Summary

Abstract

By listening to hundreds of MSMEs throughout the country, we understand better why there is often a gap between what public policies intend to achieve and what the actual outcome is. Listening is more powerful than hearing and talking, because it forces us to give our undivided attention to whatever commented by MSMEs we intend to help, and allows us to capture their mental bandwidth and the ecosystem they are in. We tend to overlook the inadequacies in our knowledge-base regarding the nature and sources of the prevailing social capital, and the institutional mechanisms for defusing tensions and diverse perceptions among MSMEs and other stakeholders. Yet, they are the stumbling blocks for narrowing the gap between intention and action.

We are now at the end of our journey.

From the outset, we took a deliberate approach of listening to hundreds of MSMEs throughout the country. That listening is more powerful than talking. Unlike hearing, listening requires us to give our undivided attention to whatever is commented on by our respondents.

But there is more. Not only that we ought to be fully present and engaged in a conversation without talking much—sometimes even not at all—but the issues we converse about, and the questions we ask, must reflect the real concerns according to MSMEs, not what they are according to us. This perspective suggests that the questionnaire, the hierarchies, and the networks we used in the survey must be based on a conversation and deep dialog with them. When about to frame the questionnaire, we were concerned whether we talked about the right issues and asked the right questions. Indeed, in many approaches to deal with development issues, including the one we are working on here, the limitations of implicit assumptions used by those approaches, reflecting in the questions being asked, are often ignored (Redman, 1980). Of those limitations, the difference in perspective and framework on time horizon, market structure, and aggregation over the issues in question is among the most common ones (Kanbur, 2001). For this very reason, we conducted the off-line and online pre-surveys to learn about the relevant issues and challenges that MSMEs had been dealing and struggling with, about their accomplishments, hopes, aspirations, and shattered dreams.

Every life, including that of millions of MSMEs, can be made better by improving our listening to them. And we can and should all do more of, if our genuine intention is to elevate their productivity and improve their welfare. The central premise of this study is that too many policy problems can be traced, in the ultimate analysis, to the gaps between intentions and actions. Listening and designing an appropriate policy application can help close the gap between what policy seeks to achieve and what it accomplishes, the gap between what MSMEs need and what the public policy offers. The ability to effectively listen and ask questions can reduce the differences in the mental bandwidth of those whom the policies are intended for and those who design the policies, so that they will understand better what causes the gaps.

Of many obstacles to listen and ask effectively, behavioral barriers are among the critical ones. Cognizant of the fact that human bias is an inevitable part of human behavior and conditions, in which respondents whom we asked and listened to may objectively perceive themselves and their environment (a type of cognitive bias), we adopted a particular approach to obtain and rank their perceptions in a way that we could capture the non-tangible parts of their feelings, the intensity of their choices, and the consistency of the rankings. It is not just about asking questions and listening to the answers. How we ask and synthesize the answers matters a lot.

For these reasons, after constructing the relevant hierarchies and networks, we specifically designed the questionnaire in a pairwise comparisons fashion. Following the Analytic Hierarchy Process/Analytic Network Process (AHP/ANP) approach, we systematized the results to generate the consistent ranking. For the AHP case, that consistent ranking was derived from the eigenvector (based on the maximum eigenvalue), and for the ANP the consistent ranking was obtained from the limit of supermatrix (based on the stochastic weighted supermatrix). The process to build, and the contents within, the hierarchies and the networks are explained in Chap. 2, and brief mathematical explanations are given in Appendices B and C.

We first adopted the above approach to generate MSMEs’ preferred ranking of social capital compatible (SC-compatible) policies. To the extent one policy may either reinforce the effect of, or create a trade-off with, other policies, we then proceeded with attempts to find and rank the policy-mix. This strategy instilled the realm of policy-making as there was hardly a case where social planners took one single stand-alone policy. Yet, to construct the hierarchies for such a strategy also requires taking account the complexity and the dynamics of several SC-compatible policies when they are paired or combined.

Since different groups had not only different interests and ranking of preferences, but also different attitudes toward being truthful or not about their state/profile when expressing their perceptions, attempts were made to identify the preferred policies that could be aligned with the interests of the overall groups. Only policies that met such a condition would be considered “implementable SC-compatible.” We adopted the mechanism design theory (MDT) approach to find which among the SC-compatible policies and policy-mix are implementable, and what mechanisms or rules could be used to implement them. For this purpose, in Chap. 3 we performed a series of monotonicity tests on the ranking results of AHP/ANP.

In many cases, the gap between what policies sought to achieve and what they accomplished was due to the institutional constraints. In such circumstances, it is more important to find the right institution or mechanism than to focus only on searching policies by making an assumption that the prevailing institutions can make those policies effective. In MDT, it is the mechanism, not the policies, that are endogenous.

Based on the results of the preference ranking using a network that includes feedback effects in Chap. 2, the one appeared consistently as most preferred SC-compatible policy was to have a network for interactions, using which MSMEs could form business linkages among themselves and with other stakeholders. Discussed in greater detail in that Chapter, such findings were derived from the responses to the carefully crafted questionnaires, the contents of which were based on direct inputs from MSMEs. Hence, not only the responses to the questionnaires, but also the questionnaires themselves capture the perceptions of the respondents. They reflect the mental bandwidth of MSMEs.

The same conclusion was uncovered from the case of policy-mix, in which the highest ranked policy was always the one that includes network for interactions. Even when the preferred policy turned out different, i.e., to provide supporting infrastructure as revealed by MSEs (but not MEs), the other preferred policy in the mix was to establish linkages, which is analogous to having a network.

Survey story: A traditional shop in Wamena, Papua, selling various handicrafts produced by local tribe that belongs to a number of related ethnic groups, the most prominent of which are Dani (in the main valley), Lani (to the west) and Yali (in the South-East). Their businesses are self-funded, not relying on bank lending. Helping the tribe community by giving them loans or money is not effective, but helping them with networking with outside parties is

On the policies implementability, it was revealed in Chap. 3 that some of the preferred policies failed to make the preference of one group (MSE) aligned with that of another group (ME). Similarly, as discussed in Chap. 4, some SC-compatible policies and policy-mix preferred by ME and MSE could not be aligned with what the social planners believed to be the right policies for MSMEs. It is therefore important to find a mechanism by which we can identify policies that are alignable with the interest of the overall group (society) to qualify them as implementable.

It was evident from the monotonicity tests that a policy of creating a network for interactions could align the preferences of ME and MSE, hence implementable. It was also revealed that an indirect mechanism could be designed by social planners without knowing the MEs and MSEs state/profile. Did the COVID shock make their perceptions change? From the test results, having a network continued to receive the highest priority during the COVID pandemic. The shock did not seem to alter MSME perceptions about the importance of having a network. In the case of policy-mix, the implementability of creating a network for interaction was even more ascertained, as it included promoting linkages among themselves and with other stakeholders in the policy pair. Such a mix can be potent as it is capable of harmonizing the network and the two related components (establishing interaction and linkages). In the language of MDT, the social choice rule (SCR) is a dominant strategy and incentive compatible.

Interestingly, the above finding about the implementability of creating a network for interactions was derived under the condition that the respondents took into account the complex interrelations among objectives, criteria, and alternatives (System 2). If they ignored such complexity (System 1), the implementable policy was to provide supporting infrastructure. That is, when approaching the problem in a simplistic way, it is easy to come up with building infrastructure as the solution. But in a more realistic situation where MSMEs must in fact deal with problems that are more complex than what they seem, infrastructure is not the answer as it does not meet the social choice rule (SCR). Creating a network is.

Comparing the preferences of social planners and those of MSMEs, the gap between the plan and the achievement became more evident. As elaborated in Chap. 4 based on the long list of policies that have been taken during the past several years, social planners seem to place the highest priority on financing. Network for interaction came only the second, followed by the provision of supporting infrastructure. Broken down the comparison into social planners versus ME’s preferences and social planners versus MSE’s, the same order of ranking applied to social planners and MSE but not social planners and ME. Indeed, looking at the array of government measures taken thus far, providing more financing tends to be the social planners’ preferred choice in responding to MSMEs’ multitude of problems. The ongoing push toward raising the share of subsidized credit known as Kredit Usaha Rakyat (KUR) from 20 to 30 percent by the banking sector is a notable example.

Although loan financing is important, many MSMEs did not see it as the main factor, certainly, it is less important than having a network for interaction. Given the same shock, if small firms could have a similar network of connections enjoyed by large firms, the productivity improvements of the two would have been comparable. Even for the second and third ranked priority, the MSMEs considered having the supporting infrastructure and the ability to resolve matters related to regulation and legal issues are more important than financing.

Since the primary sources of inputs for all the above findings were the perceptions and judgments of MSMEs, what constitute perceptions and judgments was discussed in the early Section of Chap. 4. It was also argued that the geometric mean is the most appropriate to use for averaging out group perceptions, and that adopting the dominance property in measuring the inconsistency of perceptions is preferable. The case where the results from using pairwise comparisons to generate ranking are different from those using a simple Borda count was also discussed in that chapter.

To check the implementability by contrasting the perceptions of ME and MSE on the one hand and those of the social planners on the other, the monotonicity test confirmed that creating a network for interaction is implementable. As discussed in Chap. 4, this policy will align the policy rankings of social planners and those of MSME regardless of the state/profile of MSME. Even when we broke down into ME and MSE, where the results always put a network for interaction at the top, such a policy was found implementable. The only qualification is when we compare the social planners’ ranking with that of MSE, in which the state/profile of MSE matters. In particular, if we assume that MSE responded to the questionnaire without considering the complex interrelation between the objectives, criteria, and alternatives (System 1), the SC-compatible policy that could align the preferences of MSE and social planners would be to provide more affordable financing. But when a more realistic case (System 2) holds, an indirect mechanism could be designed without knowing the MSEs and social planners’ state/profile, i.e., create a network if System 2 and provide more affordable financing if System 1.

The primacy of network and linkages revealed from the survey is not too difficult to make sense of. However, its practical meaning for policy may remain fuzzy and imprecise because networking is one component (beside trust and norms) of social capital, a concept that has an imprecise meaning due to which economists are weary of using them. Based on the ranking results of the objectives, criteria, problems, and alternatives, and also confirmed by what we learned from our respondents about what they had gone through, an important channel connecting network-linkages and productivity likely works through transaction costs. These are costs that MSMEs must incur during the process of buying, producing, and selling, including expenses for getting permits, settling legal matters, selecting inputs, identifying potential markets, accessing affordable financing, choosing modes of production, or opting for payment system. Information pooling and sharing through a network would enable them not only to lower production and marketing costs, but also to reduce the search costs significantly. The latter makes all the above actions easier to conduct, allowing productivity to improve. Ignoring search costs is equivalent to assuming zero transaction costs, which makes it impossible to understand fully why most small businesses fail to achieve higher productivity.

The importance of transactions’ costs in development performance has long been known. From Commons (1932), Coase (1937, 1960), Arrow (1969), North (1992), Williamson (2008) to Cornelisse and Thorbecke (2010), all recognized the critical role of transaction costs, the size of which could be determined by the prevailing institutions. Embedded in the institutions is social capital. As revealed from our previous study (Azis, 2022), participation and coordination in a cluster are critical for productivity improvements. They are particularly useful for information searching and sharing. But to have them and make them effective would require trust, which is another important element of social capital. With no trust, any participation and coordination would have limited effectiveness and likely failed to generate the intended benefits. Not only that trust is a necessary condition for participation and coordination, but it could also help provide the enabling environment for productivity improvements; e.g., ease to enforce contracts or to renegotiate them when problems arise.

Considering the whole eco-system in which MSMEs operate in Indonesia, the yearning for having a network could also be driven by a lack of reliable regulatory institutions. In such circumstances, the need for a network for interactions is even greater. To create a network beyond local communities where MSMEs operate requires specific investments for which a potential obstacle may arise, i.e., no certainty whether interactions and relations built through the network will last sufficiently long to make the investment worthwhile. In the absence of reliable regulatory institutions and a basis for sustained interactions or relations, the pressure on MSMEs to survive by going alone “at all costs” could prolong—if not worsen—the problems that caused the productivity low to begin with.

Survey story: A sustainable farming in Garut, West Jawa. Led by a young farmer (second to left) who wanted to produce local food products and to raise the standard of living of farmers, the farm managed to grow well and to help educate the community about various aspects of farming that adhere to the environmentally sustainable standards. One of the reasons behind its succsess was a good network it built during the process

It is a common portrayal of MSMEs being run mostly by a traditional management system, informal, located in non-lucrative areas with poor network, and having limited funds and small assets. Access to affordable finance remains a constraint to most MSMEs in Indonesia as the country’s relative position in terms of MSME credit/GDP compared to other countries is not favorable. The gap in trade finance is even more severe, preventing some MSMEs from exporting, which is another important driver of productivity. Revealed from the results of the disequilibrium model presented in Chap. 5, during normal times credit allocation to MSMEs was constrained more by credit rationing imposed by lenders rather than limited demand for MSME loans. In this context, the focus of policy to mitigate the financing gap problem should be on reducing transaction costs, including assessment and monitoring costs, high default risk, and other factors associated with the asymmetric information faced by lenders.

Moving toward using digital technologies could potentially help narrow the financing gap, either through improvements in the “know your customer” process or through mitigating the forces that caused credit rationing (beyond the interest rates). Digitalization could also widen the product and input market, including market abroad, stimulate branding and information sharing, and showcase multiple products including the “stories” behind them. The convenience of transactions and speed of payment is another direct benefit that some MSMEs could reap. The increasing use of digital payment such as QRIS is a notable example, although its potential risks and privacy issues surround it remain a concern, and hence need to be addressed. It is also a fact that not all MSMEs are able or willing to use it.

Although growing in number, the usage of digitalization in Indonesia remains low. A combination of a lack of information and regulatory support, a relatively low digital literacy, difficulty to break the old habit (e.g., using cash), extra tasks that could be costly, and problems with hardware (e.g., electricity and Internet) had played a role in this area. The uptake of such technology during COVID could have been only out of necessity as they had no choice under the imposed restrictions, lockdown, and fear of infection. Until a longer series of data are available, it is too early to make a conjecture about the post-COVID trend, or whether the increase during COVID was cyclical or structural. Our observations in the field suggest that a distinction also needs to be made in targeting experienced and inexperienced users. Making the latter aware of how other MSMEs react to digitalization are important; e.g., advertising the experience and success stories of others. Beyond that, it is unwise to force the adoption of digitalization to some segments of MSMEs who find learning and using new technology unbearable or undesirable. A deeper study needs to be done to delve into the behavior of MSMEs over the dismissal (opposition, rejection, or postponement) of the usage of digital technology.

In Chap. 6 we re-tested the primacy of network (viz., size) as being the most critical role in MSMEs’ productivity improvement by using a model that considered two issues discussed in Chap. 5, i.e., the financing conditions and the use of digitalization. In addition to other socioeconomic factors such as education, health, and poverty, we also included in the model the presence of ethnic group as a strictly exogenous (instrumental) variable to reflect the cultural factor in localities where MSMEs operate, as well as other relevant control variables. This approach was done to avoid either under or over estimated role of network in affecting productivity. Incorporating cultural factor turned out to be necessary (statistically significant) to avoid the overestimated role of network and size. Even after taking into account the cultural factor, the primacy of network remained intact. Although both network and size have a significant positive effect on productivity, the degree of significance is higher for the former, and the results are statistically robust.

Having listened to MSME speaking about their hopes, aspirations, accomplishments, problems, challenges, and shattered dreams, including during the COVID pandemic, we may have sensed their visceral fear of sounding too demanding. The fact is, they are not. Instead, they are struggling but enthusiastic, they are put in a disadvantaged position but buoyant, and they are affected by crisis but hopeful. Providing them with better connection or network involving lenders, suppliers, customers, and those in charge of making policies and regulations is what they need the most. Drawing upon diverse methods of survey-based decision-making (AHP and ANP), a reverse game theory (MDT), and an econometric approach, the analysis throughout the book clearly shows that such findings are robust to a battery of tests. The monotonicity tests and the instrumental variable (IV) strategy using cultural factors as the instrument, and local conflicts and socioeconomic variables as the control variables, have all confirmed the primacy of network. For many MSMEs, networking is potent as it can help increase productivity and sales.

For mitigating and adapting to climate change and other environmental problems, networking could also be beneficial to MSMEs. While some whom we interviewed were already aware of—and their activities had been affected by—climate change, many faced obstacles in accessing information about the changes to be made and the costs of engaging in sustainable practices. Most of these MSMEs were of the opinion that having a good network with the relevant parties including government and regulators, could be helpful for getting the necessary information and the tools to assess their climate impact, including access to green finance. Larger firms have such network, but micro and small businesses do not.

It has become more and more obvious that many problems MSMEs have to face can be rooted in the lack of networking. Measures to create a new network or expand/improve the existing one are greatly needed. Formulating policies along this line is not rocket science, albeit requiring a good understanding of the factual problems causing low productivity, and designing policies that are compatible with the prevailing social capital. For that purpose, one needs to listen directly to MSMEs first before selecting and designing policies.

We are now at the end of our journey. From the analysis throughout the book, the message is clear: the country’s overall welfare will have much to gain from shifting the strategy for MSMEs toward delivering concrete support for networking, and a lot to lose from relying most of the time on offering financial assistance or adopting policies of the past. The stumbling blocks are often the inadequacies in our knowledge-base regarding the nature and sources of the prevailing social capital, and the institutional mechanisms for defusing tensions and diverse perceptions, and for reducing distrust among the stakeholders. Overcoming these difficulties simply requires us to “Listen and Design.”

Examples of Medium, Small, and Micro Enterprise

A medium enterprise producing leather-made bags and handicrafts in Jogjakarta, owned by a local entrepreneur (third from right). The enterprise has a good network with stakeholders including foreign buyers

A small enterprise, a coffee shop, in Kupang, East Nusa Tenggara. It is run by a group of female entrepreneurs, one of whom (in the above picture) was explaining about its history and the challenges they are facing. The shop tries to innovate by using a variety of local healthy ingredients, for which a network with other parties and local government helps to facilitate the process

A microenterprise selling various local produce in Alor, East Nusa Tenggara, run by indigenous Kabola tribe. Rather isolated in a mountain area, the most serious constraints for improving their business conditions are access to market