Can Knowledge and Culture Eradicate Poverty and Reduce Income Inequality? The Evidence from Indonesia

Abstract

This research aims to investigate the potential role of knowledge and cultural in reducing income inequality and alleviating poverty in Indonesia. The method used includes panel data regression estimation with the least squares dummy variable approach, compounded with an interactive session. The results showed that knowledge and cultural have a significant effect on income inequality, poverty, and development. This research has developed a knowledge-based and cultural-based development framework, by incorporating democracy-political institution as a new proxy variable for knowledge and tourism-cultural, as well as active participation as determinants of the cultural economy. The significance of cultural heritage possesses a remarkable capacity to attract tourists, thereby generating revenue, and bears indigenous wisdom and knowledge, inherent to local communities. This research stands out from the previous due to the unique combination of democracy-political institution determinants and cultural economy, which creates a blend of knowledge and cultural. By filling the gap left by previous research that solely focused on these areas independently, this research presents a holistic understanding of the subject matter.

Introduction

The potential role of knowledge and cultural has to be leveraged to eradicate poverty and reduce income inequality, especially in Indonesia, given that this developing country has a relatively low level of enlightenment, despite having abundant cultural resources. Knowledge as an economic resource has been widely recognized, and Drucker (2011) referred to it as knowledge-based. Hadad (2017) stated that it is distinguished from the traditional economy. Knowledge economy is associated with abundance, while the traditional is associated with scarcity. Ahmed (2017) stated that there is no specific characteristic of enlightenment, but it is known that knowledge is information that increases productivity and enhances the economy.

According to Mokyr (2017), cultural is a set of beliefs, values, and preferences that affect socially inherited behavior possessed by certain groups of the community. Mokyr (2017) mentioned that cultural has shed light on progress in economic growth, technology, and industry. At the end of the nineteenth century, Europe experienced an industrial revolution that led to significant progress. Cultural enlightenment promoted intellectual forms, innovation, and knowledge, which generated the industrial revolution. The industrial revolution lifted civilization in European communities, which rapidly spread to other countries.

Previous research showed that knowledge economy significantly affects economic growth (Amirat & Zaidi, 2020; Barkhordari et al., 2019; Nurunnabi, 2017; Parcero & Ryan, 2017) and reduces poverty (Shahabadi et al., 2017). Martin (2019) stated the role of higher education in regional development, alleviation income inequality, and poverty through knowledge transfer. The higher the educational impact increases income, the more the reduction in income inequality and poverty is achieved. Affortunato et al. (2010) developed local-based knowledge economy indicators and added a cultural-based index to measure the impact of cultural factors on regional development.

Previous research also showed that technological advancements create income inequality (Peng, 2014). Antonelli and Tubiana (2020) stated that the transition to a knowledge economy brings radical changes, because technological advancements have an impact on income inequality, as a result of expertise exploitation in technology-based industries and the polarization of the workforce. Besides technology, income inequality is also caused by educational differences. The research by Mahmud and Akita (2018) in Bangladesh showed that the educational inequality between urban and rural areas led to income differences among the residents.

Meanwhile, previous research on the cultural economy showed that its role as a driver of sustainable development has been widely recognized (Gilberto & Labadi, 2022), as it directly contributes to poverty alleviation (Rosetti et al., 2022). Dobrosława Wiktor-Mach (2019) stated that there has been a paradigm shift in the relationship between cultural heritage and development. Cultural heritage is not just a legacy with high value but also has the real role in development.

This research complemented the previous by incorporating institution determinants of democracy-politics into the research model. It also incorporates cultural tourism determinants and examines their impact on income inequality and poverty. Guiso et al. (2006) showed that cultural does not directly affect the economy but through transmission. In this case, the fascinating potential of the tourism sector tends to attract both domestic and foreign tourists. Indonesia is endowed not only with nature but also with abundant cultural resources in the form of tangible and intangible assets.

This research drew motivation from the 1945 Constitution, stating that the goal of development is for the welfare of all Indonesian communities. Also, the Brundtland Commission’s report on sustainable development in 1987 stated that all forms of poverty and income inequality should be reduced.

Income inequality and poverty are the major development issues faced by many developing countries, including Indonesia; however, economic growth does not solve the totality of the problems. This is true when growth is based on strengthening capital and technology, introducing more problems, for countries that rely on low-educated personnel. When economic growth relies on reducing the workforce, it tends to face problems of efficiency and productivity, eventually decreasing economic output and impoverishing the masses.

Currently, Indonesia is developing its human resources, where the human development index (HDI) is classified as medium. According to the United Nations Development Programme (UNDP) report in 2020, it ranked 107th out of 189 countries, still behind ASEAN countries, such as Malaysia, Brunei Darussalam, and Thailand. Based on the global knowledge index (GKI) report in 2021, it ranked 87th out of 154 countries with a score of 46.3, while the world average score was 48.4. One of the essential parts of knowledge measured is education, which consists of three sectors: pre-university, technical education and training, and higher education. Based on these, Indonesia ranked 95th for pre-university, 82nd for technical education and training, and 77th for the higher institution.

The low education level among the workforce, uneven development of information and communication technology (ICT) infrastructure, low innovation systems, and the suboptimal role of democratic-political institutions are the main challenges that Indonesian government needs to address. Infrastructure development in the cultural tourism sector still needs to be focused on Java and Bali, but it continues to receive less attention from the government, as well as low-quality of human resources. Unequal development between eastern and western parts, Java advancement, and the low quality of human resources have caused poverty and income inequality to become increasingly difficult to overcome. However, eradicating poverty and income inequality remains a top priority for development in Indonesia.

Indonesia has achieved relatively fast economic growth compared to most countries in the last 50 years (Hill, 2021), but it has not been followed by an improvement in the standard of living. Poverty reduction rate is slow, and according to data from the Central Bureau of Statistics, there were still 26.50 million poor communities in 2021. According to Yusuf and Sumner (2015), the number of rural communities tends to be higher than the figures released by the Central Bureau of Statistics, considering that poverty line used is the lowest in the world and lower compared to international poverty figures issued by the World Bank.

Meanwhile, looking at the trend of the Gini index development from 2011 to 2021, it is at 4.0, which is the threshold for high inequality. Wicaksono et al. (2017) suggested that one of the causes of income inequality in Indonesia is the uneven access to education, especially between urban and rural areas. The uneven development of ICT infrastructure causes some communities not to have equal opportunities to access information, which can exacerbate income inequality.

The government is committed and giving more attention to reducing income inequality and poverty, with the implementation of policies and programs to curtail these problems. These policies have continued since independence, during the New Order period, and up to the current presidency of Joko Widodo. During the Suharto regime, Duflo (2001) conducted research on the impact of the Inpres Elementary School development project program during the periods of 1973–1974 and 1978–1979. The results showed that the community who had received education during the program experienced a significant increase in wages.

The program recommended to improve the quality of human resources was promoted by President Joko Widodo in 2019, although infrastructural development was previously a top priority. However, before the program was fully implemented, the COVID-19 pandemic had started ravaging the country. The impact caused an increase in the number of communities living in poverty and income inequality.

Based on these, this research aims to determine the potential role of knowledge and cultural in reducing income inequality and alleviating poverty. This research is divided into five parts. The first is an introduction that briefly describes the problem examined, while the second part is a literature review relevant to this research. The third discusses the methodology used, and the fourth presents the findings, discussion, and analysis, while the fifth part presents the conclusion.

Literature Review

Knowledge Economy and Cultural Economy

The first research on knowledge economy was conducted by Machlup (1962) who measured the contribution of enlightenment and distribution to the national income of the USA. Subsequently, Choong and Leung (2021) attempted to divide the evolution of knowledge economy into two waves of thought, namely, the enlightenment economy and the second related to management. The development of thought in the first stage emphasizes the importance of considering knowledge aspects in policymaking. Including the role of knowledge in establishing a good educational system, boosting the community that relies on information in creating jobs, and promoting economic growth. The second wave of thought on knowledge economy is more focused on the creation, the role, and the implementation of management science, organizations, businesses, and innovation. This era is known as the concept of knowledge management.

Knowledge and cultural are closely related, and there is a pearl of general local wisdom in Indonesia. The creation of knowledge is divided into two dimensions, namely, explicit and tacit (Borges, et al, 2018). Explicit knowledge is formally taught enlightenment, while tacit is based on experience. Local wisdom is tacit knowledge passed down through generations in certain communities and is registered as cultural heritage in the Ministry of Education and Cultural.

Throsby (2020) introduced the concept of cultural capital as one of the important components of the economy, which provides the potential value, and is divided into two forms, tangible and intangible. Tangible cultural capital includes historical buildings and sites, while intangible form includes performing arts, music, and traditions, as well as the values and beliefs held by certain communities.

The cultural economy framework issued by UNESCO in 2009 (Sung, 2014) divided the cultural economy framework based on domains and those related. Cultural domains consist of activities, goods, and services produced, including intangible traditional heritage. The related domain is an extension of the cultural aspect including social events, recreational activities, and tourism that showcases traditional content. This division of domains is used as a guideline for collecting statistical data related to cultural and its relationship to economic and social outcomes.

Theory of Inequality, Poverty, and Development

Kuznets (1955) is considered the first person to explain the relationship between economic growth and income inequality. The research emphasized that, in the early stages of development, income inequality rises and then declines as economic growth increases. Then, it advances further due to trickle-down development through workers who earn higher average wages. The relationship between growth and inequality is depicted by an inverted U-shaped curve. Although Kuznets’ hypothesis is not supported by sufficient quality data, since its publication, it has motivated other research to test its feasibility. Some research has found empirical evidence for the U-shaped curve, while others have not found clear evidence of the existence of the Kuznets curve (Buceli, 2017).

An interesting research was conducted by Arshed et al. (2019) in ASEAN developing countries. The research confirmed the Kuznets hypothesis in the context of the relationship between education and income inequality as positive. A corresponding widespread enrollment in primary education tends to decrease inequality. At the first stage of enrollment, an increase in secondary school enrollment increases the level of income inequality, while further rise in academic engagement reduces the difference in earnings, indicating a negative index. The implication is increased income inequality because individuals who achieve high levels of education tend to demand higher wages compared to elementary and secondary school graduates.

McKnight (2019) attempted to review several research on the relationship between inequality and economic growth. Some of the results are ambiguous, while others show inconsistent. The relationship between inequality and economic growth is non-linear. The idea from Kuznets hypothesis led to a broader relationship, namely, poverty, growth, and inequality, known as poverty-growth-inequality triangle.

Bourguignon (2004) was the first to propose the concept of poverty-growth-inequality triangle. Absolute poverty and the alleviation are caused by two factors, namely, the growth effect and income distribution impact. The conclusion drawn from Bourguignon’s hypothesis is that the better the change in the average income distribution, the increase in income beyond poverty line. Similarly, even growth tends to reduce absolute poverty. Bergstrom (2020) investigated the role of income inequality in poverty alleviation using the World Bank data. The findings showed that changes in income inequality by increasing average income tend to reduce poverty rates. Meanwhile, there is a trade-off between income inequality and economic growth. The greater effect of changes in inequality and growth that alleviation poverty depends on the value of its elasticity. Therefore, policymakers need to examine the impact of various policies on inequality and growth.

Research Framework

For research purposes, a framework is constructed in Fig. 1. The foundation is knowledge economy scheme, which refers to the World Bank framework (Chen & Dahlman, 2004; Chen et al. 2006; WBI, 2007). Meanwhile, the cultural economy scheme refers to Throsby’s theory (2020) and the UNESCO framework (Sung, 2014). The goal of development is to reduce income inequality, alleviation poverty, and increase per capita GRDP. There is a brief explanation of knowledge and cultural economy scheme, which are the variables used in this research.

Fig. 1.

Research framework

Education

Education is the first essential part of knowledge economy scheme. Human resources is an essential component in development, and then, the quality is improved through education and training. Duflo (2001) emphasized that individuals participating in the education program at that time had better income. This research used the same indicators as the World Bank framework (Chen & Dahlman, 2004), namely, literacy rate, school enrollment ratio, and average years of research.

Information Technology

ICT are two inseparable concepts that refer to all technical equipment for processing and delivering information. The information technology encompasses all that is related to the process, manipulation, and management of data, while communication technology covers the aspects related to the use of tools to process and transfer data from one device to another. ICT infrastructure is known to be the backbone of knowledge economy. The proxy for information technology in this research used indicators of the percentage of telephone, the Internet, and computer usage.

Innovation System

The innovation system is indicated by the number of quality scientific publications, intellectual property rights, and the amount of foreign investment. Scientific publications and patents are some indicators of the innovation system. New findings in the form of products, services, processes, or systems; increase productivity; and more effective and efficient work methods ultimately provide added value to development as shown by an increase in GRDP per capita. Meanwhile, foreign investment provides learning and improves technology through knowledge transfer.

Democratic-Political Institutions

The role of democratic-political institutions as a source of efficiency and economic progress has long been a concern, which then affects development. Poor quality has the tendency of causing economic problems, and vice versa. Banik (2022) stated that democracy is important, because of the political right to determine economic, social, cultural, and ecological needs. This aspect guarantees citizens to provide alternative directions for development. This research used a democracy index to measure the effect of democratic-political institutions on development. There are three aspects of democracy, namely, civil liberties, political rights, and democratic institutions. The main function of democratic-political institutions is to carry out the mandate of the 1945 Constitution, which focused on the welfare of all Indonesian communities.

Cultural Tourism

Cultural tourism is one of Indonesia’s leading sectors, as the country is rich in cultural heritage, spreading from east to west, both tangible and intangible. Many traditions and legacies have been registered with UNESCO, and they tend to attract tourists, thereby contributing to the revenue of the local community. Hristova et al. (2018) stated that culturally fascinating cities tend to attract the community to visit and have an impact on the economy and development of a region.

Cultural Participation

Cultural participation is the involvement of the community in activities, such as traditional ceremonies, visiting custom sites, engaging in performances, and watching. A framework was successfully formulated by UNESCO in 2009 (Sung, 2014), which serves as technical guidelines for measuring cultural participation. This social indicator shows the role of a community in expressing its identity through cultural values. The higher the cultural participation, the stronger the community maintains its behavior, habits, artistic values, local wisdom, and social relationships.

Methodology

This research used a quantitative approach with secondary data from publications by the Central Bureau of Statistics, the Ministry of Education and Cultural, and the Ministry of Law and Human Rights. The unit of analysis was the entire 34 provinces in Indonesia, from 2011 to 2020. Three models were developed, each with dependent variables of income inequality, poverty rate, and gross regional domestic product (GRDP) per capita. Meanwhile, 20 indicators used were reduced or formed into 6 variables using the primary component analysis technique. Table 1 provides the operational definition of the variables.

Table 1.

Operational definition of research variables

Variable	Data source	Indicators
Income inequality	Central Bureau of Statistics: https://www.bps.go.id/	Gini index
Poverty	Central Bureau of Statistics: https://www.bps.go.id/	Percentage of number of poor communities
GRDP per capita	Central Bureau of Statistics: https://www.bps.go.id/	GRDP per capita
Education	Central Bureau of Statistics: https://www.bps.go.id/	Literacy rate, school enrollment ratio, average length of research
Technology	Central Bureau of Statistics: https://www.bps.go.id/	The ratio of HP users, Internet access, computer access
Innovation	Central Bureau of Statistics: https://www.bps.go.id/ Ministry of Law and Human Rights: https://pdki-indonesia.dgip.go.id/; Ministry of Education and Cultural: https://sinta.kemdikbud.go.id/	Intellectual property rights, number of scientific publications, foreign investment
Institutions	Central Bureau of Statistics: https://www.bps.go.id/	Civil liberties, political rights, democratic institutions
Tourism-cultural	Cultural Statistics-Ministry of Education and Cultural; Central Bureau of Statistics: https://www.bps.go.id/	Cultural heritage, GRDP of cultural and tourism services, average length of stay of foreign tourists, room occupancy rate,
Cultural participation	Central Bureau of Statistics: https://www.bps.go.id/	The ratio of watching live shows, the ratio of live show participation, visits to cultural places, participating in traditional ceremonies

In general, panel data regression analysis uses three approaches, namely, ordinary least square (OLS), random effect, and fixed effect methods. In the fixed effect method, the intercept in the regression is differentiated between individuals, because each individual is considered to have their characteristics. To differentiate the intercepts, dummy variables are used, and the model is referred to as the least squared dummy variable (LSDV) model (Juanda & Junaedi, 2012). LSDV is a panel data regression with the least square method by adding dummy variables that allow for differences in intercept values between individuals. Meanwhile, LSDV with interaction effects is the combination of independent with dummy variables, which allows for differences in the slopes of certain factors used (Juanda & Junaedi, 2012). This model is highly effective for data with short periods (Elhorst, 2010). The use of the LSDV with interaction effects provides new contributions to the development of models for analyzing knowledge and cultural economy’s impact on inequality, poverty, and development.

To determine the heteroscedasticity, the Wald test was employed, while the Wooldridge test was used to determine serial correlation. Multicollinearity problems are observed from the value of the variance inflation factor (VIF) or high correlation between independent variables. To solve heteroscedasticity and time serial (autocorrelation) problems, regression with the robust standard error technique was performed (Baltagi, 2005; Hoechle, 2007; Stock & Watson, 2008; Wooldridge, 2016; Zhu, 2013).

The first-panel data regression estimation was conducted to determine the effect of knowledge economy, cultural tourism, and participation on income inequality. The regression estimation used the LSDV approach with interaction effects. The dummy variable D-Java was used to distinguish between provinces in Java Island (Jakarta, West, Central, and East Java, Yogyakarta, Banten) and those outside the region. Meanwhile, D2 was used as an interaction variable to determine the difference in the effect of technology on income inequality between the inside and outside of Java. The regression estimation model used is written as follows.

$${\mathit{INE}}_{\mathit{It}}=\alpha +{\beta }_1{\mathrm{lnPDRBk}}_{\mathit{it}}+{\beta }_2{\mathit{EDU}}_{\mathit{it}}+{\beta }_3{\mathit{ICT}}_{\mathit{it}}+{\beta }_4{\mathit{INO}}_{\mathit{it}}+{\beta }_5{\mathit{INST}}_{\mathit{it}}+{\beta }_6{\mathit{HTR}}_{\mathit{it}}+{\beta }_7{\mathit{CP}}_{\mathit{it}}+{\beta }_8{D-Java}_{\mathit{it}}+{\beta }_9{D2}_{\mathit{it}}+{\epsilon }_i$$ 1

where

INE_it

Gini index in province i and period t

lnPDRBk_it

logarithm of GRDP per capita in province i and period t

EDU_it

education index in province i and period t

ICT_it

technology index in province i and period t

INO_it

innovation index in province i and period t

INST_it

democracy index in province i and period t

HTR_it

tourism-cultural index in province i and period t

CPit

cultural participation index in province i and period t

D-Java_it

a dummy variable where the value is 1 for provinces in Java (Jakarta, West and East Java, DI Yogyakarta, Banten) and 0 for other provinces outside Java, province i in period t

D2_it

the interaction variable between the technology variable and the D-Java dummy variable (D-Java*ICT) in province i period t

α,

intercept or the average Gini value when the independent variables are excluded

β₁.. β₉

the coefficient value of each independent variable to be estimated

Ɛit

error term

ln

logarithm to change the variable

Subsequently, to examine the role of knowledge and cultural tourism on reducing the percentage of the poor population, the estimation equation of LSDV regression with an interaction dummy was employed. In this model, interaction dummy D10 was used to determine the difference in the average effect of culturaltourism on poverty between provinces inside compared to those outside of Java Island. The estimation regression model is written as follows.

$${\mathit{POOR}}_{\mathit{It}}=\alpha +{\beta }_1{\mathrm{lnPDRBk}}_{\mathit{it}}+{\beta }_2{\mathit{EDU}}_{\mathit{it}}+{\beta }_3{\mathit{ICT}}_{\mathit{it}}+{\beta }_4{\mathit{INO}}_{\mathit{it}}+{\beta }_5{\mathit{INST}}_{\mathit{it}}+{\beta }_6{\mathit{HTR}}_{\mathit{it}}+{\beta }_7{\mathit{CP}}_{\mathit{it}}+{\beta }_8{D-Java}_{\mathit{it}}+{\beta }_9{D10}_{\mathit{it}}+{\epsilon }_i$$ 2

where

POOR_it

percentage of the number of poor community in province i and period t.lnPDRBk_it: logarithm of GRDP per capita in province i and period t

EDU_it

education index in province i and period t

ICT_it

technology index in province i and period t

INO_it

innovation index in province i and period t

INST_it

democracy index in province i and period t

HTR_it

tourism-cultural index in province i and period t

CPit

cultural participation index in province i and period t

D-Java_it

a dummy variable where the value is 1 for provinces in Java (Jakarta, West and East Java, Yogyakarta, Banten) and 0 for others outside, province i in period t

D10_it

interaction variable between the tourism-cultural and D-Java dummy (D-Java*HTR) in province i period t

α,

intercept or the average value of the percentage of poor community when the independent variables are excluded

β₁.. β₉

the coefficient value of each independent variable to be estimated

Ɛit

error term

ln

logarithm to change the variable

One of the indicators of successful development is the even increase in the GRDP per capita in all provinces. Therefore, the role of knowledge and cultural tourism factors on GRDP per capita was also examined. Using the same model but replacing the dependent variable with GRDP per capita, the estimation regression model is written as follows.

$${\mathrm{lnPDRBk}}_{\mathit{it}}=\alpha +{\beta }_1{\mathit{EDU}}_{\mathit{it}}+{\beta }_2{\mathit{ICT}}_{\mathit{it}}+{\beta }_3{\mathit{INO}}_{\mathit{it}}+{\beta }_4{\mathit{INST}}_{\mathit{it}}+{\beta }_5{\mathit{HTR}}_{\mathit{it}}+{\beta }_6{\mathit{CP}}_{\mathit{it}}+{\beta }_7{\mathit{CP}}_{\mathit{it}}+{\beta }_8{D4}_{\mathit{it}}+{\beta }_9{\mathrm{i}.\mathrm{provincialcode}}_{\mathit{it}}+{\epsilon }_i$$ 3

where 

lnPDRBk_it

logarithm of GRDP per capita in province i and period t

EDU_it

education index in province i and period t

ICT_it

technology index in province i and period t

INO_it

innovation index in province i and period t

INST_it

democracy index in province i and period t

HTR_it

cultural tourism index in province i and period t

CPit

cultural participation index in province i and period t

D4_it

interaction variable between political-democratic institutions and D-Java variables (D-Java*INST) in province i period t

i.provincialcode

variable provincial code in province i and period t

α,

intercept or the average value of GRDP per capita when the independent variables are excluded

β₁.. β₉

the coefficient value of each independent variable to be estimated

Ɛit

error term

ln

logarithm to change the variable

D4 is an interaction variable used to observe the difference in the average effect of political-democratic institutions inside and outside of Java. D-Java model was not used, but instead, i. provincialcode was employed. The use of dummy variables helps understand the heterogeneity of the effect of independent factors for each province (Juanda & Junaedi, 2012).

Result and Discussion

Table 2 is a summary of the best model and heteroscedasticity test from each panel data regression. Based on the Hausman test where H0 = RE and H1 = FE, the result showed that the Prob > chi2 value is 0.0000 indicating that the best model used a fixed effect. In the fixed effect model (FEM), the intercept is distinguished between individuals because each is considered to have its characteristics. To distinguish the intercepts between entities and slopes for certain selected variables, the LSDV model with interactions is used (Juanda & Junaedi, 2012).

Table 2.

Summary of the best model selection test

Model	Best model selection test
OLS	The OLS model is ignored because it assumes a constant slope and intercept
Random effect model (REM)	Based on the Hausman test model, the value of Prob > chi2 = 0.0000, H0 = RE is rejected, then the best model employed the FEM
FEM	From the Wald test, the symptoms of heteroscedasticity are observed with the regression results causing bias. Modified Wald test for groupwise heteroscedasticity in fixed effect regression model, including H0: sigma(i)^2 = sigma^2 for all i, chi2(34) = 4374.56, and Prob > chi2 = 0.0000. To overcome heteroscedasticity, robust standard error is used (Baltagi, 2005; Hoechle, 2007; Stock & Watson, 2008; Wooldridge, 2016; Zhu, 2013)
LSDV with interaction	LSDV with interaction allows for different intercepts and slopes between entities. The emphasis of the analysis is on regional aspects; hence, this model does not include time-invariant effects

Income Inequality

Table 3 presents a summary of the LSDV regression estimates on the effect of knowledge economy and cultural tourism on income inequality. D-Java is a dummy variable, where the value is 1 for provinces in Java Island (Jakarta, West, Central, and East Java, Yogyakarta, Banten) and 0 for those outside the region. D2 is the interaction variable between D-Java and the technology variable.

Table 3.

LSDV regression on income inequality

Inequality	Coef	Robust Std. Err	t	P > t	[95% Conf. Interval]
GRDP per capita	−0.010***	0.003	−2.64	0.009	−0.018	−0.002
Education	−0.007***	0.001	−4.69	0.000	−0.010	−0.004
Technology	0.008***	0.002	3.12	0.002	0.002	0.013
Innovation	−0.012***	0.001	− 6.27	0.000	−0.016	−0.008
Institutions	−0.006***	0.001	−3.45	0.001	−0.009	−0.002
Tourism-cultural	0.013***	0.002	6.18	0.000	0.009	0.017
Cultural participation	−0.011***	0.001	−5.66	0.000	−0.014	−0.007
D2	0.018***	0.002	6.74	0.000	0.013	0.024
D-Java	0.047***	0.005	9.42	0.000	0.037	0.057
_cons	0.462	0.041	11.23	0.000	0.381	0.543
Observations			335
R-squared			0.4075
Number of provincial code			34

Robust standard errors in parenthesesRobust standard errors in parentheses

*P < 0.1; **P < 0.05; ***P < 0.01

Based on the LSDV regression, the variables that significantly affect the reduction of income inequality are GRDP per capita, education, innovation, democratic-political institutions, and cultural participation. Meanwhile, technology and cultural tourism have a significant positive effect. The greater the access to technology and the higher income growth in the cultural tourism sector, the higher the income inequality. Contrarily, the interaction variable between D-Java and technology usage (D2) shows a positive value, meaning that the interaction of the D2 variable indicates greater access to technology in provinces inside the Island compared to those outside. The coefficient of the D-Java shows a positive value, indicating that there is a difference in the average level of income inequality between the provinces. Provinces inside Java Island have higher income inequality compared to those outside.

The first variable that significantly affects the reduction of income inequality is GRDP per capita, presented by the regional aggregate output divided by the population. Jakarta and East Kalimantan have relatively the highest GRDP per capita compared to other provinces in 2020, but they also have relatively high-income inequality. In contrast, Yogyakarta has the highest Gini index, but a low GRDP per capita. Based on the relationship between GRDP per capita and income inequality, there is no consistent trade-off between these variables.

Education significantly reduces income inequality, as proven by some research. Coady and Dizioli (2017) found that the advancement of education continuously reduces income inequality, especially in developing countries. Qazi et al. (2016) found that higher education reduces differences in earning in the long run. As stated by Nabassaga et al. (2020), equal access and opportunity to higher education reduce income inequality.

The LSDV regression shows that technology causes income inequality because the development of ICT infrastructure is not evenly distributed. Java Island as the center of growth has better ICT infrastructure compared to areas outside the region. Indonesia as an archipelagic country poses a challenge in building an Internet network evenly. This is evidenced by the value of the dummy variable D-Java and the interaction effect of D2, showing that the development of technology infrastructure inside Java Island is better than in provinces outside, but still increases income inequality (Ndoya & Asongu, 2022). Uneven access to technology occurs between rural–urban areas, Java Island, and areas outside the region, which is one of the factors causing difference in earning.

Innovation is capable and significantly influential in reducing income inequality. The innovation proxy includes scientific publications, intellectual property rights, and foreign direct investment (FDI). Scientific publications are academic activities that produce research works not only in the field of science but also publications in the area of community service. The Tridharma (three pillars) of higher education is an obligation for every educator throughout Indonesia. The intellectual involvement of various campuses and higher education to directly engage with local communities and provide contributions, thoughts, and training increase knowledge and skills required, thereby increasing the level of income. Meanwhile, FDI, besides the influx of capital, also involves the transfer of knowledge, which absorbs workforce and generates income.

The aspects of democracy-political institutions are capable and significantly influential in reducing difference in earning, both through political decisions and policies. The function of political institutions is to enforce laws mandated by both the 1945 Constitution and Pancasila as the basis of Indonesian state. As found by Keppel (2022), good political control over development is necessary. Supervision is not only over the process but also the allocation of equal opportunities to existing resources. The role of democracy-political institutions in reducing income inequality is inneglegible, as it is the main function and duty of democratic institutions, which is to supervise the development process.

Tourism-cultural is one of the main economic sectors that generates foreign exchange, contributes to GRDP, and provides employment opportunities. Bowitz and Ibenholt (2009) found that cultural heritage investments and other forms of traditions have economic benefits, not only in terms of consumption but also in the aspect of job creation. Concerning economic development, cultural heritage is closely related to tourism.

The regression results show that an increase in tourism-cultural increases income inequality, indicating that tourism management is still dominated by large capital owners, such as hotel management and other supporting facilities located around the sites. While local communities do receive income and job opportunities, they receive a much smaller share of income compared to large capital owners. However, the priority for developing and supporting infrastructure for the tourism-cultural sector is still focused on Java and Bali. Currently, the government is developing the super priority tourism destination areas, such as West and East Nusa Tenggara, to become target areas for tourism development plans. These two provinces have natural beauty and cultural heritage assets that tend to attract tourist. The development plan for these areas is expected to have a domino effect on regional scheme.

Based on the regression estimation, cultural participation reduces income inequality. Zwell (2000) suggested that traditional values and beliefs increase competence, enhance skills, and generate revenue.

Among the 34 observed provinces, Yogyakarta has the highest Gini index, while Bangka Belitung has the lowest. Based on the research conducted by the Central Bureau of Statistics and the Regional Planning Agency of Yogyakarta in 2020, the lowest expenditure group only completed elementary school education, while the highest expenditure group had university education. Furthermore, there is also inequality between sectors.

Two other provinces that have income inequality with a Gini index above 4.00 are Gorontalo and West Java. Based on the Central Bureau of Statistics, Gorontalo’s Gini index increased from 0.403 to 0.42, between 2018 and 2019, but the percentage of revenue generated in the rural community decreased during the same period from 16.81 to 15.52% (Fig. 2). Income of the urban community increased more than that of the rural. West Java is a province with moderate education indicators, as the literacy rate is 98.63%, school participation rate is 71.17%, and average school length was 8.55 years. Half of the workforce according to education are university graduates working in urban areas, while the primary school certificate holder work in rural communities. The education inequality in this province causes income inequality between urban and rural areas. The percentage of the poor is 9.58% in rural areas and 5.98% in urban settlements.

Fig. 2.

Gini index of provinces in Indonesia in 2020

Poverty

The structure of poverty and income inequality is divided into rural and urban areas. Income inequality tends to be greater in urban areas, while poverty is more prevalent in the rural. From Fig. 3, Jakarta does not have a rural poverty line as it is the capital of the country.

Fig. 3.

Percentage of poor community and poverty lines in urban–rural areas in provinces in 2020

Provinces in the eastern region such as Papua, West Papua, East Nusa Tenggara, and Maluku have the highest percentage of poor communities compared to others. Meanwhile, the five provinces with the lowest percentage of the poor community are Bali, Jakarta, South Kalimantan, Bangka Belitung Islands, and Central Kalimantan

Based on the LSDV regression, all variables except cultural participation significantly affect the reduction of poverty (Table 4). Meanwhile, D-Java variable is not significant, indicating that poverty is the same in provinces on Java Island and provinces outside. The D10 dummy variable with the interaction effect between D-Java and tourism-cultural shows a positive and significant effect. The D10 number indicates that there is a difference in the average tourism-cultural sector between Java Island and provinces outside concerning poverty. The D10 interaction dummy shows that tourism-cultural in provinces inside causes a higher level of poverty compared to those outside. For example, Yogyakarta is one of the provinces on Java Island that is a major tourist destination, but its poverty rate is relatively high and is the highest compared to those inside. While compared to Bali province, the difference is visible, as it is the province with the lowest poverty rate, and one of the world’s cultural tourism destinations.

Table 4.

LSDV regression on poverty

Poverty	Coef	Robust Std. Err	t	P > t	[95% Conf. Interval]
GRDP per capita	−1.640**	0.690	−2.37	0.018	−299	−0.280
Education	−1.035**	0.332	−3.12	0.002	−1.689	−0.381
Technology	−0.725*	0.393	−1.84	0.066	−1.500	0.049
Innovation	−0.001*	0.001	− 2.54	0.012	−0.001	−0.000
Institutions	−1.887***	0.334	−5.64	0.000	−2.545	−1.229
Tourism-cultural	−0.638*	0.339	−1.88	0.061	−1.305	0.029
Cultural participation	0.514	0.335	1.53	0.126	−0.144	1.173
D10	1.885***	0.443	4.25	0.000	1.013	2.758
D-Java	−0.630	0.850	−0.74	0.459	−2.303	1.042
_cons	28.863	7.159	4.03	0.000	14.778	42.948
Observations			335
R-squared			0.4177
Number of provincial code			34

Robust standard errors in parentheses

*P < 0.1; **P < 0.05; ***P < 0.01

GRDP per capita reflects the added value created by each individual from production activities. An increase in added value per person is expected to improve the overall living standards through increased income and reduced poverty. Research has shown that a 10% increase in GRDP per capita reduces poverty rate by 16.40%.

It has also shown a negative relationship between education and poverty. A good education indicates better human resources, which increase income and reduce poverty. Programs to improve human resource quality through education are one of the keys to overcoming poverty in Indonesia. Hofmarcher (2021) found significant economic effects of an additional year of education, not only in increasing income but also in reducing the likelihood of basic household needs shortages, such as family tuition fee and healthcare.

As an archipelagic country, the role of technological infrastructure is crucial. The country has built the Palapa Ring as a network infrastructure; hence, the community has equal access to the Internet. Although uneven access to technology leads to income inequality, as mentioned in the previous research, the opportunity for access to information through technology helps alleviate poverty. Raeskyesa and Erica (2019) showed that information technology has a positive and significant effect on poverty. The provision of information technology infrastructure, enabling easy access to information is not enough, needs to be supported by human resources capable of mastering ICT. Sharifah Rohayah et al. (2019) stated that the technological infrastructure gap between urban and rural areas causes impoverishment among the rural populations due to a lack of access to information technology.

In the KOMINFO (Ministry of Communication and Informatics) Annual Report (KOMINFO, 2020), the base transceiver stations (BTS) have been built in 1682 villages and sub-districts to provide ICT services to remote communities, especially in the frontier, outermost, and underdeveloped areas (3 T), which is part of the total of more than 550,000 BTS scattered throughout Indonesia. The digital economy has proven to be able to support the national economy. During the economic recession caused by the COVID-19 pandemic, the economy resulted in negative growth of − 5.32% in the second and − 3.49% in the third quarter of 2020. The digital economy sector was able to slow down the decline in economic growth by contributing 10.61% yearly from the information and communication sector.

Although innovation has a significant effect on poverty reduction, its impact is very small when viewed from its coefficient, which is only 0.001. This means that a 100% increase in innovation only reduce poverty by 0.1%. This figure shows that innovation level is still low, while novel products in the form of research and intellectual property rights have not been further processed for poverty reduction programs. FDI, as one of the proxies for innovation, help reduce poverty; besides the transfer of technology to rural areas and fund flow, it also provides opportunities for local communities.

The role of democratic-political institutions has proven to significantly reduce poverty rate, although the regression estimation proves that the institutions have the strongest coefficient compared to other variables. This indicates that institutions carry out their mandate, including monitoring the development programs for creation of wealth in the community.

Tourism-cultural is able to contribute to poverty alleviation, by providing opportunities for communities to develop local products and small-scale businesses, which generate revenue and create jobs. Indonesia is known to rich in cultural heritage from Sabang to Merauke, both tangible and intangible. Many tangible and intangible cultural heritages have been registered with the international organization UNESCO; most recently, the pantun and gamelan have been recognized as intangible. Research has shown that cultural heritage sites through tourism reduce poverty (Quadri-Barba et al., 2021; Salama & Mshenga, 2018). Local communities involved in and benefiting are highly dependent on the number of tourist visits, and a decrease in tourism has an impact on their income. As proposed by Muringa et al. (2022), national institutions establish partnerships with local communities where they receive a fixed portion of tourism activities and are supported during seasons of low revenue.

Cultural participation does not have a significant effect on poverty alleviation. Cultural has two sides of significance, and with proper management, it tends to enhance social capital. However, cultural is a behavior or attitude of accepting concepts in their original form and feeling satisfied. This behavior or attitude is called cultural of poverty (McDermott & Vossoughi, 2020). Although this concept is controversial, it still occurs in a small portion of the population in rural areas of Java Island and other regions.

Based on the regression estimates, D10 shows that poverty from the tourism-cultural sector is higher on average in provinces inside compared to those outside of Java. For example, Bali, West, and North Sumatra are among the provinces that are centers of tourism and cultural. Their poverty rates are relatively low at 3.78%, 6.28%, and 8.75%, respectively, in 2020. However, poverty rates in the provinces in Java, which are also centers of tourism and cultural, such as Yogyakarta and Central and East Java, are higher than those in Bali and West and North Sumatra, at 12.28%, 11.41%, and 11.09% of the population, respectively.

GRDP Per Capita

GRDP per capita is the aggregate production of a region divided by its population, and it is one of the indicators of regional development. The higher the GRDP per capita, the better its impact on development and welfare. Based on the previous discussion, an increase in the concept has a significant effect on reducing income inequality and poverty. Table 5 shows that all economic, knowledge, and tourism-cultural variables have a positive and significant effect on GRDP per capita. Meanwhile, D4 is a dummy interaction variable between D-Java and democracy-political institutions. The regression coefficient of D4 indicates that there is a difference in the average effect of democracy-political institutions between provinces inside Java compared to those outside. The role of democracy-political institutions inside the province is smaller compared to the effect on the increase of GRDP per capita outside Java. The emphasis of this research is on spatial analysis; by including the i.provincialcode variable, the heterogeneity of the effect of independent factors is found to be absorbed by the dummy categories (Juanda & Junaedi, 2012). Aceh province is not displayed in the regression results because it is the base province and serves as a reference for dummy variable categories.

Table 5.

The results of regression estimation of GRDP per capita

GRDP per capita	Coef	Robust Std. Err	t	P > t	[95% Conf. Interval]
Education	0.065***	0.011	5.75	0.000	0.042	0.087
Technology	0.045***	0.005	8.02	0.000	0.034	0.056
Innovation	0.019***	0.004	4.45	0.000	0.010	0.027
Institutional	0.009*	0.004	1.94	0.054	−0.000	0.018
Tourism-cultural	0.012**	0.006	1.99	0.048	0.000	0.024
D4	−0.016**	0.006	−2.46	0.015	−0.029	−0.003
Province
North Sumatra	0.291	0.025	11.64	0.000	0.241	0.340
West Sumatra	0.101	0.025	3.95	0.000	0.051	0.152
Riau	1.088	0.041	26.09	0.000	1.006	1.170
Jambi	0.508	0.029	17.34	0.000	0.450	0.566
South Sumatra	0.393	0.031	12.64	0.000	0.332	0.455
Bengkulu	−0.114	0.026	−4.36	0.000	−0.166	−0.063
Lampung	0.168	0.032	5.24	0.000	0.105	0.231
Bangka Belitung	0.459	0.037	12.26	0.000	0.385	0.533
Riau islands	105	0.029	35.86	0.000	0.998	1.114
Jakarta Capital Special Region	1.478	0.032	45.49	0.000	1.414	1.542
West Java	0.046	0.041	1.13	0.261	−0.034	0.127
Central Java	0.126	0.040	3.15	0.002	0.047	0.206
Special Region of Yogyakarta	−0.170	0.027	−6.27	0.000	−0.224	−0.117
East Java	0.441	0.042	10.35	0.000	0.357	0.525
Banten	0.279	0.031	8.92	0.000	0.217	0.340
Bali	0.268	0.034	7.87	0.000	0.201	0.335
West Nusa Tenggara	−0.121	0.048	−2.48	0.014	−0.217	−0.025
East Nusa Tenggara	−0.537	0.034	−15.64	0.000	−0.604	−0.469
West Kalimantan	0.183	0.040	4.51	0.000	0.103	0.263
Central Kalimantan	0.350	0.031	11.24	0.000	0.289	0.411
South Kalimantan	0.209	0.035	5.86	0.000	0.138	0.279
East Kalimantan	1.562	0.036	43.38	0.000	1.491	1.633
North Kalimantan	1.170	0.037	31.38	0.000	1.096	1.243
North Sulawesi	0.206	0.030	6.67	0.000	0.145	0.267
Central Sulawesi	0.318	0.057	5.58	0.000	0.206	0.430
South Sulawesi	0.322	0.039	8.27	0.000	0.246	0.399
Southeast Sulawesi	0.305	0.030	10.01	0.000	0.245	0.365
Gorontalo	−0.081	0.034	−2.36	0.019	−0.148	−0.013
West Sulawesi	0.058	0.037	1.53	0.126	−0.016	0.132
Maluku	−0.454	0.025	−17.70	0.000	−0.504	−0.403
North Maluku	−0.191	0.027	−6.96	0.000	−0.245	−0.137
West Papua	1.037	0.034	29.83	0.000	0.969	1.106
Papua	1.055	0.085	12.35	0.000	0.887	1.223
_Cons	10.024	0.030	332.31	0.000	9.964	10.083
Observations			335
R-squared			0.9932
Number of provincial code			34

Robust standard errors in parentheses

*P < 0.1; **P < 0.05; ***P < 0.01

Education is the strongest determinant in increasing GRDP per capita, followed by technology, innovation, tourism-cultural, and democracy-political institutions. Although all factors contribute positively to the increase in GRDP per capita, the increase in each independent variable’s contribution is relatively negligible. The contribution of innovation only provides 1.9% to GRDP per capita for every 10% increase in innovation. Meanwhile, democracy-political institutions are even smaller, at only 0.09% to GRDP per capita for every 10% increase in the role and political decisions for regional development.

The coefficient of the province category variable describes the heterogeneity of the average PDRB per capita difference compared to the reference dummy variable, namely, Aceh province. Provinces that have a value of ( −) indicate the average GRDP value of per capita. While the sign ( +) shows that the average value of GRDP per capita is higher compared to the provincial reference.

The Special Capital Region of Jakarta is the province with the highest GRDP per capita, followed by the East Kalimantan, Riau Islands, Riau, and West Papua. The Special Capital of Jakarta is the center of economic, business, and trade activities. East Kalimantan, Riau Islands, Riau Province, and West Papua are the provinces with the largest revenues from natural resources, petroleum, and natural gas. While those with the lowest GRDP are East Nusa Tenggara, Maluku, West Nusa Tenggara, North Maluku, and West Sulawesi.

Although West Papua Province has a high per capita GRDP, in terms of the percentage of rural community, it is the second largest after Papua (21.7%), with the level of income inequality, which is 0.382 in 2020. Even though it is endowed with natural resources in the form of forest, minerals, and marine products, the residents are not satisfied.

There is an economic leak, a backwash effect where income continues to be enjoyed by capital owners and development processes are taken out of the region. The provinces of Papua and West Papua, in terms of the amount of education, technology, and innovation, are relatively low compared to several other regions in Java (Fig. 4). West Papua has the second lowest innovation index after West Sulawesi, which is only 0.28 points. Several other provinces with low innovation indexes are Bangka Belitung, Jambi, East Nusa Tenggara, and Central Kalimantan. The low innovation system index in Indonesia is a challenge to continue to improve as a prerequisite for developing countries to become developed while taking into account the problems of inequality and poverty.

Fig. 4.

Education, technology, and innovation index several provinces in Indonesia

There are several policy implications for knowledge and cultural-based development. First is equitable education not only between villages, but also among the provinces of western and eastern Indonesia. The second involves accelerating the development of technological infrastructure, especially in the foremost, outermost, and disadvantaged areas; hence, all residents have the same access to information. The third is increasing the regional innovation system based on knowledge and technology that is implemented not only to increase GRDP per capita, but to also overcome the problem of inequality and poverty. Furthermore, it is necessary to strengthen the role of political-democratic institutions both at the central and regional levels; hence, development aimed at the welfare of the community is achievable. Finally, policies are needed to improve the quality of human resources and develop infrastructure as supporting facilities for traditional activities, not only in Java and Bali, but also in provinces that have the great tourism potential. The role of community participation is aimed at enhancing social capital, which contributes to the development and reduces income inequality.

Conclusion

This research ascertains the role of knowledge and cultural in reducing inequality and alleviating poverty. The results show that education, innovation, political-democratic institutions, and cultural participation reduce income inequality, while tourism technology and cultural increase differences in earning. This is due to uneven access to information through technology between residents inside and outside of Java. The largest source of income from the cultural tourism sector is still enjoyed by high capital owners compared to residents. All economic and cultural variables except participation affect poverty. Cultural participation has two sides of significance, and when managed properly, it generates social capital, but it becomes a habit known as a cultural of poverty where the community is satisfied with their current position.

Equitable development should be a priority, and the results showed that all economic and cultural variables had a positive and significant effect on GRDP per capita. Increasing GRDP per capita tends to overcome the problem of inequality and poverty. However, equity and improvement of knowledge and cultural are key factors in the success of the development. There is no backwash effect, in which the results of development are not enjoyed by the community. The implication is that strong policies are needed at the central and regional levels; hence, equity in knowledge and cultural economy should be implemented.

These results complement the previous research, but with difference in the aspect of supporting a framework or model of knowledge and cultural-based development which is relatively novel and similar to the case of Indonesia. The research on knowledge and the cultural economy and their impact on development are still few. Most research in the context of knowledge and cultural is carried out separately using one or two determinants. In this research, the determinants of political-democratic institutions are included as new variable of knowledge economy, while that of cultural tourism and participation as variable of the cultural economy.

This research has limited data, especially in relation to cultural economy at the regional level. More in-depth research is required with a field research approach using primary data. Cultural economics should be researched from the perspective of trust, the spirit of cooperation, beliefs, values, and community preferences. In global level research, it is possible to obtain larger data which include the time effect in the research model; hence, knowledge-based development has unlimited space because it is very dynamic.

Declarations

Conflict of Interest

The authors declare no competing interests.