Avoiding the Middle-Income Trap: Renovating Industrial Policy Formulation in Vietnam

Abstract

Vietnam's growth in the last one-and-half decades has been driven by the liberalization effect and large inflows of external purchasing power. Now that the processes of systemic transition and global integration have deepened, Vietnam needs to create internal value to continue to grow and avoid the "middle-income trap". The country has reached the point where growth towards higher income cannot be secured unless policy making is renovated significantly to activate the country's full potential. The vision of Industrialization and Modernization to be achieved by 2020 must be backed by realistic industrial strategies and concrete action plans, which are currently lacking. Stakeholder involvement in policy design, inter-ministerial coordination, clear directives from the top, and incentive structure for government officials must be improved. This in turn calls for radical changes in policy administration. A new style of leadership, a technocrat team directly serving the top leader, and strategic alliance with international partners are proposed as key entry points for the renovation of Vietnam's industrial policy formulation.

I. Entering a New Era

The Vietnamese economy has grown rapidly with the average growth rate of 7.6 per cent in 1991–2008. In 1990, Vietnam was among the world's poorest countries with GDP per capita of US$98 (ADB data). By 2008, with the GDP per capita of US$1,024, Vietnam has already reached the status of a lower middle-income country by the World Bank classification method.¹ The growth has been broad-based and touches virtually everyone's life and generates profound social changes in the entire country. This is quite different from the experiences in Latin America or Sub-Saharan Africa where growth occurs in limited sectors and benefits only few people while poor farmers see little improvement in their lives. However, Vietnam's achievements up to now have been driven mainly by one-time liberalization effects and external forces associated with global integration rather than internal strengths. Despite impressive growth records and reform efforts in the last one-and-half decades, local firms remain generally uncompetitive, and policies and institutions remain very weak by East Asian standards.

From the mid-1980s to the mid-1990s, growth was stimulated by the incentive and reallocation effects of internal economic liberalization (doi moi). Subsequently, from the mid-1990s to the present, growth has been supported by new trade opportunities as well as large inflows of foreign funds. Industrial activities —especially manufactured exports —continue to be dominated by foreign firms, and value creation by local firms and workers has been limited. Now that Vietnam is nearing the final stages of systemic transition and global integration, productivity breakthrough is needed to climb further. Future growth must be fuelled by skill and technology rather than a mere injection of purchasing power.

Growth statistics presented in Table 1 are consistent with this interpretation. Until the mid 1990s, the incremental capital-output ratio (ICOR) was low and the contribution of total factor productivity (TFP) to growth was high, which indicates that growth was achieved through improved efficiency —albeit from a very low level of planning years —without much investment.² In the latter period, ICOR rose, TFP's contribution to growth declined, and capital's contribution increased significantly. That is an indication of investment-driven growth with low efficiency in capital use.

The "Washington Consensus" policy package prescribed by the World Bank and the International Monetary Fund such as liberalization, privatization, legal reforms, macroeconomic stability, and so on, may achieve middle income if they are properly executed, but that is not enough for continued growth to higher income. Vietnam's growth pattern basically follows the past experiences of East Asian neighbours whose features include openness and regional integration as an initiator of growth; deepening intra-regional trade and foreign direct investment (FDI); high savings and investment; dynamic transformation of industrial structure; urbanization and rural-urban migration; and growth-generated problems such as income and wealth gaps, congestion, pollution, financial bubbles, and so on. At the same time, a number of new elements for Vietnam, such as faster integration than ASEAN-4, must also be acknowledged.

Within this dynamic East Asian context, Vietnam must successfully conduct three crucial policies to sustain growth, namely: (i) generation of internal value; (ii) coping with new social problems caused by rapid growth; and (iii) effective macroeconomic management under financial integration. The first promotes drivers of growth while the second and the third prepare political stability and social support without which industrialization and modernization cannot be sustained. By 2008, the risks of social problems such as traffic congestion and environmental destruction as well as macroeconomic imbalance such as asset bubbles and price instability had become evident in Vietnam. Management of industrialization in this broad sense must be installed to face new challenges, or the entire process of industrialization may stall (Murakami 1992, 1994). While all three tasks are important, the present analysis focuses on the first issue of internal value creation while leaving the discussion of the remaining two to other occasions.

II. The Middle-Income Trap

A low-income country which has gone through a war, political turmoil, socialist planning, and severe economic mismanagement is usually characterized by a fragile economic structure. It relies heavily on extractive resources, monoculture export, subsistence agriculture, or foreign aid. Internal value created by traditional industries such as mining and agriculture is small, but the absence of vibrant manufacturing activities makes them loom large in production and trade shares. This is stage zero on a long road to industrialization.

From the East Asian perspective, economic take-off starts with the arrival of a sufficient mass of manufacturing FDI firms that perform simple assembly or processing of light industry products for export such as garment, footwear, and foodstuff. Electronic devices and components may also be produced this way. In this early stage (stage one), design, technology, production and marketing are all directed by foreigners, key materials and parts are imported, and the country contributes only unskilled labour and industrial land. While this generates jobs and income for the poor, internal value remains small and value created by foreigners dominates. Vietnam's industrialization up to now has been basically characterized by this situation.

Table 1.

Vietnam: Summary of Growth Performance

In the second stage, as FDI accumulates and production expands, the domestic supply of parts and components begins to increase. This is realized partly by the inflow of FDI suppliers and partly by the emergence of local suppliers. As this occurs, assembly firms become more competitive and a virtuous circle between assemblers and suppliers sets in. The industry grows quantitatively through the internal supply of physical inputs. Internal value creation rises moderately, but production basically remains under foreign management and guidance. Obviously, local wage and income cannot rise very much if all important tasks continue to be performed by foreign hands. Thailand and Malaysia have already reached this stage.

The next challenge is to internalize skill and knowledge by accumulating industrial human capital. Locals must replace foreigners in all areas of production including management, technology, design, factory operation, logistics, quality control, and marketing. As foreign dependence is reduced, internal value rises dramatically. The country emerges as a dynamic exporter of high-quality manufactured products challenging more advanced competitors and reshaping the global industrial landscape. Korea and Taiwan are such producers.

In the final stage, the country acquires the capability to create new products and lead global market trends. Japan, the United States, and some members of the European Union are such industrial innovators.

However, progress is not guaranteed for all. A large number of countries that receive too little manufacturing FDI stay at stage zero.³ Even after reaching the first stage, climbing up the ladders becomes increasingly difficult. Another group of countries are stuck in the second stage because they fail to upgrade human capital. It is noteworthy that none of the ASEAN countries, including Thailand and Malaysia, has succeeded in breaking through the invisible "glass ceiling" in manufacturing between the second and the third stage.⁴ See Figure 1. A majority of Latin American countries remain middle income even though they had achieved relatively high income as early as in the nineteenth century. This phenomenon can be collectively called the middle-income trap.

East Asian growth performance has differed significantly in depth and speed even among countries that are considered "successful". There should be a clear distinction among Taiwan and South Korea (high achievers), Malaysia and Thailand (middle achievers), and Indonesia and the Philippines (low achievers). The first group is far ahead of the second or the third in terms of income and industrial capability.

Figure 1.

Stages of Catching-up Industrialization

Figure 2 shows per capita real income of selected East Asian economies relative to the United States level. Until the mid-1960s, these economies (except Japan) showed no clear sign of catching up. However, Taiwan and Korea, which started from equally low levels, took off in the late 1960s and have improved income dramatically. In comparison, the catching up of Malaysia and Thailand looks less impressive, and Indonesia and the Philippines failed to improve their positions vis-à-vis the United States. In addition, there are economies which are not even on our radar screen —those that continue to struggle at the bottom of income ladders such as Cambodia, Laos, Myanmar, East Timor, and North Korea. Divergent performance comes from different speed of catching up rather than delayed starts (except Vietnam where wars and socialist planning prevented economic take-off until the early 1990s). ASEAN-4 are taking much longer to reach the industrial capability that Taiwan and Korea had achieved in the 1980s and 1990s.

Figure 2.

Different Speed of Catching Up

(Per cent of US real income)

Starting from a very low level, Vietnam is currently in the first stage of industrialization trying to reach the second in Figure 1. Large FDI inflows, a necessary condition for this transition, are already happening. Neighbouring ASEAN countries even fret about losing FDI to Vietnam. While Vietnam's short-term goal is the attainment of physical expansion of the industrial base, it should also simultaneously prepare to avoid the middle-income trap in the next stage. For this, front-loaded and well-targeted policy action for upgrading industrial human resources is the key.

In order to overcome the middle-income trap, a developing country needs to acquire capability to embrace an appropriate industrial vision and implement effective measures towards it. Required action is more aggressive than suggested by the Washington Consensus. Deregulation, privatization, integration, and presence of a sound business environment are good enough up to stage two in Figure 1, but insufficient to improve skill and technology and break the glass ceiling towards stages 3 and 4. This is true even in the twenty-first century when globalization has deepened and WTO rules and FTA proliferation have significantly narrowed the policy space of latecomer countries.

Even under the restricted policy space currently available, however, it is possible to design and execute meaningful strategies to accelerate industrialization. For example, the promotion of supporting industries and industrial human resources does not violate WTO rules at all. Measures to enhance infrastructure, logistics, industrial clusters, technology transfer, education and training, FDI marketing, SME finance, factory evaluators, industrial parks, and so on, are also permissible under the current international regime.

At the same time, it should also be recognized that the catching up of latecomers is becoming increasingly difficult for the following three reasons. First, because of forced early integration, they are not given temporary protection periods which were available to their predecessors. Second, today's latecomers generally lack a strong private sector comparable to Japanese industrial groups, Korean chaebols, or Chinese and Indian merchant networks. Third, their governments are often without developmental orientation or sufficient policy capability. The last two can be regarded as weaknesses associated with the losers' bias. If they initially had a strong private sector and a good government, they would have joined the flying geese much sooner and would not have stayed poor until now. How to overcome these latecomer problems in the early twenty-first century will be the topic of the remaining sections.

The point that developing countries must acquire skill and technology, rather than just offering factory space and cheap labour, can be stressed in various ways. Four such arguments are presented below to make this point from different angles.

First, at the general level, it can be argued that the only way for a country to remain competitive is to improve labour productivity faster than wage increase. Competitiveness depends on the difference between the two, not on the absolute wage level. Wage increase should be a boon to workers, and there is no reason to fear it as long as productivity is improving in tandem. In the context of Vietnam, this point has consistently been made by Professor Tran Van Tho of Waseda University since the mid-1990s. Under wage pressure, Malaysia and China have already stopped inviting labour-intensive FDI projects and turned to more "high-tech" investors. Vietnam is also experiencing rising wages as a result of large concentration of labour-intensive FDI in some areas such as northern Dong Nai as well as an inevitable response to the 2007–08 inflation. If wages begin to rise rapidly now, Vietnam may not have enough lead-time to improve productivity.

Second, the concept of "manufacturing plus plus", which governed Malaysia's Second Industrial Master Plan (IMP2) 1996–2005, is instructive because it concisely states what middle-income countries should do to climb up to stage 3. Manufacturing plus plus expresses the two dimensional desire for domestic industries to: (i) expand along the value chain to encompass higher value-added activities; and (ii) uplift the whole value chain by raising productivity (Figure 3). Since Malaysia started industrialization as a conventional assembler, which was the lowest point in the value chain, it wanted to master R&D, design, product development, distribution, marketing, and so on horizontally, and improve the skills of all these activities vertically. In principle, this is what Vietnam —and all other latecomers —should do. IMP2 selected eight industrial clusters to be thus strengthened: electronics and electricals, textiles and apparel, chemicals, resource-based industries, food processing, transportation equipment, materials, and machinery. However, Malaysia did not succeed greatly in achieving this goal during the implementation period of IMP2 (Ohno 2006).

Third, the Japanese concept of monozukuri, which literally means "making things", may give some hints on the direction to go. Monozukuri is manufacturing for the primary purpose of achieving customer satisfaction through high quality in the spirit of a proud and dedicated artisan, rather than just making profits. To achieve this, long-term relationship and internal accumulation of skill and knowledge are institutionalized within each company as well as among partner companies (between assemblers and suppliers, for example). Practical means of productivity improvement such as 5S, QCD,⁵kaizen, just-in-time method, and quality control circles have been established and available to companies in the developing world through experienced instructors and manuals. In the policy realm, the concept of monozukuri is often highlighted by the Japanese government for the purpose of upgrading domestic manufacturing capability and spreading the Japanese business model abroad (Tsai 2006).

Figure 3.

The Manufacturing Plus Plus Strategy of Malaysia

Fourth, the theory of business architecture advanced by Takahiro Fujimoto and his research team at the University of Tokyo elaborates how firms in developing countries can form strategic alliance with Japanese manufacturing firms (Fujimoto 2004, 2006; Fujimoto and Shintaku 2005). According to this theory, business models can be divided into two broad categories: modular and integral. Modular manufacturing is characterized by easy assembly of globally common parts and components (for example, a desktop computer) while integral manufacturing features unique design of parts and components for each model based on long-term collaboration among assemblers and suppliers (for example, a passenger car). The former is suitable for realizing quick profits under flexible combination of outsourced business components while the latter permits a continuous pursuit of high quality over time. Fujimoto argues that the United States and China are appropriate production partners because they both practise modular manufacturing. Meanwhile, Japan is an integral producer without an effective international partner. For developing countries, integral manufacturing is harder to learn but eventually more rewarding as production technology is internalized rather than outsourced. While none of the ASEAN countries has acquired sufficient skill and technology for integral manufacturing, Fujimoto regards Thailand and Vietnam as likely candidates for Japan's future monozukuri partner provided that they level up their internal capability (Fujimoto and Ohno 2006).

While the Malaysian experience or the Japanese business theory may not fit every country, they point clearly to the importance of internal value creation through skill and technology and the existence of concrete strategies and methods to attain it.

III. Policy Vision and Orientation

In high performing economies in East Asia, industrial policy has usually taken a goal-targeting form. The top government leader launches a long-term national vision which shows a general direction without specifying details. To realize this, appropriate government organizations are designated or newly created to draft feasible strategies and execute concrete action plans. Action plans may take the form of readable documents and matrices or may remain a process without such documentation. Strategies and action plans may be revised as circumstances change, but the long-term vision remains intact.

Japan in the 1960s had the goal of doubling income within the decade as well as competing effectively with Western multinationals as trade barriers were lifted. The Ministry of International Trade and Industry (MITI) together with the Japan Development Bank coordinated and assisted private efforts in improving productivity. In Malaysia, Vision 2020, an aspiration to become a "fully developed country" by 2020 set by former Prime Minister Dr Mahathir in 1991, remains the overarching goal. The Economic Planning Unit (EPU) of the Department of the Prime Minister directs national effort to concretize this vision under a system of overlapping policy documents and cascading organizations.⁶ Thailand under Prime Minister Thaksin Shinawatra (2001–2006) put up industrial visions which were both ambitious and ambiguous, such as becoming the "Detroit of Asia", the "Hub of Tropical Fashion", or the "Kitchen of the World", while leaving the details to be worked out among relevant ministries, private businesses, and experts. For execution, industry-specific committees and industry-specific institutes were established, and the private sector additionally had direct access to the prime minister when necessary (Ohno 2006).

This policy formulation method, which has been the hallmark of successful East Asian development policies, can be summarized as working backwards from broad goals to phased strategies and concrete action plans, while making necessary adjustments and accumulating experience and confidence along the way. This pragmatism, which we prefer to call Dynamic Capacity Development, allows the gradual building of policy capability as concrete problems and challenges are encountered over time.

At the beginning of industrialization, most East Asian countries had weak governments. In 1960, the Korean civil service was widely viewed as a corrupt and inept institution (World Bank 1993). Similarly, in 1959, Thailand was given a low mark for the absence of investment planning and an acute shortage of qualified personnel (World Bank 1959). But through trials and errors and learning by doing, their administrative capacity has greatly improved. This hands-on approach is in sharp contrast to the current global aid practice, such as the good governance drive,⁷ where all countries are urged to correct their weaknesses ex ante relative to some international norm without reference to any concrete national goal and before formulating a specific growth strategy.

From this perspective, Vietnam's industrial vision leaves much to be desired. Vietnam already has a long-term vision of attaining industrialization and modernization by 2020. The ambiguity of this vision does not worry us too much as with the case of Dr Mahathir's 2020 vision or Mr Thaksin's call for becoming the Detroit of Asia. However, the problem with Vietnam is the lack of proper strategies, action plans, and institutions to follow up on this vision. The present administrative system does not permit necessary policies to be drafted and implemented.

It is essential that Vietnam formulate as soon as possible a clear roadmap of industrialization to inform and guide its people, investors, and policy-makers. It should outline a strategic path towards the 2020 vision backed by concrete action plans. Vietnam should declare, among other things, its strong resolve and clear plan to secure an important position in the East Asian production network. It should affirm that the private sector, not the state or state-owned conglomerates, should be the agent of production and investment; that growth should be driven by the skill, technology, and hard work of the Vietnamese people; that openness and the market mechanism are defended as a matter of principle; and that the state will actively support and coordinate the private sector without dictating its business plans. Policy orientation in the areas of savings mobilization, financial development, usage of foreign resources, income gaps and other emerging social issues, and sectors under external competitive pressure should be clarified.

At present, Vietnam does not have an overall industrial master plan. The industrial chapters of the Five-year Plan and the Ten-year Strategy do not offer a consistent industrial vision. As a result, many important policy questions remain unanswered, including the future roles of SOEs, private firms, and FDI, respectively; the choice between export orientation and import substitution under deepening integration; and the scope and extent of official support to emerging as well as declining industries. Sectoral master plans for steel, automobiles, motorcycles, electronics, textile and garment, and so on, are being drafted and approved without overarching principles at a higher level. Private investments and official aid pour in without knowing exactly where Vietnam is headed in the coming decades. In this connection, it should be noted that some countries, with much lower income levels than Vietnam, already have industrial visions and action plans which are more consistent and far detailed than those of Vietnam.⁸

One of the issues in promoting a mechanical industry under globalization is the choice between direct and indirect promotion (infant industry promotion versus FDI-led industrialization). Malaysia established Proton, a national car company, in 1983 and supported it with heavy subsidies and protection. Starting from the knock-down production of Mitsubishi Lancer, Proton subsequently internalized capability in styling and design, platforms, engines, logistics, marketing, and so on. By 2005, Proton had become the largest supplier of passenger cars in Malaysia with the domestic market share of over 40 per cent and 286 local suppliers producing its parts. However, as globalization deepened, it became apparent that Proton's production volume was too small and technology not high enough to compete with global giants from Japan, Korea, EU, and the United States, and it also began to lose the domestic market as foreign models invaded. The strategy of internalizing capability under strong official support has hit a thick wall. By contrast, Thailand created a relatively free environment for FDI car makers to achieve large production volume, quality, and even exports. By not insisting on national brands, it succeeded in creating the largest automotive cluster in Southeast Asia. However, Thailand's problem is the slow pace of domestic capacity building and the continued dominance of foreign design and technology.

Vietnam has not clearly stated whether or how it wants to promote such industries as automobiles, audio-visual devices, home electronics, and general machinery. Under the current situation in which discriminatory measures are no longer permitted under WTO rules, refraining from supporting such industries and letting the market decide their fate is one option. But if the Vietnamese government wants to promote them, it must do some serious thinking to see what are realistic goals and what strategies and action plans can be adopted without violating international commitments.

IV. Policy-making Procedure and Organization

Vietnam's failure to produce effective industrial strategies and action plans comes mainly from the structural weaknesses in policy making. Vietnam's policy formulation is saddled with the legacies of planning days and cannot cope effectively with problems in the age of global competition. After the growth bout of the 1990s and the early 2000s driven by economic liberalization and large capital inflows, Vietnam has reached the point where further progress towards higher income is increasingly difficult without a radical reform in policy formulation procedure and organization.

The problems associated with Vietnam's industrial policy making are many. However, instead of presenting a long list of problems, we will highlight just two procedural problems and two organizational problems which are inter-related and constitute the main sources of formalism and the general lack of creativity and responsiveness in policy making. These problems are unique to Vietnam in the sense that they are not observable in East Asia's other high performing economies.⁹

The most serious procedural problems in designing and executing industrial strategies and action plans are the lack of involvement of the business community and the lack of inter-ministerial coordination, which together render approved policies ineffective and even unimplementable. In any developing country, policy implementation is a big challenge due to shortages of budget, human resources and proper mechanisms. However, the proportion of unimplemented policies in Vietnam is exceptionally high not only in industrial matters but also in other policy areas. It can even be said that very few policies are actually implemented as stipulated in Vietnam because of delays in preparing "implementation details"; the non-provision of necessary budget, personnel or equipment; the lack of support from the business community; and the lack of ability or interest among responsible ministries to solve these problems.¹⁰

The policy-making process in Vietnam is closed within the government with little involvement of other stakeholders. Within each ministry, an order to draft a master plan is handed down to a drafting team, which normally consists of a middle-ranking official supported by a few experts in the ministry. The team collects internal data and data from other ministries, and may commission additional analyses to experts in other ministries or research institutes. The budget for each master plan is fixed by an inter-ministerial circular and used mainly for securing external data and analyses as well as conducting domestic travel, interviews and hearings. The master plan is drafted internally by the team members and submitted to the minister or the vice-minister in charge for internal review. After that, it is circulated among relevant ministries for comment (which is rarely substantive) and then submitted to the prime minister for final approval. Significant delay may occur at internal review or final approval. Requests for revision are also common. In this process, debates on the fundamental direction or crucial issues rarely take place. The drafting team is routinely overworked with a large number of master plans to finish each year, which does not allow sufficient time (or money) to think creatively, interact with non-government stake-holders, or publicize the final result. Approved master plans are neither translated into English nor uploaded for dissemination although summary versions for the prime minister's approval, in the Vietnamese original, are usually available on the web.

If a domestic or foreign firm wants to raise its voice, it must devise its own way since the current procedure does not allow meaningful involvement of the business community. Although enterprise hearings are becoming more popular in recent years, sufficient details of the master plan draft are not revealed at such hearings and enterprises therefore can only make general requests. If a firm later finds certain points in the master plan objectionable (for example, demand forecasts, taxes and import duties, numerical targets for production or export, designation of producers for certain products, and so on), it needs to seek meetings with responsible ministries, use symposiums and media to make the point, or write a letter to the prime minister, to request a change in the already approved policy. This situation is in sharp contrast to Malaysia, where private sector participation is institutionalized as members of the steering committee and task forces in drafting the Industrial Master Plan; Thailand, where the private sector decides targets and action plans and the government merely accepts them; or Japan, where business decisions on technology, products, investment and so on are left to individual firms and the government provides only supplementary services such as trade negotiation and setting standards for quality, safety, environment, and industrial property (Ohno 2006).

Another procedural problem is the absence of inter-ministerial coordination on policy substance as well as implementation details, which in turn comes from the lack of mechanism to force different ministries to work together. Compartmentalization of the government along ministerial lines is a common problem around the world, but most governments manage to somehow ameliorate it. One solution is to have a strong top leader with a good economic mindset who directs various ministries and becomes the hub of policy making himself. In this case, policy components become mutually consistent even though ministries still fail to talk to each other (Thailand under Thaksin Shinawatra, 2001–06; Ethiopia under Meles Zenawi, 1991–present). Another way is to establish a powerful technocrat team directly serving the president or the prime minister which makes key policy decisions while ministries become executing agents of the plans emanating from this team (South Korea's Economic Planning Board, 1961–1994; also see below). Still another way is to let a super ministry, with sufficient policy authority and instruments at its disposal, lead industrial policy making and be responsible for it (Japan's Ministry of International Trade and Industry in the 1960s). Finally, it is also possible to install a mechanism to guarantee the representation of all relevant ministries and non-government stakeholders in the official drafting process as well as in informal exchange (Malaysia's drafting of the Industrial Master Plan at present). In Vietnam, though all policy documents specify a leading ministry and a list of related ministries, the mechanism to make them work as one is entirely missing.

We can go deeper to see why it is difficult to ensure involvement of non-government stakeholders and inter-ministerial coordination. Behind these problems lie fundamental issues in policy making organization. The most serious ones in this regard are the lack of clear directives from the top and the distorted incentive mechanism among government officials that causes brain drain.

It is well known that Vietnam's decision making is based on consensus. Checks and balances are in place horizontally (across ministries and departments), vertically (between central and local levels) and geographically (North, South, Middle and remote areas). There are three top national leaders, and the Party and the Government interact in a complex manner. This system can produce stability and continuity but it is not suitable for staging bold reforms or responding quickly to the changing world. Policies remain mostly reactive rather than proactive. Development effort centred on a clear roadmap towards a national vision with concrete strategies and action plans, which is the hallmark of East Asian industrialization, is missing in the Vietnamese policy process.

The Vietnamese government copes with urgent issues —be it inflation or traffic jam —in a bottom-up fashion and without a clear focal point of leadership or responsibility. When a serious problem is identified, an inter-ministerial committee is called and its chair is appointed. Each ministry proposes solutions from its perspective, which are summarized into general policy recommendations without execution details. Bureaucracy can supply broad ideas touching every aspect of the problem, but it does not lead to prioritization or selectivity for real action. This approach must be supplemented by a person or an organization that decides on a short list of actions and sequencing of measures among many proposals. There should be an interaction between the high level and the implementing level of the government to produce policies which are both realistic and sharply focused.

Another problem which is common in many countries and also becoming highly visible in Vietnam is the decline of quality and morale among government officials, prompting an exodus of talented people to other sectors. Vietnam's public service must overcome the problems of overstaffing, low salary, prevalence of second jobs, formalism, rigidity, nepotism, corruption, relation-based promotion, and ODA-related benefits (foreign travel, training, benefits associated with supervising aid projects, etc.). These were the legacies of the subsidy system existing up to the 1980s, where the public sector was the provider of jobs, minimum income and social security for all, and where no alternative employment opportunities were available in the private or foreign sectors with far more attractive salaries and rewarding duties. Under the present circumstance of market orientation and global integration, the public sector only attracts people who want stability, people who genuinely believe in the importance of public service, or people who want to take advantage of official privileges to study abroad or receive training as a stepping stone to a better-paying job in the future. As a result, highly qualified and motivated people are becoming difficult to recruit or retain.

This problem cannot be solved by minor repairs or ad hoc adjustments. ODA-supported training programmes of government officials may only worsen the brain drain without raising the average level of official competency. To reverse the hollowing-out of the Vietnamese government, far reaching reforms to completely remake the public administration is needed as soon as possible. This should encompass, among others, a significant down-sizing of the public sector through leaner organization, forced retirement, and outsourcing of non-essential services; a competitive and transparent recruitment system; a higher and performance-based salary schedule and promotion linked to transparent personnel evaluation; and clear rules regarding the conduct of public servants and their interaction with citizens, businesses, and service providers. Obviously, these are not easy because of the magnitude of required tasks and political resistance. But they are also absolutely necessary for Vietnam to move forward. Vietnam's public administration lags far behind other successful economies in the region such as Singapore, Malaysia, and Thailand. It should also be mentioned that the initiative for such reforms must come from the top rather than the bottom. No bureaucracy can transform itself so radically without the order from a strong leader.

V. How to Break a Solidified System

To propose a solution is one thing. To carry it out is quite another. Even if Vietnam knows the best policy formulation procedure and organization, how can it make sure that they are actually adopted?

According to comparative institutional analysis, a branch of institutional economics that relies heavily on evolutionary game theory, a society may get stuck in a bad equilibrium owing to institutional complementarity, strategic complementarity, and path dependence (Aoki 2001a, 2001b). Institutional complementarity means that any social system has resilience to shocks because its institutional components enhance each other. For example, Vietnam's education, recruitment, salary and promotion systems are mutually complementary to produce relation-based rent sharing. Strategic complementarity means that individuals in such an institutionally solidified society have little incentive to deviate from the dominant behaviour. Finally, path dependence underscores the importance of the beginning. Once installed by chance or design, any social system requires a large amount of political and social energy to change it. Together, these concepts point to institutional inertia and difficulty of reforming any established system.

Policy impasse arises when an inefficient method of policy formulation is set up and then solidified, and institutional components and people's attitude to support it have formed. Removing one person or reforming one organization does not improve the situation because of institutional and strategic complementarities mentioned above. Changing the policy formulation method in a fundamental way, as proposed by this paper, will surely require enormous energy and meet fierce resistance.

However, this does not mean that there is no way out. There are times when a social system jumps to another social system. Comparative institutional analysis suggests the following occasions and agents of change.

1. i. Collective mutation: A large number of people inside a society may mutate simultaneously, as if their DNA has changed. If only a few people behave differently, they are simply called "crazy" or "silly" and the system remains unchanged. But a sufficiently large mass begin to behave differently, institutional and strategic complementarities of the old type stop working, and rules and customs start to change. This is a spontaneous and internally driven change, which may occur when a large number of people feel suppressed or victimized under the existing system. In a rapidly growing economy, this may also happen when a generation with new values and behavioural patterns grow up, or when people begin to have new demands and expectations from the government as a result of successful development and higher income. A small incident may trigger a large social movement by letting accumulated public discontent to come to the open.

2. ii. Foreigners: Foreign governments, firms and individuals follow different systems and are not bound by the behavioural code of the domestic society. They bring and sometimes even force new elements, which causes friction and inconsistencies with the indigenous system. In low-income countries, bilateral donors and international organizations are particularly powerful. Foreign firms and investors as well as international migration and human exchange may also produce foreign pressure on a society. If this prompts a change in a direction that generates healthy development, such pressure is highly welcome. However, not all foreign influences are good from the viewpoint of social evolution. For this reason, the government must guide and coordinate foreign pressure to prevent undesirable changes.

3. iii. Policy: Even without domestic or foreign pressure, the government as deux ex machina can start a change from inside the system by introducing policies that upset existing calculations and complementarities. Here the key question is who will activate such policies. As noted before, it is extremely difficult for bureaucrats to initiate a bold reform. Their power within the government is miniscule compared with enormous institutional and strategic complementarities they face. Drastic policy shifts are usually introduced when a new, strong top leader comes to power. Leadership equipped with strong will and economic literacy is crucial for this to succeed. When such leadership skillfully and strategically aligns with foreign partners who want go in the same direction, even a very bold reform becomes possible.

In view of these theoretical implications, let us identify three players that may make institutional reforms possible in the Vietnamese context. They are leadership, the technocrat team, and foreign partnership.

V.1 Leadership

Crucial importance of leadership is made sufficiently clear in the discussions above. Leadership is the prime force of change while other necessary conditions can be created or reshaped by the leader if they do not already exist. In countries with advanced political systems, policy initiative can also emerge from various domestic groups such as civil society organizations, intellectuals, interest groups, and political parties because legal mechanisms to capture and reflect their opinions are firmly in place. However, in developing countries where political systems are less well developed, only a small number of channels of effective participation are available. For all practical purposes, initiative for bold change in these circumstances must come from the top leader. When such leadership is combined constructively with the aspiration of domestic groups and foreign pressure, reforms become possible. For the leader to play proper roles in development, it is not always necessary to change the existing political regime or expend social energy to change it. The Vietnamese political regime at present is flexible enough to allow a strong leader with political savvy to emerge and orchestrate policies.

V.2 The Technocrat Team

In high performing economies of East Asia, the existence of a technocrat team directly under the top leader (the president or the prime minister) has played a crucial role. This team is created from the brightest officials from various ministries as well as the smartest returnees who have studied or taught abroad. Prominent business leaders with strong policy mindset may also be mobilized. The team receives full confidence and responsibility from the top leader to concretize the policies that this leader envisions. It also acts as the command post for all ministries which are obliged to implement the policies that this team drafts. It acts as the nation's brain for development without which even excellent leaders cannot function. The Economic Planning Board in South Korea, the Kuomintang technocrats in Taiwan, the Economic Planning Unit (EPU) in Malaysia, the National Economic and Social Development Board (NESDB) in Thailand, the so-called Berkeley Mafia in Indonesia, and the National Economic Development Authority (NEDA) in the Philippines, all aimed to fill this need at certain critical points in their economic development with varying degrees of success. Japan's Ministry of International Trade and Industry (MITI), although being one of the ministries rather than above all ministries, also operated effectively to strengthen the competitiveness of Japanese manufacturing industries in the high growth period of the late 1950s and the 1960s.

Vietnam also had the Prime Minister's Research Commission (PMRC) until recently, but it was an advisory group rather than a central policy-making body entrusted with the power to lead the entire government. Its responsibility was too weak and its members were experienced but perhaps too old. Nor does Vietnam have a super-ministry such as Japan's MITI to centrally coordinate development effort; the Ministry of Planning and Investment (MPI) is not strong enough in terms of authority, capability and policy instruments to undertake this task. It is strongly suggested that Vietnam create a new dynamic technocrat team within the government as a focal point of policy-making authority and responsibility. In its design, experiences of other East Asian countries, with necessary modifications, should be referenced. Vietnam needs such a team at least for the next few decades to climb to higher income and cope with growth-generated problems and instabilities.

V.3 Foreign Partnership

Vietnam's foreign policy shifted dramatically in the early 1990s when the close ties with the Soviet bloc were replaced by multi-directional diplomatic relations and re-integration into the global economy. Since then, interaction with foreign actors has exerted indirect and subtle influences on Vietnam's development orientation although the Vietnamese government never allows foreigners to take the driver's seat (I. Ohno 2005). Bilateral and multilateral donors have registered their desire to see faster reforms and more administrative transparency and efficiency on such occasions as the semi-annual consultative group (CG) meetings, comments on the Five-year Plan and the Ten-year Strategies, policy dialogue for the Comprehensive Poverty Reduction and Growth Strategy (CPRGS) and the Poverty Reduction Support Credit (PRSC), and so on. Foreign businesses also have pressed the government to improve the legal and policy framework, the tax and import duty system, and other business-related matters through the Vietnam Business Forum, government-business dialogue, trade fairs, and symposiums. As Vietnam graduates from the status of a low-income transition country into the status of an industrializing middle-income country, the focus of foreign concern should also shift from the removal of the negatives to the creation of Vietnam's unique strengths.

As the leading economy in East Asia, Japan has also contributed significantly to Vietnam's development through trade, investment, aid, and human and knowledge exchange. Japanese businesses and officials are particularly interested in bolstering Vietnam's industrial competitiveness and have initiated a number of bilateral programmes to this end. They include the building of infrastructure especially in power and transportation, education and training of industrial human resources, and a series of action-oriented bilateral policy dialogues (Table 2).

These bilateral dialogues aim to improve Vietnamese policies where Japan has particular interest or comparative advantage. At the same time, they have the additional purpose of institutionally correcting the weaknesses of Vietnam's policy formulation by introducing new procedures and organizations. For example, concrete action plans are bilaterally agreed and

Table 2.

Vietnam-Japan Bilateral Policy Dialogue for Industrial Competitiveness

rigorously monitored to prevent non-implementation (the New Miyazawa Initiative, the Vietnam-Japan Joint Initiative, and the proposed Vietnam-Japan Monozukuri Partnership). Inter-ministerial cooperation is ensured by making the leading ministry, typically MPI, responsible for the participation of all other ministries (the Ishikawa Project, the Vietnam-Japan Joint Initiative, and the proposed Vietnam-Japan Monozukuri Partnership). Active involvement of non-government stakeholders (especially major manufacturers) was enforced throughout the joint drafting process of the Motorcycle Master Plan —perhaps for the first time in Vietnam's master plan drafting. Japanese officials and businesses are well aware of the structural shortcomings of Vietnam's policy making, and they are willing to spend time and energy to work with the Vietnamese side to solve them, without which they know their dialogue will not lead to meaningful actions.

These policy dialogues have so far been initiated mainly from the Japanese side. It is suggested that the Vietnamese government should be more pro-active in reforming its policy formulation and inviting Japan (and other countries) to participate in the effort.

VI. Concluding Remarks

While Vietnam's past achievements as a developing and transition country are great and many, this paper has focused on the future and offered candid evaluation and advice so that Vietnam might develop its potential to the fullest extent. I trust that the Vietnamese people and government are not satisfied by merely achieving MDGs or stopping at middle income. Their aspiration must be set higher, and it is surely attainable if the nation clearly identifies its present shortcomings and squarely faces its challenges. The key message of this paper can be summarized as follows.

Vietnam has reached the point where further progress towards higher income can be secured only if internal value creation is enhanced. This calls for proper government action, rather than laissez faire, to guide and complement private sector dynamism and avoid the middle-income trap. To improve policy quality, Vietnam needs to change the policy formulation process. This in turn requires a radical change in the public administration system. The scope and sequencing of reforms must be chosen carefully to minimize the political and social energy needed to change the system while maximizing their positive impacts. Enlightened and strong leadership, a new technocrat team, and strategic partnership with foreigners have been proposed as effective starting points that satisfy these conditions.