Dualism and Development

Abstract

Historical ‘source’ of dualism in Indonesia is first discussed, with a particular emphasis on the relevance of Boeke’s concept of dualism. Despite the presence of various policies to address the issue, the inequality between regions in the country is large by international standard. The productivity gap of micro-small enterprises and large businesses are also stark. As policy goals tend to be ambiguous, some proposed measures are not well received by the MSMEs.

Conversation

• A: “We LAUD the achievements in macro indicators.”

• B: “But the aggregate picture conceals inequalities that plague the DUAL system.”

• A: “To reduce interregional inequalities , we have allocated more funds and built infrastructure in periphery REGIONS.”

• B: “That is an incomplete conjecture as it IGNORES the role of agglomeration forces and the prevailing institutional arrangement.”

• A: “Although inequalities remain and dualism persists, we certainly do STRUT our policies when it comes to bringing higher growth and lower poverty .”

• B: “The effectiveness of policies and how they are received by the public is also determined by the level of TRUST as part of social capital .”

• A: “The STATE has been actively engaged in programs to support small businesses during a crisis.”

• B: “In helping the small businesses we need to understand and have a TASTE of what we miss about the complex relationships between policies and institutions and the behavioral insights of the programs.”

A Historical Source

By choosing the appropriate assumptions of the institutional framework in underdeveloped countries, Boeke (1953) provided a systematic analysis of development through his ‘dualistic theory,’ also known as the ‘social dualism.’ He elaborated the concept by using numerous examples of cases in Indonesia based on detailed descriptions of the colonial society during the Dutch East Indies period between 1910 and 1929.¹ Boeke’s investigations were stimulated by the problems of declining welfare of the Indonesian population. He lamented that, “social dualism is the clashing of an imported social system with an indigenous social system of another style. Most frequently the imported social system is high capitalism. But it may be socialism or communism just as well, or a blending or them” (Boeke, 1953, p. 4). He saw a co-existence of such two systems in Indonesia—and also India—and argued that the resulting dualistic economy would be a permanent feature of the countries’ economic structure (Boeke, 1961).

Boeke analysis essentially implies a culture-related backward-bending supply curve for labor, driven by ‘limited wants’ as well as inelastic demand behavior among the indigenous social system. Bearing in mind that institution differs across cultures, he argued that it cannot be easily implemented elsewhere in a top-down manner. Hence, a distinct economic approach is needed for the Dutch-Indonesian colonies where the indigenous population responded otherwise than expected to the economic incentives set by the western colonial institution. The implication is, if a specific policy or institutional framework is imposed onto another culture, it will only be partially adopted by the respective population.

Boeke’s work along with his concept of economic dualism was among the earliest attempts to diagnose the causes underlying the dualistic characteristics and their implications for the developmental process in underdeveloped countries. It is this dualism that served as one of the necessary reasons for the persistence of informal economy in many developing countries. It is also one that led to the marginalization and exclusion of social groups from formal economic activities (Clement, 2015). Boeke’s theory laid the foundations for a series of concepts and other theories on socio-economic duality that came later, e.g., the work of Lewis (1955), Hirschman (1957), and Geertz (1963), the labor surplus model of Fei and Ranis (1964), including the relevant comments by Eckaus (1965) and Dixit (1970).

Not all scholars, however, agreed with Boeke. Among several issues causing disagreements concerns the static nature of his theory. Boeke clearly believed that the dual society in Indonesia is “permanent at least within a measurable distance of time.“ He also wrote “that a precapitalistic society is driven further and further away into an exchange economy for which it is not fitted and which it cannot master.” In his view, the static feature of his theory could be used in the context of pre-capitalistic sector. Higgins (1966) countered that if they are static in the sense that per-capita incomes are not rising and capital accumulation proceeds at slow rates, not only the description bears little resemblance to the theoretical abstraction of static economy but it also ignores the fact that a non-growing economy is actually a special case of economic dynamics. Itagaki (1960) made further arguments that since no transitional process is considered, the static nature of Boeke’s theory undermines the dynamic aspects of the problem of structural changes. In his view, what should be important to analyze—yet missing in Boeke’s analysis—is the critical role of ‘colonial capitalism’ that caused the ‘unequalizing factors’ suggested by Myint (1964), and the ‘backwash effect’ expounded by Myrdal (1957), both of which could either preserve or reinforce the initial dualism by hindering further development of the indigenous economic sectors. But other than highlighting the important role of state and economic nationalism to eliminate those forces, Itagaki did not offer specific solutions. Nonetheless, Boeke's failure to recognize the dynamic elements in Indonesian life has been one of the sources of criticisms against his theory.

Boeke’s concept was used not only by the colonial authority but also by the Indonesian government in a post-independence period to serve as the basis of a protectionist policy. The goals were to prevent social disintegration and preserve social coherence. Sadli (1957) concurred with Boeke’s assessment: “It cannot be denied that Boeke's descriptions of the Eastern (or perhaps only the Javanese) village are in many instances true. Boeke knows a lot about the Javanese village life in the colonial period.” On the other hand, he was also critical about Boeke’s idea being thought as a new theory. In his words “Boeke explained much about the social impact of a high capitalistic system upon a precapitalistic society but this can hardly be called a separate economic theory.”

The concept of dualism continues to be relevant in today’s environment, as it is closely associated with inequality.² Not that Boeke was entirely right. He was not. It was Lewis (1955) who provided a thorough analysis and showed that the movement of labor from traditional agriculture to modern industrial activities is what makes growth and development possible. Hence, dualism is not static. The intersectoral labor migration is the source or engine of economic development. However, the notion that the gap between economic and social organizations, that is, the gap between a relatively modern economy performed by large businesses and the traditional indigenous economy of small businesses continues to persist, is hard to deny. So is the development gap between core regions and periphery. The coexistence of such contrasting economic and social organizations is a fundamental aspect of today’s growth process in many countries. The trend of income and wealth inequality has not been too encouraging in almost everywhere around the world.

Globally, over the course of twentieth century, income inequality within countries followed a more-or-less U-shape pattern, but the trend in most countries have been worsening. Unlike what was assumed in Boeke’s theory, this occurred even with the traditional sectors experiencing a process of transition. On the other hand, Boeke’s prediction that dualism will be “permanent at least within a measurable distance of time” does not seem to be off target. The overall picture of within-country dualism continues to exist, albeit with certain fluctuations. Income inequality persists and continues to be high in some countries, and the performance gap between large and small businesses (where the latter are generally traditional) remains large.

For a country like Indonesia, given the size and its archipelagic nature, inequality between regions including between rural and urban area is particularly important. Discussed in the next section, the country’s interregional inequality is indeed large by international standard.

Another reason why it is important to address the issue of dualism and inequality is because of its intricate relations with growth. The classical economic thinking (Kaldor, 1956) posited that the rich have a higher marginal propensity to save than the poor so that a higher degree of initial income inequality tends to result in higher savings and investment, hence higher growth. But the modern view asserted that greater inequality causes a lower growth through the following mechanisms (Nissanke & Thorbecke, 2007): less secured property due to unproductive rent-seeking activities, uncertainty due to diffusion of political and social instability, disincentives among the rich to invest due to redistributive policies to address inequality, underinvestment by the poor caused by imperfect credit markets, and higher fertility associated with smaller income share of the middle class. The United Nations summed up its recent report with the following message: “High or growing inequality not only harms people living in poverty and other disadvantaged groups. It affects the well-being of society at large. Highly unequal societies grow more slowly than those with low inequality and are less successful at reducing poverty. Without appropriate policies and institution, inequalities in outcomes create or preserve unequal opportunities and perpetuate social divisions. Rising inequality has created discontent, deepened political divides and can lead to violent conflict” (United Nations, 2020). Whichever forces at work, high inequality and greater dualism tend to worsen the growth prospect.

Having recognized the importance of addressing the issue of dualism and inequality, it is imperative to understand the mechanisms of how they emerge and affected by policies. Many of the sources behind dualism are institutional, not direct economic in nature. Local characteristics, culture, tradition, and social capital are parts of the institution that shape the backward-bending supply curve and the inelastic demand behavior of the traditional sectors. The incompatibility of the indigenous (receiving) culture and the colonial (giving) culture reflects the clash between imported social system with indigenous social system. To the extent that economic development must be understood as larger socioeconomic processes, not only as economic processes, recognizing the importance of local characteristics and culture is imperative. When imposed policies are designed to be compatible with the local culture and characteristics, they have a chance to work and be adopted by the local population.

From this perspective, the consequence of Boeke’s conjecture about the permanence of dualism does not have to be far-reaching in the sense that it excludes the possibilities of development and rejects any policy measures. We do not have to be defeatist. As long as the plan and policies are carefully designed by considering the prevailing institutional arrangements, we do not need to accept what Boeke lamented, “I will expose no plans, except to stress the need for a village restoration.” Yet, his position on how to conduct the restoration/policy is hard to disagree with. He specifically emphasized the need to conduct the restoration democratically and to have local leaders with strong sense of social responsibility. In his words: “This restoration will not take place through a revival of the rural gentry, but must follow more democratic ways. New leaders must spring from the small folks themselves, and must be accompanied by a strong feeling of social responsibility in the people themselves.” Thus, between Boeke’s position on the stagnancy of dualism—where no policy will work except village restoration—and the view that external interventions are needed to modernize the traditional sectors, there is a middle position. This book takes such a position.

Although dualism and inequality emphasized in concern with spatial dimension and the gap between small and large businesses, other forms of inequality are no less important. The bulk of discussions in the following chapter is on the evidence of inequality in Indonesia. To the extent various policies intended to reduce dualism have been promulgated, the observed high inequality implies that there is indeed a gap between policy and outcome. On the other hand, given the heterogeneity and size of the country, some degree of dualism and inequality should have been expected.

Evidence of Dualism

Being the world’s fourth most populous and diverse archipelago with 300-plus ethnic groups and more than 700 spoken languages, Indonesia is one of the highly heterogenous countries in the world. It is ranked 175th (out of 218 countries) in language fractionalization and 161st in ethnic fractionalization (Alesina et al., 2002). Although much of Indonesia’s diversity is explained by its peoples’ differing histories rather than their isolated development (Geertz, 1963), the inherent dualism is likely high in a country with such an attribute. No matter what the aggregate trends based on macro data show, features of dualism will always be part of them. It is only the degree that may be different from one episode (or one region) to another. Any analysis ignoring the implications of such inherent dualism is likely distorted, incomplete at best.

Most aggregate and macro data show that Indonesia has made a tremendous progress in raising people’s living standard. The relative per-capita income compared to that of the United States increased significantly albeit not steadily due to the severe crisis in 1997 (Fig. 2.1), and the poverty rate declined persistently from a double digit to a single digit rate until the Covid-19 pandemic strike in 2020 (Fig. 2.2). Reasons behind the impressive performance range from the country’s strategy and policies, richness of natural resources, strategic location, and evolving external conditions such as increased global supply chains, capital flows, and technology transfers.

Fig. 2.1

Source CEIC data

Fig. 2.2

Source World Bank and CBS data

Going deeper, however, reveals the inherent dualism. First is on income inequality. Figure 2.3a shows that the trend of inequality, measured by the Gini index, has not been too encouraging. In particular, since 2005 the index has been persistently higher than during the decade of 1990s, and the trend up to the first half of 2000s was worsening, only slightly improved since then. The worsening trend of inequality is further confirmed if we look at the recent Gini index by region as compared to the same index in 1996. As displayed in Fig. 2.3b, virtually in all regions the Gini index moved up above the 45° line over the two decade period. In 1996, only 3 provinces had a Gini index greater than 0.3 (DI Yogyakarta, Sulawesi Tengah, and Papua), and more-than 2 decades later only 1 province (Bangka Belitung) registered a Gini index of less-than 0.3, and 6 provinces had a Gini index greater than 0.4. All these trends occurred before the pandemic hit. If experts’ prediction is to be believed, the post-pandemic recovery is likely to be K-shape, implying that the inequality will get even worse since the hardship has fallen disproportionately on flexible, low income workers and young people. On the other hand, unable to go out, eat out, shop or travel during the pandemic, the middle and upper income group build up their savings.

Fig. 2.3

a Source World Bank data, b Gini Index By Regions: 2017 compared to 1996. Source Processed from CBS

Inequality between regions displays another dimension of dualism. Figure 2.4 shows that the coefficient of variations (CoV) of per-capita gross regional product (GRP/cap) was high and failed to register a meaningful decline over the last two decades. Marked improvements occurred only during the 1980s; but even that trend was due primarily to a high CoV at the base point. Compared to the interregional inequality in other countries that also have a large population, such as India, China, and Brazil, Figure 2.5 clearly shows that Indonesia is at the lower edge of the list as it has the most unequal distribution between regions. A study shows that the contribution of inequality across urban-rural and across regions or districts in Indonesia’s overall inequality is ranked the second, only surpassed by the inequitable access to education (Chongvilaivan & Kim, 2016).

Fig. 2.4

Source Calculated from CBS data

Coefficient of variations of Gros Regional Product/Capita (current price).

Fig. 2.5

Notes Indonesia (2016), Colombia and Mexico (2015), Brazil (2014), China and India (2013). Source Statistik Indonesia and OECD Regional Database

Interregional inequality in selected countries.

Part of the reasons behind the inequality can be linked to what happened with the country’s wealth distribution. Wealth as a stock concept generates flows of income. Given the rate of returns, wealth inequality generates income inequality. The implication and evidence of the relation between the two has been shown by many authors, including Piketty (2017) who used the data since the 18th century in Europe and the United States. Although using the cases of two countries, his evidence-first approach helped set off a global debate on income inequality and no longer makes one be able to assert that rising inequality is a necessary byproduct of growth and prosperity, or that capital deserves protected status because it brings growth.

During the last few years alone, the wealth share of the top wealthiest 10 percent in Indonesia had increased from 36.4% in 2014 to a whopping 74.1% in 2019. Those numbers imply that the wealthiest 1% of population owned 45% of total wealth, making the country’s wealth inequality among the worst in the world (Credit Suisse, 2019). It is not surprising that in 2019 the wealth Gini index in Indonesia (83.3) was higher than in other large countries such as China (70.2), India (83.2), and Mexico (77.7). As predicted, the fastest growth occurred in the financial asset. When we look at the ratio of the growth of market capitalization (a proxy for financial asset) over the growth of housing price (a proxy for non-financial asset) during the last 10 years before the pandemic, the number has reached 345.5%, much higher than the world average. In 2000, the share of financial wealth in the total wealth was 21.8%, and by 2019 the share had already reached 42.3% (Fig. 2.6). Hence, a considerable portion of wealth inequality in Indonesia—which eventually translates into income inequality—has been largely caused by a rapid expansion of the financial wealth which is predominantly owned and controlled by the urban-based middle to upper level income group.

Fig. 2.6

Source Processed from Credit Suisse (2019), Global Wealth Databook 2019. October

Financial and non-financial assets/gross household wealth (%).

Growing financial assets cannot be separated from financial liberalization. Although the worsening impact of liberalization on inequality has been confirmed by many studies, those using the case of Indonesia have been rare. To the extent ‘financialization’ is easier to detect and measure than financial liberalization, some studies looked instead at the impact of ‘financialization’ by using a standard indictor such as the size of financial assets on inequality. Using stock market capitalization and returns on asset (ROA) as indicators of financialization, Buhaerah (2017) used panel data of ASEAN countries including Indonesia and found a negative effect of financialization on inequality. Greater capitalization and higher ROA tend to worsen inequality. Measuring financial liberalization, let alone its impact on inequality, is far more difficult due to the diverse type of financial liberalization and the complex channels of transmissions to reach household income or consumption. By using a financial computable general equilibrium (FCGE) model with the Flow-of-Fund data to evaluate the impact of financial liberalization-driven inflows of portfolio capital, Azis and Shin (2015) confirmed that financial liberalization in Indonesia tends to worsen inequality. By exploring counterfactual experiments and tracing the channels of transmission, it was further revealed that the relatively lower growth of the non-financial sector (more employment-generating activities) and the increase of financial returns (mostly owned by and accrued to high income households) contribute significantly to the country’s growing inequality.

Behind the impressive reduction of Indonesia’s poverty shown in Fig. 2.2, two important observations need to be highlighted: the dynamic trend of the link between growth and poverty, and the regional dimension of poverty including non-income or non-consumption poverty.

Numerous factors affecting consumption poverty in Indonesia have been identified by different authors; some are macro in nature, including sectoral composition of employment, others are more of micro-social type. On the macro side, the positive link between national growth and poverty reduction has become a standard hypothesis. By focusing primarily on trends in measurable poverty and inequality indicators, Hill (2021) made it clear that the link applies for all episodes. During the mid-1970s, when the government promoted rice and other food crop production, and during the 1980s when the labor intensive along with exports program was emphasized, higher growth led to a sharp decline in poverty. Since the 1980s, by using the growth incidence curves (GICs), the link remained intact. Although the last chart of Fig. 2.7 shows that for the whole period (1980–2017) the consumption growth of all households were positive, supporting the trend of declining poverty, variations occurred during different periods. Comparing two political eras, i.e., Orde Baru up to 1996 and post-Orde Baru after 1997, the outcomes differed significantly: growth was faster and inequality was stable in the first, and growth was slower with segmented labor market and growing inequality in the second. But so was the poverty rate: it fell sharply in the first and much slower in the second, confirming the established link between national growth and poverty reduction. Variations in the inequality outcomes are clearly detected during the four periods shown in Fig. 2.7. In the early years up to 2000, a relatively egalitarian growth occurred except during the financial liberalization period (1990–1996) where higher income households benefited the fruits of growth at a much faster rate, consistent with what was discussed earlier. In the subsequent 2000–2017 period, inequality has clearly become worse. Overall, for the whole period since 1980 Indonesia’s higher income households have been benefiting the fruit of growth at an unmistakably faster rate than the lower income households.

Fig. 2.7

Source Hill (2021)

Growth incidence curves-spatial plus temporal, 1980–2017.

On regional dimension of poverty, the headcount index in Papua and Maluku was 1.5 times higher than the national average, while the index in Sulawesi failed to improve. If poverty is measured based on factors other than income, particularly health and education, the spatial gap of non-income poverty is fairly big. A study by Hanandita and Tampubolon (2016) showed that based on the health-related deprivation, including having illness for more-than 3 days and a disease for more-than 4 days (morbidity), the difference between rural and urban throughout the country was markedly large. Although it followed an inverted U-shape trajectory, the urban–rural health inequality during 2003–2013 did not show an improvement. Across regions, the gap between Nusa Tenggara and the rest of the country was particularly high, i.e., up to 2.13 times greater than the national average for illness deprivation, and up to 3.70 times greater for morbidity.

In terms of education-related poverty (having primary education and ability to read and write latin characters), variations across regions were also high. Rural deprivation was over twice of that in urban area, and the gap did not seem to narrow over the years. While the contribution of income poverty in Indonesia’s multidimensional poverty is higher than that of non-income poverty, the contribution of the latter has been increasing steadily. It is also important to note that poverty reduction in rural area was driven by improvement in both, the income and non-income components, whereas the reduction in urban area was driven mainly by improvements in the non-income (deprivations) component. As a result, nearly a quarter of Indonesian adults living in rural areas failed to complete primary school. This occurred despite a substantial reduction in poverty and the constitutional mandate for the provision of universal primary schooling and for the allocation of 20 percent government expenditure to education (the 20% rule).

But duality is particularly stark at the district level. Based on the multidimensional poverty measure (taking into account the simultaneous deprivation), a large variation is found between districts across regions. The divide between Western and Eastern Indonesia cannot be more obvious. From 346 districts, 5 out of 10 of the poorest are in Papua. Dualism also occurs within each region; for example, in East Jawa province, the multidimensional poverty in Kabupaten Bangkalan is 7 times higher than in the city of Surabaya.

The multidimensional poverty has been always higher in rural than urban areas. Most of the least-deprived districts are in urban areas including municipalities. Hence, by looking deeper at the data to include the non-income poverty across regions and rural-urban, the impressive poverty reduction over the last decade has not been complemented by strong improvements in the non-income dimensions. Consistent with the results of a study using income or consumption poverty (Sumarto, Vothknecht, & Wijaya, 2014), the spatial inequality of non-income poverty is fairly high, with a large variation between districts. Suryahadi, Rishanty, and Sparrow (2020) argued that the trust among people across different ethnic groups, being the most important social capital, plays an important role in poverty reduction. At any rate, a regional disparity of high poverty unarguably continues to characterize Indonesia’s journey to prosperity.

One of the most important institutional changes that has affected Indonesia’s intra and inter regional development was the decentralization policy implemented in early 2000s followed subsequently by direct elections at the local level (pemilihan kepala daerah, PILKADA). At the beginning, PILKADA was conducted in only few regions. By 2005, a full scale implementation began. Since then, data on socio-economic performance show that some regions experienced improvements but others did not. On the basis of per-capita gross regional product (GRP/cap), more than 60% of Indonesia’s provinces experienced a reduction in annual growth after PILKADA, and in terms of human development index (HDI) more-than 90% provinces experienced a decline in annual growth (Fig. 2.8).

Fig. 2.8

Source Based on CBS data

Annual growth of GRP/cap and HDI before and after decentralization.

Dualism in regional performance is also detected when we consider institutional factors of decentralization. By applying the ‘institutional model of decentralization’ (IMD), the detailed of which is discussed in the next chapter, a survey conducted in seven regions/districts throughout Indonesia after a full swing PILKADA—over the period of 2008–2009—reveals that the importance of peoples’ participation in determining local welfare is overwhelming. Yet, the quality and intensity of participation have been highly diverse among regions. From all seven regions combined, the results show that participation is the most critical factor followed by the size of local budget, and subsequently by the initial conditions (Table 2.1). The test of robustness of the results through a dynamic sensitivity analysis corroborates the finding.

Table 2.1 Results of IMD-based field survey: group, individuals, and combined hierarchy

While the importance of local budget is obvious (poor regions with low budget have more difficulties to improve welfare), the role of initial conditions cannot be overlooked. Evaluating the breakdown of the survey results, in some least developed regions the initial socio-economic conditions matter more than the size of local budget, indicating the presence of path dependence, i.e., regions that were poorer than others some decades ago remain poorer now. On the other hand, more developed regions tend to grow faster. The welfare effect of decentralization is greater when local people are more politically aware and actively participate in various local development programs. What this suggests is, the gap between less-developed and more-developed regions (duality) after decentralization tends to persist or even widen, consistent with the finding using the secondary data discussed earlier.

Similar to the development gap between regions, dualism in business activities is equally stark. While the number of business establishment and business unit is dominated by MSME, i.e., more-than 99% and over 96%, respectively, their productivity has been way below the large businesses. Before delineating this productivity gap, let’s first look at the trend of Indonesia’s productivity at both the national and regional level where another type of dualism exists.

Why productivity matters? Many arguments and rationales can be put forth. Krugman (1994) summarizes it well: “Productivity isn't everything, but, in the long run, it is almost everything. A country’s ability to improve its standard of living over time depends almost entirely on its ability to raise its output per worker.” Fig. 2.9a shows the productivity growth at the national level, and a set of charts in Fig. 2.9b at the regional level. In both, the productivity growth is decomposed into “within” and “structural” components.³ At the national level, a deceleration of productivity growth is clearly observed. Across regions, the deceleration is also evident, albeit varies. It is obvious that the changes in productivity have occurred predominantly in the “within” category, implying that improvements in one sector tend to confine to that same sector, hardly diffusing to the rest of the region’s economy. The absence of spillover has not only dragged down the region’s productivity growth but also widened the gap between sectors. The trend of productivity growth since 1990 varied between regions—some show an acceleration, others display a deceleration. But since 2011, the trend became more obvious, i.e., virtually all regions suffered from a declining productivity growth. Looking at the data more closely, the sharp fall of productivity occurred mostly in provinces of Eastern Indonesia.

Fig. 2.9

a Decomposition of Labor productivity growth: “within” and “structural” components. Source Calculated from APO and Conference Board, b Decomposition of labor productivity growth by regions. Source Calculated from APO, CBS, and various sources

Dualism in terms of interregional inequality and the level of productivity was stark. During the last decade, measured by the relative productivity of the agriculture and the total average, the ratio in provinces with the highest degree of economic dualism reached 3.6 times than the ratio in provinces with lowest degree of dualism. Inequality in productivity level was even starker: the highest productivity in regions like Jakarta and East Kalimantan reached 13 and 11 times more, respectively, than the productivity in East Nusa Tenggara. Since 1990, the divergence between regions also increased, where the standard deviation rose from 17.1 in 1990–2000 to 19.2 and 21.1 in 2000–2010 and 2010–2018, respectively (Fig. 2.10).

Fig. 2.10

Source Calculated from APO, CBS, and various sources

Interregional inequality in productivity.

What about the trend of MSME productivity? Based on the World Bank Enterprise Survey, the productivity per worker of micro enterprises in Indonesia was only 3 percent of that in large enterprises, and for small and medium firms the percentage was 16% and 31%, respectively. To put into perspective, these levels are equivalent to roughly only one-quarter of the OECD median value. Consequently, based on the direct and indirect exports (MSMEs supplying products to exporters) Indonesian MSMEs are less integrated into global markets compared to their counterparts in some ASEAN countries, particularly Cambodia, Malaysia and Vietnam. Despite the increasing use of internet and e-commerce, and that various measures had been taken to encourage small business internationalization, very few Indonesia’s MSME were able to penetrate the export market.⁴ Their already low contribution in total exports excluding oil and gas continued to decline (OECD, 2018). The World Bank Enterprise Survey (WBES) Database indicates that in 2016 the share in total exports for medium-size firms was 11.5%, and for small firms and micro enterprises were only 2.8% and 1.4%, respectively. Having fewer resources to meet the high costs associated with engaging in international markets, facing greater challenges than larger firms in navigating foreign markets, and having less capacity to address complex regulatory requirements, are all binding constraints.

The proliferation of free trade agreements (FTA) in which Indonesia has been actively seeking to be part of, did not seem to contribute significantly to MSME’s participation in global and regional trade. Even for those who have been actively exporting, a large portion of them did not utilize FTA facilities. Part of the reasons is a lack of knowledge regarding their use (Anas, Mangunsong, & Panjaitan, 2017). More generally, the complexity of rules and agreements, and the low margin-of-preference are among the top reasons why the utilization of FTA facilities has been low (Azis, 2019). Trade has grown due to the unilateral trade liberalization, not because of the FTA proliferation. Nonetheless, MSMEs tend to be underrepresented in international trade.

There are multiple reasons behind the dismal productivity performance of MSMEs. They range from a lack of innovation, limited quality of entrepreneurship, asymmetric information, and problems in financing, including access to trade financing, opening LC, and securing scheduled payment from importers. In terms of innovation, the dualism is reflected in the R&D spending. While in average only 2% of all firms in Indonesia invest in R&D, the share in large companies is about 10%, implying a very tiny portion of MSME’s investment in R&D. Not surprisingly, having a lack of innovation, the number of small and medium enterprises that managed to introduce a new product and/or service during the last three years prior to the survey was only 5% and 9.7%, respectively.

Across regions, the inability of MSME to enhance productivity was also due in part to the limited programs designed to boost productivity growth at the firm level. Differences in regulations and licensing between regions, as shown by large variations in the ease of doing business, also contribute to the diverse performance of MSME productivity across regions. So do variations in local government capacity, which plays a significant role in a paternalistic society like in Indonesia. Note that the decentralization-related Law 23/2014 on the role of local governments assigns the responsibility to different tiers of government: national governments are assigned to support co-operatives and medium-sized enterprises, provincial governments to support small enterprises; and cities and regencies are mandated to support micro-enterprises. While useful on paper, such a distinction exacerbates the already widened interregional disparity, and its implementation tends to confuse regulators and MSME operators. For variety of reasons, bureaucracy at the local level and the attitudes of some local officials are not sufficiently conducive for productivity improvements. Often, they behave as if they deserve respect from local residents, and MSME operators ought to obey and listen to them rather than the other way round. Their conduct tends to serve their own interest rather than the interest of the residents they are supposed to serve. Expressed by MSMEs participating in our survey, the role of local government in assisting MSME has been thus far weak. Some also complained about the practice of nepotism among local officials. Many MSMEs operating in the agricultural sector indicated that the impact of government-initiated social programs were either very little or none at all.

On the financing front, interregional variations are no less obvious. Unlike in Jawa, many MSME in remote regions of Eastern Indonesia received loans from cooperatives. In regions like Papua, many MSME receive loans from banks. Across all regions, however, middlemen remain the most active lenders, and in recent years the so-called fintech lenders have also been proliferating. According to some MSMEs with whom we had discussion, both of these financing sources imposed too-high interest rates. Another challenge is getting the payment promptly from foreign buyers. Some MSMEs in outside Jawa reported that after putting much efforts and energy to penetrate the foreign market, they were finally able to find foreign buyers. But they complained that after delivering the products they could not get the full payment promptly despite the agreed transactions. Payments were made only after a long delay, putting pressures on the business cash flow. At the end, they had no choice but abandoning the contract all together, at the cost of no more exports and sales. Having limited network and connection, those exporting MSMEs need help from the government. Assisting them should be neither difficult nor costly, yet absolutely necessary given the weak domestic demand especially during the pandemic, and the government’s repeated assertion to encourage MSMEs to reach beyond the domestic market.

Looking at the experience of other countries, aside from the continuous efforts to improve the various schemes of credits for small business and rural activities, new initiatives of financing should be continually explored.⁵ For example, a cash-flow based (instead of collateral-based) system of lending has been adopted in some countries. In others, third-party insurers are actively involved to lower lenders (banks)’ disincentive-to-lend by guaranteeing a sufficient portion of loan repayment. For MSMEs that are able to find and secure foreign buyers, some sort of guarantee per purchase-order (PO) can be explored, for example by allocating a guarantee based on the MSME’s business track record. Still another potential scheme that takes advantage of the mobile system is the so-called pay-as-you-go (PAYGO) which requires a nominal down payment to take possession of an asset electronically, followed by frequent small payments made via a mobile payment system. For the lenders, such a system is relatively cheap and easy to disable the flow of services, insuring them against default, and for the borrowers (MSMEs) who are unable to make a payment, they do not lose the asset, rather they are simply unable to consume the flow of services from the asset until they start paying again. They may lose something of value but that provides incentive to repay, or they can decline the loan offer all together. Many other options could be explored, and ideally they should involve the MSMEs as potential borrowers right from the planning, the policy design, all the way to the implementation stage.

Policy Measures

Given the trends discussed above, one may wonder if the numerous policies taken over the years had any meaningful effects on the SME performance. Early in the development plan of the New Order government, efforts were made to provide financial and technical assistance to help improve the operations of SMEs; some implemented through cooperative units, others through regular business operations. Attempts were also made to promote the SMEs through regulation and coercion, including to enforce subcontracting schemes (mainly in the automotive and electronic industries), and to use a foster father or ‘Bapak Angkat’ system where state enterprises or large firms were required to sponsor the local SMEs. In addition, the government also imposed preferential procurement programs and issued regulations allowing only firms of a certain size that can produce certain goods. Table 2.2 tabulates the key policies, programs and organizations relevant to the promotion of SMEs in Indonesia during the three decades since the first five-year plan (REPELITA) began in 1969.

Table 2.2 Policies, programs and organizations for the development of SMEs in Indonesia

Although a comprehensive analysis to evaluate their effectiveness has never been made, several studies deduced that most of these supply-side programs were not effective, having low participation rate, and often beset by problems of corruption (Berry, Rodriguez, & Sandee, 2001; Musa & Priatna, 1998; Hill, 2001; Sandee et al., 1994; Tambunan, 2007). Part of the reasons is because most of them were not designed with a clear and unambiguous framework. Where SMEs succeeded to make improvements, they did so in spite of, not because of, government programs. The growth of MSME over the years have been found to be influenced by factors other than government assistance, and the probability of receiving assistance is positively related to the firm size (Berry et al., 2001).

Since the early 2000s, the government continued to use various measures to promote micro enterprises and SMEs, hereafter MSMEs. The list of detailed measures is too long to show here, but they cover the financial, technical, and regulatory assistance. For example, in the financing front the measures taken include subsidized small credits such as those allocated through the Koperasi Unit Desa (KUD) for small farmers and village cooperatives; Kredit Investasi Kecil (KIK), Kredit Modal Kerja Permanen (KMKP), Kredit Usaha Kecil (KUK) for general purposes, Kredit Umum Pedesaan (KUPEDES) for village units, Badan Kredit Desa (BKD) for small rural development banks, and Kredit Umum Rakyat (KUR) for MSME (launched in 2007). For technical assistance, a wide range of measures have also been taken, from training and improving product design, marketing, promotion, accounting and book-keeping, and using digital technology such as e-commerce, fintech, and other internet-based activities. In the regulatory front, the government continues to require banks to allocate 20% of credits to MSME, assigns lower tax rates or grants tax exemptions for some MSMEs, streamlines the procedure to obtain license and other documents/permits, creates linkages between MSMEs and large enterprises and other related activities (subcontracting), as well as linkages among the MSMEs themselves, etc. Policy makers have also made frequent and numerous statements supporting the MSME operations especially during the pandemic.

Some of the measures taken before 2020 continued and some were expanded during the pandemic, such as providing unconditional cash transfer program or bantuan langsung tunai (BLT) for ultra-micro and micro enterprises through banking and finance company, restructuring credit and interest subsidy for micro enterprises, restructuring coop credit trough revolving fund agency known as the Lembaga Pengelola Dana Bergulir Koperasi dan Usaha Mikro, Kecil, dan Menengah (LPDB-KUMKM), subsidizing credit interest to cooperatives, and providing liquidity assistance to cooperatives with low interest rates and easy mechanisms. Attempts have also been made to entice MSMEs to take training programs by providing stimulant fund to the participants through a Pre-Employment Card Program (program kartu pra-pekerja).

Some demand-side measures have been also taken to maintain and enhance the purchasing power of consumers to buy MSME/cooperative products, for example by allocating funds for a discount to purchase MSME goods (offline and online), distributing discount voucher, utilizing stall/shop (warung) data connected to e-commerce, establishing partnerships with nine state-owned enterprises (e.g., in food cluster), and utilizing young influencers to encourage people to shop MSME products around their neighborhood. Measures are also taken to boost SME exports through virtual business match-making events.

While remains to be seen whether these programs are effective to help improve the MSME performance, a clear and important lesson from the past is that, addressing MSME problems caused by genuine market failures should not be mixed-up with other objectives. Even if the latter are important, they should be addressed by different policies designed specifically to meet them, not by policies for MSME. For example, programs to help MSME are often confused with targeting employment creation because MSMEs are seen to be more labor intensive than large firms. Yet, evidence suggests that enterprise scale is not a reliable guide to labor intensity: many MSMEs are in fact more capital-intensive than larger firms in the same industry. Policies to boost employment should instead focus on altering the pattern of demands in favor of labor-intensive industries rather than on supply-side efforts to change the size distribution of firms. Another important lesson is that, if the goals are not specific enough or too ambiguous, there is a risk that facilities provided by the policies will not be well received or even avoided all together by the MSMEs. It is hence imperative to understand the internal problems faced by MSMEs and why such a case occur.

The remaining chapters of the book discuss the above issues by way of establishing a framework of analysis capturing the interplay of policies and institutions to explain the phenomena of interregional inequality and challenges faced by MSMEs to improve their performance. The latter is validated by a micro and small enterprise (MSE) survey conducted in different regions and sectors throughout Indonesia.