Chapter 2. A new growth path: Unlocking the potential of regions

Thailand’s growth path has created large disparities that pose a risk to development. However, a closer look at the country’s peripheral provinces and cities suggests that convergence is underway and that much can be done to boost this trend. Productivity drives economic performance and enables rising incomes. Policy makers therefore need to learn lessons from Thailand’s best-performing provinces and cities. This chapter examines previous policies designed to develop Thailand’s regions and assesses the resulting productivity landscape across regions. Given the limited available resources, past policies have focused on the few areas of the country that showed rapid success. Going forward, more broad-based and innovative regional policies that put local innovation in the driver’s seat, provide flexible support and that learn from best performers will be necessary. An analysis of provinces at the productivity frontier points to superior human capital and public services as the distinguishing attributes of high-performing provinces and cities. Based on these insights, regional policies will need to support cities as the centrepieces of integrated regional development policies and focus on skills development as a tool of regional and urban policy.

    

Regional and urban disparities in Thailand present a pressing structural challenge that must be addressed by boosting convergence

Thailand has undergone a deep structural transformation over the last few decades, but has recently lost momentum. Since 1970, GDP growth per capita has averaged 4.2% per year, and in 2016 income per head stood at 42% of the OECD average. Thailand has evolved from a largely rural country into a hub of manufacturing and services, fully integrated with global value chains for automobile and electronics (Annex Figure 2.A.1). As the country changed the structure of its economy, manufacturing activities began to attract an increasingly larger share of workers. The share of the primary sector in total GDP fell from 23% in 1990 to 9% in 2017; however, agricultural activities still employ one-third of the Thai workforce, mostly in informal and vulnerable jobs. As a result of informality, as well as weak demand arising from lacklustre global trade and slow capital formation, Thailand’s recovery has not yet returned to pre-global financial crisis growth rates (OECD, 2018[1]).

Figure 2.1. GDP per capita across Thailand is so diverse that different regions can be compared to countries at all stages of development
picture

Note: For comparative purpose, GDP per capita is measured in PPP (constant 2011 international USD).

Source: Authors’ work based on national accounts as provided by the NESDB and (World Bank, 2017[2]).

 StatLink https://doi.org/10.1787/888933847543

Today, Thailand’s regions are at widely different levels of economic development, illustrated here by countries at various stages of development (Figure 2.1). The Bangkok Metropolitan Area is the best performer, with GDP per capita (in PPP (constant 2011 international USD) at levels similar to Poland and Hungary at around USD 25 000. The East follows closely with about USD 23 000 that places it between Latvia and Chile. The Centre also performs above the national average with a GDP per capita similar to that of Brazil or Algeria (about USD 15 000). Thailand’s average of about USD 11 500 per capita is similar to Peru and Mongolia. The poorer provinces are the West and South, with about USD 7000 similar to Guatemala or the Philippines. The North and Northeast are significantly poorer and with levels of per capita GDP close those of Pakistan or Honduras in the case of the North (USD 4 400) and to that of Ghana or Zambia in the case of the Northeast (USD 3 700).

Bangkok and a few industrial and touristic hubs have been driving structural transformation. Between 2001 and 2015, the share of manufacturing activities increased in the Central region and now amounts to 63% of regional GDP.1 In the same period, industry (including manufacturing and construction) minimised the role of the agricultural sector in the East, and in 2015 accounted for more than 50% of local GDP. Over time, the Bangkok Metropolitan Area has become a major pole of services, with trade, transport, real estate and business activities representing 54% of GDP – 10 percentage points higher than in 2001. In addition, some provinces in the South – namely Phuket and neighbouring areas – have become top destinations for global tourism.

Figure 2.2. Thailand’s past urbanisation and growth path reflects similar global patterns of transformation
Evolution of GDP-urban share relationship for Thailand between 1990 and 2015
picture

Note: Empty markers represent the share of the urban population and GDP per capita for countries for which data are available in 2015. The blue markers represent the same relationship but across time (from 1990 to 2015) for the three selected countries. For comparative purpose, GDP per capita is measured in PPP (constant 2011 international USD). Source: Authors’ calculation based on (World Bank, 2017[2]).

 StatLink https://doi.org/10.1787/888933847562

Thailand’s experience of high population concentration corresponds to those of other countries at similar stages of structural transformation. Structural transformation refers to the creation of new, more productive activities and the reallocation of labour towards these new activities. In the early stages of a country’s transformation, the creation of new, highly productive activities is necessarily concentrated in terms of space, as the capabilities and capital involved are very limited. Driven by these limitations, proactive development policies tend to reinforce concentration at this stage of development. Thus, these early poles of growth and transformation correspond to an urban centre where the elites of business, government and research are concentrated. This was the case in Viet Nam, South Korea and several other countries. Indeed, the profile of Thailand’s urbanisation and corresponding growth fits well with global patterns (Figure 2.2).

The same trend is observed in Thailand’s urban landscape, where urbanisation plays an important role in economic development, but remains highly concentrated in Bangkok. Because of their high density, cities are a fertile ground for economies of agglomeration and are conducive to a country’s growth. Dense urban clusters facilitate the formation of networks and interactions that allow for a more dynamic flow of knowledge, experience and information (Henderson, 2005[3]; Annez and Buckley, 2009[4]; OECD, 2015[5]). At the beginning of the 1960s, one out of four Thais lived in urban areas. In 2015, for the first time, more than half of the population lived in urban areas and in 2017, 36 million people lived in cities. While one-third of these live people in Bangkok and the vicinity, no other urban cluster has more than 500 000 inhabitants.

Figure 2.3. Global patterns suggest that the large disparities between Thailand’s regions pose a significant challenge on the path towards a high-income economy
picture

Note: The Gini index measures the degree of inter-regional inequalities. An index equal to 0 implies no inequality – all resources are equally redistributed across the country. An index equal to 1 implies extreme inequality – all resources are concentrated in only one region. The analysis is based on OECD regional typology, and in particular on regional inequalities among Territorial Level 2 (TL2) regions. TL2 broadly corresponds to the first tier of subnational government. For comparative purposes, TL2 regions in Thailand correspond to the country’s 77 provinces. GDP per capita is measured in PPP (constant 2011 international USD). Source: Authors’ calculations based on national accounts, as provided by NESDB.

 StatLink https://doi.org/10.1787/888933847581

However, global patterns also suggest that the large disparities between Thailand’s regions pose a significant obstacle to further transformation and must be addressed (Figure 2.3). As of 2015, Thailand’s Gini index of regional inequality is significantly higher than that of most other countries at comparable levels of economic development. In fact, except for the Russia, no other country for which these data are available, has achieved a higher level of GDP per capita with the same level of inequality between regions). With such high regional inequality, poorer areas struggle to develop their potential as opportunities and investments are lacking. At the same time, labour and capital markets in already advanced areas likely become saturated with an oversupply of workers and investments.

The track record of Thailand’s provinces over the last decade suggests that, when supported well, convergence can become a powerful engine of development and unlock new sources of growth and transformation. Although their levels are lower, the poorer regions of Thailand have shown consistently higher growth in both production and productivity, since the beginning of the 2000s, than Bangkok (Figure 2.4). With the right policy mix, much more could be done to support further convergence and structural transformation.

Figure 2.4. Poor regions have grown faster than richer ones, displaying potential for convergence
Correlation between real GDP per capita in 2001 and the average GDP growth rate between 2001 and 2015, taking into account the relative size of regional GDP
picture

Note: Calculations are based on chained volume measure of real GDP per capita. The Compound Average Growth Rate (CAGR) is used to measure the evolution of real GDP from 2001 to 2015. The size of the circle is proportional to the share of regional GDP to national GDP.

Source: Authors’ calculations based on national accounts, as provided by NESDB.

 StatLink https://doi.org/10.1787/888933847600

The same holds for Thailand’s secondary cities, which have grown faster than Bangkok and will play a key role in boosting regional development and Thailand’s long-term growth potential more broadly. Over the past 17 years, growth in the Bangkok Metropolitan Area has not kept up with the overall increase in the urban population. As structural transformation spread from Bangkok, the Centre and the East towards the rest of the country, current and prospective workers in rural areas migrated to secondary cities, seeking better industrial and non-farm activities (Christiaensen and Todo, 2014[6]). In most developing and developed countries, large metropoles such as Bangkok are rarely the main determinant of a country’s economic performance. Secondary cities help spread the linkages with global supply chains throughout the country, diversify the economic structure and labour force, and offer alternatives to young workers searching for new and more remunerative opportunities (Frick and Rodríguez‐Pose, 2018[7]). Their integration with surrounding rural areas plays an indispensable role in spreading the dividends of economic transformation to otherwise marginalised villages.

Policy makers need to learn from the best-performing provinces and cities, and target productivity, human capital and access to services. Productivity drives economic performance and enables rising incomes. The next section of this chapter examines past policies designed to develop Thailand’s regions and assesses the resulting productivity landscape across regions. Given the limited available resources, these policies focused on a few areas of the country that showed rapid success. Going forward, more integrated regional policies that learn from the best local performers and provide flexible support will be necessary (section 3). An analysis of provinces at the productivity frontier points to superior human capital and urban public services as the distinguishing attributes of high performance provinces. Based on these insights, integrated regional policies will need to support cities as the centrepieces of integrated regional development policies (section 4) and focus on skills development as a tool of regional and urban policy (section 5). Section 6 reviews a possible toolkit of policies that could support regional integrated policies. More details are discussed later in Chapter 3. Aspects of this analysis are shared by the 12th NESDP, which provides several important policy suggestions (Box 2.1).

Box 2.1. Regional development and the 12th National Economic and Social Development Plan

The 12th National Economic and Social Development Plan (NESDP) identifies the concentration of economic activities in Bangkok and the Central region as the main source of disparities among regions. On this basis, it provides guidelines to boost spatial development along the three main policy axes analysed by this chapter: tailor-made regional agendas, urban development and human capital formation.

Regional development. The objectives set by the Plan are: a more equitable distribution of regional growth and economic opportunities, a reduction in disparities in terms of Gross Regional Product per capita, and a reduction in the Gini coefficient among regions. The Plan also identifies development strategies for six regions: the Centre, East, North, Northeast and South regions and the Bangkok Metropolitan Area. Moreover, the Plan envisages a spatial agenda building on current place-based policies, in particular the special economic zones and the Eastern Seaboard Areas. This chapter recommends moving away from traditional place-based policies towards “integrated regional policies”. These polices are more flexible instruments to pursue socio-economic targets, resulting from consultation with local stakeholders in areas that transcend traditional administrative units.

Urban development. The 12th NESDP encourages provincial city centres to become liveable cities for all groups of people in society. It provides recommendations for the four main groups of existing urban centres: Bangkok and urban areas in the metropolitan vicinity, urban areas in Chiang Mai and Phitsanulok, urban areas in Khon Kaen and Nakhon Ratchasima, and urban areas in Phuket and Hat Yai. The current government also aims to enhance the infrastructure and capabilities of secondary cities, essentially to boost local tourism. This chapter recommends leveraging cities to unleash untapped local potential and promote integrated regional development. In addition, it proposes bypassing standard classifications of urban areas in favour of using innovative methodologies and big data to identify secondary cities.

Human capital formation. According to the 12th NESDP, the current education system does not provide pupils with the knowledge and skills that the labour market needs. Skills mismatch may ultimately lead to low-qualified and informal jobs. The Plan therefore recommends implementing improvements in educational quality through Dual Vocational Education and Co-operative Education, with a view to enabling the workforce to attain the requisite skills before entering the labour market. It also encourages the participation of regional universities and vocational institutions in the development of the local community. This chapter builds on the Plan’s existing guidelines and further emphasises the need for regional universities to “localise” the services they offer. It provides recommendations on how TVET and regional universities can help shape students’ skills and feed entrepreneurial initiative based on the socio-economic characteristics of integrated regions.

Source: NESDB (2017), National Economic and Social Development Plan, 2017-2021.

Past policies for regional development and structural transformation have created today’s productivity landscape

Past regional policies: an overview

Regional policies date back to the 1970s and initially targeted Bangkok and the East. At the time, the Ministry of Industry created the Industrial Estate Authority of Thailand (I-EA-T) to manage industrial estates, areas where factories could cluster to benefit from fiscal incentives and favourable infrastructure. The first General Industrial Zones were created in the Bangkok Metropolitan Area – Pathum Thani (1971) and Bangkok Metropolis (1973 and 1979) – and Chonburi (1976). As of 2015, the East region has the highest number of industrial estates (23) followed by the Centre and West (19).

Strategies to create alternative industrial development areas outside the East and Centre followed in the 1980s. The fifth national economic and social development plan (1982-1986) promoted a place-based approach, identifying 24 centres in the outer regions to generate non-agricultural employment opportunities through increased investments, industrial promotion and strengthening of local fiscal capacity. The plan identified a few of these centres as priority targets: Chang Mai in the North; Khon Kaen and Nakhon Ratchasima in the Northeast; and Songkhla, Phuket, Krabi and Surathani in South. In the 1980s, the Thai government launched the Eastern Seaboard Development Programme following the discovery of natural gas in the Gulf of Siam at the end of the previous decade. The programme aimed to boost industries that use gas as raw material as well as related downstream industries in Chonburi, Chachoengsao and Rayong. In the 1990s, the Thai Board of Investments increasingly sponsored local or foreign investors that chose to settle in rural areas, promote export-oriented activities and transfer technology from abroad.

Past approaches to regional strategies gave rise to path dependence in productivity growth and affected regional productivity potential. Productivity gains from economic structural transformation were distributed unevenly across Thai regions. As of 2015, the regions that benefited initially from place-based policies were those that enjoyed the highest productivity levels. Output per worker varied from THB 497 000 in the Bangkok Metropolitan Area to THB 477 000 in the East region and THB 318 000 in the Centre. The other four regions scored relatively poorly compared to the best performers and the national average. The Northeast has remained the least productive region – in 2015, its productivity levels were approximately six times lower than those in the Bangkok Metropolitan Area, and almost three times lower than the national average.

Changes in sectoral composition and labour flows explain regional productivity growth

Thai regions have undergone major structural transformations, though along different paths. An examination of the evolution of sectoral composition between 2001 and 2015 highlights three different patterns of structural transformation (Figure 2.5). The service sector in the Bangkok Metropolitan Area grew stronger. The transport and storage and real estate sectors expanded, mostly at the expenses of hotels and manufacturing. In the Centre and East, manufacturing gained an even larger role in the local economies. The other regions have a more diverse economic structure. Agriculture remains important in the Northeast and North, although manufacturing is on the rise alongside the financial sector. In the South, the tourism industry has increased its share in regional GDP by 8 percentage points, while the primary sector has shrunk but still accounts for more than 20% of regional GDP. The West is the least dynamic region: the shares of agriculture and manufacturing remain unchanged, while utility provision has assumed more importance.

Box 2.2. Breaking down labour productivity growth through shift-share analysis

Labour productivity growth between two points in time (2001 and 2015 in this chapter) can be broken down into within-sector productivity growth, the reallocation level effect (or shift-effect) and the reallocation growth effect (or cross effect).

The within-effect captures an increase in aggregate productivity due solely to productivity gains in individual sectors.

The shift-effect captures changes in overall productivity that stem from a shift of labour between 2001 and 2015 from sectors that were less productive in 2001 to sectors that were more productive (in which case the effect is positive), or from high productive to less productive sectors (in which case the effect is negative). It therefore does not factor in within-effect productivity gains.

The cross-effect combines the first two effects and measures the shift of workers towards sectors that are growing in their productivity. A positive cross-effect indicates a reallocation of workers towards increasingly more productive sectors (or away from less productive sectors). A negative effect, instead, captures the misallocation of workers in sectors that are losing productivity momentum.

Figure 2.5. Thai regions have undergone major structural transformation
Sectors that make up 80% of regional GDP, 2001 and 2015
picture

Note: Sectoral shares are based on real regional GDP (chained volume series). The definition of sectors is based on the International Standard Industrial Classification (ISIC Rev. 3.1). The category “Trade” includes wholesale and retail trade. “PA” stands for Public Administration. “Utilities” include electricity, gas and water supply. The category “Real Estate” indicates renting of machinery and equipment, and business activities. Source: Author’s work based on national accounts, as provided by the NESDB.

 StatLink https://doi.org/10.1787/888933847619

Figure 2.6. Regions that lag behind displayed major improvements in overall levels of productivity between 2001 and 2015
picture

Note: Calculations are based on regional GDP in real terms. Box 2.2 provides details about the within, between and cross effects.

Source: Authors’ own calculations based on national accounts, as provided by the NESDB.

 StatLink https://doi.org/10.1787/888933847638

Analysis of sectoral composition and labour flows can serve as a useful tool to understand patterns of regional development (Figure 2.6). The productivity of a country can be decomposed through a shift-share analysis, which isolates the effects of within-sector innovation and labour flows across sectors on productivity growth (Box 2.2). This enables the identification of roles that economic activities play in attracting investments and workers, ultimately defining the patterns of development of each Thai region.

In the Northeast, within-sector productivity growth in the agriculture and manufacturing drove regional productivity growth. In particular, the Northeast produced a notable successful story of complementarities between sectors. In 2004, General Industrial Zone One was established in Nakhon Ratchasima. The new industrial estate attracted factories and investments for the production of electronic components and, most of all, parts of vehicles made of natural or synthetic rubber. Despite the challenging climate, which is characterised by dry seasons, the Northeast has become the second largest region in terms of processed rubber production, accounting for 20% of the 2016 production - or 808 432 tonnes. Government promotion coupled with research and development fuelled the spreading of rubber in the area. Indeed, since the beginning of 2000s, Thai governments have supported the cultivation of rubber to replace traditional crops that were no longer competitive on global markets – such as rice (Sakayarote and Shrestha, 2017[8]).2 Moreover, innovative techniques in the 1980s and 1990s facilitated the cultivation of rubber even in otherwise unfriendly environments, such as the North.

In spite of outstanding progress, environmental unsustainability, informality, cross-border competition and demographic flows can undermine catch-up potential in the Northeast. The agriculture sector still employs half of the workforce in Thailand. In addition to the precarious nature of the jobs, the sector might be exerting an increasingly heavy toll on the environment. In the fastest growing province of the region – Nong Khai – 76% of new rubber plantations are located in environmentally unsuitable marginal areas and threaten biodiversity conservation, watershed functions and environmental sustainability (Sakayarote and Shrestha, 2017[8]). One-third of the workforce is employed in the trade and construction industry, two of the least productive sectors. The large majority of workers in those sectors have precarious jobs (61% of trade workers and 74% of construction workers). Nong Khai and neighbouring provinces have benefited from trade and capital from the near Lao PDR. The development of six new special economic zones (SEZ) in Lao PDR, just across the border and around Vientiane, could represent a new opportunity for cross-border trade, but also a challenge for the capacity of local enterprises to attract investments. Finally, the number of workers in the region decreased by 3% between 2001 and 2015. One possible explanation is labour outflows – especially young workers: between 2004 and 2013, the Northeast had the second highest average migrant outflow.3

In the North, special economic zones boosted manufacturing productivity in a few provinces, leaving others behind. The North exhibits similar development patterns to the Northeast. Until the 1970s, agriculture dominated in the region. Manufacturing activities were scattered across small-scale companies, while products were rudimentary and had low export-potential. Since the 1980s, the establishment and consolidation of special economic zones in the Lampun Province has catalysed investments in the local manufacturing industry, turning the area into a major regional trade hub. The Lampun Province (and in particular Chiang Mai and the surrounding areas) has attracted the majority of economic activities in the region, but has not resulted in development in peripheral areas (Glassman and Sneddon, 2003[9]).

In the West, misallocation of labour offset within-sector productivity growth. As in the Northeast, manufacturing and agriculture – which together represent 40% of regional GDP – have experienced a significant increase in labour productivity. However, the negative cross-effect shows that workers did not move into the most dynamic sectors. In the fastest growing province of the region (Ratchaburi), the share of workers in the primary sector dropped by 10 percentage points to 30%, while in manufacturing it decreased by only 2 percentage points to 19%. Conversely, the share of provincial employment increased, although only slightly, in a battery of other mostly non-tradable sectors (like utilities) and in trade services.

A lack of coherent regional policy in the West may explain the misallocation of the regional labour force. Regional policies have traditionally overlooked the West, probably because of its proximity to more important sources of growth. Only three industrial estates have been created in the region, compared to 23 in the East and 16 in the Centre and the Bangkok Metropolitan Area.4 Moreover, these estates may not have yet started to attract labour force and investments. As discussed later in the chapter, 2015 satellite data suggest that the region has the lowest number of urban clusters. One of these, Ban Phrachedi Sam Ong, stretches along the border with Myanmar. The region may then rely on cross-border local trade, fuelling further non-tradable and highly informal services. The other two urban clusters are in proximity to the new industrial estates. Should good cross-connections and collaboration between these cities and industrial estates materialise, the potential for growth and positive agglomeration effects is significant.

In the South, a shrinking primary sector has given way to a rising tourism sector and related services. As in the West, the shares of manufacturing and agriculture in regional GDP in the South have shrunk significantly (from 30% to 22% and from 13% to 10%, respectively). However, contrary to the West, the South has developed alternative sources of employment in hotels and restaurants. The share of the tourism sector in regional GDP increased from 10% in 2001 to 16% in 2015, as did sectoral productivity, while the employment share remained stable at 8% in both years.

The rise of the Southern tourism sector poses some social and environmental challenges. First, almost half of the employees in hotels and restaurants are still informal, and therefore lack systematic insurance against unemployment and the seasonality that the job in the sector entails. Second, if not regulated well, tourism may threaten the eco-sustainability and biodiversity of the area. It may moreover place a strain on local watersheds, thereby crowding out precious resources for a still large agricultural sector. Finally, the rising tourism industry has favoured the creation of ancillary activities in low-innovative and non-tradable sectors such as construction and retail trade. Both environmental costs and the proliferation of vulnerable jobs may eventually offset the beneficial effects of a thriving tourism industry.

The Centre and East regions have experienced a pronounced shift of workers towards more productive sectors. Past industrial policies in the Centre and East drove the reallocation of the labour force towards increasingly productive sectors in the regions. In the 1990s, the Centre and the East were already the most highly industrialised regions due to investments in manufacturing and natural gas industries (Kaothien, 1991[10]). Today, provincial productivity builds on those industrial policies and, most of all, on the high-quality infrastructures that followed. Regional highways link the central and eastern provinces to Bangkok, while deep-sea ports (e.g. Sattahip, Laem-Chabang and Map Ta Phut) and local airports (Utapao) connect the area to global markets and value chains. Finally, provinces enjoy broad access to electricity, water and telecommunications.

Moving towards more broad-based and innovative regional development policies

Thailand is continuing to invest in place-based policies. In accordance with its “Thailand 4.0” agenda, the government announced the Cluster-based Special Economic Development Zones Policy in September 2015. The new clusters provide fiscal advantages to resident firms that are active in agro-processing, textiles and garment industries, and rely on advanced technologies. New SEZs have been surging in strategic provinces, especially on the border with neighbouring countries.

The Thai government has moreover allocated THB 1.5 trillion to the development of the Eastern Economic Corridor (EEC) and intends to develop a Southern Economic Corridor (SEC). By 2021, the EEC will have attracted further investments in Chachoengsao, Chonburi and Rayong, all located in the East. The ultimate objective is to transform the area into a hub of technological manufacturing and services, well integrated with neighbouring ASEAN countries. The state has committed to equip firms that operate in the new clusters and in the EEC with public utility and infrastructures crucial for investment projects. These projects include industrial estates, electricity, water supply, transportation and educational institutions at all levels. The SEC aims at making four southern provinces, namely Ranong, Surat Thani, Nakhon Si Thammarat and Chumphon, the gateway to the West, the Royal Coast and the Andaman Sea. According to the government’s guidelines, the SEC would specialise in producing and processing agricultural and bio-products, with particular regard to the conservation of the local ecosystem. Tourism and transport infrastructure, including ports and railways, will be additional pillars.

Moving beyond corridors and SEZs, Thailand wants to embark on a more comprehensive agenda for innovative regional development, focusing on the particular strengths and capabilities of each region. In December 2017, the government entrusted the National Economic and Social Development Board (NESDB) with the review and rationalisation of existing spatial policies as part of more comprehensive regional development plans. These plans will focus on bringing to the fore the particular strengths of each of the six regions and coordinate local and national planning.

Innovative regional development strategies must be multidimensional, flexible and driven by local discovery and ownership

The new agenda for regional development must balance economic, social and environmental objectives. Past interventions aimed at boosting development in Thailand’s regions have strongly focused on economic objectives. While these direct objectives were often achieved, many projects paid insufficient attention to social and environmental needs and missed the opportunity to create more balanced development (Glassman and Sneddon, 2003[9]). Future strategies for innovative regional development should aim at combining a broader set of objectives, including particularly social cohesion and environmental sustainability. This will ensure that synergies and trade-offs between these different objectives are taken into account. The right mix between socio-economic and environmental objectives may vary across the country. For this reason, prioritisation of objectives cannot be an exclusivity of the state planner. Local administrators, private sector and citizens should interact to identify how to pursue economic, social and environmental objectives in combination and in service of the common interest.

Getting regional development planning right will require placing local innovation and discovery in the driver’s seat. Economic development is essentially driven by a process of discovery; discovery of new products, of new markets, of methods production, of technology, of forms of collaboration and many more. Mastering this process of discovery at each level (national, regional, local) is key to successful economic development and continued productivity and employment growth. For the regional level, the overarching lesson from the ‘smart specialisation’ agenda and past attempts at regional development in the European Union and elsewhere is that this process of discovery must be driven and mastered by local and regional actors (Box 2.3). The role of government intervention is important but it is subsidiary. Policy intervention is required not to select the areas or activities for investing public resources but to facilitate and support the discovery process (OECD, 2013[11]). Thailand’s new agenda for innovative regional development should thus focus on facilitating discovery by regional and local actors.

Box 2.3. Smart specialisation strategy in the EU and the case-study of Portugal

Specialisation and diversification are keys for regions to upgrade by levering local economies of scale and of scope. For this reason, “smart specialisation” strategies are at pillars of the European agenda of innovation, Europe 2020. Through these strategies, European institutions encourage regions to differentiate themselves and to specialise in sectors where they have a comparative advantage.

Portugal is one of the first countries to have put in place a smart specialisation strategy. The strategy had three main components: experimentation, constant involvement of local actors and complementary support of public institutions.

The country has been piloting smart specialisation policies - such as education, health or social inclusion – in selected areas. These policies target well-defined territories, pursue clear and transparent goals, are constantly monitored and evaluated, and their performance is advertised to the public. Once successful, these policies are scaled up and disseminated.

The constant involvement of local stakeholders is fundamental for the process of discovery of local potential. In the Algarve region, for example, the design of the regional smart specialisation strategy involves 90 % of the region’s enterprises, all research centres as well as other entities. Local stakeholders are organised in working groups, or “Innovation platforms”. These platforms identifies regional innovation gaps and set the goals of innovation strategies. The platforms coordinate relevant actors, facilitate the identification of projects and investments needed, and design potential action plans.

Secondary cities were particularly fertile ground for the development of innovation platforms. As urban areas, they guarantee physical proximity and clustering of entrepreneurs, workers and other association. As bridge between metropolis and the hinterland, they reach out to rural residents more easily, thereby ensuring full ownership of the specialisation strategies.

With smart specialisation strategies, the role of the State has been evolving. Central agencies do not only catalyse private investments anymore. They are instead responsible for building an ecosystem conducive to innovation. The implementation of strategy needs public authorities to identify, evaluate and support emerging lines of regional specialisation through investments in infrastructures and local capabilities.

Source: (European Commission, 2016[12]; Ramos and Rosa, 2018[13])

The process and instruments for innovative regional development should be performance-based, flexible and reflect the specific needs and capabilities of each region. Placing local discovery and ownership in the driver’s seat of innovative regional development requires an open process that is adaptable to the needs and capabilities of each region. More advanced regions might require less in terms of direct support and can be supported simply with mutually agreed performance targets and related instruments. Areas with lower capabilities might require more in terms of direct assistance and a stronger level of oversight to ensure accountability (OECD, 2018[14]). Thailand should thus aim to build flexibility into its future innovative regional development agenda. Initial experimenting with different approaches for different regions could be an option. Strong evaluation and performance measurement frameworks must be built into all approaches from the beginning and should be widely accessible to guarantee transparency, as a key building block of local ownership.

Similarly, the geographic scope of regional development policies should be flexible and focus on functionality. The current definition of regions in Thailand seems to primarily serve a classification and statistical purpose. No administrative regional layer exists. This offers the opportunity to include flexibility in the definition of regions for the new strategies. Data analysis can help to identify the most functional clusters of provinces for regional strategies. Similarly, at the level of cities, commuting and other flows may overstep existing administrative boundaries, requiring a move towards functional urban areas to overcome administrative fragmentation and provide integrated solutions to fast urbanisation (section 2.5). Functional areas may involve different provincial clusters and change over time. Neighbouring administrations that do not find co-ordination over specific issues particularly necessary today, may find it advantageous in the face of future global trends and shocks.

Local discovery processes should be supported with data analysis to profile regions and provinces, assess potential and detect bottlenecks. Identifying each region’s best performing province with regards to productivity, for example, allows to classify other provinces as “converging” or “diverging” and develop specific objectives. Alongside data analysis, central government and local stakeholders should consult continuously through surveys and workshops. Surveyed firms, workers and citizens should identify strengths, weaknesses, opportunities and threats to local development. Foresight exercises can help all stakeholders to develop a shared vision of the future.

Building on local best performers as a guide for innovative regional development policy

Identifying best performers on the basis of productivity can be a useful method to set objectives for regional economic development. The new regional development policies can benefit from a systemic assessment of the performance of provinces relative to neighbouring “champions”. What follows is a possible classification of Thai provinces based on their labour productivity growth. In each of the six regions, a regional “productivity frontier” is derived from the average output per worker of the top 10% most productive provinces within the region. The compound average growth rate (CAGR) of provincial productivity between 2001 and 2015 is the measure of convergence towards the regional frontier. A province converges towards the regional frontier if the provincial CAGR is higher than that of the frontier. Conversely, a province diverges from the frontier if the provincial CAGR is lower than that of the frontier.5

The convergence-divergence analysis shows that most provinces at the frontier have inherited industrial estates from the 1980s. Most regional frontiers are host to long-lasting and important industrial estates.6 Rayong (East) is the most productive province in Thailand, and hosts the Map Ta Phut industrial zone – the most recent of the Asian industrial estates – which focuses mostly on processing natural gas and related products, and functions as the cornerstone of the new Eastern Economic Corridor. Ayutthaya (Centre) is the second most productive province in the country and benefits from major industrial estates in the Hi-Tech sector. Rayong and Ayutthaya also attract a significant proportion of investments in the automotive industry in Thailand. Nakhon Ratchasima (Northeast), Ratchaburi (West) and Chiang Mai (North) share similar features. Moreover, all these regions have displayed a positive average growth rate between 2001 and 2015.

In some cases, agriculture, education and tourism have pushed some provinces to the frontier. Some provinces at the frontier do not directly benefit from long-lasting industrial and regional policies. Rather, their productivity levels rely on outstanding performance in the primary, education and tourist sector. Kamphaeng Pet (North) has the largest plantation of tapioca in the country and may have benefited particularly from rapid increase in the volume of trade of this root. Khon Kaen (Northeast) hosts one of the top 10 universities of the country. Tourism has been an important driver of productivity growth in the Southern frontiers of Phuket and Krabi. However, the South is the only region where the productivity frontier actually shrank between 2001 and 2015, possibly because of the rise of less productive ancillary activities.

The majority of provinces are catching up, but the gap with those lagging behind is widening. From 2001 to 2015, labour productivity in provinces at the regional frontier grew on average by 2.5% per year. Lagging provinces have fallen further behind the frontier, as their productivity grew only by 1.3% per year. The average productivity in diverging provinces was only 30% of that of their respective regional frontier province. Converging provinces, on the other hand, grew faster than the respective regional frontier provinces (around 3% per year). The gap with the best regional performers increased only by 2% and the 2015 average productivity in converging provinces was 44% of that of the frontier (up from 35% in 2001).

The Northeast and East stand out in terms of convergence of provinces. Most of the top 20% fastest converging provinces in Thailand are located in the Northeast (Figure 2.7). In the East, all provinces are converging towards the frontier. However, convergence does not happen at the same rate everywhere. Productivity in the first two fastest converging provinces – Chachoengsao and Prachin Buri – is growing at rates respectively 4.5 and 3 percentage points higher than the third best performer in the region. On top of existing place-based policies, the East may need complementary measures to ensure that productivity gains from existing engines of growth spill over onto the respective vicinities.

Figure 2.7. Eastern provinces are converging but at different rates, northeastern provinces are keeping pace with the regional frontier and the southern frontier is losing productivity
picture

Note: Calculations are based on real GPP. The regional frontier is the average output per worker among the top 10% most productive provinces in a region. Convergence (or divergence) status is based on the compound average growth rate (CAGR) of provincial productivity between 2001 and 2015. Bangkok is not included in the sample.

Source: Authors’ calculations based on national accounts, as provided by the NESDB.

 StatLink https://doi.org/10.1787/888933847657

A convergence-divergence analysis to identify the distinguishing attributes of high-performing provinces

Human capital and access to public services emerge as key differences between provinces that perform well and those that lag behind. Once regional frontiers and the potential for productivity growth are identified, integrated regional policies need to address obstacles to growth. In particular, they should address endogenous factors that prevent lagging provinces from converging towards the best regional performers. The analysis below reveals two main differences distinguishing frontier from lagging provinces and converging provinces from diverging ones: human capital and access to services (Figure 2.8).

Figure 2.8. Human capital and access to adequate services drive provincial convergence towards the regional frontier
Odds ratio of being a frontier or converging province with respect to being a diverging province
picture

Note: Each bar represents the odds ratio of being a frontier province with respect to being a diverging province (Panel A), or of being a converging rather than a diverging province (Panel B), for an increase in the variable of interest. For example, a 1-percentage point increase in the share of workers who attained upper secondary education implies a two-fold increase in the probability that a province converges rather than diverges. Contrariwise, an increase by 1 unit in the Gini index implies a decrease in the probability that a province lies at the frontier rather than being among those provinces that diverge.

Odds ratios were obtained through a multinomial logit model where the dependent variable is the logarithm of the ratio between the probability of being at the frontier (or converging) and the probability of being a diverging province, given a set of ten control variables including workers’ education attainment; share of households with access to garbage collection and drinkable water; Gini index of household inequality; and share of informal workers. Only odd-ratios that are statistically significant are shown.

Source: Authors’ calculations based on microdata from Household Survey 2015.

 StatLink https://doi.org/10.1787/888933847676

Access to water and waste disposal services sets frontier and converging provinces apart from those that lag behind. Access to adequate sources of drinking water in municipal areas are significant determinants of convergence with respect to divergence (Figure 2.8, Panel B). Contaminated water, for example, can cause water-borne diseases that threaten children’s health, their school participation and potentially their prospective professional opportunities (Miguel and Kremer, 2004[15]). Poor-quality water moreover imposes an extra burden on already limited budgets of poor households. Instead, access to adequate garbage collection and disposal in municipal areas makes provinces increase the likelihood of provinces lying at the frontier rather than diverging. Inadequate waste disposal may threaten the viability of roads and infrastructure that workers or students use every day to commute to their workplace or schools. If burnt, garbage can pollute air and cause respiratory diseases. High-density urban clusters tend to amplify these negative externalities from lagging access to services.

The level of access to basic public services is of particular importance for the performance of cities across Thailand. Thai household surveys show a positive return in terms of salary and monetary income for households living in secondary cities (Table 2.1). The average labour income for households outside Bangkok is 11% higher in urban centres than in rural areas. However, the “urban wage premium” varies across regions, and is significantly negative in the Centre and Northeast. However, access to drinking water and garbage collection smoothen the negative returns from living in urban areas in all three regions. In fact, the urban wage premium is positive (and increasing in access to services) in the North, where there are two already well-developed urban areas: Chiang Mai and Chiang Rai.

Lower attainment in upper secondary and higher education undermines convergence. Comparisons between frontier and lagging provinces show that provincial human capital endowments matter in two ways (Figure 2.8, Panel A). Provinces that converge towards the frontier have a higher share of workers who attained upper secondary education than diverging provinces, while almost 30% of the workforce in laggard provinces have never completed primary school. At the same time, provinces at the frontier have higher a share of workers who have completed tertiary education than in other provinces.

Thai education strategy should focus on better localising vocational training and universities. A secure and inclusive economic transition depends on the capacity of Thai provinces and regions to upgrade the skills of their labour forces and to generate innovation. Expanding access to upper secondary is crucial for diverging provinces to catch up. In general, 42% of the Thai labour force have completed at most lower secondary education. Thus, strengthening TVET institutes at the upper secondary level helps to build a better-skilled labour force, especially if these schools tailor their coursework to local labour market needs. Better skills would in turn lead to better jobs and higher productivity levels (OECD, 2016[16]). Well-developed universities contribute to attracting R&D investments that could bring innovation and benefit first the local community. Section 2.5 provides policy recommendations to enhance vocational training institutions and universities in Thailand.

Table 2.1. Access to services increases income among urban dwellers

(1)

All regions

(2)

East

(3)

West

(4)

South

(5)

Northeast

(6)

North

(7)

Centre

Urban = 1

0.11***

-0.15

0.03

-0.16

-0.16***

0.19*

-0.22***

Electricity

0.65***

0.91**

-0.30

0.49*

0.79*

1.21***

2.02***

Drinking water

0.31***

0.24***

0.41***

0.26***

0.24***

0.48***

0.24***

Garbage collection

0.2***

0.21***

0.13*

0.05

0.19***

0.21***

0.23***

Adequate sanitation

0.69***

0.81

0.92***

1.05**

0.21

0.88**

0.63

Urban x Drinking water

0.14

-0.11

0.26*

0.30***

-0.15

0.16*

Urban x Garbage collection

0.12

0.18

0.16*

0.11*

0.08

0.11

Regional dummies

Yes

No

No

No

No

No

No

Note: The dependent variable is the logarithm of monetary income as reported by the Households’ Socio-Economic Survey (2015). NSO does not distinguish between urban and rural areas, but rather between municipal and non-municipal areas. Hence, “Urban” is a dummy that switches on for households living in “municipal” areas. Interactions were included between the “urban dummy” and a variable capturing the share of households with access to adequate services, namely drinking water and garbage collection. A positive coefficient associated with the “urban” dummy means that salary is higher in municipal areas than in non-municipal areas. A positive coefficient of interaction between the “urban” dummy and access to services implies that urban salaries increase with access to services.

Source: Authors’ calculations based on the Households’ Socio-Economic Survey, 2015.

Secondary cities as engines of growth outside of Bangkok

Cities attract talent and economic development. Data and the relevant literature show that cities function as important engines of development across a territory. They attract firms and workers, and thereby decrease the cost of matching skills with the needs of enterprises in the labour market. Cities facilitate knowledge sharing and provide a vibrant environment to innovate and experiment. Hence, they help to retain the young and educated in regions that they would otherwise leave, leading to the depletion of human capital (Frick and Rodríguez‐Pose, 2018[7]). Therefore, although Thailand’s population is ageing and decreasing in size, more educated young people will make up the core of Thailand’s future workforce. The distribution of education across municipal and non-municipal areas in Thailand testifies to the attractiveness of cities for educated workers (Figure 2.9).

Figure 2.9. Urban areas attract educated workers
Distribution of urban population aged 15-60 by level of education
picture

Note: The category “None” includes workers that have neither started nor completed primary education. “Urban” population indicates the share of population in a region that lives in municipal areas, as defined by the National Statistical Office.

Source: Authors’ calculations based on the Informal Employment Survey (2016).

 StatLink https://doi.org/10.1787/888933847695

Secondary cities can generate new opportunities for regional development in Thailand, but need the appropriate infrastructures (Box 2.4). For decades, Bangkok has been the main destination for Thais looking for better jobs outside of rural areas. As the capital grew, the cost of life in Bangkok increased, roads became congested and services became overcrowded. More recently, Thais have begun to return to rural areas, while others seek a better life and job opportunities in secondary cities (Figure 2.10) (OECD/UNESCO, 2016[17]). Small cities may therefore play an increasingly important role in the regional economy and should be used as leverage for regional development plans – especially in provinces that are converging to or diverging from the regional frontiers. For continued urbanisation to generate growth, cities need urban infrastructures able to absorb an increasing number of urban dwellers. If not, economies of agglomeration usually associated to cities could turn into negative externalities, undermining local well-being and growth.

Box 2.4. What are secondary cities?

The term “secondary cities” (or “intermediary” cities) was first coined in the 1980s by academics investigating the interaction between urban and rural economies. The characteristics of secondary cities vary with the national context and there is no global consensus on a standard definition. Traditionally, a secondary city would have a population or economy ranging in size between 10% and 50% of a nation’s largest city.

Cities Alliance, a joint World Bank and UN-Habitat initiative, classifies secondary cities into three spatial categories. “Subnational cities” are significant economic poles outside of the capital. “City clusters” are town cities that usually locate within 50 km of large metropolitan cities and where firms tend to relocate. “Corridors cities” are cities that develop along major transport corridors.

In Thailand, the third National Economic and Social Development Plan (1972-1976) acknowledged the over-concentration of economic activities in the Bangkok area. As a result, national urban policies aimed at restraining migration from rural areas to the capital. In order to compensate for Bangkok’s primacy, the fourth plan (1977-81) stressed the development of regional cities as centres for the development of industrial and commercial activities. The enthusiasm for secondary cities was short-lived, however, and the sixth plan (1987-91) focused rather on governing the overwhelming expansion of the Bangkok Metropolitan Area and developing satellite cities around the periphery. The 12th NESDP envisages policies to strengthen the tourism capabilities of secondary cities, but does not outline a more systematic urban development plan.

Source: (Rondinelli, 1991[18]) and (Roberts, 2014[19])

Integrated regional policies can prepare secondary cities for future urbanisation flows. The contribution of integrated regional policies would be twofold. First, their scope would be independent of traditional administrative boundaries and endogenous to the policy process. For example, as a secondary city sprawls, the pool of users of urban infrastructure may stretch across several pre-existing local administrations. If policy makers base their decisions on those administrative borders, they may overlook the need for integrated networks of services. Integrated regional policies would instead address citizens in the peripheries that would otherwise be at risk of marginalisation. Second, integrated regional policies set socio-economic objectives for urban development that are context-specific. A clear definition of urban areas in Thailand and co-ordinating mechanisms across local administrations are two preconditions for integrated regional polices that well serve secondary cities.

Figure 2.10. Secondary cities are attracting more people
Comparison between the evolution of urban population in the Bangkok agglomeration and the evolution of urban population in secondary cities
picture

Note: “Urban population outside of Bangkok” measures the share of urban population in the capital. “Urban population in secondary cities” measures the share of urban population outside of Bangkok. In 2000 (reference year for the comparison), 6,395,429 people lived in Bangkok, 32% of the urban population. In 2017, the figure increased to 9,898,653 (29% of the urban population).

Source: Authors’ work based on (World Bank, 2017[2]).

 StatLink https://doi.org/10.1787/888933847714

Shaping the definition of cities with the help of satellite data

Secondary cities need a clear definition for better urban policies. Thailand divides the country into urban and rural areas according to purely administrative criteria. However, in fast urbanising countries, the boundaries of cities evolve faster than administrative borders. As a result, the traditional distinctions of urban and rural areas may fail to capture the actual urban extent of some areas (Keola, Andersson and Hall, 2015[20]). Secondary cities in Thailand should be defined as a function of the actual pool of users that use (or would use) urban infrastructures. This could be followed by a possible identification of these “functional urban areas”, using population densities and the extent of built-up area from satellite data to identify both urban cores and the hinterlands, where workers may settle. Relevant government agencies could integrate this analysis with data on travel-to-work flows to identify more precisely the peripheral labour markets that are effectively integrated with the cores (Box 2.5).

Box 2.5. Using satellite data and travel-to work flows to define “functional urban areas” in Colombia

Under certain conditions, cities can contribute to a country’s productivity growth and become its main growth engines. Successful cities provide universal access to basic public services and are efficient labour markets were the costs of matching demand and supply of skills is minimised. Importantly, urban planning of successful cities take into account both residents and daily commuters. These commuters often live outside of the city, in the periphery or in other administrative units. If cities and neighbouring towns do not cooperate or if the flows of commuters is underestimated, administrative fragmentation can jeopardise labour markets efficiency.

This is the case in Colombia. One of the reason behind Colombian cities’ underperforming productivity is the inadequate transport infrastructure that connects cities with their outskirts. Transportation costs are higher, it takes more time and resources to match demand and supply on labour markets, and well-being is ultimately affected.

The OECD, in collaboration with the European Union, proposed a new definition of Colombian cities – the Functional Urban Areas (FUAs) - that go beyond the traditional administrative borders and take into account the commuting flows. This measurement is moreover standardised across countries, allowing for international comparison of urban policies.

The identification of Functional Urban Areas in Colombia consists of three main steps.

Identification of contiguous densely inhabited city centres. Colombia’s territory is first divided into a grid with cells of 1 km2. The Global Human Settlement Layer (GHSL), combined with satellite data on built-up area and information from latest census, provides an estimation of the population in each of these cells. The identification of city centres follows three steps: i) all grid cells of 1 km2 with a density of more than 1 500 inhabitants per km2 are selected; ii) high density clusters are defined as an aggregation of continuous high density 1 km2 grid cells. Only the clusters with a minimum population of 50 000 inhabitants are kept as a high density cluster; iii) an urban core is made up of contiguous municipalities (based on 2005 boundaries) that have more than 50% of their populations living within “high density” cells.

Identification of interconnected city centres that are part of the same functional area. Based on the commuting data derived from the 2005 census, 10 cities over a total of 59 (identified in step 1 above) are highly interconnected. Based on the OECD-EU methodology, two city centres are considered integrated and thus part of the same urban system if more than 15% of the population of any of the city centre commutes to work in another city centre.

Definition of the commuting zone of the FUA, linked by commuting flows to the city centres. In order to delineate the extension of the commuting zone, municipalities were assigned to each city centre if at least 15% of the population in the municipality goes to work to the city centre.

The definition of FUAs in Colombia is helping the country to clearly identify its metropolitan areas, integrating local labour markets and overcoming administrative fragmentation. By providing an integrated set of services to both residents and commuters, FUAs are expected to increase the productivity of cities and thereby of the whole country. Moreover, the inclusion of FUAs in the OECD Metropolitan Database allows for benchmarking of Colombian cities to successful urban experiences in most of the OECD countries.

Source: (Sanchez-Serra, 2016[21])

Geospatial data can help identify Thai secondary cities, by transcending traditional administrative boundaries. Secondary cities can be defined based on Global Human Settlement Layers (GHSL), elaborated by the Joint Research Centre of the European Commission.7 GHSL divides the country into a 1 km² population grid. For each element of the grid, it estimates the share of area covered by buildings based on satellite imagery. It then crosses this information with the latest available census data. The derived estimated population density is then used to identify urban areas. The resulting borders of secondary cities transcend existing administrative borders and are rather a function of population distribution and density in contiguous built-up areas.8

Urban agglomerations are widespread in the East, while density is highest in the West. The geospatial analysis based on GHSL identifies 41 cities outside of Bangkok. A full list, including related population densities and population numbers, is given in the Annex at the end of this chapter. The East has the highest number of secondary cities, as identified from the satellite imagery. Several are located in the Chonburi province and include Pattaya-Ban Lamung cluster, Chonburi and Phanat Nikhom. Others form the Rayong-Map Ta Phut cluster. The West hosts the lowest number of secondary cities, but includes the city with the highest density among all secondary cities (8 800 people/km2): Phra Chedi Sam Ong (Figure 2.11). This is an urban agglomeration stretching across the border with Myanmar, with cross-border trade as one of the main economic activities.

Figure 2.11. Urban agglomerations are particularly widespread in the East
picture

Note: The Centre as shown here excludes the Bangkok Metropolitan Area.

Source: Authors’ calculations based on the Global Human Settlement Layer as provided by the Joint Research Centre – the European Commission.

 StatLink https://doi.org/10.1787/888933847733

Targeted surveys should support satellite data to assess the needs of these secondary cities. The Thai government needs to assess the actual nature of these urban agglomerations, as identified from geospatial data. Targeted surveys should be conducted to assess the structure of the local economy and its dynamics, as well as the state of local infrastructure. Since existing surveys and the census still rely on the municipal–non-municipal classification, Thailand may need to design new, tailor-made surveys based on the updated classification of cities. Once the new definition of urban and rural areas comes into force, the National Statistical Office should, moreover, design the next census accordingly.

Enabling secondary cities to develop in response to challenges specific to their location

Integrated regional policies need the right institutional framework to improve the management of network services successfully. Network infrastructures such as transport and water provision, as well as sewage and solid waste management, may overstep traditional administrative divisions. The redefinition of city boundaries is not enough to help mitigating network externalities. Well-functioning integrated urban areas require an institutional framework that provides existing administrations with incentives to co-ordinate. To improve co-ordination at the local level, Thailand needs to step up the current multi-level governance framework. Chapter 3 of this report discusses in more detail how better decentralisation can serve this purpose. Thai secondary cities could rely more on informal rather than formal institutions – without adding complications to an already complex institutional framework. As in Barcelona, they could promote collaboration across jurisdictions based on voluntary agreements (Box 2.6). Successful collaboration may eventually lead to more formalized structures of metropolitan governance. Chapter 4 shows how better management of water basins across jurisdictions can improve the allocation and quality of water.

Box 2.6. How voluntary cooperation between municipalities can give rise to a well-functioning secondary city: The case of Barcelona

Until 2011, local administrations within the Barcelona metropolitan area relied on voluntary co-operation to address planning, transportation and the environment. The Mancomunitat de Municipis was a voluntary association of 31 municipalities in the area, with the objective of bringing a common metropolitan perspective to planning and sector specific issues.

The Mancumitat de Municipis co-ordinated policies to improve metropolitan infrastructure and public space through interjurisdictional public companies based on voluntary participation. For instance, the Institut Metropolità de Promoció del Sòl i Gestió Patrimonial managed housing and land in the area. The metropolitan transport organisation was in charge of the subway network for seven municipalities and regulated the local taxi system. It moreover tackles traffic issues by jointly manage the road network. The organisation for the environment was responsible for the construction and maintenance of hydraulic infrastructures, water supply, drainage, and wastewater and the treatment of urban and industrial waste.

The Mancumitat paved the way for an even more integrated form of metropolitan governance. In 2011, the new Barcelona Metropolitan Area organisation was formed. It fosters increased interjurisdictional coordination by encompassing the Mancumitat, and the transport and environment authorities in a two-tier council structure.

Source: (Lozano-Gracia, Panman and Rodriguez, 2012[22])

Citizen participation can help to handle challenges, such as large seasonal tourism inflows. Thailand is a worldwide major tourist destination. Seasonal inflows are an opportunity, but can dangerously stretch the capacity of rising secondary cities. In addition to equipping secondary cities with adequate infrastructure, involving people who live and work in the city in governing the phenomenon is paramount for the internalisation of possible negative externalities. Moreover, residents play a fundamental role in shaping the appeal of their city, since they inevitably come into constant contact with tourists. In Japan, the growth in urban tourism has made local residents more conscious about their living areas, leading them to undertake spontaneous activities to manage these spaces (Horita, 2018[23]). Tourism-based secondary cities may build on the higher propensity of citizens to contribute to the public good through local forums and other informal institutions that bring them closer to local administrators.

Skills development as a tool of regional and urban policy

Secondary and tertiary education are important drivers of productivity growth at the level of provinces and must be core elements of regional policy. The preceding analysis has demonstrated the importance of human capital, specifically of secondary and higher education for productivity growth in provinces. About 40% of workers have completed at least upper secondary education in Thailand, significantly below the OECD average of 80% (OECD, 2016[24]). The share moreover varies from 54% in the Bangkok Metropolitan Area to 28% in the Northeast (Figure 2.12). Depending on the local demand for skilled workers, policies for regional development should focus on upper secondary vocational colleges and universities as core building blocks for local economic development.9

Figure 2.12. There is significant variation in educated workers across regions
Share of employees by level of education attained
picture

Note: The category “None” includes workers that have neither started nor completed primary education. Employees included all Thai citizens aged between 15-60.

Source: Authors’ calculation based on Labour Force Survey (2016) and Socio-Economic Survey (2015).

 StatLink https://doi.org/10.1787/888933847752

Enhancing technical and vocational education and training as an engine of skills development for local economies

Work and skills-oriented training benefits both students and regions more than general education for students that do not attend university. Employment surveys show that, across all regions, students that decide not to pursue higher education obtain better salaries if they complete upper secondary vocational training, rather than upper secondary general education (Figure 2.13) (Tangtipongkul, 2015[25]). The wage premium of TVET over general secondary education is above 20% in all regions. It is highest in the North and Northeast, where salaries for TVET graduates are even higher than in Bangkok. Moreover, TVET graduates also show higher rates of insertion. At the national level, nine out of ten skilled workers find qualified jobs, and the share is above 90% in every region. Higher salaries and more qualified jobs not only mean that it is more profitable for students to pursue TVET, but also indicate that they are more in demand by firms and more productive and useful to the local economy once employed. This makes sense, as TVET focuses on providing students with practical technical skills that otherwise would have to be acquired on the job. In the strongest TVET systems, firms play an important and active role in defining the curriculum and providing opportunities for practical training.

Figure 2.13. The hourly wage of TVET graduates is higher than the salary of workers with general upper secondary education
Hourly salary for employed TVET graduates, by region
picture

Note: Fitted values of hourly wage by region were obtained by first controlling for age, the squared value of age, occupation status and sectors of economic activity.

Source: Authors’ calculations based on the Informal Employment Survey (2017).

 StatLink https://doi.org/10.1787/888933847771

Despite these clear advantages, TVET do not always match the needs of the labour market, or fulfil students’ aspirations. Employers often lament a mismatch between the skills that vocational institutes teach and those that industry needs. In the automotive sector, almost half of manufacturers face challenges in hiring skilled workers to fill their vacancies (Pholphirul, 2014[26]). Many of their employees lack the know-how to operate sophisticated, high-technology equipment. At the same time, students’ ambitions do not focus on TVET because of a cultural bias towards a more academic education and against vocational training (Tan and Tang, 2016[27]). As a result, 34% of upper secondary school students were enrolled in vocational programmes in 2015 – down from 36% in 2011 and below the government’s 45-55% target (OECD, 2018 and MOE, 2017) (Figure 2.14, Panel A). Between 2001 and 2017, the share of workers with a high school degree (and no further education) more than doubled, while that of workers with a TVET degree only rose by one percentage point on average across all regions (Figure 2.14, Panel B). The Northeast and North regions account for the smallest shares of TVET graduates among workers, while the more industrialised regions of BMA, the East and Centre have shares between 4% and 6%.

Figure 2.14. Enrolment in vocational training is below the government target
picture

Note: Panel A shows the share of upper secondary students enrolled in vocational programmes across selected countries in the latest available year: 2015, Thailand; 2014, China, Colombia, Malaysia Indonesia, South Africa and the OECD average; 2013, Korea, Poland and Turkey; 2012, Mexico; 2009, Singapore. Panel B shows the share of workers who attained at most upper secondary schools, by stream (general vs. vocational). The upper secondary general track includes academic schools and teachers’ training.

Source: Panel A: (World bank, 2017[28]); Ministry of Education (2017). Panel B: Authors’ own work based on the Labour Force Survey 2001 and 2017, as provided by NSO.

 StatLink https://doi.org/10.1787/888933847790

Thailand can encourage the success of vocational training by ensuring that curricula build on local labour needs. Provincial authorities, local private sectors and schools can join forces to tailor the content of curricula to regional needs, as they differ across the country. They are indeed best positioned to identify untapped productivity potential and to incorporate current and future needs into schools’ coursework. The Thai government, under the aegis of the 12th Educational Development Plan, is expanding co-operation between vocational institutions, the private sector and academia to develop courses that better meet industry needs. Thailand is further establishing vocational education schemes on a bilateral basis between local Chambers of Commerce and relevant public institutions in an effort to meet labour demands in the agricultural and services sectors. However, neither provinces nor schools enjoy particular freedom in adapting technical education to the local economy – with few exceptions (Box 2.7).

Box 2.7. The Thai-Austrian Technical College: A role model highlighting the importance of independence and close collaboration with industry

The Thai-Austrian Technical College was established in 1963 in Eastern Chonburi province in close proximity to the industrial hub of the Eastern Seaboard. An agreement between the Thai Government and the Federal Government of Austria grants the college a special status and the freedom to devise its own curriculum and teaching methods within the broader confines of national education policy. The Austrian government offered technical assistance, machinery, tools and expert advice to assist with the installation of mechanical equipment, while the Thai Government provided land for the construction of school buildings. Today, the college is recognised as a “Super model for technical colleges” and recognised as a national best example. It is self-sufficient and hosts 6 078 students in five fields: mechanics, electronics, automotive, construction and hospitality. It benefits from partnerships with nearby multinational enterprises (especially in the automotive sector) and from proximity to some of the most successful special economic zones in Thailand.

Several lessons emerge from the success story of the Thai Austrian College. First, proximity to and close collaboration with large firms has proven very important. These firms provide opportunities for learning not only for students but also for teachers, and collaborate actively with the college on the continuous development of curricula. Second, to be able to respond and collaborate with firms, the college needs and benefits from a special degree of independence from nationally set curricula and teaching methods. Third, an entrepreneurial and driven leadership has been key to making the most of the given opportunities and continually adjusting the homogeneous national education framework to the needs of local students and business.

Source: Interviews with staff and the website of the Thai-Austrian Technical College: www.tatc.ac.th.

Matching curricula with market’s needs requires a close partnership between employers, social partners and TVET institutions. Their active participation in curriculum development helps to ensure that the skills taught correspond to the needs of modern business. For instance, in Latvia the Ministry of the Economy carries out forecasts for medium and long-term skills in the labour market. The Agency for Employment supports intermediation in the labour market, and develops education and training programmes for the unemployed. The participation of employers and unions in curriculum development helps to ensure that the skills taught correspond to the needs of modern business (OECD, 2018[29]). In Germany, regions develop their curricula within a framework agreed with the Ministry of Education, which in turn ensure that 20% of the syllabus reflects local needs. Regions, moreover, determine the number of training programmes and available places in consultation with local authorities and ad-hoc committees (Hippach-Schneider and Weige, 2012[30]).

Work-based training is another essential component of vocational training to form better-skilled workers. The workplace offers a dynamic framework for training, while trainees actively contribute to the fulfilment of the company’s objectives. Moreover, the hiring companies can propose practical training to trainees that would then facilitate hiring. The promotion of practical training is also cost-beneficial for the state. Trainees have access to machines and other equipment that already exist in firms and that would be too expensive for public schools to obtain. Several countries have successfully implemented work-based training. In Colombia, for example, the Servicio Nacional de Aprendizaje integrates coursework and classes with internships and practical training hours in affiliated firms (OECD, 2018[29]). In Thailand, the 12th NESDP promotes “Dual Vocational Training” to encourage vocational schools to complement coursework with traineeships in the private sector.

An appropriate framework and system of incentives is necessary to motivate entrepreneurs to participate in vocational education. In Tunisia, payroll tax credits and funding encourage enterprises to organise training, especially if they invest in new technologies. There have also been notable efforts to accredit apprenticeships in the informal sector. Such approaches play a dual role as rewarding learning and skills acquired in the informal economy could help to support social inclusiveness (OECD, 2015[31]).

Counsellors can contribute by informing prospective students about the opportunities that TVET may offer. In spite of the higher returns and entry rates, TVET still suffers from a poor reputation among students and their families. In this context, policies should widen access to information about the coursework and career perspectives of TVET schools. To reach out to prospective students, local authorities can organise local career fairs where TVET and local private sectors showcase the ways in which they collaborate. Schools at primary and lower secondary level should engage counsellors to help students discover their potential and attitudes from an early age. Counsellors should therefore be trained professionals, with an accurate knowledge of the local economy and the capacity to analyse labour data to forecast future market needs. They should also guide pupils through application procedures, the necessary paperwork and deadlines to apply to TVET institutes (McCarthy and Musset, 2016[32]).

Strengthening provincial universities to boost productivity and entrepreneurship on the periphery

Tertiary institutions such as universities and research institutes play an important role in boosting local potential. They contribute to regional and provincial productive systems through four channels: First, they provide the local labour market and knowledge-based industry and services with a highly qualified workforce. Second, they attract investments in research and innovation. Third, they cultivate local entrepreneurial spirit and new business opportunities (OECD, 2007[33]); and, fourth, they can help spreading innovation and growth beyond a region’s capital.

To fulfil this role regional tertiary institutions need the right conditions, flexibility and support. Many countries struggle with optimising the role of higher education and research for subnational development. Policies for the sector are often determined at central level, primarily focussing on national objectives and funding gets skewed towards the top institutions. The most successful examples from countries like Korea and Japan show, that individual institutions need the right amount of freedom and flexibility and on that basis take entrepreneurial initiative. Importantly, to attract students and retain graduates in the local labour market universities and cities need to offer an attractive environment (Box 2.8).

Box 2.8. Strengthening the link between higher education and local development: Challenges and possible solutions in OECD countries

OECD countries have been facing several obstacles to the regionalisation of higher education institutions for the sake of local development.

Higher education policies often lack a regional dimension, focusing instead on meeting national aspirations. As with the Rajabhat universities, tertiary and community colleges in OECD countries bear a responsibility to apply research and development to the needs of local industry and labour market. However, these institutions have neither a well-established tradition in research, nor the resources or infrastructure to support this goal.

In Korea, from 2004 to 2008, the central government funded the New University for Regional Innovation project to enhance regional innovation outside the Seoul metropolitan area. The project helped local higher education institutions to attract and retain talent in the regions. The project encouraged the development of programmes to help students acquiring skills in line with the local economy. Participating institutions built productive partnership with local authorities as well as business and industry. They also provided skilled workers and advanced technologies to industrial clusters in the regions. Not all project-affiliated universities succeeded in attracting pupils and training potential high-skilled workers. Successful regional universities required comprehensive and coherent urban policies that provided students with appealing services and amenities.

Imperfect decentralisation excludes local authorities from the design of higher education policies and biases the distribution of research funds. In countries with a centralised higher education system, the capital city and some large metropolitan areas generally have the largest universities and account for a considerable share of higher education institution research. Education policies that target only this handful of spatially concentrated institutions undermine the development of human capital at the periphery. A side effect is that research funds flow consistently towards the same institutions in the same areas, further capping the resources of universities outside already thriving areas.

Some countries have devolved authority in higher education policy to regional governments. This has enabled them to contribute actively to the establishment of higher education institutions and better respond to the needs of the local community. Mexico, for instance, has designed a set of educational policies that aim to improve greater decentralisation. In this framework, the State Commission for Higher Education Planning (COEPES) functions as a co-ordination body managing higher education planning at the regional level. Meanwhile, the National Agency for Science and Technology (CONACYT) provides mixed federal and state funds tied to the development of the regional cluster.

School management may lack sufficient power to shape the mission of local institutions. In 2004, Japanese national universities were transformed into National University Corporations with the authority to own land and buildings and hire staff. Since the members of the faculty were no longer civil servants, the institutions could propose more flexible contracts and salary schemes. Following these changes in governance, institutions found it easier to attract funds from industry organisations rather than individual companies. The scope of their research thereby broadened and university-industry collaborations often evolved into small start-up firms.

Source: (OECD, 2007[33])

Thailand’s best-performing provinces host well-developed university poles that attract investments and innovation. The analysis in Section 2.3 shows that the share of workers who complete university degrees is higher in most productive provinces than in those lagging behind. Moreover, the most productive provinces in Thailand host universities that are among the top ranking Asian academic institutions (Buasuwan, 2018[34]). These research institutions – as defined by the current government – have contributed to fuelling regional economic growth on several occasions. For instance, back in the 1990s, researchers experimented with systems to cultivate rubber in the Northeast and North, areas usually unsuited for this type of plantation given their frequent droughts. As seen in section 2.2, innovation helped to boost agricultural productivity in the area. Researchers from Thai universities are still looking for methods to improve yields of rubber plantations (see for instance (Chantuma et al., 2011[35]).

Table 2.2. Provincial universities still attract a large share of students and hence represent an opportunity for local productivity growth

Types of universities

Purpose/Degree focus

Number of institutions

Number of students

Students (% total)

Closed public university

Bachelor to graduate

10

120 917

6%

Autonomous university

Bachelor to graduate

23

562 489

28%

Public open university

Bachelor to graduate

2

322 462

16%

Private university

Bachelor to graduate

41

258 132

13%

Rajabhat university

Provincial university offering bachelor degrees to graduate

38

522 535

26%

Ratchamongkol Technological University

Technological university offering high diplomas to graduate

9

151 811

8%

Private college

Mainly bachelor

19

28 505

1%

Private institute

Mainly bachelor

9

21 067

1%

Community college

Community college

20

16 075

1%

Total

 

171

2 003 993

 

Source: (Buasuwan, 2018[34]) based on data from the Office of the Higher Education Commission.

Regional institutions of higher learning could help lagging provinces catch up, but need attention. Thailand’s top universities are expanding across the country through secondary campuses, helping to boost the capabilities of regional centres.10 To boost the potential of less connected and productive provinces, Thailand builds on its 38 provincial institutions of higher education (Rajabhat) (Table 2.2). These schools have historically had a strong focus on teacher education. In 1995, they evolved into higher learning institutions with a more articulated education offer, in the process becoming local poles for training high-qualified workers. However, the quality of education in these schools remains a concern. For instance, the Office of Higher Education Commission (OHEC), which is in charge of monitoring university standards, has recently expressed concerns over the ways in which regional universities select and assess students. Enrolment is trending downward and some institutions – especially in the Northern provinces and those provinces where the population is ageing fast – may have to close. Provinces that are already lagging behind therefore risk losing potential engines of innovation and productivity growth.11

Thai provincial universities can regain importance by tailoring their education offer to local needs. The adaptation of coursework to provincial economic structures might not be enough. The teaching paradigm should also change. Up to 50% of the study work in Aalborg University (northern Jutland, Denmark) is problem-oriented project work. Students work in teams to solve problem areas often defined in co-operation with local firms and administration. The Aalborg model provides students with transferable skills and authentic work experience; enterprises have a clearer picture of the actual knowledge accumulated and how the students might fit in as prospective employees. The university gains feedback and access to instructive cases and ideas for research and teaching. To further tighten the link with the private sector, universities can hire high-skilled personnel from industry and society as part-time teachers and adjunct professors (OECD, 2007[33]).

Local higher educations are crucial to enhancing local entrepreneurship and creating new business opportunities. Universities often provide the right environment for business ideas to incubate and develop. Following the example of some OECD countries, provincial universities should provide generic start-up advice and guidance, training, one-to-one advice, legal start-up costs, business competitions and incubation (Box 2.9). An even more effective option is to embed entrepreneurial learning into their core curriculum. During these classes, students learn about business disciplines – such as planning, marketing and finance (OECD, 2007[33]).

Box 2.9. Speed MI up: The city of Milan and Bocconi University join forces to develop local entrepreneurship

Since 2005, Italy has allocated funding to universities and research institutes willing to join the “Incubator for start-ups” programme. Under this programme, higher education institutes provide high-level technical assistance to entrepreneurs during the start-up phase (OECD, 2007[33]). In 2013, as part of this programme, the City of Milan, the local Chamber of Commerce and Bocconi University created an incubator to help entrepreneurs build and strengthen business ideas, entitled “Speed MI Up”.

Unlike other incubators, Speed MI Up does not enter the capital structure of the start-ups, but rather provides them with a wide range of services. Upon selection through a public tender, selected entrepreneurs gain access to several services including: tutorship and training by the Bocconi faculty; periodical business reviews conducted by an Advisory Board; and a Consulting Bureau on legal, financing, marketing and digital media. Office spaces are clustered to ensure network building and knowledge spill-overs between entrepreneurs. In addition, past, present and future promoted start-ups can sell their products in a market place on the incubator’s website.

A selection committee consisting of representatives of the three institutions select aspiring start-up projects based on an elevator pitch, a business plan, candidates’ resumes and interviews. Entrepreneurs can access short video courses on how to draw up business plans and directions about delivering the elevator pitch on the incubator’s website. Selected candidates can then enjoy the full range of services for two years and, in some cases, for the following five years.

Since 2013, Speed MI Up has incubated several start-ups that have subsequently become thriving local small and medium-sized enterprises in the food-processing industry and trade services and personal services sectors.

Source: Authors’ work based on “Speed MI Up Incubator Seeks Innovative Startups” (6 March 2017): www.viasarfatti25.unibocconi.eu.

Effective and innovative regional development necessitates fiscal and institutional reforms

The new regional development plans require a new fiscal relationship between the different layers of government. Thailand’s provinces have traditionally relied on inter-governmental transfers, the size of which was based mostly on population thresholds. As outlined in Chapter 3, this system placed less populated provinces at a disadvantage. As part of the new regional development plans, the current government has linked the distribution of grants to a set of socio-economic characteristics in the provinces. Since December 2017, the new “Regional Development Policy Integration Committee”, chaired by the Prime Minister and coordinated by the NESDB, is empowered to set and periodically update a formula that regulates the distribution of general grants.12 The new formula will come into force from the next fiscal year (2018-19) and will take into account the provincial population, the number of poor, household income and the gross provincial product in each province (see Chapter 3 for more details).

Performance-based transfers could be a useful tool. Part of the allocation of the existing general transfers could be conditional on the achievement of socio-economic targets that would close the gaps between frontier converging and diverging provinces. The choice of targets should be the result of a high-level political debate involving representatives from the central government and governments from each region. They should also suit the characteristics and needs of different places in each region. Moreover, targets should be clearly identified and their achievement measured through a series of indicators (Barca, 2009[36]). The way targets are set can facilitate interaction and comparative performance measurement across provinces (Box 2.10).

Once targets are set and grants distributed, the government should monitor the achievement of targets on a constant basis. Systems of performance monitoring depend on four main criteria. First, they should rely on a strong system of indicators and targets and a nationally and internationally accessible database. Second, based on the above indicators, relevant government agencies should undertake periodical in-depth performance assessments with provinces. Third, following the assessment, the system should envisage rewards and sanctions linked to targets. Finally, the systems should encourage public debate around the achievement of targets through the publication of progress scorecards (Barca, 2009[36]).

Box 2.10. Performance-based grants in Italy

Italy has been reforming its approach to regional development policy since the 1990s. Changes concern not only underlying principles, but also policy delivery mechanisms. As in Thailand, the trend towards decentralisation to lower levels of administration has required new ways of co-ordinating a growing number of actors in the field of regional development. In this context, at the beginning of the 2000s, Italy embraced a result-oriented approach to planning and expenditures: the National Performance Reserve.

Table 2.3. Performance-based grants are redistributed according to clear and measurable social and economic objectives

Objective

Indicator

Education: Improve students’ competence, reduce drop-outs and broaden population’s learning opportunities.

% of early school leavers

% of students with poor competencies in reading

% of students with poor competencies in math

Child and elderly care: Increase the availability of child and elderly care to favour women’s participation in the labour market

% of municipalities with child care services

% of children (age 0-3) in child care

% of elderly people benefiting from home assistance

Urban waste management: Protect and improve the quality of the environment, in relation to urban waste management.

Amount of urban waste disposed in refuse tip

% of recycled urban waste

% of composted waste

Water service: Protect and improve the quality of the environment in relation to integrated water services.

% of water distributed

% of population served by waste water treatment plants

Initially, the National Performance Reserve had three main objectives: (i) the simplification of public administration, (ii) improvement of spending efficiency, and (iii) the promotion of projects that require co-ordination among local administrators (also known as “Territorial Integrated Projects”). To achieve these objectives, a new mechanism envisaged the redistribution of 4% of 2000-06 EU Structural Funds among Italy’s 21 regions conditional on the achievement of a series of targets. The targets were the result of a two-year negotiation that involved high-profile political representatives of the central and regional governments

The National Performance Reserve evolved throughout the 2000s. The Italian National Strategic Framework 2007-13 introduced a new set of targets to improve citizens’ quality of life and increase the propensity of business to invest in the south of Italy.

Source: (OECD, 2009[37])

The new regional policy framework requires dynamic actors with the freedom to experiment at each level of government. Local and provincial layers of government need to be given the flexibility and means to experiment, as opposed to implementing top-down policies designed at the central level. Experimentation allows policy makers to learn from the “small-step” interventions they pursue to address local issues. These necessary experimental processes require mechanisms that capture lessons and ensure that these are used to inform future activities (Andrews, Pritchett and Woolcock, 2013[38]). Chapter 3 explores best practices that foster inter-administrative co-operation and further decentralise fiscal and political power.

Policy recommendations

Goal to reach

Recommendations of the Multi-Dimensional Country Review of Thailand

1. Moving towards more broad-based and innovative regional development policies

1.1. Innovative regional development strategies that are multidimensional, flexible and driven by local discovery and ownership

1.1.1. Ensure that targets of regional development plans and results-based allocation measures balance economic, social and environmental objectives. Targets and objectives should be informed by sound data analysis and continuous consultation between central and local governments, stakeholders and citizens.

1.1.2. Place local innovation and discovery at the centre of regional development plans. Focus on facilitating discovery by regional and local actors. The central government should not select the areas or activities for investing public resources but rather facilitate and support the discovery process.

1.1.3. Build flexibility into regional development instruments and initially experiment with different approaches adapted to regions’ capabilities.

1.1.4. The geographic scope of regional development policies should be flexible and focus on functionality. Data analysis can help to identify the most functional clusters of provinces for regional strategies.

1.1.5. Strong evaluation and performance measurement frameworks must be built into all approaches from the beginning. Data on results should be widely accessible to guarantee transparency and enable public scrutiny.

2. Supporting secondary cities as the centrepieces of regional policies

2.1. Define secondary cities in Thailand.

2.1.1. Identify secondary cities as functional urban areas. The combination of geospatial data and local micro-data would allow for an exact definition of the pool of users of urban infrastructure, both in the core of cities and their hinterlands.

2.1.2. Carry out local surveys to assess the needs of secondary city residents to deepen the above analysis. These surveys should complement a database on the financing, costs, availability and quality of subnational government services, as proposed in Chapter 3.

2.2. Design and implement a new urban policy agenda specific to secondary cities.

2.2.1. Equip local authorities and secondary cities with the power and fiscal tools to address local needs. Effective decentralisation reforms will be key. Chapter 3 discusses detailed policy recommendations along this line.

2.2.2. Given the cross-boundary nature of secondary cities, permanent consultation among local administrations should guide strategies of investment in infrastructure. Local co-ordination can be enforced through formal institutions and more effective decentralisation policies, as well as informal institutions and voluntary interjurisdictional agreements.

2.2.3. Citizen participation should be ensured to create ownership of new urbanisation agendas. Local authorities should further promote informal fora and other physical and virtual places where citizens can interact with local administrators and issues.

3. Skills development as a tool of regional policy

3.1. Invest in better and more relevant skills to adapt the workforce to the needs of the place-specific and regional labour market.

3.1.1. Identify the place-specific and regional productive sectors and incorporate their current and future needs into the curricula. TVET institutions need to provide their students with the necessary skills to access the best job opportunities and outcomes. Therefore, the curricula of training programmes must target the most productive sectors and work to identify the key competencies those sectors require.

3.1.2. Ensure that TVET programmes are reactive and can adapt to the changing needs of the local labour market. Structural transformation has an impact on the skills that employers look for in the workforce. TVET institutions should therefore continuously update their curricula to adapt to evolving labour market needs.

3.1.3. Increase the involvement of local entrepreneurs and private sector in the design of education curricula, technical programmes and workplace education. Through discussions with the private sector, TVET institutions will be able to develop education curricula that respond directly to labour market needs. Partnerships between TVET institutions and the private sector are also crucial to understanding the current and future needs of the labour market. Consider fiscal incentives to encourage private sector participation in TVET.

3.1.4. Complement traditional coursework with work-based training. TVET institutions should include in their coursework two- to six-month mandatory training and internships in affiliated firms.

3.2. Ensure that the TVET sector becomes more attractive for young people.

3.2.1. Guarantee each vocational institution a certain degree of autonomy in order to better match students’ ambitions and market needs.

3.2.2. Introduce counsellors that can help students explore different schools and programme options. Counsellors should have extensive and accurate knowledge of the characteristics of the local labour market and enterprises.

3.3. Invest resources in the development of provincial universities, corresponding to integrated regional policies.

3.3.1. Tighten the relationship between provincial universities, local authorities and the private sector. Universities, local firms and government management should exploit synergies by designing, enhancing and monitoring a common long-term agenda for local skills development.

3.3.2. Promote provincial universities as centres of entrepreneurship. In collaboration with local authorities and Chambers of Commerce, universities can become incubators of local start-ups, by providing general advice and guidance, training, one-to-one advice, legal start-up costs, business competitions and incubation.

4. Support the implementation of regional policies with fiscal and institutional reforms

4.1. An institutional and fiscal environment conducive to experimentation at all levels of government

4.1.1. Pursue decentralisation reforms and thereby allow local decision makers to experiment through “small-step” interventions. Chapter 3 discusses more specific recommendations, as well as policy tools to decentralise fiscal and political power efficiently in Thailand.

4.1.2. Redesign the formula for the distribution of general grants to complement population thresholds with socio-economic criteria. Chapter 3 discusses more specific recommendations as well as policy tools to establish local fiscal capacity.

4.1.3. Complement the new formula for general grants with a move to results-based transfers conditional on the achievement of socio-economic targets that would close the gaps between best-performing, converging and diverging provinces.. Targets should match the characteristics and needs of different places. The data used to measure results should be publicly accessible to encourage public scrutiny and debate.

4.1.4. Boost local capacity of co-ordination across layers of governance. Local authorities should develop their capacity to interact both vertically (state-province-districts-municipalities) and horizontally (between municipalities). Chapter 3 provides specific recommendations on how effective decentralisation policies can serve this purpose.

References

[38] Andrews, M., L. Pritchett and M. Woolcock (2013), “Escaping Capability Traps Through Problem Driven Iterative Adaptation (PDIA)”, World Development, https://doi.org/10.1016/j.worlddev.2013.05.011.

[4] Annez, P. and R. Buckley (2009), “Urbanization and Growth: Setting the Context”, in Spence, M., P. Annez and R. Buckley (eds.), Urbanization and growth.

[36] Barca, F. (2009), Towards a place-based social agenda for the EU, Independent Report, Prepared at the Request of the European Commissioner for Regional Policy, Danuta H¨ ubner, European Commission, Brussels..

[34] Buasuwan, P. (2018), “Rethinking Thai Higher Education for Thailand 4.0”, Asian Education and Development Studies, Vol. 7/2, pp. 238-254, https://doi.org/10.1108/aeds-07-2017-0072.

[35] Chantuma, P. et al. (2011), “An Innovative Tapping System, the Double Cut Alternative, to Improve the Yield of Hevea Brasiliensis in Thai Rubber Plantations”, Field Crops Research, http://dx.doi.org/10.1016/j.fcr.2011.01.013.

[6] Christiaensen, L. and Y. Todo (2014), “Poverty Reduction During the Rural–Urban Transformation–The Role of the Missing Middle”, World Development, Vol. 63, pp. 43-58, https://doi.org/10.1596/1813-9450-6445.

[12] European Commission (2016), “Smart Guide to Cluster Policy”.

[7] Frick, S. and A. Rodríguez‐Pose (2018), “Big or Small Cities? On City Size and Economic Growth”, Growth and Change, Vol. 49/1, pp. 4-32, https://doi.org/10.1111/grow.12232.

[9] Glassman, J. and C. Sneddon (2003), “Chiang Mai and Khon Kaen as Growth Poles: Regional Industrial Development in Thailand and its Implications for Urban Sustainability”, The Annals of the American Academy of Political and Social Science, Vol. 590/1, pp. 93-115, http://dx.doi.org/10.1177/0002716203257075.

[3] Henderson, J. (2005), “Urbanization and growth”, in Handbook of economic growth, Elsevier, https://doi.org/10.1016/s1574-0684(05)01024-5.

[30] Hippach-Schneider, U. and T. Weige (2012), “VET Qualifications versus Bachelor Degrees? Recruitment at the Intermediate Qualification Level - Case Studies from Germany, England and Switzerland”, in The future of vocational education and training in a changing world, VS Verlag für Sozialwissenschaften, Wiesbaden, https://doi.org/10.1007/978-3-531-18757-0_15.

[23] Horita, Y. (2018), “Urban Development and Tourism in Japanese Cities”, Tourism Planning & Development, Vol. 15/1, pp. 26-39, https://doi.org/10.1080/21568316.2017.1313774.

[10] Kaothien, U. (1991), “Regional and Urbanisation Policy in Thailand: The Tertiary Sector as a Leading Sector in Regional Development”, Urban studies, Vol. 28/6, pp. 1027-1043, https://doi.org/10.1080/00420989120081181.

[20] Keola, S., M. Andersson and O. Hall (2015), “Monitoring Economic Development from Space: Using Nighttime Light and Land Cover Data to Measure Economic Growth”, World Development, Vol. 66, pp. 322-334, http://dx.doi.org/10.1016/j.worlddev.2014.08.017.

[22] Lozano-Gracia, N., A. Panman and A. Rodriguez (2012), “Interjurisdictional Coordination”, in Samad, T., N. Lozano-Gracia and A. Panman (eds.), Colombia Urbanization Review : Amplifying the Gains from the Urban Transition, Washington, DC: World Bank, https://doi.org/10.1596/978-0-8213-9522-6.

[32] McCarthy, M. and P. Musset (2016), A Skills beyond School Review of Peru. OECD Reviews of Vocational Education and Training, OECD Publishing, Paris, https://doi.org/10.1787/9789264265400-en.

[15] Miguel, E. and M. Kremer (2004), “Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities”, Econometrica, Vol. 72/1, pp. 159-217, https://doi.org/10.1111/j.1468-0262.2004.00481.x.

[29] OECD (2018), Examen multidimensionnel du Maroc : Volume 2. Analyse approfondie et recommandations, Les voies de développement, Éditions OCDE, https://doi.org/10.1787/9789264298699-fr.

[1] OECD (2018), Multi-dimensional Review of Thailand (Volume 1): Initial Assessment, OECD Development Pathways, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264293311-en.

[14] OECD (2018), Rethinking Regional Development Policy-making, OECD Multi-level Governance Studies, OECD Publishing, https://doi.org/10.1787/9789264293014-en.

[16] OECD (2016), Multi-dimensional Review of Peru: Volume 2. In-depth Analysis and Recommendations, OECD Publishing, Paris, https://doi.org/10.1787/23087358.

[24] OECD (2016), PISA 2015 Results (Volume I): Excellence and Equity in Education, PISA, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264266490-en.

[31] OECD (2015), Investing in Youth: Tunisia: Strengthening the Employability of Youth during the Transition to a Green Economy, OECD Publishing, Paris, https://doi.org/10.1787/24126357.

[5] OECD (2015), The Metropolitan Century: Understanding Urbanisation and its Consequences, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264228733-en.

[11] OECD (2013), Innovation-driven Growth in Regions: The Role of Smart Specialisation, OECD Publishing, Paris.

[37] OECD (2009), Governing Regional Development Policy: The Use of Performance Indicators, OECD Multi-level Governance Studies, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264056299-en.

[33] OECD (2007), Higher Education and Regions: Globally Competitive, Locally Engaged, https://doi.org/10.1787/9789264034150-en.

[17] OECD/UNESCO (2016), Education in Thailand: An OECD-UNESCO Perspective, Reviews of National Policies for Education, OECD Publishing, https://doi.org/10.1787/9789264259119-en.

[26] Pholphirul, P. (2014), “Job Vacancies, Skill Development and Training in Workplace: Evidence from Thai Manufacturers”, Journal of Applied Sciences, Vol. 14/17, pp. 1898-1908.

[13] Ramos, A. and F. Rosa (2018), “Empreendendo descoberta inteligente: Uma abordagem aos modelos de implementação da especialização regional em Portugal.”, Public Policy Portuguese Journal, Vol. 3/1, pp. 57-74.

[19] Roberts, B. (2014), Managing Systems of Secondary Cities, Cities Alliance/UNOPS, Brussels.

[18] Rondinelli, D. (1991), “Asian Urban Development Policies in the 1990s: From Growth Control to Urban Diffusion”, World Development, https://doi.org/10.1016/0305-750x(91)90133-3.

[8] Sakayarote, K. and R. Shrestha (2017), “Policy-driven Rubber Plantation and its Driving Factors: A Case of Smallholders in Northeast Thailand”, International Journal of Sustainable Development & World Ecology, Vol. 24/1, pp. 15-26, https://doi.org/10.1080/13504509.2016.1143410.

[21] Sanchez-Serra, D. (2016), “Functional Urban Areas in Colombia”, OECD Regional Development Working Papers, No. 2016/8, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jln4pn1zqq5-en.

[25] Tangtipongkul, K. (2015), “Rates of Return to Schooling in Thailand”, Asian Development Review, Vol. 32/2, pp. 38-64, https://doi.org/10.1162/adev_a_00051.

[27] Tan, K. and J. Tang (2016), New Skills at Work: Managing Skills Challenges in ASEAN-5, http://ink.library.smu.edu.sg/soe_research (accessed on 06 September 2018).

[2] World Bank (2017), World Development Indicators (database), https://data.worldbank.org/data-catalog/world-development-indicators.

[28] World bank (2017), Education Statistics (database), https://data.worldbank.org/data-catalog/ed-stats.

Annex 2.A. Thai secondary cities, as identified through geospatial data

Secondary city

Province

Region

Location

Population density

(inhabitant per km2, 2015)

Population (2015)

Lop Buri

Lop Buri

Centre

Thailand

3 484

128 924

Kamphaeng Phet

Kamphaeng Phet

Centre

Thailand

3 444

72 315

Phra Nakhon Si Ayutthaya

Phra Nakhon Si Ayutthaya

Centre

Thailand

3 338

126 849

Phitsanulok

Phitsanulok

Centre

Thailand

3 000

165 001

Saraburi

Saraburi

Centre

Thailand

2 977

89 309

Nakhon Sawan

Nakhon Sawan

Centre

Thailand

2 904

95 821

Paoy Pet

Sa Kaeo

East

Border with Cambodia

8 778

122 897

Prachin Buri

Prachin Buri

East

Thailand

4 794

86 287

Phanat Nikhom

Chon buri

East

Thailand

3 508

52 621

Chanthaburi

Chanthaburi

East

Thailand

3 394

67 880

Chachoengsao

Chachoengsao

East

Thailand

3 201

57 622

Pattaya

Chon buri

East

Thailand

2 989

242 133

Chon Buri

Chon buri

East

Thailand

2 565

392 467

Bang Lamung

Chon buri

East

Thailand

2 374

168 554

Map Ta Phut

Rayong

East

Thailand

2 253

110 390

Rayong

Rayong

East

Thailand

2 006

138 423

Ban Mae Tao

Chiang Mai

North

Border with Myawadi (MMR)

7 291

131 230

Mae Sai

Chiang Rai

North

Thailand

5 133

153 994

Chiang Rai

Chiang Rai

North

Thailand

3 199

83 166

Chiang Mai

Chiang Mai

North

Thailand

2 818

493 149

Lampang

Lampang

North

Thailand

2 509

85 299

Nong Khai

Nong Khai

Northeast

Border with Lao PDR

3 337

407 100

Udon Thani

Udon Thani

Northeast

Thailand

3 000

149 983

Ubon Ratchathani

Ubon Ratchathani

Northeast

Thailand

2 811

126 488

Khon-kaen

Khon Kaen

Northeast

Thailand

2 656

154 057

Nakhon Ratchasima

Nakhon Ratchasima

Northeast

Thailand

2 533

159 579

Savannakhet

Mukdahan

Northeast

Border with Lao PDR

2 449

66 124

Ranong

Ranong

South

Thailand

4 742

118 553

Sungai Kolok

Narathiwat

South

Border with Malaysia

4 255

55 312

Trang

Trang

South

Thailand

4 078

89 717

Yala

Yala

South

Thailand

3 835

111 221

Phuket

Phuket

South

Thailand

3 707

426 293

Nakhon Si Thammarat

Nakhon Si Thammarat

South

Thailand

3 658

146 307

Hat Yai

Songkhla

South

Border with Malaysia

3 545

297 792

Surat Thani

Surat Thani

South

Thailand

3 064

150 127

Pattani

Pattani

South

Thailand

2 835

76 558

Songkhla

Songkhla

South

Thailand

2 388

162 380

Phrachedi Sam Ong

Kanchanaburi

West

Border with Myanmar

8 792

79 132

Ban Phrachedi Sam Ong

Kanchanaburi

West

Thailand

4 105

102 628

Phetchaburi

Phetchaburi

West

Thailand

3 513

91 335

Ratchaburi

Ratchaburi

West

Thailand

3 341

76 841

Source: Authors’ work based on the Global Human Settlement Layer, which is publicly provided by the Joint Research Centre of the European Commission.

Annex Figure 2.A.1. Manufacturing and agriculture are the predominant sectors
picture

Note: The figure maps the highest share of real GPP in 2015.

Source: Authors’ work based on national accounts, as provided by the NESDB.

Notes

← 1. Due to data availability, most of the analysis in this chapter focuses on the period from 2001 to 2015.

← 2. Between 2010 and 2013, the export value of Thai natural rubber averaged USD 9.3 billion a year. Over the same period, rice had an average export value of USD 5.3 billion a year (Sakayarote and Shrestha, 2017[8]). The decline in the price of rice on global markets motivated rice-growing farmers to produce other crops, such as rubber.

← 3. This figure may be compatible with the fact that Northeast is predominantly a rural and ageing society. In this case, a decrease in the absolute numbers of workers would be physiological. The neighbouring North shares similar socio-economic features. However, contrary to the Northeast, it witnessed an increase in the absolute number of employees by almost 5% between 2001 and 2015.

← 4. The most recent of these are the Bangsaphan Steel Industry Real Estate (created in 2011) in Prachuab Kiri Khan, and the Khao Yoi Industrial Park (2014) in Petchaburi.

← 5. It should be noted that the gap in productivity is only a rough indicator of untapped growth potential. This methodology does not take into account other dimension of local development and can therefore be improved. Relevant govenrment agencies could measure convergence with respect, for instance, the Human Achievement Index.

← 6. The convergence-divergence analysis identifies the following provinces as frontiers: Chiang Mai and Kam Phaeng Phet (North), Khon Kaen and Nakhon Ratchasima (Northeast), Rayong (East), Ayutthaya (Centre), Ratchaburi (West), and Krabi and Phuket (South). The analysis does not include the provinces that constitute the Bangkok Metropolitan Authority.

← 7. The GHSL dataset is public available here: http://ghsl.jrc.ec.europa.eu/degurba.php. More information about the project is available here: https://ghsl.jrc.ec.europa.eu.

← 8. The OECD has been developing an even more accurate definition of so-called functional urban areas (FUA). FUA comprises a set of contiguous local administrative units that are a function of population density as well as commuting times. In Europe, these administrative units correspond to municipalities. In the rest of the OECD countries, the local units composing FUAs are the smallest administrative areas for which national commuting data are available. FUAs can be defined in Thailand as well upon the collection and release of micro-data on commuting times or commuting behaviours (e.g. through updated Population Censuses).

← 9. Throughout this section, post-secondary education is considered part of university and tertiary education.

← 10. For example, Mahidol University (the hightest-ranked university in Thailand according to the Times Higher Education and World University Rankings) has 6 campuses, 4 of which are in the Northeast, East, West and Centre. Thammasat University (3rd-higest ranked according to QS World University Rankings, 2018) also has 6 campuses, 4 of which in the North, Northeast, East and South.

← 11. “Rajabhat enrolments shrinking”, Bangkok Post (2 February 2018): www.bangkokpost.com/news/general/1406186/rajabhat-enrolments-shrinking (accessed August 2018).

← 12. The overall mission of the committee is to draft new integrated regional policies for all regions.

End of the section – Back to iLibrary publication page