Journal Archive

Lee and Lee (2014) identified the establishment or closure of companies based on the names of individual companies cited in the Rodong Sinmun and official North Korean media and estimated the number of companies by region in North Korea, including Pyongyang. The status of manufacturing, mining, and energy industries was examined, and it was noteworthy that the manufacturing sector was investigated after it was subdivided into 12 sub-categories. The investigation was carried out for each province and Pyongyang as a city. Furthermore, as a follow-up study continues by Lee et al. (2018), time series changes in some industries can be investigated. Hong et al. (2016) identified the status of the official markets in North Korea by a rough assumption of the location of the official markets based on interviews with North Korean defectors and then comparing it with satellite images of Google Earth.

By estimating the number of official markets, location, regional distribution, and the number of workers by province and city, the status of wholesale and retail business and the logistics and distribution structure in North Korea were identified. Lee (2017) analyzed the classifications of service industries in North Korea and the status of management authorities using North Korean literature and interviews with North Korean defectors. Yoon (2019) used the data from the National Statistical Office to examine the status of North Korea's electric power industry. It was pointed out that the shortage of electricity supply was the main cause of North Korea's economic hardship, and predicted the electricity demand for each scenario type by applying the scenario methodology. Kang (2020) analyzed the changes in the tourism industry using North Korean tourism-related literature, North Korean laws, and media coverage. It discovered that the tourism industry is growing as a counter-strategy to the deteriorating foreign trade balance due to the sanctions against North Korea, and analyzed the spatial characteristics and limitations of the tourism development zone. As such, the individual industry analysis has in common that the scope of research is focused on a small number of specific industries and that the status of industries was investigated at a microscopic level.

Shin (2009) conducted research based on the Bank of Korea's estimates of the economic growth rate in North Korea and the data from the Ministry of Unification on the comprehensive evaluation of the North Korean economy. The input-output tables of North Korea were estimated by applying the principle of cross-entropy minimization. The social accounting matrix was estimated and the accounting multiplier for investment in North Korea was identified based on the estimated input-output tables. Shin et al. (2014) estimated the input-output tables in North Korea and analyzed the forward and backward linkage effects by industry. As a result, it was found that both the light industry and primary metal industry in North Korea have high forward and backward linkage effects. On the other hand, the machinery and electronics industries showed low forward and backward link effects. Consequently, it was concluded that the investment priority in the former is high whilst the priority is low in the latter.

Kim and Lee (2008) found that the industrial structure of North Korea was mainly centered on the heavy industry from the establishment of the regime to right before the end of the Cold War, however, since 2000, the agriculture and forestry, light industry, trade and service industries have grown, diversifying major industries. Jung and Yoon (2004) pointed out that North Korea has pursued an imbalanced growth strategy that prioritizes the heavy industry related to the military industry over other industries since the Korean War, and the resulting industrial structure acts as an obstacle to hamper the economic growth of North Korea. North Korea's heavy industry-oriented policy that both Kim and Lee (2008) and Jung and Yoon (2004) pointed out, helps understand the background in which the manufacturing industry takes up a large share of the industries in Pyongyang and North Korea.

Lee et al. (2014) selected Nampo, Cheongjin, Sinuiju, Rason, and Wonsan as major base cities in North Korea based on city population, city location, and socioeconomic characteristics. Furthermore, considering the development cases of transition countries abroad, the study proposed the industries to be strategically fostered and the urban development promoters. Unlike previous studies, it analyzed the industrial structures considering the size and location of cities, and understood the implications from the transition cases of the base industry among the development cases of transition countries abroad. Park et al. (2019) examined the industrial status of North Korea based on macroeconomic indicators such as GDP, GDP per capita, population, and GDP contribution rate by industry, and presented macroscopic indices and industry forecasts based on the scenarios. Further, the study predicted the economic growth rate according to the growth scenario of North Korea's GDP and projected the development stage by industry to present the demand for the labor force by industry.

Previous studies had common limitations as follows. First, the reliability of the analysis results. Although the interview data used in individual industry studies provide detailed information on industries in the region where North Korean defectors resided, it is likely to be inaccurate for non-resident areas. The study of industrial structure conducted estimation such as a scenario analysis in addition to the use of macroeconomic indices estimation data as the basis, indicating poor reliability of the analysis result.

Second, it is impossible to determine the industrial status by region. There was no information at the city and provincial level in North Korea in the data used in the industrial structure analysis based on macroeconomic indicators. This made it impossible to analyze the industrial structure at the city and provincial level, and only the analysis results at the national level were presented. Therefore, information such as the level of agglomeration and the diversity of industries in each region could not be identified.

Third is the difficulty of generalization. In the study of individual industries, it is difficult to generalize the industrial status due to a small number of interview samples, and impossible to compare the study results with other cities and countries as the industry classification developed for the study is different from the Korean standard industry classification.

This study enhanced the reliability of the estimation with quantitative indicators derived from the North Korean census. Furthermore, this study is significant in that it generalizes the industrial structure of Pyongyang using data on working population by region and industry and population with tertiary education in North Korea, and compares the industrial structure with other countries and areas other than Pyongyang to present the analysis results on the industrial characteristics of Pyongyang.

In 2008, North Korea conducted the second census. There is no information on the working population by region or industry in the 1993 census data, and there is no census data open to the public from 2008 to 2021. Therefore, a time series analysis of changes in the industrial structure of Pyongyang cannot be conducted. Although the analysis is based on single-year data, this study helps address the difficulties of generalizing the industrial structure which previous studies faced.

As shown in <Table 1>, the 2008 North Korean census data shows 12,184,720 North Korean working population aged 16 and over, by grouping them into 18 industries and regionally into 9 provinces and Pyongyang city. The 18 industrial classifications used in the census are consistent with the 21 major classifications of the Korean Standard Industrial Classification. “Real estate industry”, “activities of households as employers & undifferentiated producing activities of households for own use”, and “activities of extraterritorial organizations and bodies” are not included in the 18 industry classifications, and “membership organizations, repair and other personal services” in the Korean Standard Industrial Classification is denominated as “Other service industry”.

Pyongyang city and nine provinces of North Korea were the focus of this study. As such, the industry classification of the North Korean census data is almost akin to the Korean standard industry classification, and can be used in the industry comparisons with other countries including South Korea since the Korean standard industry classification is built based on the International Standard Industry Classification of the United Nations Statistics Office (Statistics Korea, 2021). The Central Bureau of Statistics (2009) states that the census data applied the UN Principles and Recommendations on Census of Population and Housing (2^nd Revision) with necessary modifications to prepare the questionnaire. However, due to the social and cultural background of North Korea, there are some limitations that the industry classification used in the census may be interpreted in a way that is different from other countries. For example, the working patterns and the details of the job between North Korean and South Korean workers in professional, scientific and technical activities may differ from each other.

Data on population with tertiary education by region and field of study included in the 2008 census were also used. Population with tertiary education is an indicator that shows the status of human resources that can be mobilized in a city and a variable that shows the potential of regional economic development. This study identified the status of human resource agglomeration by region by applying the same formula to estimate the location quotient (LQ) of industries to the population with tertiary education.

As shown in <Table 2>, 2,972,852 North Korean residents aged 16 and over with at least 3 years of tertiary education are classified into Pyongyang city and nine provinces and 23 fields of study. Furthermore, the UN statistics data, census data by country, and international statistical data from Statistics Korea were used to analyze the industrial status of Pyongyang and North Korea.

Location Quotient (LQ) is an indicator showing how much a specific industry is agglomerated in a region. It can be calculated with Equation (1) (Choi, 2004). Equation (1) divides the proportion of the industry i in region j by the proportion of the industry i in the country. Variables for location quotient calculation may include the number of businesses, production amount, and the number of employees depending on the purpose of the study (Cho et al., 2019).

(1)

When analyzing the location quotient, a value of 1.0 indicates that the level of agglomeration of the corresponding industry is equal to the average. There are no clear criteria to determine whether an industry is agglomerated, but in general, if the location quotient is 1.25 or higher, it is regarded that the corresponding industry is reasonably more agglomerated than other regions (Segarra-Oña et al., 2012; Mans et al., 2008; Park et al., 2015). Choi (2004) pointed out that it should be carefully interpreted when the overall industrial scale of the analysis region is small. In this case, even a small concentration of a specific industry can be classified as an agglomerated industry. To overcome the limitations of the location quotient interpretation as mentioned above, Cho et al. (2019) identified an agglomerated industry of Seoul if that industry satisfies both “relative and absolute” criteria. The relative criteria were set for the location coefficient exceeding 1.0 based on the number of businesses or employees. Three absolute criteria were set for the number of corporations of 100 or more, the ratio of the working population in Seoul to that of South Korea exceeding 2%, the ratio of the working population of a specific industry to the industrial size of Seoul exceeding 3%. Satisfying two or more of the three criteria was set as the conditions of satisfaction the absolute criteria.

In this study, if both the relative and absolute criteria in <Table 3> were satisfied, the industry was identified to be significantly more agglomerated than other regions considering the criteria adopted by previous studies. As for the relative criteria, the location quotient of 1.25 was used which is more stringent than 1.0 that has been widely used in other studies. Unlike Cho et al. (2019), the number of corporations was not taken into account in the absolute criteria, since the North Korean census data does not contain the number of companies by region. The reason why the ratio of the industry to the total industrial size of the region was reduced from 3% to 2% is that in most developing countries the industries are less diversified compared to developed countries and some industries account for a large proportion of the economy. Therefore, if the conditions of satisfaction for the ratio of the industry to the total industrial size of the region is considerably increased, it will make it difficult to identify agglomerated industries by region.

To measure the level of agglomeration of human resources by region, the working population by industry in Equation (1) was replaced with the population with tertiary education by field of study of <Table 2>, and <Table 3> was used as the criteria to determine the level of agglomeration.

The working population per 1,000 people can be used when comparing the absolute value of the industrial size for each region by standardizing the industrial size based on 1,000 people in a situation where population varies by city and province. The working population per 1,000 people is calculated using Equation (2).

(2)

The level of industry diversity was estimated using the entropy index. The measurement of entropy index is based on the laws of thermodynamics of physics. It is about measuring the increase in disorder of each element constituting a system, and it is used to measure income inequality, diversity, and level of specialization in industrial structures in social science. The entropy index formula to measure the industrial diversity can be found in Equation (3). S_n represents the employment ratio of industry n in the region. N refers to the number of total industrial classifications in the region. In Equation (3), E_i shows the maximum ln(N) when the employment ratio for each industry is in uniform, and shows the minimum value of 0 when the employment ratio is 100% in one industry (Hong, 2015). In this study, the entropy index was used in two different ways. First, the industrial diversity of the region was measured. This is to find out whether the industrial ecosystem of a region is concentrated on ‘some industries’ or whether 18 industries are evenly distributed. In the measurement of industrial diversity, S_n from Equation (3) represents the employment ratio of industry n in the region. Second, the entropy index in relation to inequality of industrial distribution was measured. This is to check whether the corresponding industry is concentrated on ‘some regions’ in North Korea or whether it is evenly distributed across North Korea. In the measurement of the entropy index in relation to inequality of industrial distribution, S_n from Equation (3) represents the employment ratio of region n within the industry.

(3)

<Figure 1> shows the agglomeration of the industry and human resources that satisfy the criteria of <Table 3>. As a result of comparing the agglomeration status of Pyongyang and non-Pyongyang regions, two distinctive characteristics were found.

Figure 1. 
Agglomeration of industry and human resources in North Korea by each region

First, there is a wide gap between the agglomeration levels of industry and human resources. Eight industries and human resources in eight fields of study are concentrated in Pyongyang. It can be assumed that the urbanization economy effect, an agglomeration effect between different industries, would be at play since various industries and human resources are evenly distributed. Furthermore, it is considered that the human capital externalities (Enrico, 2014) created by a high level of human capital will be greater than that of other regions. The location quotient of 6 out of 8 agglomerated industries hovered above 1.9, indicating a high level of agglomeration.

<Table 4> shows the inequality of industrial distribution among 18 industries. The minimum value of the entropy index in relation to inequality of industrial distribution is 0, and the closer to 0, the more the industrial distribution is concentrated in a small number of regions. The maximum value is ln(N), and considering that N is 10, the number of North Korean cities and provinces, the maximum value is ln(10), which is equal to 2.3. It shows that the closer to 2.3, the more evenly distributed the industry is across North Korea. <Table 4> indicates that the industries such as professional, scientific and technical activities, information and communication industry, and construction industry agglomerated in Pyongyang are highly concentrated in a small number of regions compared to other industries.

In other words, it can be assumed that the industries with a low entropy index in <Table 4> among the agglomerated industries in Pyongyang are underdeveloped in other regions of North Korea. Only one or two industries are mostly agglomerated in regions other than Pyongyang, and no industry shows the high level of agglomeration with the location quotient of above 1.9, except for the mining and quarrying industry in South Phyongan Province. Whilst the agglomeration of human resources in three fields was found in some regions, some other regions did not show any agglomeration. As a result, it can be seen that there is quite a big difference between Pyongyang and non-Pyongyang regions in terms of the size and level of agglomeration of industries and human resources.

Second, the nature of the agglomeration field is different. As can be seen in <Figure 1>, many of Pyongyang's agglomerated industries are knowledge and technology-intensive industries. Professional, scientific and technical activities, information and communication industry, construction industry, and arts, entertainment and recreation industry are typical industries that require knowledge and technologies. The agglomeration of knowledge and technology-intensive industries outside of Pyongyang was only found in the education sector in Ryanggang Province and the manufacturing industry in Jagang Province. Most agglomerated industries in regions other than Pyongyang are labor-intensive industries such as agriculture, forestry and fishing, transportation and storage, and mining industries, whose nature are different from the agglomerated industries in Pyongyang. The agglomerated knowledge and technology-intensive industries in Pyongyang are also highly related to the agglomeration of human resources. In Pyongyang, the agglomeration of human resources takes place in eight fields of study - engineering and natural science such as computer science, physics, architecture, and mathematics and statistics. These fields of study are highly related to the agglomerated industries such as professional, scientific and technical service, information and communication, and construction industries.

Agglomeration of human resources in arts is also associated with the arts, entertainment and recreation industry, and also the human resource in the welfare services is related to the agglomeration of the human health and social work activities industry. This shows that there is a close relationship between the supply of human resources and the agglomerated industry in Pyongyang. Outside of Pyongyang, the only agglomeration of human resources in the fields of engineering and natural science focused on mathematics, statistics and physics in South Hamgyong Province and mathematics and statistics in Ryanggang Province, and no other agglomeration of human resource was found in the field of engineering and science.

Third, the ratio of industrial composition and the level of industrial diversity differ from other regions. The industrial size was compared based on the working population per 1,000 people. In Pyongyang, there were 510 workers per 1,000 people, and 147 of them were engaged in the manufacturing industry with the highest working population. This is equivalent to 28.8% of all working population. The industry with the second-highest number of workers was agriculture, forestry and fishing, with 69 workers. This represents 13.5% of all working population which is very different from the fact that the working population in the agricultural, forestry and fishing industry exceeds 30% in most regions other than Pyongyang. The combined working population in the two industries above accounts for 42.3%, which does not surpass half of all workers in Pyongyang. Other industries showed relatively even industrial size. Unlike Pyongyang, all regions except for Pyongyang had common features as follows. The industry with the largest number of workers is the agriculture, forestry and fishing industry, followed by the manufacturing industry. Furthermore, the combined ratio of working population in the agriculture, forestry, fishing and manufacturing industries exceeded 48%, nearly half of the total number of workers in the region. As such, using the entropy index, whether some industries dominate the entire industrial ecosystem in the region and whether various industries are evenly distributed can be quantified and compared. The entropy indices for each region of Pyongyang and North Korea calculated by Equation (3) are shown in <Table 5>.

Here, the minimum value of the entropy index is 0. Closer the value to 0 indicates that a small number of industries dominate most of the industrial ecosystem in the region. The maximum value is ln(N), and considering that N is 18, the number of industries, which is equal to 2.89. The closer the value to 2.89 indicates that the industrial ecosystem in the region is evenly filled with various industries.

Pyongyang's industrial diversity index is very high compared to other regions. This shows that various industries are evenly distributed in Pyongyang. The entropy index in Ryanggang, North Hamgyong, South Phyongan, and Kangwon exceeded 2.0, indicating a higher level of industrial diversity when compared to other regions. It is assumed that the above results were obtained as the manufacturing industry in Ryanggang, North Hamgyong, South Phyongan, and Kangwon Province is developed at a level similar to the agriculture, forestry and fishing industry and other industries are evenly distributed despite the high proportion of the agriculture, forestry and fishing industry.

This study conducted a comparative analysis with other countries to investigate the industrial structures of Pyongyang and North Korea. The World Bank classifies countries into four groups based on GNI per capita. North Korea belongs to a low-income economy, along with Burkina Faso, Afghanistan, and Somalia. Southeast Asian countries, such as Vietnam, Cambodia, and Laos, are classified as the lower middle-income group, a group right above North Korea (World Bank, 2021b). In terms of GNI and GDP per capita, North Korea can be considered as a typical developing country. Comparison countries were listed based on GDP per capita. By setting North Korea's GDP per capita as the median, the comparison group in <Table 6> consisted of the top 15 countries with GDP per capita higher than that of North Korea and the bottom 15 countries with GDP per capita lower than that of North Korea. The base year was set as 2008 when the census was conducted in North Korea, and the industrial ratio in Pyongyang was estimated based on the working population in the 2008 census data, instead of GDP. The average GDP per capita of the comparison group is $1,043, similar to North Korea's GDP per capita of $1,029, indicating that the comparison group was appropriately selected.

Since only countries are included in the comparison groups, it may not be appropriate to include Pyongyang, a city, in the comparison target in <Table 6>. However, Pyongyang was included in the analysis to identify the industrial status in the city in the absence of industrial composition data by city and industry of each country. Among various industries, only the ratios of the agriculture, forestry and fishing, manufacturing, and service industries are selected and listed in <Table 6>, hence, the total ratio of the three industries is less than 100%.

As a result of comparing the industrial ratio between countries, Pyongyang and North Korea are characterized by their high proportion of manufacturing. As can be seen from the average values of 30 countries in <Table 6>, the comparison group mainly had a low proportion of manufacturing and a high proportion of service industry. Before explaining this in detail, it is worthwhile to point out a few things regarding the interpretation of the indicators. In the analysis results above, it was explained that there are a number of knowledge and technology-intensive industries in Pyongyang's agglomerated industry. Further, it was explained that Pyongyang's industry had a high proportion of manufacturing compared with other countries. These descriptions may be seen as contrary to each other. However, the index used to examine the characteristics of a knowledge and technology-intensive industry is location quotient, and the data used to explain the high proportion of manufacturing is the ratio of agriculture, forestry and fishing, manufacturing, and service industries to the total industries for each country. The location quotient represents the ‘relative degree of specialization’ of an industry by comparing the proportion that a specific industry in a given region with the proportion of the specific industry nationwide (Cho et al., 2019). On the other hand, the ratios of agriculture, forestry and fishing, manufacturing, and service industries simply show the size of each industry rather than a comparison between the ratios of a region to the country. Therefore, it is not contradictory that Pyongyang's agglomerated industry and large-size industry are found to be knowledge and technology-intensive industries, and manufacturing industries, respectively.

The manufacturing ratios of North Korea and Pyongyang are 22.5% and 28.8%, respectively, higher than the averages of other countries with similar economic levels. North Korea's manufacturing ratio of 22.5% is the highest among the 30 countries in the comparison group, and Pyongyang's manufacturing ratio is higher than that of the average of North Korea. It can be said that GDP per capita represents the labor productivity reasonably well in North Korea in which the added values of natural resource-based industries and tourism-based industries are low (Lee, 2019). Considering its high proportion of manufacturing among countries with similar GDP per capita, a proxy indicator for labor productivity, and the accumulation experience while maintaining 17-31% of a high manufacturing ratio between 1990 and 2020 (Korean Statistical Information Service, 2021e), North Korea's manufacturing sector is expected to have greater productivity than countries with similar economic level. Szirmai and Verspagen (2015) found that the manufacturing industry has a positive (+) impact on economic growth, and that the interaction term between manufacturing and education has a significant effect on the growth. Based on these, it is considered that the high proportion of manufacturing in Pyongyang and North Korea is expected to have a positive impact on North Korea's economic growth going forward. It was the year 1977 when South Korea's GDP per capita was similar to that of North Korea ($1,055) (World Bank, 2021a). As for the working population by industry in Seoul in 1977, 1.2% of the population worked in the agriculture, forestry and fisheries, 31.3% in the mining and manufacturing industry, and 67.5% in SOC and other service industries (Seoul Institute, 2021). Although the aforementioned industry classification is different from the industry classification in <Table 6>, it was discovered that when the per capita GDP of Pyongyang and Seoul were similar, the proportion of manufacturing was relatively high in the total industries.

The high manufacturing ratios in North Korea and Pyongyang stem from the industrial structure of socialist countries in the early 1990s. <Table 7> shows the proportion of each industry as well as GDP in the early 1990s in socialist countries that have undergone regime transition. The base year in <Table 7> is the year at which statistics were collected after the regime transition took place, which is within 5 years after the transition. Thus, the base year of some countries is different from the time of transition. Countries with a point of transition and base years of available data exceeding 5 years were excluded from the table.

The average manufacturing ratio in all transition countries is 19.6%, which is higher than that of high-income countries, OECD member countries, the global average, and least developed countries. At that time, the manufacturing ratio of transition countries in Europe was 21.2% and the manufacturing ratio in transition countries in Asia was 14.3%. In Asia, only North Korea (24.6%) and China (28%) had a high proportion of manufacturing, which is considerably higher than the average manufacturing ratio of transition countries in Europe. As such, as a socialist country, North Korea is assumed to have maintained a high manufacturing ratio as a result of pursuing a ‘heavy-industry-first’ policy for a long time which is similar to Stalin's industrialization model (Yang, 2015).

The industrial diversity of North Korea was compared with other countries by estimating the entropy index using the UN data (United Nations Statistics Division, 2021).

In <Table 8>, the possible maximum entropy index for a country with 17 industry classifications is 2.83, and a country with 21 industry classifications records a possible maximum entropy index value of 3.04. The higher the entropy index, the less unevenly developed industries in the country, indicating that various industries are evenly developed. Although it is difficult to compare on an equal basis since the number of industrial classifications as well as the time of census by country are different, a rough comparison between countries can be made using <Table 8>.

With South Korea as a basis, all countries with high per capita GDP have an entropy index above 2.5 and it can be seen that they have reached a state of even development of various industries. The Jacobs’ School of Thought argued that the more diverse the local industry, the more positive impact it has on the local economy (Kim and Song, 2019), and Moon et al. (2014) found that there is a positive (+) relationship between industrial diversity and economic growth in the region. In this context, it can be interpreted that the higher the entropy index that represents industrial diversity, the more positive impact it can have on a country's economy. The entropy index of the developing countries with a smaller GDP than South Korea did not surpass 2.5, and the entropy index that showed the industrial diversity of North Korea was 2.08.

Although the GDP per capita of Ghana, Vietnam, Kenya, and Laos was higher than that of North Korea, their industrial diversity was poorer than North Korea. In short, it can be said that the level of industrial diversity in Pyongyang and North Korea is relatively high compared to the GDP per capita.

<Table 9> is the result of investigating the industrial diversity, the most agglomerated industry in the city, and the level of agglomeration of information and communication, professional, scientific and technical activities based on the working population by industry in Beijing, Seoul, Pyongyang, Ho Chi Minh City, and Phnom Penh. As for Vietnam, the working population by industry in the census data was not formulated by city, thus the value of the southeastern region including Ho Chi Minh City was used. The maximum entropy index also varies because of the different number of major industry classifications for each country. The possible maximum entropy index values are 2.94 in Beijing and Seoul, 2.89 in Pyongyang, 3.04 in Ho Chi Minh City, and 2.89 in Phnom Penh. It indicates that the closer to the maximum value, the more evenly distributed the industrial ecosystem is with various industries in the city. The level of industrial diversity in Pyongyang was estimated to be lower than that of Beijing or Seoul. Based on the year of the survey, Vietnam's per capita GDP was higher than that of North Korea, but Pyongyang's industries were more diverse than that of Ho Chi Minh City. For Cambodia, 2019 data was used, where its GDP per capita was higher than that of North Korea in 2008. When Phnom Penh and Pyongyang were compared, North Korea's per capita GDP was lower than Cambodia's, but Pyongyang's industrial diversity is higher than that of Phnom Penh. As such, Pyongyang showed higher industrial diversity than other capitals, compared to its economic size, which can be the basis for the assumption that the urbanization effect within Pyongyang would be greater than that of other cities with a similar economic size.

According to <Table 9>, the information and communication industry was the most agglomerated industry in Seoul while it was the professional, scientific and technical activities in Pyongyang and Phnom Penh, and the manufacturing industry in Ho Chi Minh City. Beijing’s most agglomerated industry was the leasing and business services followed by the information and communication industry. As such, the agglomeration of knowledge-intensive industries such as professional, scientific and technological activities, and information and communication in the capital is a universal characteristic of industries that can be found in many countries. However, the agglomeration high enough to satisfy both the absolute and relative criteria in <Table 3> was not found in Ho Chi Minh City. In Phnom Penh, the agglomeration of the professional, scientific and technical activities satisfied the criteria in <Table 3>, but no agglomeration was found in the information and communication industry. On the other hand, Pyongyang, like Beijing and Seoul, showed the agglomeration in all the professional, scientific and technical activities, and information and communication industries, indicating that Pyongyang is relatively more specialized in knowledge-intensive industries than other developing countries.