Degentrification? Different Aspects of Gentrification before and after the COVID-19 Pandemic

Abstract

The purpose of this study is to explore the aspects of “gentrification” and “degentrification” other than economic factors. To this end, this study focused on the gentrification situations occurring before and after the COVID-19 pandemic in the Itaewon area, South Korea, by using semantic network analysis. We analyzed news articles on the gentrification phenomenon in the Itaewon area reported in South Korea. As a result, gentrification in the Itaewon area is divided into four stages. The first stage of gentrification (2010~2014) is initial stage of gentrification. Gentrification stage 2 (2015~2017) is the period of commercialization as a gentrification growth stage. The first stage of degentrification (2018~2019) is the maturation period of gentrification. The second stage of degentrification (2019~30 June 2020) is the period of the COVID-19 pandemic. The results confirm the existing theoretical frameworks while building a more nuanced definition through operationalizing gentrification and degentrification. As with the etymology of the term, the degentrification phenomenon can only be revealed when the gentrification phenomenon is prominently displayed. This study has an implication in that it tried to phenomenologically examine the specific phenomenon of the next stage of gentrification through the term “degentrification”.

1. Introduction

1.1. Background and Objective

Over the last few decades, the issue of “gentrification” has spread to many academic disciplines. Gentrification is a term originally proposed by Glass [1] which usually is used to describe the processes and spatial changes in many metropolitan cities, with the suburbanization reversing around the old city center, the return of the middle class to the city center, and the push of the working class and the common people [2,3,4]. In other words, gentrification was defined as the process of social, economic, cultural, physical, and demographic transformation of a neighborhood with the influx of capital [5].

As this theory has been around for a long time in the academic field, there also have been proclamations about the demise or death of gentrification since at least the early 1990s. In Western countries, it has led to the deidealization of gentrification over time, which is also defined a “super gentrification”, “rural gentrification” and “commercial led gentrification”, “cultural led gentrification”, and “tourism gentrification” [6,7]. This suggests that gentrification is a complex intertwined problem of macroscopic perspectives, based on a micro approach to behavior among relevant stakeholders and social, economic, and spatial restructuring.

In relation to gentrification, the following five harmful phenomena are mainly explained: residential relocation; exclusion; transformation of public, social, and commercial spaces; polarization; and homogenization [8]. Gentrification is mainly used in a negative sense both semantically and analytically. In this context, how should the opposite phenomenon of gentrification appear and be explained?

The purpose of this study is to explore the aspects of “gentrification” and “degentrification” beyond the economic factors. To this end, this study focused on the gentrification situations occurring before and after COVID-19 pandemic in Itaewon area, South Korea. The world is embroiled in the COVID-19 pandemic with unprecedented uncertainty. The climate change emergency lurks right behind and has great potential to further disrupt the travel and tourism industry or spawn new forms of privileged transnational travel and mobility. Restrictions on travel mean that many of the gentrification processes in this special issue have been suspended or are questionable [9,10,11].

The Itaewon area was formed by various political, economic, and cultural contexts of Korean society. Especially, Itaewon has a parallel with the process of town-centered development of Seoul city government after the 1960s, and it has established the identity as the consumption space by neoliberalism formally introduced after the 2000s. Moreover, based on the historical practice called the “US military government”, Itaewon obtains its own unique placeness as a foreigner’s place and a place combined with various cultures. In particular, during the COVID-19 pandemic, there was a time when the number of confirmed cases was concentrated mainly on visitors to the LGBTQ (lesbian, gay, bisexual, transgender, queer or questioning) club located in Itaewon area. This has also caused a misconception about LGBTQ people, and in addition, the number of visitors has decreased rapidly compared to other areas due to the COVID-19. By examining the several phenomena related to gentrification and degentrification before and after COVID-19 in Itaewon area, we would like to take a brief look at how the COVID-19 pandemic, which is a completely different aspect from the past pandemics and financial crisis, has affected gentrification.

To examine phases and aspects of how gentrification affects the Itaewon area, this study focused on “semantic network analysis” that applied big data and social network analysis. This provides a useful perspective for explaining micro behavior and macrosocial structures by examining the behavior of individual actors in the context of the macroscopic structure constructed through the connections and interactions between the micro behaviors. Through analysis of big data, this study has an implication in comprehensively examining the macro and micro aspects of rise and demise of the gentrification phenomenon. By doing so, we delineate how the gentrification and degentrification phenomenon was broadly reflected in the landscape of South Korea, how it is being developed, and what the differences between each stage are. By entering into critical debates on gentrification, we expect to present implications for the conceptual definition of “degentrification”.

1.2. Historical and Geographical Landscape of Itaewon

The Itaewon area was formed by different authorities and is organized as an urban space by the effect of the United States, government, and capital, while the media plays a role in producing governmental ideology, interpellating minorities as others and excluding them. The discourses on the urban space tend to mask the problems of the urban space for the reproduction of capital [12,13].

The area’s commercialization has been accelerated by speculative real estate agents, who are hunting for empty spots and reselling them at higher premiums to tentative leaseholders who anticipate the commercial success of the new business operations. Cultural entrepreneurs, who have imbued cultural aspirations into Itaewon, now face pressure from speculative investments. In the three neighborhoods examined above, rents for commercial properties have spiked rapidly in recent months [14,15]. The beginning of speculative real estate is heralded not only by the increase in rents but also by the creation of premiums, which did not exist before. As the burgeoning commercial facilities penetrated into these neighborhoods, the engagement of speculative real estate agents has instigated landlords to increase rents or to create premiums [13,16].

The urban cultural regeneration models detailed above do not explain Itaewon’s experience. As a “disgraceful” place within national territory, cultural regeneration models of beautification of sites with an emphasis on historical legitimacy and nationalistic sentiment cannot be applied to Itaewon [12,14]. It was once designated as a cultural district to boost international tourism during the 1980s and 1990s, as Korea hosted international events such as the 1988 Seoul Olympics and the 2002 Korea–Japan World Cup [14,17]. However, the effects on urban regeneration were limited due to the lack of supporting programs. In addition, with the rise of other shopping districts and tourist sites, Itaewon lost its competitiveness to other tourist districts.

Since 2010, groups of younger artists have started settling in Itaewon, not as implanted artists but as residents and entrepreneurs. These groups have appeared as important cultural producers for mediating spatial and economic regeneration in Itaewon. There are two important pull factors that attract younger artists and entrepreneurs to Itaewon. First, cheap rents have provided them with affordable spaces for their shops and workshops. Next, urban authenticity, in terms of the neighborhood’s gritty character resulting from deferred urban redevelopment, as well as the area’s cultural diversity, has attracted them. In Itaewon, three different locations have been transformed into culturally vibrant neighborhoods around the same time by groups of cultural entrepreneurs (see Figure 1).

Given the fact that each group named its place by reflecting locational features such as road name and address, the locations in which the groups resided implicate historical connections between places and characters of cultural entrepreneurs: Kyungridan-gil in the proximity to military facilities and middle-class residential area has small restaurants and cafes; the nearby upper-middle-class residential area comprises high-end and stylish clothes shops managed mainly by professional designers; and Usadan-ro, where a central mosque is located and many Muslim migrant workers and lower classes reside, has many artists and cultural entrepreneurs pursuing hippies’ loose lifestyle and dissident cultures. Beyond locational differences, however, there are common socioeconomic structural backdrops behind those young cultural entrepreneurs’ activities, the reasons as to why they entered Itaewon, and the similar consequences they face as a result of their newfound popularity.

Since 2017, reports have started appearing that retail businesses in Itaewon are struggling with soaring rents and the economic downturn [18,19]. The situation in the Itaewon area has changed more radically after the COVID-19 pandemic. In South Korea, after the first patient was discovered on 20 January 2020, the social crisis of the first epidemic prevailed between February and April due to the group infection of the Shincheonji Church. On 11 May, there were 79 cases linked with the clubs in Itaewon. Seoul’s LGBTQ community feared an antigay backlash as several media outlets identified at least one of the bars as a gay bar. Social media users have been tracking and threatening to stop “immoral activity” in some clubs [20].

2. Factors of Gentrification and Degentrification

The gentrification phenomenon has been appearing in completely different forms and phenomena by various actors in different urban living environments [21]. Since today’s gentrification has a wide variety of content and spatial scope, it is necessary to consider complexity and chaos rather than looking at it from a fragmentary perspective. Therefore, it must be emphasized again that it is very difficult to explain the gentrification phenomenon in a concise and simplified way [22,23].

There are several explanations among scholars as to what triggers gentrification. The first factor is demographic change. Because this is by migration, it is important to distinguish between voluntary and involuntary migration to a place of residence [24]. This means that the lower the income level [21,25,26,27,28,29,30], the more involuntary migration occurs, and involuntary migration is inevitable depending on race [23,26,28,29] and social class [25,31,32,33]. In addition, changes in education [23,26,28,29,33,34], occupation [2,29,34,35,36,37,38], period of residence [25,34,38], household composition [25,26,34,35,37,39], population density [21,37,39,40,41,42], number of residents, number of households, and average family size can be seen as major factors affecting population change [43]. The essence of residential decentralization is a phenomenon that arises from the desire of urban residents to find a homogeneous group in various aspects within a heterogeneous urban society. Accordingly, the high-income class wants to move back to a specific area of the city center in consideration of the improvement of the city’s physical environment, fame, and investment value. In this way, the distribution of housing among the classes occurs because the access to the housing market differs according to socio-economic status [24,44]. Furthermore, Marcuse [41] classified displacement in the phenomenon of gentrification into involuntary displacement of direct residence, chain displacement, exclusive displacement, and migration pressure.

Second, deindustrialization and changes in industrial structure are also factors. When the industrial structure of a specific region changes to a cultural center [45,46,47,48], the working class, such as manufacturing, shifts to culture and services, finance, and office work, and enters the middle class, and as cultural and leisure activities increase, their housing demand begins to appear [49,50,51]. Zukin [30] and Bridge and Dowling [52] argue that the change and ripple effect of gentrification is the reallocation of space according to the unique personality of the floating population and the advanced consumption preferences of the middle class. The shift from manufacturing to service industry leads to changes in the industrial structure of cities [47,48] and eventually leads to postindustrial cities, where spatial structural changes due to industrial structural changes are transformed to suit the preferences of the middle and upper classes and their consumption patterns [27,42,53].

The third factor is changes in housing selection priority. Newman and Wyly (2006) [37] found that rent and residential environment affect migration due to gentrification. In addition, house ownership, industrial location and residential area [53], self-ownership rate and rent rate [43], the rate of change in the official land price [43,53], and the sale price and area [24] are housing factors that are closely related to the rent gap. The rent gap [54,55] is a theory that has contributed to the explanation of traditional gentrification, which is the gap between the level of potential land rent and the actual land rent capitalized in current land uses. Tullock [56,57] argued that as the economy turned into a monopoly due to rent-seeking behavior, potential profits were generated and unproductive methods such as looting and lobbying were mobilized to sustain them. Smith [58] also argued that the government should intervene in the market to form an appropriate price for the land market, but the rent-seeking actors involved (from landlords to urban planners, bankers, and real estate investors) are the ones that cause the rent gap.

Lastly, changes in policy are also main factors. Newman and Wyly [37] pointed out that local government support and housing policies affect migration due to gentrification, and Kim [43] found that gentrification is formed according to changes in the housing supply rate, local tax revenue, and financial independence [24,59]. These policies are implemented under the tacit belief that economic activities and high-income classes can be clustered in the downtown area by promoting redevelopment in the downtown area, thereby realizing long-term agglomeration benefits [60]. Lee [61] argued that gentrification would not be able to bring about rapid changes in speed and scale without the intervention of the government. Shaw and Hagemans [62] questioned whether gentrification, which is ultimately the adjustment of capital and class, can lead to actual social mixing by inducing gentrification policies to promote redevelopment or urban regeneration. Social mix is rooted in the region’s historical geography and is difficult to understand as the main concept of a form of redevelopment disguised as gentrification. In this respect, Bridge et al. [63] viewed gentrification as a problem of class and space and emphasized social ethics. Therefore, it can be seen that the gentrification policy must ultimately be linked to social mixing.

Summarizing the above discussion, it can be seen that the factors that form gentrification are diverse and intricately intertwined. In this respect, many scholars believe that a complete explanation of gentrification is impossible, that these factors are a part of gentrification, and that gentrification can be defined as factors such as production and consumption, supply and demand, economy and culture, and structure and subject.

On the other hand, studies on the factors and phenomena of degentrification are remarkably lacking. Smith [58] pointed out the processes of gentrification in the late 1980s and early 1990s as “a ruthless shakeout of small landlords, developers, marginal real estate agencies, and other gentrification-related businesses between 1989 and 1993” came to be characterized as “degentrification”. In the 2008 global financial crisis, there were predictions of degentrification in North America and Europe, but as Hochstenbach and Mustered [64] argue, the crisis impacted gentrification processes but did not lead to degentrification. In 2020 with the COVID-19 pandemic, the media outlets began proclaiming the occurrence of the “degentrification” phenomenon in several cities such as New York City and London [65].

Most of the literature dealing with the phenomenon of “degentrification” regards the standard scale as an economic phenomenon. “Degentrification” has been accepted as the fact that after the inflow of capital reaches its peak due to the gentrification phenomenon in a certain area, facing various economic situations, the rent in that area falls again. However, since the process of gentrification itself has been based on many factors such as economic and cultural phenomena, “degentrification” cannot be also regarded as a simple economic phenomenon which is the opposite of one of the many factors of gentrification. The question is becoming: how can we define and explain the “degentrification” phenomenon? Is this only embracing the meaning of the death of gentrification?

3. Methods

3.1. Data Selection

In this study, data selection and text mining were performed using the Textom program to generate text frequency calculation and semantic network analysis matrix. Textom is a big data analysis solution using text mining technology that collects and refines data in the web environment and can even process matrix data generation. It supports data collection, purification, and analysis in various languages such as Korean, Chinese, and English. Textom provides 1-way and 2-way purification/analysis methods that can process not only collected data but also retained data. Analysis results can be applied to various statistical analysis programs such as SPSS, UCINet, NodeXL, NetMiner, Pajek, and Gephi. Textom is a program developed by a Korean-based company and is useful and meticulous for analyzing Korean-language articles and big data. Therefore, Textom has been used for big data-based trend analysis when Korean is used as the main keyword in various fields [66,67,68,69]. Using text frequency lists, Textom identifies keywords in the text, selects the necessary words, and generates a matrix. It follows the social network view that special meanings will occur when keywords are combined into specific behaviors and identifies the frequency with which keywords occur at the same time [70]. In addition, Textom can delete data that are not related to the research subject from the collected data and refine words that combine words that have been used but have the same meaning.

We analyzed news articles on the gentrification phenomenon in the Itaewon area reported in South Korea. The mass media, including newspapers, play an important role by reflecting the way the public perceives public issues, which in turn leads to an understanding of the population’s issues [71,72]. The unit of analysis is an individual news article. Random sampling was used to select articles that refer to the gentrification in relation to the Itaewon area by titles and texts. The data analysis period was limited to eight years from 1 January 2010 to 30 June 2021, to examine recent aspects of the region of Itaewon. The year 2010 is when the issue of gentrification began in earnest, starting with Kyungridan-gil in the Itaewon area. Kyungridan-gil was the place where chefs who had been studying and working abroad in search of a relatively cheap rent compared to the main commercial area of Itaewon since around 2010 started their restaurant. Along with this, several popular celebrities opened restaurants here; Kyungridan-gil became the hottest place on social media [73,74]. The search terms were “Itaewon”, “gentrification”, and their derivatives. As a result, we retrieved a total of 1324 news articles. The exclusion criteria were articles not directly related to spaces and articles that overlapped with other articles. Finally, a total of 510 articles were included in the analysis.

After the data collected through the Textom package were firstly subjected to hypotheses in the system, they were repeatedly purified by the researchers using MS-EXCEL. The refinement criteria were first deleted when there was an unnecessarily long postposition or modifier. Second, keywords with similar concepts such as “measures”, “alternatives”, and “plans” were integrated into one word and refined. Third, keywords that can be used as a single word in terms of content, such as “contract renewal application”, were used without spaces, and finally, unnecessary symbols and special characters were deleted for core keyword analysis.

3.2. Semantic Network Analysis

Semantic network analysis enables structural analysis of how words are used and placed on specific topics [67]. It is advantageous to extract formalized information from unstructured data to visualize the semantics and patterns of the communication process.

Based on the final data, keyword frequency was extracted by using Textom, and term frequency (TF)–inverse document frequency (IDF) and degree centrality were analyzed together. Degree centrality is linkage centrality, which is a statistical figure that examines key keywords by identifying how many keywords (nodes) are associated with other keywords (nodes). TF-IDF is a product of TF multiplied by IDF, which is a statistical value that weights how important a keyword (node) is in a particular document [75]. $T{F}_{i,j}$ is the value obtained by dividing the number of times keyword $t_i$ appears in article $d_j$ by the number of times occupied by all keywords. $ID{F}_i$ is the total number of articles divided by the number of articles with keyword $t_i$, and $TFID{F}_{i,j}$ is the product of $T{F}_{i,j}$ and $ID{F}_i$. This is formulated as follows:

$$T{F}_{ij}=\frac{n_{i,j}}{{{\displaystyle \sum }}_k{n}_{k,j}}$$

${\displaystyle \sum }_k{n}_{k,j}$: number of times all keywords appeared in article $d_j$.

$n_{i,j}$: number of times keyword $t_i$ appeared in article $d_j$.

$$ID{F}_i=log\frac{\left|D\right|}{\left|d_j: t_i \in d_j\right|}$$

$\left|d_j:t_i \in d_j\right|$: number of articles in which $t_i$ is suggested.

$\left|D\right|$: total number of articles.

$$TFID{F}_{i,j}=T{F}_{i,j }\times ID{F}_i$$

3.3. Phases of Gentrification and Degentrification

In this study, related articles on gentrification and degentrification were collected annually from 2010 to June 2021, and the process of reviewing the contents presented as key keywords was repeated. Through this process, each year was divided into years in which similar key keywords were presented, and the four-stage formation period was set as shown in

Table 1. Phases of gentrification and degentrification aspects of Itaewon area.

	Gentrification		Degentrification
	Phase 1 (2010~2014)	Phase 2 (2015~2017)	Phase 1 (2018~2019)	Phase 2 (2020~2021 June 30)
Articles	30	257	182	42

.

4. Results

4.1. Gentrification

4.1.1. Gentrification Stage 1 (2010~2014)

Itaewon’s first stage of gentrification is from 2010 to 2014. Itaewon has long been known as the most exotic space in Seoul [76]. The change of place in the Itaewon area has accelerated since 2010 and has rapidly expanded not only along the roadside but also along the local alleys. Usadan-ro, Hoenamu-ro, and Kyungridan-gil in Itaewon-dong became popular among young people in their 20s and 30s, and the scenery of the alleyways changed rapidly. While Itaewon’s 1 and 2 dongs are the center of restaurants serving a variety of cuisines from around the world, the Hannam-dong area is characterized by high-end commercialization [77].

During this period, a total of 30 articles related to Itaewon gentrification were searched. The most mentioned keywords were Cities (106), followed by Space (81), People (68), Cafe (64), Market (53), Kyungridan-gil (45), Restaurant (43 times), Life (39 times), House (38 times), Culture (37 times), Project/Society (34 times), Usadan village (30 times), and Merchant/World/Gentrification (27 times). The others were listed in the order of Progress, Rearrangement, Exploitation, Street, Itaewon-ro, Redevelopment, and Merchants Association (

Table 2. Keyword frequency, TF-IDF, and connection centrality for gentrification phase 1.

Keyword	Frequency	TF-IDF	Degree Centrality
Cities	106	159.49979	0.0574
Space	81	151.32899	0.04487
People	68	116.0029	0.04284
Cafe	64	120.86908	0.03912
Market	53	130.62735	0.0337
Kyungridan-gil	45	111.09374	0.02625
Restaurant	41	99.19598	0.03065
Life	39	95.0371	0.02761
House	38	105.69581	0.02761
Culture	37	88.79877	0.0249
Project	34	82.56045	0.0249
Society	34	82.56045	0.02795
Usadan village	30	89.29343	0.02185
Merchant	27	85.6054	0.02151
World	27	76.92106	0.02287
Gentrification	27	74.25238	0.02355
Progress	25	75.75643	0.02151
Rearrangement	24	71.33878	0.01847
Exploitation	23	70.47831	0.02185
Street	23	68.77383	0.01982
Itaewon-ro	22	72.11069	0.01779
Redevelopment	21	69.3571	0.01813
Merchants Association	20	65.2002	0.01881

).

In addition, the major weighted keywords of TFIDF show that Cities (159.49979) had the highest weight, followed by Space (151.32899), Market (130.62735), Cafe (120.86908), People (116.0029), etc.

Figure 2 visualizes the semantic network between key keywords for the first phase of gentrification. First, the key keywords with the highest connection-centricity were Cities (0.04976), Spaces (0.03723), People (0.03520), and Cafe (0.03148). The keywords with the lowest connection density and the outermost network were Workshop (0.00947), Deterioration (0.01015), Program/Merchants Association (0.01049), and Refurbishment/Redevelopment/Neighborhood (0.01083). Keywords that act as mediators were Art (0.02335), Artist (0.02301), and Culture/Project (0.01726).

4.1.2. Gentrification Stage 2 (2015~2017)

The second phase of gentrification is from 2015 to 2016. The change of place in Itaewon began to be explained as a gentrification phenomenon around 2015, and the mass media rushed to pour out negative reports about gentrification [78]. The gist of the report is that the gentrifier, the party who brought change to the alleyways in the Itaewon area, is rather expelled from the space created by him because of the market logic [79,80].

During this period, a total of 257 articles related to Itaewon gentrification were presented. The most mentioned keyword was Gentrification (714 times), followed by Kyungridan-gil (579 times), Rent (572 times), Legal System (380 times), Region (361 times), People (338 times), Landlords (325 times), Resident (289 times), Cities (245 times), Tenant (214 times), Shop (197 times), Building (181 times), Solution (159 times), Restaurant (146 times), Business District (124 times), House (105 times), Space (99 times), Village Community (95 times), Merchant (85 times), Business (84 times), Development (74 times), Neighborhood (72 times), Artist (59 times), Cafe (47 times), Usadan-ro (46 times), Itaewon-ro (43 times), and Tourists (36 times). Other keywords were mentioned less than 10 times, including Impression, Real Estate, Alley, and Capital (

Table 3. Keyword frequency, TF-IDF, and connection centrality for gentrification phase 2.

Keyword	Frequency	TF-IDF	Degree Centrality
Gentrification	714	1320.42446	0.06768
Kyungridan-gil	579	1337.84999	0.06601
Rent	572	1328.14593	0.05047
Legal System	380	2191.35427	0.02416
Region	361	1162.13754	0.0519
People	338	1057.52408	0.04917
Landlords	325	1015.81554	0.03883
Resident	289	848.32343	0.03611
Cities	245	941.40322	0.03549
Tenant	214	873.28925	0.03264
Shop	197	875.55199	0.03382
Building	181	800.78084	0.03153
Solution	159	752.4419	0.03085
Restaurant	146	789.09596	0.03128
Business District	124	736.34226	0.02664
House	105	715.49348	0.03153
Space	99	700.58404	0.03079
Village Community	95	732.8817	0.02806
Merchant	85	686.92663	0.02305
Business	84	663.31044	0.02856
Development	74	638.46138	0.02961
Neighborhood	72	672.18338	0.02751
Artist	59	637.87564	0.02472
Cafe	47	621.79934	0.02497
Usadan-ro	46	647.253	0.02113
Itaewon-ro	43	618.92086	0.02119
Tourists	36	590.58873	0.02509
Impression	9	551.39004	0.01661
Real Estate	8	553.16989	0.02206
Alley	6	609.35231	0.0207

).

In addition, when looking at the key keywords with high weight of TFIDF, Legal System (2191.35427) was the highest. The next highest were Kyungridan-gil (1337.84999), Rent (1328.14593), Gentrification (1320.42446), Region (1162.13754), People (1047.74408), Cities (941.40322), Shops (875.55199), Tenants (873.28925), etc.

This stage can be said to be the stage leading to commercialization as Gyeongridan-gil became famous and the influx of floating population and tourists was frequent. As the rent rises and the residents move to another place, it is crowded during the day, but at night it becomes an unpopular neighborhood, and the vitality of the city slowly disappears. Therefore, in the third stage, it can be seen that the keywords “Legal System”, “Kyungridan-gil”, and “Rent” are significant.

Figure 3 visualizes the semantic network between key keywords for the second phase of gentrification. First, the core keywords with the highest connection-centricity were gentrification (0.06004), Kyungridan-gil (0.05837), Region (0.04426), Rent (0.04283), People (0.04153), Landlords (0.03119), and Residents (0.02847). The keywords with the lowest connection density and the outermost network were Impression (0.00897), Alley (0.01306), Usadan-ro (0.01349), Itaewon-ro (0.01355), Real Estate (0.01442), Merchant (0.01541), Legal System (0.01652), Artist (0.01708), Cafes (0.01733), Tourists (0.01745), and Business District (0.01900). Keywords that act as mediators between the center and the edge of the network were Solution (0.02321), Buildings (0.02389), and Tenants (0.02500).

4.2. Degentrification

4.2.1. Degentrification Stage 1 (2018–2019)

The first phase of degentrification in the Itaewon area is from 2018 to 2019. This period is also the maturity stage of gentrification, and conflicts between stakeholders are prominent due to the increase in rent. Accordingly, key keywords such as Win–Win Agreements, Commercial Rental Protection Law, and Ordinances appear in this period [80]. In 2018, Itaewon was the only place in Seoul with a vacancy rate of 30% or higher for small shopping districts. This is mostly because the U.S. Forces Korea and the United Nations Command relocated from Yongsan in 2018. The excessive increase in rents of building owners accelerated gentrification and reduced vitality.

A total of 181 articles related to Itaewon gentrification were published during this period. The most mentioned keyword is overwhelmingly Gentrification (674 times), followed by Rent (541 times), Policy System (272 times), Urban Regeneration (187 times), Conservation (184 times), Region (167 times), Mall Building (111 times), Win–Win Agreement (101 times), Landlords (100 times), Tenants (95 times), Seoul (86 times), Residents (75 times), Self-Employed (71 times), Business District (69 times), Increase (68 times), Solutions (53 times), Development (52 times), Merchant (49 times), Cites (47 times), Kyungridan-gil (46 times), Space (45 times), Shopping Mall Lease Protection Law (44 times), Citizen (43 times), Restaurant (42 times), Contract Request for Renewal (39 times). Other keywords, such as Ordinance, Franchise, Government, Periphery, Prevention, are referred to at a frequency of 20 times or less (

Table 4. Keyword frequency, TF-IDF, connection centrality for degentrification phase 1.

Keyword	Frequency	TF-IDF	Degree Centrality
Gentrification	674	729.16495	0.11275
Rent	541	714.71068	0.06702
Policy System	272	1571.44059	0.03173
Urban Regeneration	187	1022.1241	0.04366
Conservation	184	915.89354	0.03124
Region	167	766.69672	0.04679
Mall Building	111	798.40457	0.04391
Win–Win Agreement	101	869.82454	0.03741
Landlord	100	761.3875	0.04251
Tenants	95	769.48601	0.03354
Seoul	86	753.36017	0.04448
Residents	75	697.81563	0.03774
Self-Employed	71	688.85683	0.0407
Business District	69	725.87493	0.03667
Increase	68	631.52053	0.02524
Solutions	53	595.49895	0.03231
Development	52	616.97693	0.03535
Merchant	49	587.86474	0.02754
Cites	47	668.28026	0.03486
Kyungridan-gil	46	650.88591	0.0277
Space	45	628.52175	0.03116
Shopping Mall Lease Protection Law	44	555.49266	0.02318
Citizen	43	543.97974	0.03108
Restaurant	42	534.91906	0.02787
Contract Request for Renewal	39	558.70392	0.02047
Ordinance	14	529.67429	0.02195
Franchise	9	505.59233	0.02326
Government	8	445.82139	0.02466
Periphery	7	445.77608	0.02376
Prevention	6	410.91671	0.02063

).

In addition, when looking at the key keywords with high weight of TFIDF, Policy System (1571.44059) was the highest, followed by Urban Regeneration (1022.1241), Conservation (906.11354), Win–Win Agreement (869.82454), Mall Building (798.40457), Tenant (769.48601), Landlord (761.3875), and Seoul (753.36017).

South Korea has been promoting urban regeneration policies as part of urban renewal and revitalization of alleyways, which also affects the Itaewon area. Accordingly, if gentrification in the extension of urban regeneration proceeds, the area will become livelier, the average income of the residents will increase, and the residents’ sense of belonging and self-esteem for the area may increase. However, the question is whether the participation of local residents, the main body of urban regeneration, is actually taking place, and the land price rises excessively due to the influx of the middle class, tourists, and the participation of large external capital, which inevitably destroys the existing local ecosystem. The above proves that “Policy System”, “Urban Regeneration”, and “Preservation” were derived as keywords with high weight in TFIDF during this period.

Figure 4 shows the semantic network between key keywords related to the third stage of gentrification. The core keywords with the highest connection-centeredness were Gentrification (0.10511), Rent (0.05938), Region (0.03915), Seoul (0.03684), Mall Building (0.03627), Urban Regeneration (0.03602), and Landlords (0.03487). The keywords with the lowest connection density and the outermost network were Contract Renewal Application (0.01283), Prevention (0.01299), Ordinance (0.01431), Mall Lease Protection Law (0.01554), Franchise (0.01562), and Periphery (0.01612). The keywords that act as mediators between the center and the edge of the network were Business District (0.02903), Win–Win Agreements (0.02977), Development (0.02771), and Solution (0.02467).

4.2.2. Degentrification Stage 2 (2020~30 June 2021)

In the Itaewon commercial district, vacancy increased due to a rise in rent (gentrification). In 2018, the U.S. military base in Yongsan faced a crisis, and in 2020, it faced the worst situation with COVID-19. Entering this period, small business owners and self-employed people demanded diversified and fundamental measures such as tax and rental benefits for them, saying that the difficulties caused by the government measures so far have been aggravating and the economic damage is huge [81].

A total of 42 articles during this period were searched. The most mentioned keyword was Gentrification (272 times), followed by Merchants (218 times), Rent (203 times), COVID-19 (154 times), Vacant (139 times), Young People (119 times), Demise (112 times), Confirmed Case (103 times), Club (99 times), Policy System (48 times), Kyungridan-gil (35 times), Alley (28 times), Yongsan (24 times), and Real Estate (20 times). Other keywords, namely LGBTQ, Real Estate, Gallery, Entrance, Substitution, and Situation, were mentioned fewer than 10 times (

Table 5. Keyword frequency, TF-IDF, connection centrality for degentrification phase 2.

Keyword	Frequency	TF-IDF	Degree Centrality
Gentrification	272	120.82718	0.20406
Merchants	218	140.09776	0.09692
Rent	203	142.56889	0.06121
COVID-19	154	259.48406	0.06121
Vacant	139	195.43662	0.03442
Young People	119	59.71147	0.04781
Demise	112	107.82718	0.03889
Confirmed Case	103	93.01781	0.03889
Club	99	81.94703	0.03442
Policy System	48	183.99034	0.03442
Kyungridan-gil	35	169.57873	0.03442
Alley	28	157.21789	0.03889
Yongsan	24	78.91408	0.02996
Real Estate	20	150.16922	0.03889
LGBTQ	8	49.21775	0.02549
Gallery	7	38.15527	0.02103
Entrance	7	35.21872	0.03442
Substitution	5	29.09983	0.03442
Situation	4	25.11274	0.03442

).

Looking at the key keywords with high weight of TFIDF, COVID-19 (259.48406) showed the highest weight, followed by Vacant (185.65662), Policy System (174.21034), Kyungridan-gil (169.67873), Alley (157.21789), Real Estate (150.16922), Rent (142.56889), Merchants (140.09776), Gentrification (120.82718), Demise (107.82718), Confirmed Case (93.01781), Club (81.94703), and Yongsan (78.91408). Unlike the keyword frequency, it can be seen that the keywords for “COVID-19” and “Vacant” are the most important.

Figure 5 shows the semantic network between key keywords for the fourth stage of gentrification. First, the key keywords with the highest connection-centeredness were Gentrification (0.19642) and Merchants (0.08928). The keywords with the lowest connection density and the outermost of the network are Club/Policy System/Kyungridan-gil (0.01339) and LGBTQ (0.01785). Keywords that play a mediating role between the center and the edge of the network were Rent/COVID-19 (0.05357) and Demise/Confirmed Case/Alley/Real Estate (0.03125).

In this period, merchants complained that gentrification was the root cause of the real threat to the alley economy before COVID-19. Most of the self-employed are still suffering from gentrification and being hit directly by the COVID-19, but still cannot afford the high rent. In fact, a survey found that the most difficult thing for small business owners is rent. According to the results of a survey conducted by the Federation of Small Businesses on the management situation of small business owners after the COVID-19 crisis, 69.9% of the respondents chose “rent” as “the most burdensome of business expenses” [81].

5. Discussion

In this study, the gentrification phase of Itaewon was analyzed by dividing it into a gentrification phase and a degentrification phase. It can be seen that there is a significant correlation between the frequency of media and SNS exposure as a sign of commercialization, popularity, and attractiveness of a region as a sign of gentrification. Based on the analysis, an empirical analysis of the Itaewon area was reviewed to summarize the list of phenomena and corresponding indicators found in the process of commercial gentrification and degentrification (

Table 6. Gentrification and degentrification characteristics and possible indicators of Itaewon area.

			Characteristics	Possible Indicators
Gentrification	Signs of escalation		First appearance in the press in 2001 2009: appearance of foreigners entering shops 2013–1015: article exposure frequency surge	News articles and SNS exposure frequency
	Factors in escalation		Expansion of Itaewon commercial district, Haebangchon urban regeneration area selection, Itaewon Special Tourism Zone Project, Yongsan Park Construction Plan	Creative producer Public project
	Progress in earnest	Rent Entitlement	Rapidly expanding commercial area since 2011 Rent has risen significantly since 2012 (50% increase in rent from 2011 to 2016)	Increase in official land price and rent
		Real estate transaction	The transaction volume of detached houses and apartments has surged since 2013 (160% increase in actual transaction price in Itaewon-dong in 2013–2015) Increase in rental demand (2011~2016)	A surge in real estate transactions Change of store tenants
		Business change	Business increase in 2011~2016	Increase in the number of businesses
		Industry change	Decrease in neighborhood stores Significant increase in general restaurants (2010~2015, 12 times) The trend of expanding the scope of the site to the back road	Decreased in the number of neighborhood stores Increase in restaurants
		Resident population	Decline in the resident population (January 2012–May 2014 decreased by about 480 people)	Decline in the resident population
		etc.	Increase in foot traffic	Increase in foot traffic
Degentrification	Maturity stage of gentrification		Vacancy rate increasing trend Increase in absentee landlords (change to nonresident owners of buildings)	Public policy Increased in vacancy rate Composite/new construction Tourist Franchise
	Demise of gentrification		COVID-19 Order to ban gathering (no business) for all entertainment facilities in Seoul	Rapid decrease in foot traffic Public policy

).

Based on the above analysis results, the characteristics of degentrification are summarized as follows: First of all, the COVID-19 pandemic is likely to be the most important cause of degentrification, which is ultimately the root cause of the rent increase during the previous maturity of gentrification. After all, it is said that degentrification due to the COVID-19 pandemic is possible only in areas where there has been a phenomenon of maximization of gentrification. The pandemic situation would accelerate degentrification, but there is a limit to addressing the economic recession caused by the COVID-19 pandemic in places where there has been no gentrification as “degentrification”. In particular, in the case of Itaewon, in 2020, the so-called “degentrification” aspect was noticeable due to the COVID-19 confirmed cases of visitors to the clubs including for LGBTQ in the area. Some media outlets referred to this club as a “gay club” or described the confirmed person as a homosexual in articles, and speculation about the gender identity of the COVID-19 confirmed person arose.

Second, looking at the analysis results of this study, the first stage of gentrification is a step inducing urban revitalization, and there is a positive aspect of providing a better environment to underdeveloped areas. The period of entering the degentrification stage is also the period when the vacancy rate increases. Looking at the semantic network between key keywords by stage, policy-related keywords to solve gentrification start to appear as central keywords from 2018, when degentrification begins. This is related to the culture-driven gentrification by gentrifiers, and it can be seen that the first and second degentrification periods had negative aspects as an issue about real estate as major keywords such as “rent” and “demise” appeared. New small business owners (tenants), a new cultural class lacking economic capital, are being highlighted as another victim of gentrification due to rising real estate prices and rents brought about by large developers who have begun to enter as the mass media pays attention to the area.

Third, what should be noted is that the target area of gentrification is “transitioned” to the adjacent area. In the case of the Itaewon area, as the gentrification phenomenon accelerated from 2013, and rents went up indefinitely, the vacancy rate in Itaewon increased and the surrounding areas suffered gentrification. This suggests that just as cities show cycles of growth and decline, gentrification also has cycles as an extension of downtown revitalization. In other words, gentrification is transferred in the process of moving to a new space or an adjacent area by displaced residents to the site where gentrification occurred and settling elsewhere, and the gentrifier also loses interest in the area and voluntarily migrates to a new space or adjacent area.

Lastly, the phenomenon of gentrification appeared in a wide variety of ways, as Beauregard [3] illustrates this as a “confusing and complex phenomenon”. The concept of gentrification is now more complex and widespread on a global scale as time and space expand under the inevitable effects of neoliberalism. There is also the peculiarity of South Korea. In South Korea, the term gentrification, which has been disputed, has led to definitions that reflect various characteristics [24]. In response, the Korean National Institute of Korean Language (KMA) proposed the term “being driven out of a nest (둥지내몰림)” in Korean to substitute for the term “gentrification” [78]. Most of the characteristics of gentrification in South Korea are commercialization, in which residential facilities are converted to commercial facilities such as cafes and restaurants. The Korean-style gentrification phenomenon was initially due to the revitalization of the downtown area through the input of public policies such as urban regeneration and commercialization of existing commercial areas and neighboring facilities through rapid capital inflow in the process of urban regeneration. Tenants and residents who contributed to the revitalization of the commercial area were displaced. In addition, the residential area has lost its meaning as a place of life for local residents and has been turned into a tourist attraction for outsiders. Moreover, voluntary migration of gentrifiers and indigenous peoples occurred simultaneously over time. In the end, in order to prevent the decline of the local community and the deterioration of the quality of life of the residents due to the collapse of the local community, the government invests in the urban regeneration project in connection with the leading area in the vicinity of the area.

6. Conclusions

This study attempted to improve the consciousness of issues requiring a clear conceptual approach to degentrification by a case study of the Itaewon area. The Itaewon area has been formed by different authorities and has been organized as an urban space by the effect of the United States, government, and capital. The discourses on the urban space tend to mask the problems of the urban space for the reproduction of capital. Therefore, it is necessary to try to reveal an ideological illusion hidden behind those discourses. Using semantic network analysis, this study analyzed the main keywords and the meaning network difference between main keywords related to gentrification in national newspapers in South Korea. In doing this, what matters is how the keywords are connected, what keywords are located in the center of the overall network, and which keywords play a mediating role between the keywords [82].

As a result, gentrification in the Itaewon area is divided into four stages. The first stage of gentrification (2010~2014) is the initial stage of gentrification, and the main keywords are “Space” and “Life”. Gentrification stage 2 (2015~2017) is the period of commercialization as a gentrification growth stage, and the main keywords are “City”, “Space”, “Art”, “Market”, “Region”, “People”, etc. The first stage of degentrification (2018~2019) is the maturation period of gentrification, and the main keywords are “Legal System”, “Seoul”, “Rental”, etc. The second stage of degentrification (2019~30 June 2020) is the period of the COVID-19 pandemic, and the main keywords are “Gentrification”, “COVID-19”, and “Demise”.

The results have confirmed the existing theoretical frameworks while building a more nuanced definition through operationalizing gentrification and degentrification. As with the etymology of the term, the degentrification phenomenon can only be revealed when the gentrification phenomenon is prominently displayed. Although various factors of gentrification have been studied, in terms of degentrification, in the end, rent was found to be the most important factor even in the COVID-19 pandemic situation.

Meanwhile, since the study site is located in South Korea, there is a limitation to generalizing the results. This should have to be proven by criticism and disproving it through future follow-up studies. The gentrification phenomenon in South Korea has been revisited through the input of public policy such as urban regeneration, and the existing commercial and neighboring facilities were commercialized by the rapid capital entry in the process of urban growth. The tenants and indigenous people who contributed to the revitalization of business were displaced. In addition, housing lost its meaning as a place of life for local residents and became a tourist destination for outsiders. In the process, space has changed and the identity of the area was damaged. Moreover, over time, the spontaneous migration of the gentrifiers and the local residents took place simultaneously. Eventually, in order to prevent the collapse of the local community and the deterioration of the quality of life of the residents, the government began urban regeneration projects.

This study has an implication in that it tried to phenomenologically examine the specific phenomenon of the next stage of gentrification through the term “degentrification”. The point is that when gentrification is used politically for urban regeneration, another policy will be created to alleviate the negative phenomenon of gentrification. Urban regeneration literally means revitalizing the life of a city, and it is not only improving the visible physical environment but also revitalizing the functions of the invisible city as a whole. Rather, deep reflection on whether urban regeneration has induced gentrification and contributed to its decline under the plausible cause of urban regeneration is required.

This study also provides overall implications on how to measure the process of gentrification/degentrification and reflect the results on gentrification policies and programs. The case of the Itaewon area shows that gentrification researchers need to provide a way to reflect empirical results practically in gentrification policy and programs. Given that gentrification and degentrification occur very rapidly and in a complex way, careful interpretation of empirical results should be emphasized by researchers.