FACTORS INFLUENCING INFORMAL WORKERS’ REGISTRATION FOR SOCIAL SECURITY: A COMPARATIVE ANALYSIS BETWEEN INDONESIA AND TAIWAN

: Social security should be mandatory for all members of society to protect them from social risks, including the informal workers who are particularly vulnerable to such risks. However, the coverage of social security for informal workers in Indonesia remains very low. Therefore, the study aims to identify the factors that drive informal workers' desire to enroll in social security programs. The Theory of Planned Behavior will be utilized as a tool to uncover these factors. The study will compare the findings with the policies implemented in Indonesia and Taiwan as a comparison for countries with extensive social security coverage. The research sample is determined by using purposive sampling method with 100 respondents participated in this study. Data are examined by using structural equation model - partial least square (SEM-PLS). The results show that Attitude Toward Behavior and Perceived Behavioral Control have a significant impact on the intention of informal workers to join social security programs, while subjective norms have not been proven to have a significant impact. In conclusion, Indonesia needs to review its current policies, which primarily focus on subjective norms, and learn from Taiwan's successful implementation of broad social security coverage. Transforming informal labor into formal employment can be an effective strategy for achieving this goal.


INTRODUCTION
The primary objective of social security is social protection, which includes addressing issues like family and children, old age, disability, and poverty. Unfortunately, not all citizens, especially workers, are currently protected by the social security system, particularly in Indonesia. This is unsurprising, considering achieving universal coverage is one of the main obstacles to social security implementation today (ISSA, 2019). The majority of workers in the informal sector frequently go unnoticed by the social security system, and the existing social security programs primarily only focus on and target workers in the formal sector, which inflates the number of uninsured workers in developing nations. As shown in Figure  1, the fact that over 90% of the labor in developing countries is employed informally is also complicating matters (Kyengo, 2010;Dorfman, 2015).
In general, unregistered small businesses, independent contractors, and part-time employees are considered informal workers (ILO, 2000). Informal workers are typically not as adequately protected by established social security schemes as formal workers, who typically enroll in either mandatory or voluntary social security programs (Masanyiwa et al. 2020). Many workers in the informal sector lose access to training and safety resources. Due to a lack of safety resources and training, this industry is more vulnerable to occupational illnesses, fatalities, and other adverse outcomes (Suthakorn et al. 2020).
Some nations, especially those in Asia, struggle to provide social security protection for their informal workers. This is caused by the difficulty of calculating and determining contributions for informal workers that have fluctuating income over time. In Indonesia, it has been observed that the majority of the workforce accomplishes their financial necessities in the informal sector, a decentralized, small independent economic sector with a high degree of entrepreneurial flexibility. However, there is little social protection to go along with the huge informal sector in society. Not many informal workers participate in the social security scheme, and the social security scheme does not cover all of them. Recent studies have been conducted to close the social security gap for informal workers (Suryahadi et al. 2014;Rekson, 2017;LIPI, 2020). Although theoretically advanced, the application of empirical research is needed to obtain a comprehensive and indepth strategy by discovering the factors that influence the intention to participate in informal social security schemes. Thus, by analyzing the factors experienced by Indonesian informal workers to participate in social security programs, an effective strategy for achieving universal coverage can be found. In the end, this can be a reference for evaluating the strategies and efforts that have been made by BPJS Ketenagakerjaan to achieve universal coverage. Figure 1. Informal laborers in developing nations (ILO, 2018) behavior explanation. By obtaining easily accessible behavioral outcomes, normative references, and control factors, the use of TPB enables the measurement of the factors that directly influence customer decisions.
The attitude toward behavior (ATB), perceived behavioral control (PBC), and subjective norms (SN) are discussed in the theory of planned behavior as the antecedents of behavioral intention (BI). The intention of informal workers to participate in social security programs may be determined by circumstances that can be explained by planned behavior.

Attitude towards behavior (ATB) is an action that
indicates what customers prefer about certain things. ATB refers to a consu'er's evaluation of their preference toward a particular incitement. Attitudes develop reasonably from the beliefs people hold about the object. Subjective norms (SN) are perceived expectations that an individual feels they should meet when those around them (such as family, friends, or coworkers) agree on certain actions and encourage them to do so (Ajzen, 1977). It is a technique for evaluating what a person should do based on consideration from their closest societies. The term "perceived behavioral control" (PBC) refers to a person's decision-making that is influenced by their sense of how difficult it will be to carry out a certain activity. According to Ajzen (2005), peo'le's perceptions of behavioral control are influenced by the availability of resources such as equipment, compatibility, expertise, and opportunities that support the conduct that is expected. The higher the sense of personal control over an activity, the more strongly people believe that their resources are available for them to engage in it.
After identifying the reasons that motivate Indonesian informal workers to register in the social security scheme, this study moves on to determine the measures that the country has implemented. Whether or not the policies are in accordance with the findings of this study. In addition, the policies will be contrasted with the ones Taiwan has implemented in order to draw lessons for Indonesia. Lesson from Taiwan needs to be studied and needs to be analyzed to find out whether it can be adapted and applied by Indonesia, if so then Indonesia needs to adopt what Taiwan has done.
On the other hand, unlike Indonesia, Taiwan has its own way of dealing with the difficulty of protecting informal workers with social security. Taiwan is now one of the Asian nations that joined the rich during the second half of the 20th century, along with South Korea, Singapore and Japan. They all have one thing in common: none of them have a huge number in informal economy. Though this does not necessarily mean that Taiwan has absolutely no informal workers. Tai'an's social security law allows independent contractors, employers with less than five employees who are not members of a labor union, and self-employed individuals who also hire others to voluntarily enroll in social security insurance. This is possible because managing employees who work in the informal economy is highly challenging. The ability to calculate the contribution that must be made is one of the largest challenges because the income earned might often fluctuate wildly. Despite this challenge, Taiwan has managed to provide all of its workers with social security coverage.
The low and poor contribution of social security is a big hurdle for Indonesia, particularly for the social security organization specialized for workers, BPJS Ketenagakerjaan. Participation in BPJS Ketenagakerjaan by law is mandatory, but based on BPJS Ketenagakerjaan internal data (2021) there are still many workers who have not yet become participants (Pasaribu et al. 2022). BPJS Ketenagakerjaan work'rs' social security coverage has barely reached 55% of the entire 93 million laborers, with only 3.7% of informal workers covered (LIPI, 2020). Current researches have been carried out in order to address the gap in social security coverage for informal workers (Suryahadi et al. 2014;Rekson, 2017;LIPI, 2020). Despite theoretically sophisticated, empirical approaches are required to get completely comprehensive and deep strategies by identifying variables influencing the willingness to engage in an informal social security program.
Given the fact that there are various difficulties in acquiring and protecting informal workers with social security, it is interesting to investigate what influences informal work'rs' decisions to sign up for social security, particularly in Indonesia, where the coverage is still low. The Theory of Planned Behavior (TPB) will be applied in this study to identify those factors. TPB can help to predict customer intention and It is critical for Indonesia to progressively transition from a social protection system that exclusively targets the poorest through non-contributory public assistance to a system that includes people with missing middle income. Those who can contribute to the contributory system will benefit from social insurance (contributory). Progressively, the government must guarantee that every person is covered, whether through contributory or non-contributory programmes. With a large proportion of its working age or productive age group population laboring in the informal sector, having contributing social protection is critical, especially given that Indonesia is expected to become an aging society over the next 25 years.
Theory of Planned Behavior (TPB) (Ajzen, 1991) states that activity is generated due of intention / interest, where interest is impacted by attitude toward the behavior, subjective norms, and perceived behavioral control. These factors are connected and impact one another. Intentions are significantly influenced by attitudes toward behavior, subjective norms, and perceived behavioral control. As a result, the TPB method may be applied to studies concerning the factors that impact participation intentions in social security.
This research serves as advice and recommendations for BPJS Ketenagakerjaan, the organization that administers social security for informal worke, to broaden its coverage. When developing a strategy to expand membership coverage, consideration should be given to factors that have been found to be significant in this study. To ensure that Indone'ia's social security system can protect all societal layers as well as employees, lessons learned from other nations, particularly Taiwan, must become the focus of membership expansion.

METHODS
The studys focus is on workers who fall into the "missing middle" category rather than informal workers who are below the poverty line ( Figure 2). Through a number of non-contributory programs, the Indonesian government already offers social aid or assistance to those who are in need. However, the "missing middle" receives little government assistance in meeting their fundamental economic requirements, particularly those in the informal sector (TNP2K, 2018). The labors who are referred to as "the missing middle" may contribute benefits to the underprivileged by paying into social security and giving back a tiny fraction of their contributions to help others in need. perceived behavioral control (PBC). These three aspects are important because they represent determinants for measuring the intention of informal workers to join social security schemes. Are informal workers interested in registering because of the quality of the BPJS Ketenagakerjan social security (ATB aspect)? Or, do they want to register because others closest to them are doing so (SN Aspects)? Or are they interested because they feel they have sufficient resources to make it easier for them to join social security (PBC Aspect)? These questions make the research interesting to conduct.
Directly from informal workers, a series of structured questionnaires is used to collect primary data. Each variable requires indicators to determine its value (Table 1). Then to measure the variables, this study uses questionnaires as the indicators. The questionnaires are designed with a Likert scale ranging from 1 to 5, representing an interval scale (1: Strongly disagree, 2: Disagree, 3: Neutral, 4: Agree, 5: Strongly agree).
The factors behind the intention of informal workers to join social security are important, because these factors should be the main considerations in how BPJS Ketenagakerjaan needs to do to attract informal workers' intentions to join social security schemes. Since the informal workers have full control over their decision to enroll in social security, their intentions also play a significant influence in the informal workers' final decision to join the program. In contrast to formal workers, who are often registered by their employer, in this situation their employer has more control over whether or not they participate in the social security program. Figure 3 explained the research framework of this research. Based on the framework that has been shown above, the hypotheses of this research are as follows: Hypothesis 1 (H1): Attitude will influence intention to participate in social security scheme. Hypothesis 2 (H2): Subjective norm will influence intention to participate in social security scheme. Hypothesis 3 (H3): Perceived behavioral control will influence intention to participate in social security scheme.
The relationships between the variables created for this study are shown in the above figure. The independent variables that affect participating intention are attitude toward behavior (ATB), subjective norms (SN), and Figure 3. The research framework a. I am knowledgeable enough to pay into and sign up for social security. b. I don't need anyone's assistance to register for and make contributions to social security. c. Without assistance from anybody, I am able to contribute to social security. d. I have the opportunity and time to make social security contributions.

Dependent Variables
Participating Intention (INT) a. I want to make social security contributions. b. I will advise people to make social security contributions. c. I In the future, I'll make voluntary Social Security contributions.

Assessment of Measurement Model (Outer Model)
The measuring model or outer model is evaluated to determine the model's validity and reliability. The validity test is measured by conducting convergent validity test and for the reliability test, it is measured by conducting convergent validity test. According Ghozali (2015), the loading value is the common rule of thumb for measuring convergent validity. The loading value must be greater than 0.7 for confirmatory research and between 0.6 -0.7 for explanatory research. Based on the calculation results shown on Table 2, all indicators used are valid and can be used to explain each variables. Furthermore, as stated by Nunnally and Bernstein (1994), the average variance extracted (AVE) value in the outer model must be larger than 0.5 and the cronbach's alpha value must be greater than 0.6. From Table 3, it can be seen that the convergence validity result has AVE values above 0,5, and also Cronbach alpha values above 0.6.
The respondents must meet the following criteria: a) they must be "the missing middle" of informal workers, b) informal workers who are neither government or other private institution recipients of social assistance, and c) who do not have any other form of social insurance except public social insurance.
The majority of informal workers are employed in Jakarta, where this survey was conducted. Informal workers' population in Jakarta is 3.036.133 workers (BPS, 2019). The definition of informal workers in this study uses the definition from the Central Statistics Agency (BPS), where informal sector workers are the main employment status of people which include self-employed, working assisted by temporary workers, working assisted by permanent workers, freelance workers in agriculture, casual workers in the non-agricultural sector and family/unpaid workers. However, unpaid workers are not included in research since they are not the intended receivers of social security benefits, while they receive social assistance. Purposive sampling is used to select respondents who have not joined the social security system, do not receive social assistance, and do not have commercial insurance. This is done to ensure that the informal workers involved fall into the category of "the missing middle". There were 100 respondents who responded and met the requirements out of the 124 who were given a set of questionnaires. The preparation and data collection procedure began in October 2021 until February 2022. Furthermore, data analysis and processing are finished by the end of March 2022.
For data analysis, the structural equation model-partial least squares (SEM-PLS) approach is used in this study. According to Henseler et al. (2016), the two sets of linear equations that officially characterize SEM-PLS path models are the measurement model (also known as the outer model) and the structural model (also called the inner model). After the intention has been measured and significant factors have been found in the intention, then a comparison will be made between the policies that have been taken by Indonesia and Taiwan as a comparison country. This is done to get lessons from Taiwan that Indonesia can do to achieve universal coverage on the social security system, especially for informal workers. In this method, various sources will be obtained through social security laws in Indonesia and Taiwan, and to strengthen the existing information. To establish the significant value between constructs, the t-Statistic (bootstrapping) is applied. The hypothesis is rejected if the t-statistic value is between -1.96 and 1.96. The P-value could also be used to test the proposed hypothesis, if the p-value < alpha (α = 0,05) the hypothesis is accepted. From Table 5 we can see that ATB and PBC have a positive and significant correlation with the intention, so the results are reliable. However, for SN which has a negative correlation value with the intention, it cannot be proved because it does not have a high significance value.

Comparison with Taiwan
The analysis of theory of planned behavior to determine informal workers' intention to join the social security using SEM-PLS, shows that attitude towards behavior and especially perceived behavioral control influence informal workers to join in social security system. Table  6 shows the policy comparison between Indonesia and Taiwan on TPB aspects.

Evaluation of the Structural Model (Inner Model)
The structural model (Inner model) must be tested once the estimated model fits the requirements for the outer model. According to Ghozali (2015), the structural model (Inner model) assessment seeks to anticipate the link between latent constructs or variables. Hair et al. (2017) in Ramayah et al. (2017) recommends paying attention to the coefficient of determination (R2), path coefficients, and t-value statistics while evaluating the inner model (bootstrapping). When evaluating the inner model with SEM-PLS, to assess how well a research model is used, we can see and assess the value of R-square. The R-square (R2) value on Table  4 indicates how well exogenous factors (independent variables) explain endogenous variables (dependent variables). The R-square value for intention to join is 0.840 which means that 84% is explained by the model (ATB, SN, PBC) and the remaining 16% is explained by other variables not used in this study.
The path coefficient is used to examine the proposed link between constructs. The path coefficient values vary from -1 to 1, according to Helm et al. (2009) in Hair et al. (2014). A path coefficient value close to 1 implies a substantial positive relationship, whereas one close to -1 indicates a substantial negative relationship. Although results near 1 or -1 are frequently statistically significant, the standard error must be determined using bootstrapping to assess for significance.
From Figure 4, it can be seen that PBC has the highest correlation with intention to join with path coefficient value of 0.771. Therefore, thoughts about financial ability and knowledge about social security might influence the intention to join in social security. We can also see that SN has a negative coefficient value, which means that the higher the SN value is, the less likely it is that workers have the intention to contribute to social security. However, this cannot be concluded directly, because the significance must be tested to prove that it is indeed a significant negative correlation. High statistical significance supports the fact that the results are real and not due to luck or chance. Simply put, if a statistic has high significance, it is considered more reliable.   In addition, there is the aspect of perceived behavioral control, in which the financial capabilities and knowledge of potential participants play a role. Informal workers who have problems with their financial capabilities may not have the intention to do so. This is similar with the findings of Beck and Webb (2003) and Lee et al. (2010), who discovered that the most In the aspect of attitude towards behavior, public opinion regarding social security needs special attention in order to be positive. Positive opinion can be increased by improving the quality of services. BPJS Ketenagakerjaan is generally the comparable to other institutions in other Asian countries that already has a clear process flow from upstream to downstream, starting from registration to claiming. This process is also supported by various electronic channels such as websites, social media, call centers, and applications. Therefore, we can conclude that all efforts related to the ATB aspects have been carried out by Indonesia. If we take a look at Taiwan, this country is doing the same. Taiwan has been constantly expanded the scope of its there is no positive and significant correlation under the aspect of subjective norms. Taiwan also has no policy in this aspect, but the number of participation is almost perfect, reaching almost 100% of the total workers.

Lesson Learned from Taiwan
Taiwan today, compared to the 1980s, has undergone various transformations, especially in the field of employment (Mehrotra, 2020). Table 7 shows the estimated proportion of total labor forces in the informal sector in Taiwan. The transformation from the informal economy to the formal economy is the key to how Taiwan can achieve universal coverage in social security. The great employment transformation in Taiwan is the result of a reasonably well conceptualized strategy of industrialization with the appropriate sequencing of policies within a planning framework. It is this strategy of employment-intensive industrialization that lies behind the success at formalization. The informal economic activities in Taiwan appear to have contracted dramatically over a period of 43 years (Wong, 2015). In addition, the economy has been able to maintain a relatively equitable income distribution and achieve high levels of well-being. Lee and Torm (2017) investigate how social security provision -a key determinant of formality -impacts on small and medium-sized firm performance. That means there needs to be a transformation process from the informal economy to the formal economy.
significant predictors of demand for social insurance are price and financial ability. This means that BPJS Ketenagakerjaan has no other option but to support the government effort to increase the welfare of the citizens, as the financial issue is the main factor to the intention to join. Another possible measure currently being implemented by BPJS Ketenagakerjaan is to levy a series of contributions based on the worker's income, so that low-income workers do not have to pay high contributions as well. In Taiwan, it is actually the same, the contribution amount is based on the income of the worker. For people with middle or low income, there are subsidies or they are even exempt from paying contribution. All levels of society from rich to the poor can be fully protected. However, Indonesia cannot imitate this immediately. Special calculation is needed to ensure that the measures taken are financially viable.
According to the results of this study, this financial aspect is the main driving force behind the interest of informal workers to enroll in social security.
The next aspect of subjective norms must be considered as well. Indonesia, especially BPJS Ketenagakerjaan, is currently focusing its efforts and strategies to expand its membership through this aspect. Though the PERISAI program, BPJS Ketenagakerjaan provides rewards for agents who can enroll relatives, friends, or those around them to enroll in the social security system. There is a need to revisit its effectiveness, as the number of acquisitions through the PERISAI program is reported to have not encouraged the acquisition of new informal workers. In accordance with this study,  Amin (2002) to allow for greater flexibility and to facilitate the transfer of jobless persons into the formal economy. Competition policy, by reducing monopolies, can assist small enterprises in entering specific industries. Excessive constraints and bureaucratic processes should be removed. 5. Workers in the informal economy must join or create their own labor unions. Trade unions can promote awareness of the need of collective representation among employees in the informal sector through training and community initiatives. They can also make an attempt to include workers in the informal economy in collective bargaining agreements. Because women make up the majority of employees in the informal economy, trade unions should create or modify internal structures to increase women's involvement and representation, as well as to address their unique needs. Workers in the informal economy may benefit from the specific services provided by trade unions, such as legal information, educational and advocacy programs, legal assistance, medical insurance, credit and lending schemes, and cooperative formation.

Managerial Implications
Based on the results of the study, the results show that attitude towards behavior and perceived behavioral control affect the intention of informal workers to join the social security system. Thus, there are several things that need to be done by BPJS Ketenagakerjaan to attract the intention of these informal workers. The first thing that BPJS Ketenagakerjaan needs to do in the aspect of attitude toward behavior is how BPJS Ketenagakerjaan needs to ensure the quality of its services. To show the quality of this service, it can be done by means of positive reporting about the success of BPJS Ketenagakerjaan in carrying out perfect services on mass media, newspapers, social media and other news media. Furthermore, BPJS Ketenagakerjaan also needs to ensure that social security benefits have been conveyed to the right and entitled people, so that the attitudes and beliefs of informal workers towards BPJS Ketenagakerjaan and social security can be maintained. The next step in enhancing perceived behavioral control is to offer additional benefits to workers who have enrolled in the social security program. If we look at the results of the study, these results show that financial ability is one of the things that influence the intention to join social security. So there needs to be a strategy Taiwan does not have a clear method for covering these informal laborers. Because of the difficulty in collecting data and implementing the legislation on these individuals, Taiwan eventually only enforces voluntary registration for workers who are not members of a worker union. However, this has little effect on Taiwan's social security coverage rate, which is nearly 100%. As a result, the advantages derived from the contribution will be even larger with a big and broad coverage amount. Because informal workers outnumber official workers in Indonesia, workers in this sector do not have the same protection as formal workers, such as pensions and unemployment benefits.
To summarize, Indonesia might adopt Taiwan's features by formalizing the informal sector. This can be accomplished by developing effective policies to combat informality. However, it is complex by the fact that it has numerous causes and manifestations, both across and within nations. Informality is a response to a unique collection of traits and institutions in each country, and there is no one-size-fits-all answer. By undertaking extensive study and policy experimentation in both emerging and advanced economies, a common set of guiding principles for policy formulation has emerged. In this regard, the ILO (2002) has offered recommendations on how a nation might accomplish the transition, with the following aspects standing out: 1. The most effective strategy to eliminate informality is to improve education quality. Education initiatives that aim to promote access and guarantee that children stay in school are especially important so that they may gain skills and information to pursue job opportunities outside of the informal sector. 2. The tax system should prevent unintentionally boosting incentives for people to stay in the informal sector. It is widely acknowledged that simpler tax systems (with no or few exemptions and loopholes) with lower rates, as well as low payroll taxes, help in the reduction of informality. 3. Policies that promote wider access to formal (or bank-based) financial services can assist to reduce informality. Lack of access to capital is a major limitation for informal sectors and entrepreneurs, limiting productivity and company growth.
Countries with better access to finance grow quicker and have lower income inequality. 4. A number of structural strategies can be used to increase incentives and lower the cost of formalization. Labor market laws may be changed looking at policies in other countries, especially Taiwan in the expansion of social security for informal workers. This country focuses its strategy on aspects of attitude towards behavior and perceived behavioral control.
Finally, Indonesia needs to transform from the informal economy to the formal economy. The informal economy, which tends to operate in a gray area, is very difficult to be supervised by social security. So this shift is critical to the success of extending social security coverage. The Indonesian government must make this effort to protect the well-being of workers throughout the country.

Recommendations
This study contributes significant findings to social security studies. The government and BPJS Ketenagakerjaan should consider the psychological role, such as attitude towards behavior, subjective norms, and perceived behavioral control to increase the number of informal workers who join social security.
Comprehending the thought processes of informal workers is the first and foremost as well as a significant step in expanding the coverage of social security.
Psychological role provides a valuable advantage to predict and analyze a worker's decision-making process. Over time, it helps in framing the consumers' mindsets and establishing a loyal customer base. This could be done through education, advertising or other channels. For readers, it is possible to develop the topic of this research by raising other variables, so that broader research can be conducted by using other variables or methods, for example by conducting direct interviews (in-depth interviews). By doing so, more clear and concrete problems can be identified and addressed in the future, so that social security with universal coverage in Indonesia can be realized.
that shows that this social security is financially and economically profitable. This additional benefit can be in the form of providing business loans for informal workers who comply with paying contributions on a long term basis. In addition, this additional benefit can also be in the form of financial advice where informal workers can consult financially to develop their businesses, so that with these informal workers can grow and eventually transform into formal businesses that are getting bigger in scale.
The long term goal is that the transformation from the informal economy to the formal economy can be achieved as happened in Taiwan. Thus, this transformation is the key to success in expanding the coverage of informal workers in social security. This is an effort that needs to be done by the Indonesian government to ensure the welfare of workers throughout Indonesia.

Conclusions
The demand for informal social security is driven by the attitude towards behavior and the perceived behavioral control. This can be seen through the positive path coefficients of these two variables and the significant test results of both ATB and PBC variables. So that it can be interpreted that positive opinions and attitudes towards social security as well as knowledge and financial ability are the most influential factors on the intention to join social security for informal workers.
The SN variable in this study has no significant effect, this can be seen in the significant test carried out. So it can be interpreted that the recommendation from the closest person, family, or relative is not a factor that influences the intention to join social security for informal workers.
Looking at the various policies that have been implemented by BPJS Ketenagakerjaan, this research sees that the effectiveness of the strategies implemented in the subjective norms aspect to expand social security coverage for informal workers needs to be reviewed, for example the PERISAI program. This can be seen from the research results which show that subjective norms do not have a significant effect. Coupled with