 ment and Information Services, consultant), Ilsa Meidina (social protection specialist), 3.2 Data Collection Gap on Job Vacancies 29 Thelma Seoeun Choi (skills specialist, consultant), Petra Wiyakti Bodrogini (education 3.3 Data Collection Gap on Job Seekers 30 specialist, consultant), and Kilsang Yoo (professor emeritus at the Korea University 3.4 Data Validation Gap 32 of Technology and Education, consultant). The team thanks Camilla Holmemo, Javier 3.5 Accessibility of Services and Information Gap 34 Luque, Ririn Salwa Purnamasari, and Laura Ralston for useful discussions. 3.6 User Experience Gap 35 3.7 Customer Response Service Gap 36 The report team is grateful for the leadership of the Ministry of National Development 3.8 Targeted Service Delivery Gap 38 Planning (BAPPENAS) and the Ministry of Manpower throughout the course of the 3.9 Strategic Marketing Gap 39 assignment. 3.10 Interoperability with Other Systems Gap 41 3.11 Human Resources Gap 42 The work was conducted under the general guidance of Rodrigo Chaves (former 3.12 Technical Infrastructure Gap 44 country director for Indonesia) and Philip O’Keefe (practice manager for social pro- 3.13 Data Management Gap 45 tection and jobs, East Asia and Pacific Region). The team is grateful for the excellent 3.14 Information Security Gap 47 advice provided by three peer reviewers—Aylin Isik-Dikmelik (senior economist), Jo- 3.15 Career and Skills Guidance Gap 48 hannes Koettl (senior economist), and Harry Moroz (economist)—and for very useful 3.16 Government Support Gap 50 comments provided by other World Bank colleagues. The report was made possible 3.17 Labor Market Information and Analysis Gap 51 by a grant from the Korea–World Bank Group Partnership Facility (KWPF). 4.1 Action Plan 56 A.1 Work-net Reform Projects, 2005–17 62 We also thank Corinne Bernaldez and Rizky Fitriany for providing excellent administra- A.2 Work-net Staff within KEIS Corporate Organization 64 tive support and Peter Milne for editing the report. The findings, interpretations, and conclusions expressed in this document are those of the authors and do not necessarily reflect the views of the executive directors of the World Bank, the governments they represent, or the counterparts with whom they consulted or engaged during the study process. VI Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice VII Executive Summary information to flow regularly between a wide variety of key stakeholders, while con- stantly integrating both data services (data collection, processing, and dissemination) and labor market services. By providing relevant and timely information, an advanced Building a Labor Market Information System to LMIS can help stakeholders to make informed decisions on a variety of issues, includ- Improve Human Capital Development ing job search, career planning and skills upgrading, business investment and hiring, public or private workforce and skills investments, and optimal employment and skills Human capital development is at the top of Indonesia’s economic development agen- policy designed to meet future demands. da. The National Long-Term Development Plan (RPJPN 2005–2025) identifies human resource development as one of the key drivers of the eight national development This report proposes a framework with which to define and identify the key elements goals to be achieved by 2025. As part of this push, the government of Indonesia has of an advanced LMIS. The framework includes the following: taken several steps to build a skilled and competitive workforce building on the coun- try’s demographic strengths, strategic position, and sustained economic growth. An • Stakeholders, including users and producers of labor market intelligence generat- important milestone was the launch in 2016 of a national initiative known as Revital- ed by the LMIS ization of Secondary Vocational Schools, which the Ministry of Education and Culture • Functions that the system should offer, such as job matching, career and skills has updated recently. This initiative focuses on strengthening the quality and rele- guidance, government support, and labor market information and analysis vance of secondary vocational schools. • Key elements for the provision of those functions, including data collection and validation, accessibility and user experience, targeted service delivery, strategic Strengthening the labor market information system (LMIS) is central to Indonesia’s marketing, system interoperability, human resources, information technology (IT) efforts to build a skilled and competitive workforce. Labor market information fail- infrastructure, data management, and information security ures contribute to skills mismatches. When labor market information is incomplete, • Essential features to produce relevant information and to coordinate the over- job seekers lack information on the skills demanded by employers, leading to the all system, including institutional arrangements, partnerships, and technological misallocation of educational investments. At the same time, employers experience solutions. higher search costs due to limited information on job seekers’ skills or short supply of relevant skills in the labor market. Labor market information failures can lead to To date, the most developed function of Indonesia’s LMIS is AyoKitaKerja, a publicly inefficient job matches, limit workers’ economic prospects, lower firms’ productivity, managed job-matching platform launched in 2016. AyoKitaKerja provides job seekers and ultimately weaken the long-term competitiveness of the country. National labor with basic information on job vacancies and firms with basic information on potential force statistics, enterprise surveys, and educational assessments indicate that skills candidates to facilitate employment matches. While other initiatives also form part mismatches are building up in Indonesia, in part due to these information deficits. A of the country’s LMIS, they are more limited in scope. recent assessment of the Indonesian workforce development system confirms these findings and identifies the development of an advanced LMIS as one of the interven- To guide the reform efforts focused on developing a modern, effective, and ef- tions needed to overcome these information failures.1 ficient LMIS, the government of Indonesia asked the World Bank to undertake a benchmarking exercise between AyoKitaKerja and the Republic of Korea’s Work- An advanced LMIS produces intelligence and analysis to improve the functioning net. Both platforms are just one part of each country’s system overall. They were of the labor market. An advanced LMIS has institutional arrangements that allow selected for this exercise given the similarity of their trajectory. Work-net—devel- oped by the Korea Employment Information Service (KEIS) under of the Ministry 1  . The assessment identifies other types of interventions to strengthen Indonesia’s workforce development sys- of Employment and Labor (MoEL)—also started as a basic job-matching platform, tem: (1) reforms to the institutional architecture of the system, (2) stronger emphasis on a demand-driven ap- similar to what AyoKitaKerja is now. It has evolved to become a more complex proach, (3) more attention to continuous technical and vocational education and training (TVET), (4) faster de- system, offering all functions expected of an advanced LMIS either through the velopment of competency standards and national qualification certification schemes, (5) improved accreditation platform or through its interoperability with other systems. For this reason, Work- practices, (6) stronger employment services (including the establishment of an LMIS), (7) a comprehensive set of targets to improve the quality of public training, (8) a national strategy for the development of instructors and net offers a practical example of how the government of Indonesia could develop administrators of TVET institutions, and (9) stronger focus on assessing the labor market outcomes of training its LMIS from its current state. programs (World Bank, forthcoming). VIII Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice IX Work-net vs. AyoKitaKerja: Gap Analysis © Background Photo The benchmarking exercise suggests that the key elements of an advanced LMIS are not fully developed in AyoKitaKerja. While AyoKitaKerja has made some progress re- lated to data collection and validation, the development of these elements is more lim- ited than those of Work-net. Work-net collects information on job vacancies from a wide variety of sources through well-established partnerships with the private sector (including both firms and job portals) as well as national, regional, and local govern- ments and institutions. As a result of these partnerships and the consequent relevance of information on the platform, job seekers rely on Work-net to find jobs. In addition, the Korean government requires job seekers to register on Work-net to be eligible for unemployment benefits or to participate in ALMPs. Through public employment service (PES) offices, Work-net provides support to small and medium enterprises (SMEs) posting their vacancies and to vulnerable job seekers requiring more person- alized help. Work-net has a four-step validation process to ensure that all information posted on the platform is reliable, including local PES offices, computer algorithms, monitoring experts, and system linkages validations. Different from Work-net, AyoKi- taKerja only reports data collected from firms posting vacancies on its portal; it does Korea was selected for this benchmark exercise because its LMIS is considered to not report vacancies posted on other job portals. Most vacancies are directed to low- be a world-class system.2 Work-net, the online platform of labor market services es- skill workers. Staff at PES offices, although limited in number, are available to assist tablished in 1998, is at the core of Korea’s LMIS. Among other things, Work-net pro- job seekers and SMEs. Data validation is manual and infrequent: regional offices of the vides the key functions that all advanced LMIS systems should have: (a) job matching, Ministry of Manpower (MoM) perform some job-vacancy checks, while PES offices (b) information on skills available in the workforce, (c) current and future jobs and validate job seekers’ data if and when workers obtain an AK1/job-seeker card. skills in demand, (d) education and training sought by employers for different careers, jobs, and skills, (e) personal fit to specific careers and jobs, vacancies, and job seekers, AyoKitaKerja lacks several elements needed to provide client-centric, comprehensive and (f) support from the government for acquiring skills through different measures, services such as those offered by Work-net. Work-net is accessible through multiple including scholarships and other active labor market policies (ALMPs). Work-net is devices (personal computer, tablet, mobile web, and app). The platform is easy to linked to other relevant online systems, including social security and human resource navigate, with a user-friendly screen design and interface (which was designed using development networks. Its functions are supported by strong institutional arrange- User Interface/User Experience [UX/UI] technology) and interoperability with other ments involving the private sector, central and local governments, the education sys- subsystems in the Korean LMIS (with a unique log-in). The platform also offers cus- tem, a strong technological base for sharing information, and advanced-technology tomized services and information based on job seekers’ profiles, including services solutions (big data, artificial intelligence) for efficiently analyzing data and disseminat- tailored to vulnerable workers. In addition, Work-net operates a 24/7 call center and a ing the available information. feedback system for rapid client support as well as a customer complaint resolution mechanism. AyoKitaKerja, although accessible through various devices (not mobile This report presents the key findings of a benchmarking exercise focused on job-match- app), is not user friendly and lacks linkages with other subsystems in Indonesia’s LMIS. ing platforms, with broader conclusions for building up an effective LMIS. In evaluating Furthermore, it does not have a responsive client support mechanism or a feedback the performance of the Korean and Indonesian systems, the report recognizes differ- system, and it does not tailor services to a job seeker’s profile. ences in the context in which they operate as well as in their trajectories. The func- tions of Work-net highlighted in this report correspond to the characteristics of an AyoKitaKerja has limited human resources and technical infrastructure, which has an advanced LMIS and provide a potential example of what the government of Indonesia impact on the provision of reliable and efficient services. Work-net has a dedicat- could aim for in the process of revamping its LMIS. ed marketing team to attract job seekers; the team conducts active marketing cam- paigns and events, targets specific groups (for example, youth), fosters partnerships 2 . The terms “world class” and “advanced” are used interchangeably throughout the report. (for example, with universities), and disseminates research findings. A large number X Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice XI of skilled counselors in PES offices use the platform. Work-net has a research team Table ES.1 Summary of Gaps between AyoKitaKerja and Work-net dedicated to providing high-quality analysis and labor market intelligence, which are later disseminated through the marketing team. Work-net’s technical infrastructure Area Work-net AyoKitaKerja benefits from services available round-the-clock due to its redundancy server and Data collection Collection of vacancies from a wide Limited to some firms looking for low- replication system. A daily and weekly backup system and a disaster recovery system variety of sources, including firms and skill workers are in place, with data centers at different locations to prevent loss of data. In con- other platforms trast, AyoKitaKerja has very limited human resources for developing and operating Support for SMEs to post vacancies Limited support provided by PES staff the system: it only has four dedicated staff at MoM headquarters; about half of PES Direct registration of job seekers Limited to a small % of the workforce offices use the platform, which has a limited number of staff and no marketing or re- search team. AyoKitaKerja’s technical infrastructure is also relatively limited: although Strong links with ALMPs and other No services are available round-the-clock, they are slow, and, in event of failure, there is government programs no safeguard redundancy system; backup systems are in place, but they are stored on Support with registration for vulnera- Limited support provided by PES staff MoM servers, and there are no disaster recovery procedures. ble groups Data validation  Data verification through multiple steps Limited to basic and manual verifica- AyoKitaKerja’s management of data quality and information security is not sufficient and sources tion by PES staff to provide reliable services. Work-net has a data quality management system, which Accessibility and user Accessible through multiple devices Limited to personal computer, tablet, allows for periodic monitoring and evaluation of data, a master data management sys- experience (personal computer, tablet, mobile and mobile web; very limited assistance tem, which allows for personalization of services, and quality standards. Strong secu- web, and app) with assistance services rity policies and systems are in place to protect users’ information, including adminis- trative security (guidelines and trainings), physical security (data access), and technical Interconnected systems No security (network, server, and user). In contrast, AyoKitaKerja has no quality control User-friendly design and unified log-in No guidelines or correction devices for monitoring and performing data quality checks, for all subsystems and its information security is limited to firewalls. Feedback system, client support, and No customer complaint resolution mecha- Finally, AyoKitaKerja’s functions are not comprehensive. Work-net either directly pro- nisms vides functions besides job matching or is linked to subsystems within the Korean Tailored service de- Tailored job recommendations, ser- No LMIS that do so. For example, the platform links users to the Korea Network for Oc- livery vices, and information, including vulner- cupations and Workers for career and skills guidance and to HRD-net for vocational able population training. For government support, the platform provides information on employment Strategic marketing Dedicated marketing team No and welfare policies at the national and local levels; it has partnerships and links with other ministries and public agencies. In the area of labor market information and anal- User trainings No ysis, Work-net feeds information to the Korean LMIS, which KEIS uses to evaluate and Interoperability with System linked to other subsystems No produce analyses and forecasting that it makes available to a wide variety of users. other systems Work-net also shares information through open application programming interfaces Human resources Used by skilled counselors Limited to PES staff (APIs). In contrast, AyoKitaKerja’s functions are limited mostly to providing job-match- serving the system Research team No ing services for low-skill workers and basic tools for career guidance. and users  LMIS management team No A full summary of the existing gaps in service is presented in table ES.1. Technical infrastruc- Backup system Yes, but on MoM server ture Disaster recovery system No Integrated monitoring system No Big data platform for analysis No XII Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice XIII Figure ES.1 Vision of a World-Class Labor Market Information System Data management Data quality management systems, No team, standards, and checks in place Five key characteristics of an advanced LMIS Area Work-net AyoKitaKerja Information security  Information protection (system and Limited to firewall; NO training training) Relevance Reliabilitiy Efficiency Clients at the Comprehen- Center siveness Security measures in place No Attracts, collects Is built on a strong Uses technology Is centered on Is Linked to Functions other than Career and skills guidance Limited to basic guidance and applica- and coordinates data management cost-effectively to client needs and is broader national a critical mass of and information connect users to continually development job matching tion material templates high-quality security system the system and reviewed to policies and to Government support No information that guarantees leverages on M&E provide user- other relevant round-the-clock to improve friendly services information Labor market information and analysis No availability of services and and to respond to systems, serves the services programs client feedback needs of different Note: SMEs = small and medium enterprises. ALMPs = active labor market programs. PES = public stakeholders employment service. LMIS= labor market information system. MoM = Ministry of Manpower. Technology solutions Collaborative partnership Institutional arrangements Going Forward: A Vision for Indonesia’s LMIS The job-matching platform and the broader LMIS managed by the MoM in Indonesia • Relevance. It attracts, collects, and coordinates large volumes of trustworthy data, have many potential users and uses, but careful reforms are needed for Indonesia to actively gathers data, both qualitative and quantitative, from a variety of sources, move from a basic-to-intermediate LMIS toward an advanced LMIS. The reform will and transforms these data into relevant and reliable information for different users. require continuously innovating and improving the services available in response to • Reliability. It uses strong data management and IT and security systems that guar- changing client needs and in accordance with changes in local, national, regional, and antee the validity and continuous accessibility of data. global labor markets. At the same time, the job-matching platform should integrate • Efficiency. It applies technology to connect users to the system, improves the ser- current advances in technology in order to raise both efficiency and effectiveness vices provided across various dimensions, and monitors outcomes to improve ser- of the system. Broadening services and reaching out to a more diverse set of clients vices and policies. should be a transversal theme. The timing of reform is advantageous because recent • Clients-at-the-center. It responds to client feedback, targets individual needs, and is advances in artificial intelligence, the overall digitization of government services, and easily accessible, navigable, and user friendly. the experience of countries with more developed systems can help Indonesia to be- • Comprehensiveness. It forms part of, and feeds into, broader development policies come a top-performing and resource-efficient system. and serves a wide variety of stakeholders, offering various services and attending to different needs. Five key characteristics are at the core of the vision that could take Indonesia’s LMIS to an advanced level. Figure ES.1 presents a vision for what Indonesia’s LMIS could The entire system needs to be underpinned by the smart use of technology, col- aspire to be. laborative partnerships, and institutional arrangements. The system should rely on advanced-technology solutions (including advanced algorithms, artificial intelligence, The vision centers on improving the job-matching capacity and building the founda- and big-data techniques) and integrate traditional and nontraditional sources of data. tions of a coordinated and integrated system of services. A world-class LMIS has five The system should also foster collaborative partnerships with both public and private key interrelated characteristics: sector actors (that is, workers, firms, policy makers, researchers, and practitioners), who will use and produce information that feeds the system. Finally, the system should XIV Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice XV Abbrevations have institutional arrangements (policies, systems, and resources) for managing the LMIS and coordinating effectively with all stakeholders. These essential features will strengthen the availability, coherence, and use of data needed to inform key deci- sions and improve coordination of the overall system. AI artificial intelligence The report also presents a set of concrete actions to arrive at this strategic vision. ALIO All Public Information in One (Rep. of Korea) Over the short term, the actions focus on improving the job-matching platform by at- ALMP active labor market program tracting a critical mass of users, improving the effectiveness of job matching, and set- API application programming interface ting the basis for a broader LMIS. This improvement can be accomplished by making BLK work training center the platform more user friendly through the application of technological solutions, BNSP Professional Certification Authority strengthening data quality and security, identifying key stakeholders, strengthening BPS Central Bureau of Statistics (Indonesia) the connection with offline services in PES offices, and linking the platform with exist- COL Critical Occupations List ing information databases for education and training, labor market information, gov- EPS Employment Permit System (Rep. of Korea) ernment ALMPs, and social programs. Over the medium and long terms, the different GDP gross domestic product segments of an advanced LMIS can be developed by deepening and tailoring the ser- HRD-net Human Resources Development Service (Rep. of Korea) vices available for providing career guidance, labor market information, and links to ICT Information and communication technology relevant labor market government programs. Once these building blocks are in place, IT Information technology they could be integrated further so that users can move easily across functions. KBJPI Indonesian Standard Classification of Occupations KEIS Korea Employment Information Service KNOW Korea Network for Occupations and Workers LMIS labor market information system LMO labor market observatory Organization of the Report LVM German Virtual Labor Market MoEL Ministry of Employment and Labor (Rep. of Korea) Chapter 1 proposes a framework for defining an advanced LMIS that includes stake- MoM Ministry of Manpower (Indonesia) holders, functions, key elements, key characteristics, and essential features that go NIK citizen identification number (Indonesia) well beyond those of an online job-matching platform. Chapter 2 discusses the cur- OEVS Occupational Employment and Vacancy Survey rent state of Indonesia’s AyoKitaKerja, which is the main focus of the analysis, is the PES public employment service most developed LMIS function and is considered the building block of Indonesia’s PIAC Program for the International Assessment of Adult Competencies LMIS. Chapter 2 also presents Indonesia’s LMIS-related initiatives in addition to AyoK- PISA Programme for International Student Assessment itaKerja and introduces some comparison with LMISs in other countries. Chapter RPJPN National Long-Term Development Plan (Indonesia) 3 focuses in more detail on the factors that are essential for building up the five Sakernas National Labor Force Survey key characteristics of a well-functioning LMIS. The analysis benchmarks AyoKitaK- SKKNI Indonesia National Work Competency Standards erja against Korea’s Work-net in each of these areas, with a particular focus on the SMEs small and medium enterprises job-matching function. Finally, chapter 4 provides a vision and action plan for devel- SMS short message service oping a comprehensive LMIS. SSC sector skills council TKA Foreign Workers System TVET technical and vocational education and training UX/UI User Experience/User Interface VOC Voice Of Customer WLKP Mandatory Employment Reporting XVI Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice XVII As Indonesia’s economy has grown in recent years, pov- erty has fallen. Over the past two decades, gross domes- tic product (GDP) per capita increased fivefold, from USD823 in 2000 to USD3,932 in 2018, and the national poverty rate fell by more than half, reaching 9.8 percent 1. Introduction: Improving Labor Market in 2018.1 These achievements were accompanied by the significant structural transformation of the labor mar- Efficiency in Indonesia ket. Between 2001 and 2018, the share of employment in the agriculture sector declined from 44 percent to 29 percent and wage employment rose from 29 percent to 40 percent. However, about 26 million Indonesians are still poor, and another 50 million are vulnerable to falling into poverty. Indonesia also faces many jobs-re- lated challenges. Informality remains the norm, with 57 percent of the workforce either being self-employed or working for wages without regular employment. About 22 percent of workers are in elementary occupations. Recent estimates suggest that the low-skill, low-pro- ductivity nature of many jobs in Indonesia makes them vulnerable to automation (Patrinos 2017). Meanwhile, although the labor force is expanding rapidly, women remain an untapped resource, as only slightly more than half of them are active in the labor market, compared with slightly more than 80 percent of men. Indonesia is looking to improve the quality of available jobs, make rapid progress toward eliminating poverty, and ensure sustainable improvements in welfare for its large pop- ulation through a strategy centered on human capital development. There are indications that skills mismatches are building up in Indonesia due to labor market failures. If the la- bor market was functioning well, matches between job seekers and employers would be the norm. However, this is not happening in Indonesia at the moment. While unemployment rates are low overall, at 5.4 percent, young people and those with higher levels of educa- tion are overrepresented.2 At the same time, according to the 2015 Enterprise Survey conducted by the World © Joko P 1 . See https://www.worldbank.org/en/country/indonesia/overview. . Estimates for 2018, based on National Labor Force Survey (Sakernas) 2  data. Bank, employers complain about the difficulty of filling not benefit from its investments in human capital and Figure 1.1 Potential Costs of Labor Market Failures high-skill jobs, suggesting that not only the level but also is unable to compete effectively due to low productivity the relevance of skills in the population do not match and weak economic performance (figure 1.1). what firms actually need.3 More than 90 percent of In- donesians looking for work rely on informal networks Governments around the world use active labor market (family and friends) rather than on publicly provided programs (ALMPs) to address these labor market fail- employment services, suggesting that such services are ures. ALMPs are usually categorized into three groups: Employers • Reduced competitiveness due to The Economy not sufficiently developed. 4 interventions that increase employability (for exam- higuer costs and lower outputs • Lower productivity and returns to ple, targeted vocational training and self-employment/ • Below-capacity performance human capital investments -Constrained investments • slower economic Labor market failures in the areas of information, coor- entrepreneurship programs), interventions that spur performance dination, and credit constraints contribute to skills mis- labor demand (for example, wage subsidies, labor-in- matches in Indonesia.5 When labor market information tensive public work), and interventions that facilitate is incomplete, job seekers lack information on the skills the job-matching and placement process (for example, demanded by employers, potentially resulting in the mis- job-search assistance and information on job seekers). allocation of education investments. At the same time, However, ALMPs are not a silver bullet for addressing all employers experience higher search costs due to limited market failures;6 complementary interventions focused information on job seekers’ skills or short supply of rele- on the demand side (for example, programs helping Employees vant skills in the labor market. Coordination failures also firms to innovate, grow, and create more jobs) and on • Lower wages, job satisfaction and job-security contribute to skills mismatches, with employers underin- labor regulations (for example, minimum wages and la- • Limited career vesting in training if they expect that workers will change bor laws) are also important (McKenzie 2017). opportunities jobs after acquiring the additional skills. Finally, failures in credit markets imply that poorer households might This report examines how the development of a be unable to finance appropriate training, job search, or well-functioning labor market information system self-employment options, thus ending up either unem- (LMIS) can improve the information provided to dif- Source: Adapted from CMEA and WB 2019. ployed or in jobs that do not match their skills. ferent stakeholders in Indonesia. The services provided by a well-functioning LMIS fall under both the ALMPs Ultimately, these labor market failures lead to inefficient and the demand-side type of policy interventions. All of job matches and limit the economic opportunities for these services seek to improve labor market outcomes 2008). The overall objective of an LMIS is, therefore, to to strengthen its job-matching platform, AyoKitaKerja.8 less well-off job seekers. On the employer side, poor job by providing better information. Providing informa- produce relevant information to improve human capital MoM launched AyoKitaKerja in 2016 to serve a growing matches raise costs by lowering productivity and out- tion and guidance is particularly important in low- and management and labor outcomes so that the country number of firms and job seekers (both unemployed and puts, reducing competitiveness, and limiting long-term middle-income countries where access to labor market can meet its economic development and growth poten- employed who are looking to improve their working sit- performance and investments. On the employee side, information from other sources is limited and in com- tial. To achieve this objective, the LMIS needs to provide uation). The platform has some basic elements of an in- workers receive lower wages, have less job security, and munities where socioeconomic status and social norms accurate, timely information and services that facilitate tegrated LMIS but is not yet a world-class job-matching are less satisfied when their skills are poorly matched affect aspirations (; Dizon-Ross 2014). For example, pro- the decision-making processes of various stakeholders platform with a broad-based set of users, a critical mass to their job. As a result, the country’s economy does viding students and job seekers with information about in the labor market.7 of high-quality information, and a high success rate in skills and career prospects, own abilities, and avenues terms of jobs placements.9 3 . A Program for the International Assessment of Adult Competencies for financing has been shown to have positive impacts The Indonesian Ministry of Manpower (MoM) is seek- (PIAAC) survey carried out in 2015 in Jakarta also found mismatches in on education and employment outcomes (see, for exam- ing to develop a more modern, comprehensive LMIS. 8 . A job-matching platform that connects the demand for and supply of the literacy skills of workers and the level of proficiency demanded in their workers is a critical, but not sufficient, element of a well-functioning LMIS. ple, Dinkelman and Martínez 2014; Jensen 2010; Nguyen As a first step toward achieving this goal, it is seeking jobs (OECD 2015). 9 . On September 29, 2019, MoM launched the online platform named Sis- 4 . Estimates for 2018, based on Sakernas data. 7 . A well-functioning LMIS does not crowd out privately managed job naker, which aims to offer an entry point to all MoM databases, whether 5 . A forthcoming report by the World Bank provides more evidence on 6 . McKenzie (2017) shows that ALMPs are not always highly impactful and portals; instead, it makes them active players. The government plays an used by the public or by government practitioners. Sisknaker also incor- these market failures and on how the government is addressing them that the most effective ALMPs are usually designed to overcome spatial and important role as a coordinator of information so that all stakeholders in porates a job-matching platform—KarirHub—that includes some of the (World Bank, forthcoming). sectoral mismatches. the labor market have accurate, timely information for decision making. desirable characteristics discussed in this report. Yet that platform is too 2 Toward a World-Class Labor Market Information System for Indonesia 1. Introduction: Improving Labor Market Efficiency in Indonesia 3 actors; instead, it complements and promotes them Stakeholders © Kzenon by linking to and incorporating functions that are un- dertaken by private institutions. Stakeholders are the potential users and producers of la- bor market intelligence generated by the LMIS. Stakehold- LMISs can be classified according to the quality of infor- ers include actual or potential workers (for example, un- mation they produce, the involvement of stakeholders, employed, inactive, or current and future students), firms, and the strengths of the connections between institu- policy makers, researchers, and practitioners (for examples, tional partners. Table 1.1 presents the distinctive char- those in skills development systems, public and private em- acteristics of a basic, intermediate, and advanced LMIS ployment services, and other public service systems, such (chapter 2 elaborates more on this classification; see So- as social protection). renson and Mas 2016). A key element of an LMIS is an online platform that is Key Functions managed by the government and used to disseminate labor market information and provide online services. The key functions of an advanced LMIS are to provide direct While being an important component, the online plat- services to different users as well as to provide data and in- form managed by the government is only one element formation analyzed and processed by users. The various of an LMIS. LMISs also include the information and ser- functions serve the needs of stakeholders but also rely on vices provided by the private sector, such as private in- them for their input and processing. The LMIS includes the termediation services or private career guidance pro- following key functions: viders. The government-managed platform evolves in In order to identify the actions needed to develop AyoK- Labor Market Information Systems response to and jointly with the information and services • Job matching. Core services that help unemployed or itaKerja, the government of Indonesia asked the World of the whole system. Government-managed platforms employed workers—looking to improve their working Bank to benchmark the functions and features of AyoK- Most countries produce labor market information, rely on several sources of information, including other situation—to identify job opportunities, help firms to itaKerja against an advanced job-matching platform. but many lack accurate, relevant, and up-to-date in- government agencies such as national statistics offices, close their skills gaps, and assist practitioners in public or The Republic of Korea’s Work-net system was chosen formation that can assist key labor market actors. 10 line ministries, education institutions, and private inter- private employment services to conduct job matching. for this exercise. Work-net is the job-matching platform An advanced LMIS that provides relevant, timely in- mediation services, among others. These platforms are AyoKitaKerja is, above all, a job-matching platform. managed by the Ministry of Employment and Labor and formation can help stakeholders to make informed often connected to the face-to-face services of other • Career and skills guidance. Services that guide current Korea Employment Information Service, a public agen- choices on a variety of issues, including job search, organizations, such as public employment agencies and and future workers in identifying the skills that are, or cy at the center of Korea’s LMIS. Work-net, which, sim- career planning and skills upgrading, business invest- schools and training institutions. This report focuses will be, in demand, their own aptitude and interest in ilar to AyoKitaKerja, started as an online job-matching ment and hiring decisions, migration policies, public mainly on this element of the LMIS; hereafter, the term different types of jobs, and the skills development paths platform, now provides comprehensive employment or private workforce and skills investments, and op- LMIS refers to the public interface and its underlying needed to fulfill their ambitions. Successful career guid- information and employment support services. By high- timal employment and skills policy designed to meet support services. ance services connect users with education and training lighting some important gaps between the functions, future demands. An advanced LMIS is an integral part opportunities. organization, and implementation of Work-net and of modern employment services that help to reduce • Government support. Services that help individuals and AyoKitaKerja, the benchmarking exercise is intended to unemployment, boost the productivity of firms, man- practitioners to identify potential government programs provide practical information on possible actions for de- age and prevent risks related to labor markets, and (ALMPs, other skills-building programs, business sup- veloping AyoKitaKerja and improving its performance. maximize the impact of skills and labor market pro- Framework for an Advanced LMIS port, social welfare programs) from which they could grams (Sorensen and Mas 2016; Woltermann 2012; benefit. Yoo 2018). The LMIS does not compete with private This report proposes a framework defining an advanced • General labor market information and analysis. Infor- LMIS (figure 1.2). The framework includes LMIS stake- mation that provides policy makers, practitioners, and young to be evaluated comprehensively, as only a few job seekers and 10 . Typical examples are labor force data collected by national statistics holders, the functions it offers, the underlying key ele- researchers with comprehensive information on labor firms were registered at the time this report was written. offices. ments, and the essential features. market outcomes and the impact of different policies 4 Toward a World-Class Labor Market Information System for Indonesia 1. Introduction: Improving Labor Market Efficiency in Indonesia 5 Table 1.1 LMIS Classification Figure 1.2 Framework of an Advanced LMIS Stakeholders Element Basic Intermediate Advanced Workers Research Information Provides basic information Provides market intelligence Provides several subsystems (current and potential) Student Firms Policy Makers Practitioners Community via statistical indicators using different data sources of information and digital using national surveys and beyond surveys tools to collect, produce, administrative data evaluate, and disseminate Functions updated (often real-time) information using big data Labor Market Information Job Matching Career and Skills Guidance Government Support and Analysis Services None Provides basic services to Provides several services different stakeholders— in tailored to different users particular, job seekers and Key Elements firms Data Data Accessibility Targeted Strategic Interoper- Technical Data Actors Involves a small number of Involves more public sector Involves both public and Collection Validation and User Service Marketing ability with Strong HR Infrastructure Management Security public sector actors actors, including education private sector actors as Experience Delivery Systems systems, job-counseling and users and producers of data job-matching services, social and services security, statistics offices, Features and business registries Institutional Arrangements Collaborative Partnerships Technology Solutions Source: World Bank adaptation from Sorenson and Mas 2016. and programs, largely with the purpose of informing • Technical infrastructure In addition, five characteristics are key to the success data management and information security systems. policy formulation. Usually, a labor market observatory • Data management of an advanced LMIS. These characteristics are based • Ensure the provision of effective and efficient services (LMO) is the entity in charge of producing the informa- • Information security. on the benchmarking exercise described in chapter 3, by applying technology to facilitate the use of online tion and intelligence of the LMIS and disseminating it to good practice observed in other countries, and a recent services (by individuals, officers in public employ- stakeholders (Rutkowski, de Paz, and Levin 2018). LMOs Three features are essential to support these functions. cross-country comparison of LMISs (Sorenson and Mas ment service offices, and career counselors) and by can be separate agencies, a department in a ministry First, the system should build on strong institutional ar- 2016). These key characteristics allow the system to un- regularly monitoring all aspects of system efficiency (Labor/Manpower, Economy/Planning, etc.) or contract- rangements involving key stakeholders such as policy dertake the following: and effectiveness to improve services continuously. ed out to think tanks or universities. makers and the education system. Second, the system • Be client centered by providing users with compre- should enjoy an active partnership with private sector • Provide relevant information by attracting, collect- hensive, targeted, and relevant information and other Key Elements and Features actors, which is, in itself, an indicator of the value of the ing, and coordinating a critical mass of trustworthy, services through an appropriate design of the online LMIS (Sorenson and Mas 2016). Third, the system should updated information that improves job matching and experience, prompt resolution of user inconvenienc- A well-functioning LMIS has the following key elements: rely on a strong technological base and advanced- tech- by actively linking to other sources of data, both pri- es, several channels of communications, and analysis nology solutions (including traditional surveys and re- vate and public. using both traditional surveys and real-time informa- • Data collection al-time data such as big data and artificial intelligence) • Be reliable and trustworthy by providing frequent up- tion (artificial intelligence, big data). • Data validation for sharing data efficiently and for analyzing and dissem- dates, establishing quality control systems—manual • Strive for comprehensiveness in providing a broad • Accessibility and user experience inating the available information.11 and information technology based—and ensuring range of stakeholders with a diverse set of services to • Targeted service delivery the necessary technical infrastructure that guaran- improve their basis for decision making and operate in 11 . Big-data analysis should be used to complement, not replace, traditional • Strategic marketing tees service availability and information integrity. connection with other public information systems (so- surveys because information available online might be biased toward spe- • Interoperability with other systems cific groups. For example, the jobs posted online are usually high-skill jobs Technical infrastructure should include hardware cial protection, education, and training) and as a fully • Human resources serving the system and its users in the formal sector; they are rarely jobs in the informal sector. capacity, backup and disaster recovery systems, and integrated part of national development policies. 6 Toward a World-Class Labor Market Information System for Indonesia 1. Introduction: Improving Labor Market Efficiency in Indonesia 7 2. The Indonesian Labor Market Information System As presented in chapter 1, the ultimate goal of a labor market information system (LMIS) is to improve human capital management and labor market outcomes in or- der to help the country meet its economic development and growth objectives. An advanced LMIS is defined as a set of stakeholders, functions, key elements, key char- acteristics, and essential features that produce relevant information and provide services that facilitate the deci- sion making of diverse stakeholders. To this end, the Indonesian Ministry of Manpower (MoM) is seeking to develop a more modern and comprehen- sive LMIS.1 In Indonesia, the job-matching platform, AyoKitaKerja, is the most developed function within the country’s system, and the Indonesian LMIS overlaps al- most fully with it. This chapter examines the AyoKitaKer- ja platform in detail and identifies its current users and functions according to the LMIS framework presented in chapter 1. It describes additional initiatives that are part of Indonesia’s broader LMIS, discusses international ex- amples of advanced LMISs, and suggests a classification for Indonesia’s system. © Akhmad Dody Firmansyah 1 . See note 9 in chapter 1. 9 Table 2.1 AyoKitaKerja Job Vacancies and Job Seekers, 2016–19 Figure 2.1 AyoKitaKerja Job Placements, 2016–19 25,210 Year Average number of job vacancies Average number of job seekers 2019* 69,732 511,131 18,500 2018 500,480 539,170 2017 55,466 170,853 12,000 2016 161,573 468,672 7,000 Source: Ministry of Manpower and AyoKitaKerja website. Note: AyoKitaKerja was initially designed to cover domestic vacancies only. The significant rise in vacancies in 2018 was due to the incorporation of international vacancies as well as to the establishment of a team focused on posting vacancies from 2016 2017 2018 2019 various sources. The team was dissolved in 2019. Source: Ministry of Manpower (2016–18) and AyoKitaKerja website (2019). * January 1, 2019 to November 4, 2019. * January 1, 2019 to November 4, 2019. AyoKitaKerja: Overview of Its Since its inception, the number of users of AyoKitaK- based in Java (73 percent) and are young (81 percent are of the labor market. Moreover, to be representative of Functions and Users erja has fluctuated significantly. The platform current- 25 years or younger); men and women are represent- the Indonesian labor market, the platform would have ly covers a relatively small portion of Indonesia’s total ed almost equally; the vast majority of those registered to expand the services offered to firms seeking workers In 2016 MoM launched AyoKitaKerja as a job-matching unemployed. Table 2.1 shows indicators for AyoKitaKerja (86 percent) have completed no more than senior high with higher-level skills and to capture additional informa- platform catering primarily to job seekers and firms but since its inception. In 2018, the latest year for which la- school. From the demand side, vacancies are directed tion on more qualified jobs and job seekers. The limited also to practitioners in public employment offices who bor force data are available for comparison, AyoKitaK- largely to job seekers with low or medium education (97 use of AyoKitaKerja among the total number of unem- assist these groups in the job-matching process. The plat- erja had around half a million active registered job seek- percent require no more than a senior high school diplo- ployed and other job seekers (and, presumably, total job form provides basic job-matching services and some job- ers, compared with a pool of unemployed of nearly 6 ma); as many as 81 percent of vacancies are posted by offers) affects the services it can offer. The quality of the search guidance but lacks career guidance as well as links million and a labor force of 130 million.2 From the labor firms in the manufacturing sector.5 job matching and other LMIS functions will depend on to relevant public and private information sources and la- demand side, some 12,000 companies were registered. the volume and representativeness of the labor supply bor market analysis. This section assesses AyoKitaKerja’s By November 2019, the annual average number of active Since it was first launched, AyoKitaKerja has had a pos- and labor demand reached. The current numbers do not performance in each of the four key LMIS functions. job vacancies was just below 70,000, far fewer than the itive trend in terms of job placements (figure 2.1). The reflect national employment trends; as such, they have number of job seekers, at more than half a million. AyoK- total number of job placements has kept up with the limited use for policy making and for public employment itaKerja staff recognize that these figures may overesti- increasing number of job seekers. The best year for services beyond the current job matching in the system. Job Matching mate the actual number of vacancies and job seekers, as total job placements was 2019. Between January 1 and data are not updated regularly. By November 2019, the November 4, 2019, AyoKitaKerja made 25,210 job place- AyoKitaKerja currently works as an online employmentZ average daily number of visits was just below 40,000.3 ments, corresponding to about 5 percent of job seekers Career and Skills Guidance service intended to improve labor market intermedia- and one-third of vacancies on the platform. tion, reduce unemployment (and duration of unemploy- According to the most recent information available, AyoKitaKerja currently provides basic job-search guid- ment), and help enterprises to find qualified workers to AyoKitaKerja serves a specific segment of the labor mar- To become part of Indonesia’s broader human capital ance on its homepage. The platform is not operated fill their needs. The service matches job vacancies with ket. 4 From the supply side, a majority of job seekers are policy and a building block of the country’s LMIS, AyoK- by career guidance counselors; it includes generic tem- job seekers using an online tool that connects job seek- itaKerja would have to target and reach other segments plates for résumés and job-search letters as well as gen- ers and companies. The job-matching platform enables 2 . AyoKitaKerja considers a user to be active if she or he has logged in at eral job-search tips. Through an online form, users can 5 . As of 2018, a labor regulation requires firms wishing to hire a foreign job seekers to find information on current job vacancies least once during the previous six months. post questions related to job-search approaches, trends worker to also appoint a “domestic companion worker” to facilitate skills and requirements of the jobs offered and enables em- 3 . Information accessed on the AyoKitaKerja website on November 4, 2019. transfer. This regulation may encourage firms to post vacancies for very in job vacancies, information on the requirements of ployers to find information on job seekers. 4 . Produced by the AyoKitaKerja team on September 12, 2019. low-paying jobs in order to meet this requirement. different jobs, and training needed. The Counseling and 10 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 11 Job Guidance Group in MoM (outside of AyoKitaKerja) job seekers who need help registering on AyoKitaKer- Table 2.2 Indonesia’s LMIS Initiatives, by Function responds to questions by e-mail. No information is avail- ja. However, given the limited information provided by able on the volume or quality of this service. the platform, counselors cannot draw on AyoKitaKerja Labor market Career and skills Government to identify training opportunities to enhance the skills Initiatives Job matching information and guidance support analysis AyoKitaKerja does not connect its users to other relevant needed for a chosen career path or to identify govern- and currently available sources of information. For exam- ment programs and support that can help job seekers Ministry of No Limited to e-Bursa No Limited to infor- Manpower Kerja mation on TKA, ple, MoM’s website provides very basic information on to succeed in the labor market. WLKP, and SKKNI training opportunities and e-learning platforms,6 but not on general or vocational training through the education The platform is not linked to other government ser- AyoKitaKerja Yes Limited Limited No system. This information is not linked to AyoKitaKerja. vices such as social protection programs or to training BNSP No Limited to No No opportunities and scholarships. However, it is linked certifications Although it is planning to expand relationships, AyoKita- with the Ministry of Home Affairs population database BPS No No No Yes, basic statistics Kerja does not take full advantage of career centers at through a unique identification number that allows reg- based on Sakernas educational institutions. The platform is advertised in ca- istered users to retrieve some basic data. Civil Servant Basic and limited Basic and limited No No reer guidance centers at some universities and training Selection System to civil servants to civil servants institutions, but it is not linked to any of their online ser- Ministry of No No Yes No vices and is not operated by career guidance counselors. Labor Market Information and Analysis Education Education and No Yes No No Overall, given its narrow focus on job search, AyoKitaK- Although AyoKitaKerja provides information on job training platforms erja cannot provide information or counseling services seekers and vacancies, its broader use as a source of to students or persons wishing to upgrade their skills. labor market information is limited and not tailored Privately managed Yes No No Yes to different stakeholders. Moreover, as AyoKitaKerja platforms covers a very small share of the labor market, the un- Government Support derlying data are not yet sufficiently informative about Note: TKA = National Work Competency Standards. TKA = Foreign Workers System. WLKP = Mandatory Employment Reporting. general labor market trends, job-seeker profiles, or the BNSP = Professional Certification Authority. BPS = Central Bureau of Statistics. Only about half of local public employment service job-matching process. While Indonesia’s Central Bureau (PES) offices use AyoKitaKerja to assist job seekers be- of Statistics (BPS) produces labor market data, it does cause they have limited staff, computers, and budget not present information from AyoKitaKerja on its web- to cover internet costs as well as unreliable internet site, and AyoKitaKerja does not share information with • Ministry of Manpower (kemnaker.go.id). The website hiring foreign workers in the process of connection, particularly in the eastern region. MoM other users through open application programming in- provides information on different services offered by obtaining a work permit. Information includes has 514 regional offices; of these, 262 offices use the terfaces (APIs). the ministry, including: 8 the process, documents to download and fill AyoKitaKerja platform., In each regional office, on aver- o Links to the Indonesia National Work Compe- out, requirements, frequently asked questions, age, only about two staff are responsible for delivering tency Standards (Standar Kompetensi Kerja and contact information. employment services.7 Counselors at PES offices assist Nasional Indonesia [SKKNI]), which provides o Link to Mandatory Employment Reporting basic information on the objectives of the (Wajib Lapor Ketenagakerjaan di Perusahaan 6 . An example is the Workers’ Capacity Building and Job Training Search, Additional Initiatives That Are competency framework and standards and on [WLKP]), which gathers information on com- which provides information on training and courses offered by national Part of Indonesia’s LMIS ALMP training courses. panies fulfilling the Mandatory Report on Com- and subnational work training centers (BLKs). See http://elearning.pus- o Links to the Foreign Workers System (Tenaga pany Employment.9 Firms can enter specific diklat.kemnaker.go.id/ as well as https://pelatihan.kemnaker.go.id/. Besides AyoKitaKerja, the government of Indonesia has Kerja Asing [TKA]), which provides information details on the company as well as labor-related 7 . Job placement services make up a very small part of PES work in Indone- several other platforms that, although not currently in- and services to foreign workers and employers information. sia. For example, MoM’s Bogor Regional Office has 65 staff members, and only 6 staff are in charge of job placement services. Nationally, about 524 terconnected, can be considered part of Indonesia’s PES staff use the platform. MoM officers provided the estimates during an LMIS (table 2.2): 8 . See note 9 in chapter 1. 9 . The online submission of the report is mandatory according to interview on May 23, 2019. Regulation 18/2017. 12 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 13 LMIS Classification security or education systems, are not centralized or © Odua Images analyzed. LMIS can be classified into basic, intermediate, and ad- vanced systems according to the quality of information Intermediate LMISs involve more public actors, includ- they produce, the involvement of stakeholders, and ing, among others, education systems, job-counseling the strengths of the connections between institutional and job-matching services, social security, statistics partners. This classification is based on the framework offices, and business registries. Intermediate systems presented in Sorenson and Mas (2016) and built on provide both data and some basic services to diverse a landscape review of international best practices of stakeholders, in particular, job seekers and private LMIS models. firms. Different data sources beyond surveys are used to produce labor market intelligence. The private sec- Sorenson and Mas (2016) argue that an LMIS is not just tor is still far from an active user or producer of infor- the technical infrastructure that gathers, processes, mation or services, however. and disseminates labor market data (what they call a data-driven system) or just the services that provide Advanced LMISs have fully developed the five char- information to workers, employers, and intermediaries acteristics outlined in the framework in chapter 1. (what they call a service-oriented system). Instead, a Specifically, they are relevant, efficient, reliable, and comprehensive LMIS has institutional arrangements comprehensive, and they put clients at the center. that allow for constantly revolving information flows They involve both private and public actors as pro- between key stakeholders (as well as policy makers, ducers of information and users of data and services. researchers, and other institutions), integrating both These information systems are complex and compre- data-driven and service-oriented subsidiary systems. hensive, consisting of several subsystems of informa- A comprehensive LMIS produces labor market intelli- tion that successfully serve different users and help o Link to e-Bursa Kerja, which is a webpage with Culture contains information on scholarships for ed- gence and analysis needed to improve the functioning to reduce information gaps. Advanced LMISs collect, job fair announcements. ucation and training programs, both degree and non- of the labor market since “data, policies, and services produce, evaluate, and disseminate labor market–re- • BNSP (bnsp.go.id). The website of the Profession- degree, as well as courses and training information. are interlinked and mutually reinforcing.” In addition, lated intelligence (historical, current, and projections). al Certification Authority (BNSP) provides infor- • Education and training platforms (for example, the proposed framework provides useful guidance on They provide both data for different users as well as mation on regulations and processes to job seek- https://forlap.ristekdikti.go.id/perguruantinggi and how to add stakeholders progressively and enhance direct services tailored to different users. They con- ers who wish to be professionally certified. It also http://sekolah.data.kemdikbud.go.id/). These plat- their contributions and functions to help countries to nect the demand and supply sides of labor markets and provides links to professional certification agencies forms include government websites with detailed move toward a more advanced system.10 provide policy makers and researchers with data for (listed and licensed by BNSP). information on degree and nondegree educational policy evaluation and labor market analysis. Data and • BPS (www.bps.go.id). The national statistics insti- opportunities. Basic LMISs are mostly data driven. They provide sta- services are both inputs to and outputs from the sys- tute provides basic indicators and publications re- • Privately managed platforms. Platforms such as Ru- tistics on the labor market and involve only a few public tem: reliable data are needed to inform services and lated to the state of Indonesia’s labor force based angguru, JobStreet, LinkedIn, Glassdoor, and Global actors as users or producers. They focus on statistical interventions, and information emanating from ser- on the analysis of Sakernas (Indonesia’s Labor Force Salary Calculator are also part of the LMIS, provid- indicators produced from surveys and registries (labor vice functions provides input for generating data (for Survey). ing job-matching services and labor market infor- force surveys, establishment surveys, registry data), example, input from employment offices). To provide • Civil Servant Selection System (sscn.bkn.go.id). The mation that complement government-managed and they do not integrate employment services, to updated information and customized services, several website is the official site for the national registra- subsystems. the extent that such services exist. Other sources of data sources need to be combined, including both tra- tion of civil servants and provides an integrated sys- labor market data, generated from, for example, social ditional surveys as well as real-time data analyzed using tem for their recruitment, including job postings for big-data techniques (box 2.1). all government institutions and ministries. 10 . The landscape review examined the LMISs of Australia, Botswana, • Ministry of Education and Culture (kemdikbud.go.id). Cameroon, Denmark, France, Jamaica, Rwanda, South Africa, the United The official website of the Ministry of Education and Kingdom, and the United States (California). 14 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 15 Box 2.1 Data Sources for an Advanced LMIS An advanced LMIS gathers, analyzes, and disseminates infor- Since real-time labor market data have not been designed mation based on diverse and complementary data sources. for analytical purposes, big-data techniques are used to transform the data into a usable form for their analysis. Traditional labor market sources are typically worker-level The structure and content of real-time labor market data and firm-level surveys, censuses, and administrative data- (text, videos, links) vary significantly. Big data also provide bases. Surveys and censuses have been designed explicit- the raw material for many of artificial intelligence applica- ly to capture information on the labor market and to be tions, where machines “learn” from data how to mimic hu- representative of a specific population (country, regions). man behavior (for example, chatbot services, which auto- These attributes make traditional data relevant for analyz- mate customer interaction mimicking a real person). ing labor market and occupational trends, demand for skills, supply and shortages of skills, and so forth. Administrative Big data play a central role in advanced LMISs by providing data can also provide detailed information, and although new approaches to labor market analyses and improving © Maharani Afifah not all administrative data may be relevant, the informa- job matching. The following are examples of this role: tion can be disaggregated at the local level. Examples of traditional sources include labor force surveys, enterprise • The Administrative Data Research Network in the surveys, surveys on salary and wages, vacancy and occupa- United Kingdom (https://esrc.ukri.org/), which links tional surveys, population censuses, administrative vacan- data from universities, government, and statistical au- cies, job seekers, firms, and tax records. thorities’ research centers, among others. In a recent project, big-data analysis provided an improved under- Based on the current profile of information and ser- International Examples of Real-time labor market sources are an increasingly import- standing of the links between employment, benefits, vices provided, Indonesia’s system can be classified as Advanced LMISs ant source of data with two key characteristics. They are and committing an offense. The results of the analysis a basic-to-intermediate LMIS. The system has not yet continuously available and provide granular information on informed policies that help ex-offenders to reintegrate developed the quality and variety of data and services This section presents examples of advanced LMISs in the current state of the labor market due to the large vol- into society through gainful employment. or the institutional interconnectivity of an advanced other countries.11 These systems have different pub- ume of data generated by humans and machines. For ex- • The German Virtual Labor Market (LVM) (www.job- system. Within the current platforms of the Indone- lic interfaces tailored to different users; they provide ample, analyzing online job vacancies can provide specific boerse.arbeitsagentur.de), which provides a match- sian government, AyoKitaKerja is the most developed, customized services at the individual level, including information on desired skills, worker attributes and char- ing engine. This matching engine can provide tailored which is why it was selected for the benchmarking ex- through applications. Job-matching, career guidance, acteristics, job titles, and occupations demanded as well as search results based on big-data analysis even when ercise in chapter 3. The platform provides job-match- and skills development services are sophisticated and up-to-date information about job-related activities. How- not all search criteria are fulfilled. The matching ing services to private actors, provides limited career integrated and include advanced tools such as big-da- ever, given their nature, real-time data should be used as process works both from the job-seeker side and guidance, government support, and information, and ta analysis. Data and services are provided at the local a complementary source of information, not as a replace- from the employer/job-vacancy side. is used, to a limited extent, by job seekers, PES offices, level, and industry-specific data and services are avail- ment for traditional labor market data. Data from online • The Republic of Korea’s Job Compass (https:// and firms. Still, to be considered an advanced system, able. Research and analyses on different labor market– sources is likely to be biased toward specific groups or to www.work.go.kr), which uses information on mil- the platform needs to be targeted and developed fur- related topics are produced and accessible. Real-time cover a specific subsample of the population. Online data lions of clients who found jobs via Work-net to ther: a limited number of private actors use the plat- data, survey-based information, and case studies are have methodological issues related to quality, representa- provide tailored recommendations to current job form; AyoKitaKerja does not produce information for collected from a variety of private and public sourc- tiveness, and generalizability (Kureková, Beblavý, and Thum seekers. other users or purposes; data sources are limited; and es and then analyzed, transformed, and communicat- 2014). In the case of online job vacancies, for example, post- AyoKitaKerja is not linked with other producers or us- ed through a variety of channels and tools, including ings are likely to be related to formal job opportunities, to ers of data, including the BPS, education and training downloadable reports and briefs, illustrations, games, be directed to highly skilled workers, and to cover oppor- institutions, social security agencies, or private job 11 . The international examples presented in the tables include (1) a list tunities in the main job centers, larger firms, and specific Sources: Cedefop 2019; WCC Development 2017; World portals. The benchmarking exercise in chapter 3 high- of subsidiary systems, (2) detailed descriptions, and (3) their functions. sectors of the economy (Zhu, Fritzler, and Orlowski 2018). Bank 2019. lights gaps that the platform needs to overcome to be International examples from Australia, Denmark, France, and the United considered an advanced LMIS. Kingdom were adapted from Sorensen and Mas (2016). 16 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 17 Table 2.3 Subsidiary LMIS Systems in Australia Table 2.4 Subsidiary LMIS Systems in Denmark Subsidiary systems Details J-M C&S Gov S LM I&A Subsidiary systems Details J-M C&S Gov S LM I&A Labor market Provides up-to-date information on local, state, Job Indsats Provides information on different employment information portal national, and regional labor markets (jobindsats.dk) services and policies (unemployment insurance,    welfare benefits, training opportunities) (lmip.gov.au) Displays labor market indicators and publishes analyses and background reports Job Effekter Provides research from private and public sources, Employment re- Provides policy makers, researchers, and others (jobeffekter.dk) analyses of labor market trends, and evaluation of  search and statistics with relevant research and analyses on labor the effectiveness of various labor market policies (employment.gov.au) market and industry trends, skills shortages and   Jobnet Provides tools for job matching, finding informa- needs, and recruitment experiences to support (jobnet.dk) tion on employment services, searching for jobs, policy development and finding career guidance information    JobActive Provides job-matching services throughout the Serves both job seekers and employers with tai- (jobactive.gov.au) country and career and skills guidance lored platforms Includes detailed information on vacancies, salary,   Has specific interfaces for vulnerable groups job prospects, and location Jobindex Private job-matching portal Also has a mobile application  (jobindex.dk) JobOutlook Provides career guidance to help individuals to Employment sta- Provides labor market–related statistics and anal- (joboutlook.gov.au) match skills, aptitude, and ambition with jobs tistics (dst.dk/en/ ysis using a wide variety of data sources (munic- Provides local information   statistik/emner/) ipalities, job centers, business registries, social Identifies future growth industries and occupa-  welfare, and immigration) tions Allows time series to be customized by, for exam- Myskills Provides students, job seekers, workers, career ple, gender, age, and regions (myskills.gov.au) counselors, and employers with information on Note: J-M = job matching. C&S = career and skills guidance. Gov S = government support. LM I&A = labor market information and  education and training opportunities across the analysis. country MyFuture Provides career-related tools and information for (myfuture.edu.au) students, job seekers, workers, and career coun-  cludes up-to-date information from private agencies on Denmark selors, including pathways to apprenticeships Is targeted specifically to youth job vacancies and skills in demand, continuously feeds the MyFuture and JobOutlook platforms for analysis Denmark has interconnected subsidiary systems that Note: J-M = job matching. C&S = career and skills guidance. Gov S = government support. LM I&A = labor market information and and intelligence. MyFuture also partners with large pri- combine specific services for labor market supply and analysis. vate firms for providing information on apprenticeships. demand with data for decision makers and policy mak- At the same time, the labor market information portal ers (table 2.4). Labor policies are based on labor market extracts data from the JobActive platform and then intelligence and analysis provided by the country’s LMIS. provides information on skills, wages, and employment and videos. These systems provide information and Australia prospects per occupation at several geographic levels. services over both the short term (job matching) and Because of this connectivity, the LMIS is able to produce France medium to long term (career guidance, skills fore- Australia offers a clear example of how subsidiary sys- several up-to-date reports to users on the state of the casts). Private and local government actors provide tems feed into one another and the importance of ac- labor market. A wide variety of stakeholders use France’s subsidi- input to the system and obtain information from the tive engagement with the private sector (table 2.3). For ary systems (table 2.5). Employers widely use the Pôle system. example, the JobActive job-matching platform, which in- emploi’s database to find and hire workers though a 18 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 19 Table 2.5 Subsidiary LMIS Systems in France Table 2.6 Subsidiary LMIS Systems in the Republic of Korea Subsidiary systems Details J-M C&S Gov S LM I&A Subsidiary systems Details J-M C&S Gov S LM I&A National Institute Provides highly detailed national labor market Korea Employment Provides targeted information to customers and pol- for Statistics and statistics and analysis using a variety of sources Information icy makers so that they can continuously improve the  Services employment service tools and subsystems  Economic Studies (surveys, census, and administrative data) (insee.fr) (eng.keis.or.kr) Provides linkages to the other subsystems DARES Employ- Provides labor market data analysis on the Work-net Provides job-matching and customized advice, informa- ment dashboard themes of work, employment, professional train- (work.go.kr) tion, and support on employment services and pro- (travail-emploi. ing, and social dialogue    grams, government support for training and education, gouv.fr) Also has PoEm, a monthly dashboard and series social welfare programs, and lifelong career guidance (including vocational psychological tests)     on employment policies Compte Personnel Links personal training accounts accompanying Provides comprehensive, targeted information on occupa- Formation (mon- the worker through working life to online career tions, fields of studies, and labor market research and anal- compteformation. counseling and training services and tracks certifi-    ysis based on a variety of data sources, including big data gouv.fr) cations received HRD-net Provides information on training and tracks an indi- Orientation pour Information on training opportunities (hrd.go.kr/hrdp/ma/ vidual’s training history through an individual training Tous (Orienta- pmmao/indexNew.do) account  tion-pour-tous.fr) Supports training providers Provides employers with services such as workplace    Pôle emploi Contains detailed information on vacancies, sala- information management, application for training (Pole-emploi.fr) ry, and firm information Tracks trainee attendance at government-funded vo- Allows employers to access job seekers’ profiles, cational skills development projects to prevent fraud interact with an employment agency, and sched-    ule interviews Employment Provides information, statistics, and trend analysis of Also provides information on hiring, training, and Insurance Network employment and unemployment insurance public benefits (ei.go.kr/ei/eih/cm/ Provides services to individuals, employers, and PES   hm/main.do) staff to integrate unemployment insurance and other Note: J-M = job matching. C&S = career and skills guidance. Gov S = government support. LM I&A = labor market information and benefits analysis. Employment Provides foreign workers, employers, and govern- Permit System ment staff with integrated EPS-related information, (eps.hrdkorea.or.kr/ such as workplace transfer history, labor employment   main/intro.do) contracts, and training history customized employer portal. Job seekers use it to find wide variety of traditional and nontraditional sources dis- jobs. This portal includes almost 7 million résumés that aggregated by gender and with geographic visualizations. Note: J-M = job matching. C&S = career and skills guidance. Gov S = government support. LM I&A = labor market information can be filtered easily using sophisticated engines; it also Table 2.5 presents examples of France’s LMIS subsystems. and analysis. PES = public employment service. EPS = Employment Permit System. allows employers to add selected candidates to a pool and then to interact with employment agencies to receive help with the recruitment process. Intermediaries—such Republic of Korea as employment, career, and training centers—use Orien- different users, including job seekers, employers, research- outlooks, working conditions, and demand for skills in KEIS. tation pour Tous to help youth to find jobs. Orientation Korea offers another example of how interconnecting all ers, and policy makers. Data sources include national statis- HRD-net, a job-training platform, provides training opportu- pour Tous provides information on qualifications and subsystems adds value to the system (table 2.6). Korea tics, surveys related to employment and skills, and databases nities in conjunction with job information on Work-net. Both training needs, complementing information provided on Employment Information Service (KEIS) uses information in from KEIS networks, which include Work-net, HRD-net, and systems are connected and interoperable. The Employment the Compte Personnel Formation. Policy makers have ac- the LMIS to monitor and evaluate public policies and create la- the Employment Insurance Network. Work-net provides data Information Integration Analysis System analyzes data col- cess to detailed, updated labor market projections from a bor market intelligence (such as analysis and forecasting) for to KEIS and provides users with information on occupational lected from all of the subsystems in real-time. 20 Toward a World-Class Labor Market Information System for Indonesia 2. The Indonesian Labor Market Information System 21 Table 2.7 Subsidiary LMIS Systems in the United Kingdom Subsidiary systems Details J-M C&S Gov S LM I&A National Online Provides detailed, up-to-date UK labor market Manpower Infor- statistics from several official sources and subsys- mation System tems  (nomisweb.co.uk) Is run by the University of Durham Office for National Partners with the University of Durham to pro- Statistics vide labor market statistics based on surveys and (www.ons.gov.uk/ administrative data  employmentandla- Contains detailed local data by occupation, indus- bourmarket) try, gender, and employment projections National Occupa- Contains data on expected tasks, standards of tional Standards performance, knowledge, and skills for different database occupations  (www.ukstandards. Is linked to (employer-led) sector skill councils org.uk) setting these standards StatXplore Provides a tool to explore statistics on benefits (stat-xplore.dwp. administered by the Department for Work and gov.uk) Pensions, currently holding data relating to 16   benefits/programs Universal Job Provides information on job vacancies through a Match (findajob. publicly managed job-matching platform with an dwp.gov.uk) advanced search function Allows job seekers to search with an optional log-   in and to apply for jobs through the portal (if the applicant has an account) Note: J-M = job matching. C&S = career and skills guidance. Gov S = government support. LM I&A = labor market information and analysis. United Kingdom national system provides customized statistics and anal- yses to regional systems. The private sector is heavily in- The United Kingdom offers a clear example of the val- volved through UK sector skills councils (SSCs), employ- ue of partnerships (table 2.7). The University of Durham, er-led organizations designed to boost labor force skills through a partnership with the Office for National Sta- at the sectoral level. SSCs provide research and advice tistics, runs the National Online Manpower Information to employers and policy makers on skills shortages and System. The platform provides web-based databases needs through sector skills assessments, which are dis- and labor market intelligence that integrate data and in- seminated though official LMIS channels. formation from the Department of Work and Pensions, the UK Commission for Employment and Skills, and na- tional statistics, among others. At the same time, this 22 Toward a World-Class Labor Market Information System for Indonesia In order to strengthen the efficiency and effectiveness of the Indonesian labor market information system (LMIS), the government of Indonesia asked the World Bank to benchmark AyoKitaKerja against the Republic of Korea’s system. Work-net, the platform managed by the Korea Employment Information Service (KEIS), is an advanced LMIS platform and a central feature of the 3. Benchmarking AyoKitaKerja against Work-net: What Korean LMIS. Given AyoKitaKerja’s current focus on job matching, the benchmarking exercise presented in this Can Indonesia Learn from the Republic of Korea? chapter focuses mainly on the job-matching functions of the two systems. However, Work-net also provides access to other LMIS services, which could yield useful lessons for AyoKitaKerja and for the Indonesian LMIS in general. Work-net hails back to 1998, when it was established as a publicly managed job-search portal by Korea’s Min- istry of Employment and Labor (MoEL). The platform has undergone several transformations since its launch, with the purpose of improving the quality, scope, and targeting of services. It first offered only web-based services, but, following technological advancements, it incorporated mobile services (2010), big-data services (2018), chatbot services (2019), and, most recently, arti- ficial intelligence (AI) for job-matching services (2020).1 In order to provide these services, MoEL implemented several reform programs, including establishing an infor- mation hub, improving technical infrastructure and se- curity, creating a user-friendly environment, improving the interoperability of the platform with other systems, creating a master database, fully integrating systems, and incorporating AI techniques into job-matching ser- vices.2 Since 2011, Work-net has provided consolidated job postings in strong partnership with local govern- ments and the private sector (for example, Incruit, JOB- KOREA, LinkedIn, Albamon, and Jobplante) in addition to a variety of relevant, high-quality employment-relat- ed information. 1 . Figure A.1 in the appendix presents a chronological timeline of Work- © Sahlan net’s developments since its inception. 2 . Table A.1 in the appendix presents a detailed description of the various phases of reform that Work-net experienced between 2005 and 2017. 25 Table 3.1 Select Development and Labor Market Indicators Comparing Indonesia and the Republic of Korea Indonesia’s GDP per capita is eight times lower than although the specific “how-to” could differ between Korea’s (USD3,894 vs. USD31,363). Korea also outper- countries. Ultimately, Indonesia will have to design its Development and labor market indicators Indonesia Korea, Rep. forms Indonesia in several human capital indicators, reform process in accordance with its own develop- Population (millions) 267 51 including the human development index (0.85 vs. ment objectives, opportunities, and challenges. 0.54), labor productivity (USD72,000 vs. USD26,000), GDP per capita (USD current) 3,894 31,363 percentage of high-skill employees (40.33 percent vs. GDP growth (average annual %), 2011–18 5.4 3 11.91 percent), percentage of the workforce with an Human development index (0–1), 2017 0.54 0.85 advanced education (56.15 percent vs. 12.7 percent), as Labor force participation, 2018 67.7 63.4 well as performance on Programme for International Data Collection Female labor force participation 53.2 53.27 Student Assessment (PISA) assessments in all three Labor productivity (USD, thousands, per employee), 2019 26 72 domains—math, reading, and science. In addition, the Sufficient jobs data are the foundation of high-quality High-skill employees (% of employees), 2019 11.91 40.33 penetration of information and communication tech- job-matching services. A well-functioning LMIS needs Working-age population with advanced education (%), 2019 12.7 56.15 nology (ICT) in infrastructure and information techno- a critical mass of data on both job vacancies and job PISA performance (score), 2018 logy (IT) literacy are different in these countries. While seekers. To achieve the volume needed requires infor- Math 379 526 95 percent of the population in Korea use the internet, mation from various sources. Reading 371 514 only 40 percent do so in Indonesia. Science 396 519 The framework in chapter 1 outlines five key character- Job-Vacancy Data Collection Individuals using the internet (% of population), 2018 39.9 95.0 istics that define advanced LMISs: relevance, reliability, Sources: World Development Indicators database, ILOSTAT database, PISA. efficiency, clients-at-the-center, and comprehensive- Work-net Note: PISA = Programme for International Student Assessment ness. Several key elements underpin these character- istics. The next section presents a benchmarking ex- Work-net collects information on job vacancies ercise and gap analysis of AyoKitaKerja for each of the through three sources: Work-net is widely considered to be a well-functioning projects for the whole structure.3 About 175 employees following key elements in an advanced LMIS: service and is credited with helping to reduce the in- have specific roles and responsibilities related to Work- • Direct postings from firms. By far, most vacancies dividual and societal costs of the 2007 global financial net, including human resources, information technolo- 1. Data collection in Work-net come from companies directly posting crisis. It has received various customer satisfaction and gy, customer services, employment support services, 2. Data validation job openings on the system. brand-related awards (App Award Korea, Good Brand lifelong career development services, and labor mar- 3. Accessibility and user experience • Public employment service (PES) support to small Trusted by Consumers, Top Public Service Award of ket research, analysis, and forecasting. Work-net em- 4. Targeted service delivery and medium enterprises (SMEs). SMEs with limited Korea, National Service Satisfaction Award of Korea, ployees are highly specialized staff, including engineers 5. Strategic marketing to users human resource management capabilities are often and Top 10 Public App Loved by the People). In 2018 (54 percent), master’s degree holders (15 percent), and 6. Interoperability with other systems unfamiliar with Work-net as a recruitment service. the number of registered users reached about 16 mil- PhD holders (14 percent). 7. Human resources serving the system and its users PES offices provide services to assist SMEs with lion (set against a labor force of just under 30 million). 8. Technical infrastructure registering vacancies on Work-net by e-mail, by fax, In 2017, 2.2 million jobs were placed through Work-net, Korea and Indonesia differ in several ways, which implies 9. Data management systems or by direct visits to PES offices. indicating significant turnover. that the approach followed to upgrade the Indonesian 10. Information security • Job postings collected from other platforms and LMIS should be tailored to its specific context. Table 11. Additional functions beyond job matching. private job portals. Since 2011, Work-net has in- The success of Work-net is due to its solid organization- 3.1 presents a set of select indicators for the two coun- creased the quantity and improved the quality of al structure. Work-net operates under KEIS, which has tries. Although gross domestic product (GDP) growth Given the differences between Korea and Indonesia, posted information on job vacancies by fostering 407 specialized employees who plan and manage KEIS rates have been greater during the past decades in the gap analysis should be seen as a systematic re- partnerships with other public and private job por- information strategies, operate the system, and collect Indonesia than in Korea (5.4 percent vs. 3 percent), view of areas that need to be considered to strength- tals. Currently, there are 172 aligned organizations: and analyze information on the labor market, among en the job-matching platform in Indonesia. The 25 public institutions and local governments (min- other duties. In addition, KEIS has about 100 out- 3 . Appendixes C and D present KEIS’s organizational structure and staff- areas highlighted here characterize other advanced istries, the Chamber of Commerce and Industry), 11 sourced employees who work on system development ing information related to Work-net. job-matching platforms—and LMISs more broadly— private job portals, and 136 local Work-nets. Since 26 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 27 Figure 3.1 Work-net Information Sharing Table 3.2 Data Collection Gap on Job Vacancies Desirable sources Work-net AyoKitaKerja Gap Direct postings from Yes Yes, but limited  Share information Job Seeker firms with private job Public Public Public employment Yes Yes, but service is not portals service support to small very active  Foster partnerships Institution/Local Employment with the private and medium enterprises Governments Portals sector and portals to Postings from other Yes, 172 aligned No increase the registra- Job Information Job Information platforms organizations, including tion of job vacancies public institutions, pri-  Provide a link to civil vate, and local Work-nets servant job postings  Provide a link to civil service human re- 36 Institutions 104 local 9 Portals sources management Local Worknet Govermments system and supporting, SMEs with registering job postings. Job-Seeker Data Collection Due to the limited number of PES offices and staff Companies in each office, however, support for SMEs is scarce. Work-net Source: Korea Employment Information Service. AyoKitaKerja does not collect job postings from other Work-net collects job seekers’ information primarily platforms. According to interviews with MoM staff and from the following sources: private job portal companies, no other central minis- tries manage a job portal. There are, however, more • Direct postings by job seekers. By far, most job seek- 2013, Korea has provided regional Work-net ser- AyoKitaKerja than 20 private job portal companies in Indonesia, but ers’ data come from individuals who post their pro- vices by building regional versions of Work-net for they do not share job-posting and job-seeker informa- files and use Work-net to apply for jobs. local governments. AyoKitaKerja collects information on job vacancies tion with AyoKitaKerja. Some private job portals (for • PES support for job seekers. Vulnerable job seekers through the following sources: example, Jobstreet.com, Querja, Robsjobs) provide and those who are unfamiliar with Work-net may Institutional arrangements allow Work-net to obtain in- input on the number of jobs they match successfully. need support for registration. PES offices assist formation on job vacancies, including partnerships with • Direct postings from firms. Less than 20 percent of Government jobs are not posted on AyoKitaKerja. members of these groups with registering profiles 172 organizations (figure 3.1) comprising national and posts come from direct postings. Most companies, and uploading job application documents to Work- public providers of job information, private providers, especially local ones, prefer to submit vacancies to Indonesia has a legal framework to support the collec- net by e-mail, by fax, or in person. regional (provincial) governments, basic local govern- PES offices, which post them on the system. tion of job-vacancy information. Discussion with MoM • Compulsory registration for participants in active and ments, and public employment centers under the MoEL. staff confirmed that, by law, all job vacancies should be passive labor market programs. KEIS provides many ac- In addition, Work-net gathers detailed information on • PES support to firms. Firms that visit PES offices to posted through AyoKitaKerja. The regulation provides tive and passive labor market programs, including un- firms and corporate reviews through the private sector submit vacancies are registered on the platform by a framework to encourage firms to register vacancies, employment benefits and vocational training. In order and other agencies. Because of these strong institution- PES officers. PES office staff also visit the private and providing relevant services to employers may fur- to claim unemployment benefits, an unemployed per- al arrangements, Work-net has a comprehensive source sector to identify and collect information on job va- ther motivate them to post openings voluntarily. See son is required to register at Work-net as a job seeker. of information on job vacancies. cancies. PES offices are charged with reaching out to, table 3.2. Similarly, an unemployed person wishing to participate 28 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 29 Table 3.3 Data Collection Gap on Job Seekers © onyengradar Desirable sources Work-net AyoKitaKerja Gap Direct registration Yes, covers all official Yes, but not validation  Provide a link to unemployed and reaches a small part future unemployment of the workforce insurance system Linked to active labor Yes, through compulsory No  Provide a link to market program (ALMP) registration ALMP participation and unemployment benefits Support for vulnerable Yes Yes, but limited staff at groups public employment service offices in vocational training or an active labor market pro- Data Validation gram (ALMP) also needs to register with Work-net. Consequently, all individuals who are registered as un- Job-matching platforms must provide updated, verified employed in Korea are registered on Work-net. employment and recruitment information, as this will re- duce search costs, improve job matching, and build trust in the platform’s effectiveness. The PES should verify Work-net’s validation system monitors and verifies in- The algorithm verifies the legal obligations of em- AyoKitaKerja whether the content of vacancies and profiles posted vi- formation through four key steps, using both artificial ployers such as minimum wage, work hours, dispatch olates laws and regulations and whether it has discrimi- and human intelligence. work, and industrial safety, among others. All data AyoKitaKerja collects job seekers’ information primarily natory elements. In addition, users need to trust that the that Work-net receives from other public employ- from the following sources: information is up-to-date, that vacancies have not been • First step: monitoring and verification by local PES of- ment portals or private job portals are automatically filled, and that job seekers have not found employment fices. The local PES office is responsible for verifying checked and verified by the computer algorithm pro- • Direct postings by job seekers. Individuals can reg- and are still interested in finding a job. With a large vol- the legal validity of job vacancies and job seekers’ in- gram. ister online, but those applying for jobs requiring a ume of information, an efficient means of weeding out formation. When PES officers find suspicious infor- • Third step: additional monitoring and verification by job-seeker card (AK1 card) need to present them- irrelevant information or updating information is need- mation as a result of the verification process, they experts at the KEIS call center. KEIS manages a call selves in person at their local PES office. ed, including technical tools. A user-friendly design and inform employers and job applicants of the issue and center, which checks Work-net’s job-vacancy infor- • PES support for job seekers. PES staff can help job interconnectivity with other systems can also reduce er- guide the user to correct the problem and register mation and corrects any violations and errors that seekers to register on the platform. rors related to user-generated information. again. PES staff review vacancies for authenticity as the first and second steps missed. well as discrimination (whether gender, age, resident • Fourth step: verifications through system linkages. The AyoKitaKerja’s services are not linked to ALMPs or to any registration number, religion, and nonrecruitment PES office periodically checks and updates the status other government support program. There is no unem- Work-net occupation are included in the job title and job de- of users. The connectivity with other systems is vital. ployment benefit in place.4 As described in chapter 2, the scription), working conditions (violation of minimum When a registered job seeker finds employment, he website caters to a small proportion of persons defined Work-net provides reliable job information authenticat- wage, working hours, or other labor rights prescribed or she is registered in the employment insurance da- as unemployed in the labor force survey. See table 3.3. ed by the government. Because of Work-net’s thorough by law), and goods or services that are prohibited by tabase, which is connected with the PES computer monitoring system, only 60 percent of all job vacancies law from production, distribution, and consumption. system. This connectivity allows PES offices to con- 4 . The draft JOBS omnibus law (sent to parliament on February 12, 2020) submitted are posted, which means that the govern- • Second step: monitoring and verification by computer. firm offline the new employment and recruitment proposes to introduce a job-loss insurance program that will have an ment guarantees high-quality jobs that fulfill legal re- Work-net also runs a special computer algorithm to and to update the person’s Work-net status from job unemployment cash transfer component. quirements. check systematically the reliability of the information. seeker to employee. 30 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 31 Table 3.4 Data Validation Gap Figure 3.2 Responsive Web Design of AyoKitaKerja Desirable characteristics Work-net AyoKitaKerja Gap PC PC TABLET TABLET PHONE PHONE Data verification Four-step information  Provide systematic Basic and infrequent monitoring and scalable de- Sources of verification Human intelligence, com- Manual and in person vice for monitoring puter algorithms, and in- through public employ- and verification of teroperable government ment service (PES) offic- information databases es, accounts deactivated  Enable e-mail and if not used within two mobile-based verifica- years tion at registration  Ensure that job-va- cancy registration is not discriminatory Source: Ministry of Manpower. AyoKitaKerja include characteristics unrelated to an individual’s ability Accessibility of Services and Information Service, which targets young job seekers. Additional fea- to perform a job satisfactorily. For example, it is com- tures, such as the Mobile Employment Compass Service, AyoKitaKerja lacks a systematic quality control process to pulsory to fill out information on gender, marital status, Work-net use big-data analysis to provide job seekers with person- support the desired increase in the volume of applicants age, and disability, information that would be grounds alized employment information. and job vacancies. Staff in regional MoM offices verify job for discrimination. See table 3.4. Work-net services are fully accessible on personal com- vacancies and company profiles manually in the system. puters, smartphones, and tablets. The mobile job ap- In order to validate job seekers’ profiles, persons who reg- plication service permits job seekers to handle the full AyoKitaKerja ister online still must visit their local PES office in person. job application on a mobile device. Job seekers can Although this reduces the risk of invalid information, it is download job application forms onto their computer Mobile services can be a very efficient way of sharing in- not consistent with attracting a high volume of applicants Accessibility and User Experience or smartphone and can upload application forms or formation in a large archipelagic country with poor ICT or vacancies and automating counseling services. Once résumés onto the system. Employers can similarly up- infrastructure in some areas, such as Indonesia. AyoKitaK- job seekers register their job application in AyoKitaKerja, A well-functioning LMIS provides users with easy, contin- load the job-posting information by computer or smart- erja users find slow internet speed to be a critical problem. they are considered active for six months. Accounts that uous access to all available functions. The system needs to phone. Users who face difficulties registering or access- AyoKitaKerja has a web-based responsive design in which have had no activity (no log-in) for six months are asked allow different users to access its services reliably through ing information can request assistance by printing out the contents are rearranged according to the screen size to register again by e-mail; otherwise they are transferred various channels suited to their individual needs. As such, the relevant forms in the system and submitting them of the browser, so the content can be viewed on devices to a pool labeled as inactive. There is no contact verifica- a user-friendly, attractive design with well-organized con- to the PES office by fax, by post, or in person. Employers such as smartphones, tablets, and computers (figure 3.2). tion at registration, either automatically—for example, by tent, screen interface, and tools for sharing information is and job seekers needing assistance retrieving relevant However, a collapsed view in a small screen seems im- e-mail or SMS (short message service) requiring verifica- extremely important. User-friendly services support the information can receive assistance at the PES office. practical. A mobile web version also lacks the advantage tion of account—or manually by PES staff. The majority following functions: (a) customized support, (b) a “one- of mobile apps such as a push service5 and easy access. of enterprises posting job-vacancy information in AyoK- online-stop shop” for vacancies and labor market infor- Work-net’s mobile services have many user-friendly fea- itaKerja are SMEs, and no strict quality control system is mation and services that are comprehensive and easy to tures, including a “search within results” function based At the same time, a multiprong approach to digital tools is need- in place on the enterprise side, as updates and checks are access and understand, and (c) a customer response sys- on job type or region, location-based services, and active ed, taking into account the characteristics of AyoKitaKerja’s infrequent. tem that can deal with the problems of specific users and push notifications (alerts for new applicants, questions 5 . A push notification is a message that pops up on a mobile device. App publish- draw attention to systemic problems. for hiring companies, customized recruiting information, ers can send them at any time; users do not have to be in the app or using their Moreover, in contrast to Work-net, some of the com- and job placements for job seekers). They also provide device to receive them. Push notifications look like SMS text messages and mobile pulsory fields in the registration process for vacancies customized services such as the Mobile Work-net Youth alerts, but they only reach users who have installed the relevant app. 32 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 33 Table 3.5 Accessibility of Services and Information Gap Table 3.6 User Experience Gap Desirable characteristics Work-net AyoKitaKerja Gap Desirable characteristics Work-net AyoKitaKerja Gap Interactive through dif- Yes, accessible through Limited, accessible  Develop a light version Screen design, composi- Yes Not systematically used  Develop interface ferent devices computer, tablet, mobile through computer, tablet, of portal for slow-speed tion, and interface based based on UX web, and mobile app (iOS/ mobile web-version, but connections on User Experience/User  Create a feedback Android). Anytime, any- design needs to be im-  Introduce mobile app Interface (UX/UI) system (UX survey, where, anyplace mobile proved. No mobile app service Unified log-in Yes, a single identification No, log-in information is client response service  Improve the design of and password used for specific to the platform services) Interconnected system Yes, prefills information (for No mobile web and tablet different subsystems  Coordinate system example, job applications) access log-in as other sys-  Provide personalized Feedback system Yes No tems are developed Assistance services Yes, at registration and Yes, at registration at PES information and notifi- when retrieving relevant in- office, but PES staff is cation by text message formation at public employ- limited  Have an interconnected ment service (PES) offices system suggest that the platform is slow in responding. Dis- opinion (Voice Of Customer, VOC) and integrated man- cussions with users of AyoKitaKerja confirm that they agement; (c) a customer support call center; and (d) have difficulty logging in, registering, and searching for chatbot system content. information. A large number of compulsory fields and main clients on the job-seeker side. The majority of job seek- Team and the User Experience Team of KEIS continuously slow speed are among the key weaknesses. The log-in/ Work-net’s chatbot service provides a 24-hour consulta- ers are young, which augurs in favor of IT literacy. Whereas analyze Work-net use and manage a user survey on UX/ password system is unique to AyoKitaKerja and not co- tion service for customers with inconveniences as they cellular penetration is high (70 percent of the population UI (User Experience/User Interface). Results from these ordinated with other public systems in Indonesia, with arise. The local PES office responds to customer incon- uses a mobile phone), access to the internet is limited (40 analyses are applied to upgrade the system. one exception. AyoKitaKerja is linked to population data veniences by telephone or e-mail and PES offices receive percent), and only 24 percent of the population own a smart- of the Ministry of Home Affairs via the citizen identifica- regular Work-net training. KEIS operates a call center to phone.6 As such, concentrating on app technology alone or The Social Network Services identification as well as a tion number (NIK).7 This allows AyoKitaKerja to retrieve respond to customers’ complaints by phone or e-mail. If depending on e-mail as the system’s main communication single identification and password system are used to individual data automatically from the population data- local PES offices and call center consultants encounter channel (as is currently the case) are not suitable approach- log onto Work-net, HRD-net, the Employment Insurance base. However, the information that is automatically en- difficulties, they contact the KEIS Work-net team, and es given the capacity and profile of AyoKitaKerja’s end users. Network, and the Employment Permit Service (EPS) tered is limited. See table 3.6. KEIS experts will resolve them. AyoKitaKerja could consider developing SMS for communi- network. cating some of the key basic information and services pro- Work-net staff regularly make onsite visits to clients vided, for example, job notifications. See table 3.5. By linking up with other public information systems, such Customer Response Service and users, such as PES offices, universities, businesses, as the employment insurance database, Work-net can municipal employment centers, and private employ- prefill part of an applicant’s work history (job title, tenure) Work-net ment service organizations. They also meet monthly User Experience in job application forms, saving time and reducing oppor- with primary client agencies. Representatives of specif- tunities for errors in user-generated information. For Work-net, KEIS has a customer consulting team that ic target groups are recruited to panels that investigate Work-net incorporates recommendations from users and corrects group-specific issues. and eliminates issues related to customer complaints. Work-net continues to enhance the design and conve- AyoKitaKerja The team is responsible for managing and operating nience of the website. The Employment Service Promotion (a) an information system for customer feedback and AyoKitaKerja AyoKitaKerja does not have a systematic UX develop- user education; (b) an information system for customer ment system for enhancing the configuration and de- PES officers are trained on AyoKitaKerja and provide 6 . International Telecommunications Union World Telecommunication/ ICT Indicators database (cellular and internet) and https://www.statista. sign of the platform based on the needs of users. Test 7 . NIK (Nomor Induk Kependudukan) is the unique citizen identification num- registration services for AyoKitaKerja. PES offices pro- com/ (smartphone usage). runs (not logged in) of the website by the report team ber provided to every citizen and registered resident in Indonesia. vide useful feedback on AyoKitaKerja, although not in 34 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 35 © Yermia Riezky Santiago Table 3.7 Customer Response Service Gap Desirable characteristics Work-net AyoKitaKerja Gap 24-hour call center and Yes, chatbot service No  Develop response client support and Korea Employment system to resolve Information Service user inconveniences (KEIS) call center  Develop feedback Customer consulting Yes, job consultants No system to VOC, tar- team at public employment get specific groups service (PES) offices respond by e-mail, by telephone, or in person Customer complaint res- Yes, KEIS experts and No olution mechanisms customer consulting team listen to Voice Of Customer (VOC) and correct and eliminate customer complaints a systematic manner. However, AyoKitaKerja does not that matches these criteria. For recruiting companies, Baro-One, the system is linked to other data systems elderly Work-net service was established to provide re- have a systematic response system to resolve user in- Work-net similarly recommends suitable job seekers such as HRD-net and others. Employment Compass also employment, start up, vocational training, and welfare conveniences or receive general feedback on how to im- and provides updated information. offers a personalized service called The Work, which information to middle-age and elderly persons. prove the system. See table 3.7 provides content and job information favored by similar In addition to the systematic job and candidate notifica- groups derived through big-data analysis. Introducing AI in a job-matching platform reduces the tions, Work-net also helps job counselors in PES offices time and cost to help job seekers find a list of available to provide customized services tailored to individual cir- Work-net also provides differentiated services accord- jobs matching their relevant work experience, skills, cumstances and needs, through a service called Baro-One. ing to the characteristics of the users and provides roles, and educational background. From 2016 to 2018, Targeted Service Delivery Baro-One applies big-data analysis of information on job specialized recruitment information, job guides, and Work-net was reorganized into a national platform-ori- training, employment insurance, qualifications, and more, job-related news for each target customer. Youth Work- ented employment service hub, and a chatbot service Defining job-search or job-candidate conditions can be using Work-net, HRD-net, the Employment Insurance Net- net provides information on employment that may in- was introduced in 2019. Work-net is currently develop- complicated, which slows down the search process. work, and the Korea Workers’ Compensation and Welfare terest young people. For example, it selects job infor- ing a system based on the job data dictionary that uses Service. Baro-One permits PES counselors to provide tai- mation suitable for young people according to certain AI to match job seekers’ competence with job informa- lored advice on jobs, companies, and career paths. criteria. Likewise, Work-net provides job information tion and job requirements. The service is expected to Work-net tailored to women such as programs on returning to the launch in 2020. The Employment Compass service is similar in design labor market for women whose career was suspended Work-net addresses this problem by systematically rec- and objective to Baro-One, but job seekers can use it for personal reasons. It also offers programs related to ommending jobs that are closest to the search condi- directly. When a job seeker provides key input (gender, personalized services and job-training programs by in- AyoKitaKerja tions when a job-seeker creates his/her desirable job age, academic background, and desired industry or ca- tegrating various employment-related services. Services profile. If no job satisfies such conditions at the present reer track), the Employment Compass provides wage targeting the elderly also provide job information, pro- AyoKitaKerja still has to integrate advanced-technolo- time, Work-net notifies the person by smartphone or information, certificate information, and employment grams for transfer and life-career support, as well as gy tools such as AI and big-data analysis to provide tai- the Internet whenever a new job posting is registered cases according to the job-seeker’s situation. Similar to offline organizations to support their employment. The lored job/job-seeker recommendations in a systematic 36 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 37 Table 3.8 Targeted Service Delivery Gap Table 3.9 Strategic Marketing Gap Desirable characteristics Work-net AyoKitaKerja Gap Desirable characteristics Work-net AyoKitaKerja Gap Tailored job Yes, based on job No  Develop a tailored Marketing team Yes, has a team dedicat- Limited to public em-  Step up marketing recommendations seekers’ profile service system ed to marketing activi- ployment service (PES) activities to the gen- Tailored services Yes, based on job No ties and conducts active offices, ministerial eral public and target  Introduce AI seeker and job-post- marketing activities websites, educational groups job-matching service ing information through multiple chan- institutions, and job fairs  Offer user trainings nels (including public Tailored information Yes, based on job No ads, educational institu- seekers’ personal tions, research dissemi- characteristics and nation) and to circumstances as well as different target groups information that other job seekers with similar User training Yes, through PES officers No backgrounds were interested in or used Tailored services to Yes, for youth, women, No vulnerable population and the elderly • Dissemination of research focusing on high-quality, public. Interviews with private job-matching compa- Other Expected to launch No relevant analysis and research raising public aware- nies, job seekers, and employers confirm that potential artificial intelligence (AI) ness of KEIS and Work-net through media reports stakeholders are not familiar with AyoKitaKerja. In con- job-matching service in • PES officer user training. trast, potential AyoKitaKerja users are well aware of the 2020 different private job portals that market their activities broadly. As an example, Jobstreet, a leading company in AyoKitaKerja this industry, hosts 60 job fairs annually for both em- ployers and job seekers. These job portals also partner manner. It does not provide automatic job matching • Active advertising in public places such as railway sta- AyoKitaKerja is marketed in the following manner: with top universities, and most regional office staff fo- through the system (as would a PES counselor offline) tions and bus terminals, active promotion of Work- cus on the marketing and sale of services. MoM has no and is limited to helping companies and job seekers nar- net on the online portal websites that many people • In government offices and centers, above all local active sales or marketing activities for the AyoKitaKerja row their search results by way of keywords and filters. use, and a newsletter circulated through e-mail and PES centers but also in social security agency offices system. See table 3.9. See table 3.8. SMS to Work-net members, detailing the availability • With other partners, in particular, marketing activ- of both basic and new services ities through university career centers, and busi- • Target group strategy, with recruitment and selec- ness associations tion of supporters for strategic promotion activities • Advertisements placed on the MoM webpage, in Strategic Marketing to Users and specific target groups (youth Work-net sup- public ministries, and on the Ministry of Social Se- Interoperability with Other porters) curity website Systems Work-net • Promotion through events, online and offline, to in- • Local job fairs conducted at the local and national form users whenever new content is launched on level (local universities, national job fairs) under The national job-matching platform can reach more KEIS applies intensive strategic marketing and research Work-net the supervision of the Directorate of Domestic users when it is linked with other systems, such as job activities for Work-net services to alert potential users. • Campaigns to increase Work-net membership and Labor Market Development information, employment insurance information, and The payoff to doing so is significant: a marketing initia- provide customized information • PES officer user training. vocational training system networks. Interoperability tive launched in 2014 rendered a doubling in daily visits • Partnerships through memorandums of under- with other systems can also facilitate the verification of in 2015 compared with 2013. The following are major standing with various organizations and universities As these activities have limited and selective outreach, job-posting and job-seeker information. strategic marketing activities of KEIS: to expand the use of Work-net AyoKitaKerja is not widely known among the general 38 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 39 Figure 3.3 Interoperability of Work-net with Other Systems Table 3.10 Interoperability with Other Systems Gap HR D - N et Em pl oyem en t Desirable characteristics Work-net AyoKitaKerja Gap I n su ra n ce System linked to other Yes, interlinked with No, few systems exist  Develop linkages subsystems vocational training, em- for linking the platform with other systems ployment insurance, and to to enhance rele- migrant worker permit vance and validity of However, there are systems data plans to link it to social Training information Job vacancy insurance system Job vacancy Work-Net (WLKP). Training institutions could access AyoKitaKer- (employing some 154 researchers as of July 2019). Re- Employment ja to promote their training services to job seekers and searchers use the data in the overall LMIS to conduct Job vacancy employers and be linked to the Indonesia National Work research and provide labor market intelligence; the information Competency Standards (SKKNI) platform. At the same results of this research are transformed into various time, AyoKitaKerja could partner with a broader set of types of content and made accessible to the public job boards and big employers to post vacancies on the through Work-net. platform as well as share information with other agencies, E m pl oyem en t Em pl oyem en t such as MoM and the Central Bureau of Statistics (BPS) For continuous improvement and development of the Pe r m i t S y stem i n f orm a t i on System for labor market intelligence. See table 3.10. LMIS content and user friendliness, KEIS employs more than 170 staff to operate the system. At the call center, 46 employees assist with customer consultation and with Source: Korea Employment Information Service. monitoring and verification of Work-net information. Human Resources Serving the Work-net limited benefits with respect to labor market informa- System and Its Users AyoKitaKerja tion and services. Work-net was reorganized as a full-scale service in 2006 Sufficient human resources, with adequate skills, are AyoKitaKerja is used by about half of PES offices, but by by connecting the platform with vocational training AyoKitaKerja also could be interconnected with several necessary to ensure the active and effective use of the very few staff in each office. Regional autonomy implies information and the EPS. The linkages currently in place additional government initiatives. For example, job seek- information produced on the job-matching platform that local offices (PES/provincial offices) have their own are shown in figure 3.3. ers using AyoKitaKerja could gain access to education and and to improve services continuously. regulations and that, while they provide information to training platforms with either information on opportuni- MoM, the ministry does not have the power to regulate ties or e-learning platforms for capacity building and job them directly. AyoKitaKerja training. They also could be linked to platforms with infor- Work-net mation on government support for training (for example, There is no research infrastructure or significant pool AyoKitaKerja is not linked to other employment, social scholarships) managed by the Ministry of Education and For PES service delivery, Work-net is used by local job of staff for developing and operating the system. In protection, or human resource systems, partly because Culture. Civil servants or those willing to become civil ser- centers (232), by nearly 1,800 private agencies, by job MoM headquarters, only four staff are in charge of op- such systems are so rare in Indonesia. There are plans vants could be linked directly to the Civil Servant Selection centers on college campuses, and by job centers for var- erating AyoKitaKerja. See table 3.11. to develop linkages with the social insurance agency to System, avoiding duplication of profiles and information. ious target groups. verify employment status. AyoKitaKerja is linked with Employers using the platform could be linked directly to the population register at the Ministry of Home Affairs MoM subsidiary platforms such as the Foreign Workers For labor market research, KEIS draws on a significant through the NIK, although registration provides only System (TKA) and Mandatory Employment Reporting pool of researchers to provide high-quality analysis 40 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 41 Table 3.11 Human Resources Gap Figure 3.4 Software Architecture of AyoKitaKerja Desirable characteristics Work-net AyoKitaKerja Gap SERVER DATABASE System used by skilled Yes, widely used at differ- Limited, only used by half  Extend use of AyoKi- counselors ent levels of local public of PES regional offices taKerja in PES offices OPERATION SYSTEM LINUX SERVER employment service with only a few counsel- Enhance services to POSTGRESOL DATABASE (PES) offices as well as ors in charge of deliver-  specialized agencies and entice private offline ing employment services PgAdmin 10 private agencies agencies to use AyoK- itaKerja Research team Yes, Korea Employment No, limited capacity Information Service  Consider how to de- (KEIS) undertakes velop human resourc- research on key labor es strategically to SERVER APPLICATION 1 SERVER APPLICATION 2 market indicators (na- build up AyoKitaKerja OPERATION SYSTEM CENTOS 7 OPERATION SYSTEM CENTOS 7 tional workforce and and provide input to employment demand LMIS APACHE WEB SERVER APACHE WEB SERVER projections) Ayokitakerja.kemnaker.go.id Webservuces of Ayokitakerja Labor market infor- Yes, KEIS is an indepen- No, very limited staff and mation system (LMIS) dent central employment capacity management team information manage- a. PHP 7 ment center to plan, b. Codelgniter 3.1.5 implement, and manage LMIS Source: Ministry of Manpower. Technical Infrastructure system, the modified program is tested in a system sim- AyoKitaKerja Finally, AyoKitaKerja is not included in MoM’s national di- ilar to the operating system. The redundant server is in- saster recovery center (figure 3.5), located an hour from Work-net tended to enhance test quality and minimize operational The AyoKitaKerja system is not equipped with server re- Jakarta. In the event of a disaster, the AyoKitaKerja ser- errors due to program modifications and revamps. dundancy safeguards in the case of service failure. Each vice may be discontinued. AyoKitaKerja does not have a KEIS has a business continuity plan for Work-net to en- system node has to be monitored separately (server, disaster recovery manual to guide disaster recovery and sure continuous service 24 hours a day, 365 days a year. Through the traffic and server resource monitoring sys- network, application). The system consists of one web response in the event of a disaster. MoM does not per- Work-net also runs a remote disaster recovery system tem, Work-net regulates usage and logs, such as net- server for web services, one application programming form periodic disaster recovery training according to a for system monitoring and backup that addresses vari- work traffic, disk usage, central processing unit/memory, interface (API) server for data provision, and one da- disaster recovery manual. See table 3.12. ous types of disasters. database management system status, and application tabase (figure 3.4). However, the AyoKitaKerja system errors. The Work-net data system is prepared to per- only monitors traffic on the network, making it difficult The Work-net system consists of a Java-based three-tier form daily and weekly backups to prevent data loss and to respond promptly to service failures. For immediate (Web-Was-DB) system, and its operating system is distrib- restore lost data occurring from, for example, failures or response, a monitoring system is needed for each sys- uted redundantly. In case of failure, downtime, or exces- neglect in storage, transmission, or processing. A sep- tem node, and it must be monitored regularly. There Data Management Systems sive traffic at the primary server, a redundant server can arate Hadoop system-based platform handles big-data is no window for system users to alert AyoKitaKerja take the primary server’s place and continue to provide analysis. The KEIS headquarters data center and the re- of difficulties or to receive assistance in the event of a Work-net reliable services. In addition, the development system, test mote disaster recovery center are located separately, problem. The system is backed up whenever changes system, and operating system are managed separately. As and the disaster recovery plan is established periodically are made to it. However, the data are backed up on a To improve the use and openness of data, KEIS orga- developers extend the program through the development and internalized through training. general MoM server. nizes data quality management committees at the 42 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 43 Table 3.12 Technical Infrastructure Gap Table 3.13 Data Management Gap Desirable characteristics Work-net AyoKitaKerja Gap Desirable characteristics Work-net AyoKitaKerja Gap Round-the-clock availabil- Yes, with a replication Available, but services  Build a system redun- Data quality manage- Yes, established systems No  Establish a system for ity of services system that does not are slow dancy, recovery, and ment systems in place to manage data data quality manage- disrupt operations backup system for quality and obtain quality ment and for policy deci- stable operation of data from job-posting and sion support Backup system Yes, daily and weekly Yes, but on the Ministry the system job-seeker information  Improve data quality of Manpower’s (MoM)  Use a cloud system control guidelines and general server Data quality team Yes No to expand the system tools Disaster recovery system Yes No and use resources ef- Data quality standards Yes, established and No  Code input data in a ficiently in the future enforced standardized format Cloud system for Yes No resource utilization  Institute integrated Data quality checks Yes No with menus control of server  Provide a system user Integrated monitoring Yes, monitoring over No, only monitors traffic on resources and ap- Big-data platform Yes No manual and a description system network, server, and the network using the Net- plications for active of each information item application work Management System response to a failure  Encrypt personal infor- Big-data platform for Yes No mation and create user analysis privacy policies Figure 3.5 Hardware Architecture of MoM’s Data Center Kemnaker organizational level, establishes enterprise data stan- AyoKitaKerja dards and quality control plans, and monitors and re- fines data quality through a data quality management The AyoKitaKerja system has no quality control guide- INTERNET METRO E system. It enforces database standards, considers lines for the data it collects or any data correction Disaster Recovery Center them to be essential for database design, and periodi- devices for evaluating data quality and providing feed- Load Balancer Appliance NG FIREWALL cally evaluates data standards, refinement, and quality. back. Currently, input data (for example, skills) are free- KEIS is continuing to expand Work-net’s data open- text based, which increases the chances of errors and 10G 10G ness. Currently, 16 types of Work-net data are open, the completion of profiles and complicates analysis for including job-posting information. job matching. Personal information is not encrypted. See table 3.13. Nexus SW01 Nexus SW02 MAG MAG A master data management system enables customers 10G 10G 10G (firms and individuals) of Work-net, the Employment Virtual Machines Virtual Machines Virtual Machines Insurance Network, HRD-net, and EPS to integrate in- PC formation based on key personal identification infor- mation (resident registration number, business regis- Information Security ESX Server ESX Server ESX Server tration number). The integrated database is used for Web Server Infokerja Application Server Infokerja Database Server Infokerja Storage personalized employment services as verification of Cyberattacks remain a significant risk, posing intelli- eligibility to receive certain benefits such as the Em- gent, complex, and diverse threats to operations. Ad- ployment Success Package at the Employment Welfare ministrative, technical, and physical means for ensuring Physical Server Plus Center; it is also used to provide in-depth counsel- the security of information are required to protect in- ing for job seekers. formation from cyberattacks and to ensure business Source: Ministry of Manpower (MoM). continuity (figure 3.6). 44 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 45 Figure 3.6 Information Security Means Table 3.14 Information Security Gap Administrative • Guideline Desirable characteristics Work-net AyoKitaKerja Gap Security • Manual System in place to Yes, strong system at Limited to firewall  Enhance awareness, • Training protect information the national level and education, and train- Physical • Building / Office internal ing for officers’ and Security access control employees’ security Security measures in Yes, administrative, No Technical • Firewall place physical, and technical as a whole Security • IDS / IPS  Build a security solu- • VPN Information security Yes No tion for each node • Encryption training  Establish user pri- Information (Asset) vacy policies as the system develops Work-net the outside. Also, a security check of developed pro- labor market information and analysis. These addi- In 2003 the Korea Network for Occupations and Work- gram sources is essential, and security activities are per- tional functions can be provided by the job-matching ers (KNOW) was established and linked to Work-net. Work-net has measures for three levels of security: ad- formed from the program source. portal itself or by other subsystems in Indonesia’s LMIS KNOW provides information in various areas, including ministrative, physical, and technical. that are integrated in the portal. 8 jobs, college departments, job search, career counsel- ing, and others to help Work-net to enhance the job Administrative security requires the establishment of AyoKitaKerja search and career design of youth and job seekers. security guidelines for carrying out information securi- Lifelong career guidance and employment support for ty activities. KEIS conducts regular information security AyoKitaKerja security consists of firewalls to prevent Career and Skills Guidance both young and older adults have been integrated into training and employee training, monthly security checks, intrusion on the network. There is no solution for the Work-net services. For example, Work-net provides 23 and management and inspection of outsourced staff. physical security of data center access, program source Work-net vocational psychological tests for adolescents (10 test Legal, institutional, and regulatory rules on the laws re- security, technical security for intrusion detection, and programs) and adults (13 test programs) to help to lated to information protection, personal information data exposure. In addition, there is no general guidance Korea’s Work-net provides job information using infor- match an individual’s personal preferences and aptitude and user protection, internal information security reg- or control regarding security. Information security train- mation from KEIS, which publishes various types of oc- with jobs and training. ulation, and encryption and deidentification of personal ing is lacking. See table 3.14. cupational information, including occupational outlooks information are established. and occupational dictionaries. The annual survey of the KEIS operates a separate job training site called HRD- employed provides information on 800 or so jobs, in- net to provide vocational training courses, qualification Physical security protects facilities through the authority cluding current information on working conditions, ca- tests, and key statistics related to vocational training, management and access control system for controlling pabilities, education, training, and qualifications. which are provided in conjunction with job information building access. Major protection facilities have strong Additional Functions beyond on Work-net. This information includes the content of authentication systems such as fingerprint and vein au- Job Matching training courses, training objectives, application require- thentication. ments, related qualifications, training period, training 8 . As explained in note 13 in chapter 1, MoM launched a new online plat- As discussed in chapters 1 and 2, an advanced LMIS has form, Sisnaker, on September 29, 2019. This platform is more comprehen- costs, government grants, training personnel, training Technic al security performs program operations and functions in place to provide information and services sive than AyoKitaKerja and includes information on the supply of training facilities, and employment outcomes after training. Peo- inspections from program source to network securi- in several areas, not just basic labor market informa- by BLKs (which is part of the career guidance function) and entry points ple can search vocational training information at Work- to other government support programs. While this is a move in the right ty, server security, and user (end point). In particular, it tion or job matching, as in the case of AyoKitaKerja. net and apply for support for their training by computer direction, Sisnaker still does not integrate information from other line separates the internal network from the external net- Additional functions beyond job matching include ministries (like the Ministry of Education and Culture) and would benefit or smartphone. The systems and data of Work-net and work to prevent hacking and information leakage from career and skills guidance, government support, and from further developments based on recommendations in this report. HRD-net are connected and interoperable. 46 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 47 Table 3.15 Career and Skills Guidance Gap Box 3.1 Indonesia’s Skills Monitoring System Desirable information Work-net AyoKitaKerja Gap Initiatives, supported by the World Bank, are un- and constraints, current and expected vacancies, Vocational and career Yes, information for Basic job search guid-  Develop occupation- der way using complementary labor market in- and planned occupational growth and declines. information career exploration ance al and career guid- formation techniques—traditional and nontradi- The questionnaire was designed based on two (reports and videos), ance information tional—to provide a thorough understanding of well-established surveys carried out by the US vocational psychological and link it to other Indonesia’s supply of and demand for skills: Department of Labor: the Occupational Employ- test for youth and adults sources of informa- ment Statistics and the Jobs Openings and Labor tion Indonesia’s Critical Occupation List (COL). This Turnover Survey. Detailed occupational Yes No  Develop a vocational outlooks initiative was implemented in coordination with training information the Coordinating Ministry for Economic Affairs. INDO*NET. This initiative is at the design stage. It Vocational training infor- Yes, links to HRD-net Very basic. Indonesia system and link it mation containing the content, plans to introduce a The 2018 COL draws on international best practice consists of a pilot survey of selected modules of with AyoKitaKerja objectives, and informa- preemployment voca- from Australia, Malaysia, and the United Kingdom the US O*NET survey adapted to Indonesia’s con-  Introduce e-card tion on training courses, tional training voucher application and to develop a list of shortage occupations that can text. The pilot will collect information on tasks online application and system (e-card system) management system be used to inform labor market policies and pro- performed and skills needed in 15 occupations. information on require- and manage the e-card at AyoKitaKerja grams. To be included on the list, an occupation Selected occupations will be based on COL and ments, related qualifi- through an app that can  Link AyoKitaKerja to must meet two criteria: (a) it must be in shortage OEVS work. cations, government be linked to the AyoKita- the Critical Occupa- and (b) it must be strategic for the Indonesian grants, and employment Kerja system tion List (COL) economy. The 2018 COL contains 35 occupations, Online Skills and Vacancy Outlook. This initiative rate after training which represent jobs from sectors such as man- is in the data collection stage and is based on ufacturing, telecommunications and IT, accom- real-time labor market information techniques. modation and food services, construction, ICT, The initiative collects online data on vacancies and other professional scientific services. These and skills demanded at the occupational level AyoKitaKerja Government Support occupations are derived through both top-down available on up to 25 job boards and 100 employ- analysis (national survey data indicating changes er sites in Indonesia. The data collection started AyoKitaKerja provides a basic career guidance ser- Employment and welfare policies are closely linked in in supply and demand) and bottom-up analysis in November 2019 and will continue until Decem- vice menu (Layanan Bimbingan Karir). As discussed in Korea, and PES offices have been operating under the (evidence collected from enterprise surveys and ber 2020. Data collected for each vacancy will chapter 2, the services provided are limited and lack name of the Employment and Welfare Plus Center since consultative processes). Internationally, COLs include job title, location, classification into KBJI the inputs and technology to provide relevant, updated 2014. have been used to create targeted education and 2014 and O*NET scheme, required education, information. Existing information systems on vocation- migration policies that address critical skills gaps. skills, and experience, as well as the employer’s al or educational training institutions are not linked to name and sector. This survey was designed to AyoKitaKerja. The government is planning to introduce Work-net Occupational Employment Outlook. This initia- complement, not replace, other initiatives, since a preemployment vocational training voucher system tive is currently under way. It analyzes the pilot more highly skilled, formal jobs are more likely to through an e-card (Kartu Pra Kerja). This could provide Work-net also provides information that enables job of the Occupational Employment and Vacancy be found online. an impetus to strengthening career guidance services seekers to search for employment and welfare policies. Survey (OEVS) carried out between December overall and linkages with other services. In addition, the Through Work-net, 15 employment policies and 558 wel- 2019 and March 2020. The OEVS is a firm-level All of these initiatives can serve as a platform for launch of the 2018 Indonesian Critical Occupations List fare policies can be reviewed. A simulation service en- survey that collects information on the struc- monitoring skills imbalances and helping policy (COL) and other initiatives represents a positive step to- ables users to identify e-services appropriate to them. ture of employment, the roster of occupations in makers in Indonesia to determine where invest- ward creating infrastructure that can provide relevant By entering personal characteristics and job status, as firms using 2014 Indonesian Standard Classifica- ments should be made in training programs, how stakeholders with information on the skills in demand well as basic information on income and wealth, users tion of Occupations (Klasifikasi Baku Jenis Peker- incentives should be adjusted for apprenticeship (box 3.1). See table 3.15. can identify whether they qualify for different benefits jaan Indonesia [KBJPI] 2014), and occupational programs, and which skills job seekers should try to or supportive policies such as retirement payments, pa- dynamics, including shortages, hiring attempts develop to increase their value in the labor market. rental allowances, unemployment benefits, basic living 48 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 49 Table 3.16 Government Support Gap Table 3.17 Labor Market Information and Analysis Gap Desirable services Work-net AyoKitaKerja Gap Desirable services Work-net AyoKitaKerja Gap Information on govern- Yes, links to employment No  Map employment and Data used for providing Yes No  Integrate data sys- ment support and welfare programs welfare program in- labor market information tems as the quality Application to programs Yes No formation and make and analysis and quantity of data available online improve  Enable online application Data used for designing Yes, including for specif- No  Establish strong labor market policies ic target groups monitoring and Real-time analysis Yes, information tech- No evaluation system nology (IT) system ana- for public employ- pensions, and single-parent family allowance. Work-net national statistics, the integrated ne tworks database lyzes data collected on ment services and also offers information on welfare policies provided by (Work-net, Employment Insurance Information Net- the labor market infor- labor market policy mation system (LMIS) in broadly local governments. In order to provide this information, work, HRD-net), and KEIS-administered surveys in areas real time  Use integrated data Work-net partners with ministries and public agencies related to employment and skills. The Work-net data, for policy evaluation such as the Korea Workers’ Compensation and Welfare specifically, provide input to analysis of, for example, Sharing of labor market Yes, from Work-net to No and for dissemina- Service and the Information Openness Unit of the Min- labor market imbalances, emerging skills gaps, and job- information 1,000+ institutions* tion to other actors istry of the Interior and Safety. search activity.  Share information through open appli- The Employment Information Integration Analysis Sys- cation programming AyoKitaKerja tem is an IT system developed for the purpose of an- interface (API) alyzing data collected in the national employment in- * As of December 31, 2018. AyoKitaKerja does not currently provide information on formation networks in real time. Aside from Work-net employment policies and labor market programs. There data, the system includes data from the Employment is no separate online portal providing information on Insurance Information Network and HRD-net, among employment policies and labor market programs. There others. example, while BPS produces labor market data (as services and for disseminating labor market informa- are no links to welfare policies or services. AyoKitaKerja shown in table 2.2), it does not present information tion and analysis more broadly. See table 3.17. is planning to link up with the social welfare agency to Work-net also allows a wide variety of users to access from AyoKitaKerja on its website or vice versa. Addi- coordinate information. See table 3.16. its information through open APIs. As of December 31, tionally, AyoKitaKerja currently serves too few users to 2018, Work-net provides information on recruitment, provide a representative view of the Indonesian labor career tracks, labor market outlook, job fairs, and other market and, as such, is not yet suited to provide input to Labor Market Information and Analysis relevant employment information to 1,036 institutions, policy making. Building the capacity to identify sources including 312 schools, 327 public institutions, and 397 of data, integrate them, and use them for analysis and Work-net private institutions. evaluation of labor markets and labor market policies will be a major step toward creating a more advanced KEIS uses the information in the LMIS for evaluation, LMIS. Box 2.1 in chapter 2 provides a detailed descrip- analysis, and forecasting, with the objective of helping to AyoKitaKerja tion of data sources for an advanced LMIS. design, monitor, and evaluate public policy at the nation- al, regional, and local levels. KEIS also provides high-fre- As part of a basic-to-intermediary LMIS, AyoKitaKerja AyoKitaKerja does not provide information through quency employment data on its webpage (employment does not have the technological tools, infrastructure, open API. As the volume of data increases and becomes trends, labor supply and demand, vocational training) or platforms needed to integrate AyoKitaKerja data more relevant (by virtue of volume and validation pro- for other users such as active labor market participants, with data from other information systems or to ana- cedures), sharing information with other actors would firms, and researchers. Sources include, among others, lyze, evaluate, and disseminate this information. For provide a vehicle for expanding use of the platform’s 50 Toward a World-Class Labor Market Information System for Indonesia 3. Benchmarking AyoKitaKerja against Work-net 51 Going Forward With growing demand for job-matching services and in- terest from policy makers in strengthening labor market functioning, there is strong support for AyoKitaKerja to 4. Toward a World-Class Labor Market increase its relevance and evolve from a relatively un- derutilized job-matching platform into a world-class la- Information System bor market information system (LMIS). Despite its short existence, AyoKitaKerja already has some of the basic foundations for an effective job-matching portal. None- theless, the benchmarking exercise shows that handling a modern jobs platform and processing large amounts of information require a more ambitious setup. In AyoK- itaKerja, the amount and quality of information for job matching is still limited, and links to relevant informa- tion are insufficient, including career and skills guidance, government support, and labor market information and analysis. AyoKitaKerja has many potential users but is currently caught in a vicious circle: it has too few users to provide good-quality job matching, which discourag- es potential users from accessing the services. There is a need to revamp AyoKitaKerja quickly to achieve a critical mass of users, including current and future job seekers, firms, policy makers, and practitioners. Reforming and stepping up the outreach and quality of AyoKitaKerja’s services are crucial not only for strengthen- ing the job-matching platform but also for laying the foun- dations for an advanced LMIS. Clearly, this effort will require significant human and financial resources for reform. Benchmarking against the Republic of Korea’s Work-net system provides excellent insights on aspects of an LMIS beyond the job-matching function. Based on these in- sights, the government of Indonesia could identify a way forward that is best suited to its own objectives, oppor- tunities, and challenges. Developments in global technol- ogy and the experience of other countries are important © Odua Images assets to facilitate the reform process. This concluding chapter presents a vision and action plan for Indonesia to consider. Vision and Action Plan for Reform various dimensions, and monitors outcomes to im- Figure 4.1 Vision of a World-Class Labor Market Information System prove services and policies. The proposed vision is for Indonesia to move from • Clients-at-the-center. It responds to client feedback, Five key characteristics of an advanced LMIS a basic-to-intermediate LMIS to an advanced system is targeted to individual needs, and is easily accessi- involving public and private actors, integrated ser- ble, navigable, and user friendly. vices, and data provision. While at the moment AyoK- • Comprehensiveness. It forms part of, and feeds into, Relevance Reliabilitiy Efficiency Clients at the Comprehen- itaKerja is mainly a job-matching platform, the two- broader development policies, serves a wide variety Center siveness prong strategy proposed in this chapter would allow of stakeholders, and attends to their diverse needs. Attracts, collects Is built on a strong Uses technology Is centered on Is Linked to and coordinates data management cost-effectively to client needs and is broader national AyoKitaKerja both to improve its performance as a a critical mass of and information connect users to continually development job-matching platform and to lay the foundations for These five key LMIS characteristics are strongly in- high-quality security system the system and reviewed to policies and to an advanced LMIS. terlinked. For example, a user-friendly system will in- information that guarantees leverages on M&E provide user- other relevant crease access and respond better to client needs; this round-the-clock to improve friendly services information availability of services and and to respond to systems, serves the To remain relevant and operate efficiently, AyoKita- improvement will attract more users, which, in turn, services programs client feedback needs of different Kerja would benefit from establishing a platform for may improve the relevance of the information in the stakeholders continuously innovating and improving the services system. available in response to changing client needs and in accordance with changes in international, national, re- The entire system needs to be underpinned by the gional, and global labor markets. The reform will also smart use of technology, collaborative partnerships, Technology solutions Collaborative partnership Institutional arrangements require AyoKitaKerja to integrate current advances in and institutional arrangements. The system should technology into the system in order to raise both its rely on advanced-technology solutions (including efficiency and its effectiveness. Finally, broadening ser- advanced algorithms, artificial intelligence (AI), and vices and reaching out to a more diverse set of clients big-data techniques) and integrate traditional and non- will be a transversal theme. traditional sources of data. The system should foster development. By providing linkages to privately managed information on ALMPs, and information sources that collaborative partnerships with both public and private job intermediation portals, government-managed plat- underpin the flow of information into the system Five key characteristics are at the core of the vision sector actors (workers, firms, policy makers, research- forms reinforce those businesses. As such, the govern- • Improving user friendliness and developing customer that could lead Indonesia’s LMIS to reach an advanced ers, and practitioners), who will use information from ment plays an important role in coordinating informa- response services level. Figure 4.1 presents a vision for what Indonesia’s and produce information for the system. Finally, the tion to ensure that all stakeholders in the labor market • Using technological solutions and implementing AI LMIS could aspire to be. system should have institutional arrangements in place have accurate, timely information for decision making. techniques (policies, systems, and resources) to manage the LMIS • Improving and expanding the technical infrastructure The vision highlights five key characteristics of a world- and coordinate effectively with all stakeholders. These Achieving the proposed strategic vision will require fol- of the platform, including server capacity, cloud sys- class LMIS: essential features will increase the availability, coher- lowing a set of concrete actions. The proposed action tem, and monitoring systems ence, and use of data needed to inform key decisions plan presented in table 4.1 is organized around three • Strengthening the quality and security of data by • Relevance. It attracts, collects, and coordinates and to ensure coordination of the overall system. stages: short term, medium term, and long term. encrypting personal information, controlling access, large volumes of trustworthy data, actively gathers and establishing security protocols. data, both qualitative and quantitative, from a vari- Partnerships should not be limited to collaborations Over the short term, actions should focus on attract- • Over the medium term, the LMIS could tailor services ety of sources, and transforms them into relevant, with public sector bodies but also should include the ing users, improving the efficiency of job matching, and to specific labor market needs: reliable information for diverse users. private sector. A well-functioning LMIS managed by setting the basis for a broader LMIS. These goals can • Develop tailored services for specific groups, including • Reliability. It has strong data management and infor- the government does not crowd out privately man- be accomplished by the following: small and medium enterprises and vulnerable job seekers mation technology (IT) and security systems that aged job portals; instead, it makes them active play- • Develop a career guidance function, including the in- guarantee the validity, relevance, and continuous ers. The most successful LMISs recognize that the • Establishing institutional arrangements with key pub- troduction of new online tools accessibility of data. private sector is uniquely positioned to communicate lic and private stakeholders • Analyze and disseminate information on the labor • Efficiency. It applies technology to connect users to the skills needed, identify the skills in short supply, • Mapping and integrating existing labor market data market, future jobs, and skills needs. the system, improves the services provided across and help to formulate opportunities for human capital (traditional and nontraditional), economic analyses, 54 Toward a World-Class Labor Market Information System for Indonesia 4. Toward a World-Class Labor Market Information System 55 Over the long term, the functions of the LMIS can be on career guidance and labor market information in gen- Table 4.1 Action Plan (continued) developed further. With accompanying infrastructure eral, connecting with relevant information systems such and IT upgrades, the LMIS should continue deepening the as government programs, and using the LMIS to evaluate Key characteristics services and analysis available for all stakeholders. This ef- labor market policies and develop policy tools. Features and policy actions for an advanced LMIS Relevance Reliability Efficiency Clients-at- Comprehen- fort would include introducing more in-depth information the-center siveness 13. Develop user-friendly templates using UI/UX design and including tabs for different services, that Table 4.1 Action Plan is, job matching, career guidance, data analytics,       n n government program (to be populated with in- Key characteristics formation sequentially and/or in parallel and to be Features and policy actions for an advanced LMIS Relevance Reliability Efficiency Clients-at- Comprehen- linked to one another later) the-center siveness 14. Design a light version of the portal for locations   n n n   Short term with poorer ICT connectivity 1. Expand server capacity and establish redundant 15. Establish SMS as an alternative   n       communication channel and provide servers for backup       n   2. Establish a functional disaster recovery system   n       personalized information through SMS 3. Establish integrated control of server resources 16. Develop mobile applications and im-   n       prove mobile and personal computer services       and applications for active failure response n n simultaneously 4. Introduce validation procedure (e-mail or text   n       17. Establish an institutional arrangement for the message) for registration         n LMIS 5. Establish quality assurance protocols   n       18. Map and provide links to relevant information 6. Use a cloud system to anticipate system and data sources developed by government and         n expansion and efficient use of resources in the   n       nongovernment stakeholders future 19. Provide a subsystem or link to a list of existing 7. Build a monitoring system for each system node training and education institutions with search           n n     n and monitor it regularly filters by area and location 8. Implement measures to improve security, includ- 20. Provide a subsystem or link to the COL         n ing control of access, training, and regular program   n n     inspection 21. Provide a subsystem or link to the preemploy- n       n ment card program (Kartu Pra Kerja) 9. Encrypt personal information and create and main-   n       22. Develop human resources strategically to build tain user privacy policies as the system develops n n n n n up AyoKitaKerja and provide input to the LMIS 10. Implement AI technology to improve job n   n     23. Provide training to PES staff to intensify use matching n n       and improve quality of inputs 11. Improve the input system for job vacan- cies and job seekers, replacing free-text with     n n   24. Raise awareness of job-matching services predefined templates among target groups such as private sector em- ployers, government institutions, and job seekers n         12. Ask for renewal of inactive vacancies after three at different levels of qualification, including senior months to remove inactive jobs (time period to be vocational high school students   n n     calibrated over time and in consultation with job 25. Sign memorandums of understanding with pri- seekers and employers) n         vate job portals to share information 56 Toward a World-Class Labor Market Information System for Indonesia 4. Toward a World-Class Labor Market Information System 57 Table 4.1 Action Plan (continued) Table 4.1 Action Plan (continued) Key characteristics Key characteristics Features and policy actions for an advanced LMIS Relevance Reliability Efficiency Clients-at- Comprehen- Features and policy actions for an advanced LMIS Relevance Reliability Efficiency Clients-at- Comprehen- the-center siveness the-center siveness 26. Provide information and support services to Long term n n   n   SMEs to help them to register vacancies 41. Introduce more in-depth corporate information 27. Strengthen collaboration with career for assisting job seekers in assessing employer and n   n n   guidance counselors and universities, schools, workplace characteristics n     n n and training centers on use of the job-matching 42. Introduce IT solutions to verify reliability of job platform   n n     vacancies and job seekers 28. Make enrollment mandatory for (unemployed) 43. Introduce specialized surveys to monitor the n         beneficiaries of government programs quality of job matching based on administrative     n n   29. Provide information and support services to data vulnerable groups to help them to register as job n n   n   44. Connect with other relevant information sys-     n   n seekers in the system tems 30. Develop a customer response service through 45. Use job-matching and other information to       n           n several channels evaluate the effectiveness of ALMPs Medium term 46. Develop other policy tools (besides the COL)         n 31. Ensure that public job postings at the central tailored to policy makers and local levels are posted on the job-matching n         Note: LMIS = labor market information service. AI = artificial intelligence. UI/UX = User Interface/User Experience. ICT = infor- platform mation and communication technology. SMS = short message service (text messaging). COL = Critical Occupations List. PES = public employment service. VOC = Voice Of Customer. SMEs = small and medium enterprises. ALMP = active labor market 32. Develop a feedback system to VOC and target n n       program. API = application programming interface. IT = information technology. specific groups 33. Tailor specific services to SMEs n     n   34. Tailor specific services to vulnerable groups of n     n   job seekers © SVRSLYIMAGE 35. Establish a monitoring and evaluation system     n     to track key indicators 36. Develop online tools for individual career       n n guidance including tests, videos, and games 37. Develop an information base on future jobs     n n n and skills 38. Map existing social welfare programs and ALMPs by purpose and target group and n       n provide information on how to apply and access benefits 39. Assist potential beneficiaries in assessing their n     n n eligibility for different programs 40. Share job information under open API         n 58 Toward a World-Class Labor Market Information System for Indonesia 4. Toward a World-Class Labor Market Information System 59 Appendix: KEIS Materials Figure A.1 Timeline for Work-net, 1980–2020 Source: Korea Employment Information Service 2016. World Bank Group’s Social Protection & Jobs Global Practice 61 Table A.1 Work-net Reform Projects, 2005–17 Figure A.2 KEIS Corporate Organization and Divisions Related to Work-net Project reform and details Year President Establishment of comprehensive employment and recruitment services: Office of Audit information hub & Inspection • Strengthened job-matching by expanding services based on users’ characteristics (age, gender, type of 2005–06 Planning & Informatization bureau work, type of enterprise) • Established a hub system in which job seekers could obtain job information distributed among various Office of Employment Office Employment Office of Informatization Office of agencies. Applied content management system, integrated search, and web service technologies Information Service Platform Management Planning & Coordination Work-net System Innovation Project: Informatization Employment Platform Informatization Strategy & Planning infrastructure, security, user friendliness Planning Division Management Division Management Division Division WN • Improved quality of the system on three areas: (a) improved quality of employment and job-search in- 2007 formation; (b) improved information structure that allowed for customized search results and personal- Job Portal Development Division Employment Platform Development Division IT Infrastructure Division Organization & Budget Division ization of services; (c) improved system speed and established the foundation for quality enhancement WN WN Work-net Mobile Enhancement Project: WorkNet Data Standard IT Security Performance mobile web and app services Employment Support Division Division Division Managenent Division WN WN • Improved user friendliness and accessibility, by (a) building mobile web and app services; (b) introducing 2010 Employment Data Analysis Customer Human Resource & recruitment information, search, and direct call services; (c) mapping jobs and linking them to public Insurance Division Division Counselling Training division transportation information to provide localization services; (d) enabling notifications on job openings, WN Division WN closings, and placement; and (e) sharing job postings on social networking sites Human Resources Benefit Fraud International Development Division Cooperation Division Next Generation Total Employment Service System Development Project: WN Monitoring Division overhaul of services, interoperability, Baro-one Employment Permit Employment • Integrated and linked public and private sector job information 2011–12 System Division Informatization Center WN Construction Division • Established a system for real-time monitoring of jobs • Provided local governments with access to local job information on Work-net Research Planning • Established a master database by standardizing data on employment-related system management tar- Research Planning And Coordination Bureau Coordination Division gets (for example, employment insurance, job placement, and job training) and establishing an integrat- ed employment history database (master data management system) Office of Employment Center for Employment Center for Employment Policy Center For youth Policy Service Strategy Information Analysis Evaluation & Monitoring Research & Monitoring • Introduced Baro-One, a customized employment service system that relied on the master database, and established an integrated employment and industrial accident analysis system Employment Service Labor Market Employment Policy Youth Policy • Further improved Work-net’s user friendliness by reorganizing internal and external network services Innovation Division Analysis Division Evaluation Planning Research & Develop- WN WN Division ment Division • Improved Work-net’s mobile services and standardized the internal system Establishment of Job Portal Site Project: Employment Service Quality Managenent Employment Forecasting Division Job Programs Evaluation & Analysis Youth Policy Monitoring Division LMIS with full integration of systems, artificial intelligence, infrastructure Division WN Division • Established a national job information platform that could integrate and manage employment infor- 2017 Lifelong-Career Employment Statistics Local Economy & Online Youth Center mation and labor services and provide analysis of results to the public. This required the integration of Development Division Surver Division Employment Monitoring Division Planning & Managenet Division WN channels and the provision of online integrated employment information to the public through personal computers and mobile phones Future Work Research Division • Carried out a detailed analysis of users’ information using big-data techniques with the goal of providing WN personalized employment services. Specifically, the system categorization of information was improved to Employment Service classify characteristics and personality types and employment-related content on Work-net and HRD-net UX Division WN • Replaced outdated equipment and infrastructure needed for the national job information platform and established a new job portal site service Source: Korean Employment Information Service. Note: WN = Work-net–related units. 62 Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice 63 Table A.2 Work-net Staff within KEIS Corporative Organization a. Planning and Information Bureau b. Research Planning and Coordination Bureau Division Role Staff Educational requirements Initial need Division Role Staff Educational requirements Initial need Job Portal Work-net online and mo- Engineer 17 Information technolo- Essential Employment Strategies for developing PhD 2 Doctoral degree Essential Development bile system planning and PhD 1 gy (IT) developer and Service Inno- online and offline employ- Master’s 4 Master’s degree Division operation Planner 2 related license owner vation Team ment services Administra- IT major tive 2 Informatiza- Informatization budget and User Expe- IT major Optional Lifelong-Ca- Research and content PhD 5 Doctoral degree Optional tion Planning organizational manage- rience (UX) UX design major Replaceable reer Develop- development to support Master’s 6 Master’s degree Replaceable Division ment designer 2 ment Division lifelong career develop- Administra- Planner 2 ment policy tive 3 Engineer 3 Future Work Occupational research and PhD 5 Doctoral degree Optional Work-net Management of Work-net Engineer 15 IT major Essential Employment internal network oper- Monitoring For monitoring, no require- Research Divi- job-related content devel- Master’s 4 Master’s degree Replaceable Support Divi- ations for employment 17 ments sion opment Administra- sion centers tive 1 Employment Planning and program Engineer 20 IT developer and related Optional Employment Online-offline employment PhD 1 Doctoral degree Optional Insurance development and man- license owner Service UX service user experience Master’s 2 Master’s degree Replaceable Division agement of employment IT major Division and environmental design Administra- insurance system services tive 1 Human Re- Planning and program Engineer 19 IT developer and related Optional Labor Market Collection and analysis of PhD 5 Doctoral degree Essential source De- development and manage- license owner Analysis Divi- employment information Master’s 4 Master’s degree velopment ment of HRD-net IT major sion Administra- Division tive 1 IT Infrastruc- Introduction and operation Engineer 12 IT developer and related Essential Employment Mid- to long-term employ- PhD 5 Doctoral degree Optional ture Division management of hardware license owner Forecasting ment forecast and model Master’s 6 Master’s degree Replaceable network IT major Division development Administra- IT Security Establishment and manage- Engineer 7 IT developer and related Optional tive 1 Division ment of information securi- license owner Infrastructure ty policies for IT systems Profession security major can be re- placeable Customer Management of Voice Of Counselor — Optional Counseling Customer (VOC) system Division 64 Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice 65 References Betcherman and Moroz, 2017. “Employment pro- Kureková, L. M., M. Beblavý, and A. Thum. 2014. Sorensen, Karen, and Jean Michel Mas. 2016. “A World Bank. Forthcoming. “SABER Workforce grams as a social protection instrument in “Using Internet Data to Analyse the Labour Roadmap for the Development of Labor Development Country Report, Republic of East and South Asia.” Mimeo. Market: A Methodological Enquiry.” IZA Dis- Market Information System.” African Union/ Indonesia.” World Bank, Washington, DC. cussion Paper 8555, Institute for the Study of FHI 360, US Agency for International Develop- Cedefop. 2019. “The Online Job Vacancy Market in Labor, Bonn. ment, Washington, DC. World Bank. Various years. World Development Indi- the EU: Driving Forces and Emerging Trends.” cators database. Washington, DC: World Bank. Publications Office of the European Union, McKenzie, David, 2017. “How Effective Are Active La- Statistics Indonesia. Various years. National Labor Luxembourg. bor Market Policies in Developing Countries? Force Survey (Sakernas) database. BPS Statis- Yoo, Kilsang. 2018. “A Study on the Development A Critical Review of Recent Evidence.” Policy tics Indonesia. of Employment Service through Central-Lo- CMEA (Coordinating Ministry of Economic Affairs) Research Working Paper 8011, World Bank, cal-Private Collaboration.” Korean Ministry and the World Bank. 2019. Indonesia’s Criti- Washington, DC. WCC Development. 2017. Managing Workforce Po- of Employment and Labor and Korea Labor cal Occupations List 2018 Technical Report, tential: A 20/20 Vision on the Future of Em- Economics Society, Seoul. December 2019. Nguyen, Trang. 2008. “Information, Role Models, ployment Services. Utrecht, the Netherlands: and Perceived Returns to Education: Experi- WCC Development. Zhu, Tingting J., Alan Fritzler, and Jan A. K. Orlows- Dinkelman, Taryn, and Claudia Martínez. 2014. mental Evidence from Madagascar.” Working ki. 2018. “World Bank Group–LinkedIn Data “Investing in Schooling in Chile: The Role of Paper, Massachusetts Institute of Technology, Woltermann, Silke. 2012. “The Labor Market Infor- Insights: Jobs, Skills, and Migration Trends Information about Financial Aid for Higher Ed- Boston, MA. mation System as an Instrument of Active Methodology and Validation Results.” World ucation.” Review of Economics and Statistics Labor Market Policies.” Beiträge aus der Praxis Bank, Washington, DC. 96 (2): 244–57. OECD (Organisation for Economic Co-operation der beruflichen Bildung 11, GIZ, Bonn. and Development). 2015. “Skills Matter: Fur- Dizon-Ross, Rebecca. 2014. “Parents’ Perceptions ther Results from the Survey of Adult Skills.” World Bank. 2019. “Using Labor Market Bulletins to and Children’s Education: Experimental Evi- Jakarta Indonesia Country Note, OECD, Paris. Improve Labor Market Information in Viet- dence from Malawi.” Working Paper, Harvard nam.” World Bank, Washington, DC. University, Cambridge, MA, February 28, 2014. OECD (Organisation for Economic Co-operation and Development). Various years. Programme ILO (International Labour Organization). Various for International Student Assessment. Paris: years. ILOSTAT database. Paris: ILO. OECD. ITU (International Telecommunications Union). Patrinos, Harry. 2017. “Automation and Its Implica- Various years. World Telecommunication/ICT tions for Developing Countries and Education Indicators database. Geneva: ITU. https://www. Systems.” Presentation at ASEAN+3 Confer- itu-ilibrary.org/science-and-technology/data/ ence, Seoul, November 2017. world-telecommunication-ict-indicators-data- base_pub_series/database/2a8478f7-en. Rutkowski, Jan, Carmen de Paz, and Victoria Levin. 2018. “Labor Market Observatories: Critical Korea Employment Information Service. 2016. “Ten- Success Factors.” Job Note 4, World Bank, Year History of Korea Employment Informa- Washington, DC. https://openknowledge. tion Service.” KEIS, Seoul. worldbank.org/handle/10986/29594. 66 Toward a World-Class Labor Market Information System for Indonesia World Bank Group’s Social Protection & Jobs Global Practice 67