5.1. Theoretical Analysis of Designer Allocation
With the rise of the knowledge economy, human capital has begun to replace physical capital in demonstrating its importance in economic growth, drawing both businesses and governments into a “talent war” aimed at securing workers with scarce human capital [
50]. At a time when China’s construction industry is transitioning from quantitative growth to qualitative improvement, designers are key talents who can influence the high-quality development of the construction economy [
51]. Under a perfectly competitive market structure, designers freely relocate to regions with higher marginal productivity in accordance with market forces, attaining Pareto optimality in the designer labor market. Mismatch signifies that designer allocation deviates from this optimal state, suggesting that targeted interventions can rectify the imbalance, thereby boosting construction sector economic output or elevating industrial development quality [
52]. The architectural design market operates in a state of dynamic fluctuation, where designer oversupply and undersupply are common phenomena. Only by accurately identifying the root causes, types, and industrial impact levels of designer mismatch can we “prescribe the right remedy” and propose reasonable corrective strategies. Through extended analysis of previous findings, we categorize designer mismatch into three theoretical types: cyclical mismatch, structural mismatch, and frictional mismatch.
Cyclical mismatch refers to designer misallocation caused by fluctuations in the construction economic cycle, characterized by unemployment rates inversely correlating with economic prosperity [
53]. During rapid development phases of the construction industry, total demand for designers increases. However, due to the cyclical nature of designer training, supply–demand disconnection leads to mismatch where designer supply falls short. In recessionary periods of the construction industry, total demand for designers declines, yet the market retains a surplus of designers who face difficulties transitioning careers, resulting in oversupply mismatch. Currently, the growing divergence in construction industry development across provinces in China has created significant disparities in total designer demand generated by construction economies, with simultaneous growth and decline further exacerbating spatial mismatch characteristics where designer supply exhibits coexisting surpluses and shortages. Provinces such as Neimenggu, Shandong, Henan, Yunnan, Chongqing, and Sichuan have long experienced a Positive Mismatch in designer supply, indicating that the total demand for designers in these regions has consistently remained unmet. Meanwhile, Beijing, Shanghai, and Zhejiang have long been in a state of Negative Mismatch, reflecting weak total demand for designers. In contrast, the market development in other provinces is better, resulting in a serious spillover of designer space. Overall, cyclical designer mismatch is rooted in the macro-development environment of the construction economy, focusing more on the alignment between the total supply and total demand of designers. The key to addressing cyclical designer mismatch lies in counter-cyclical interventions in the total demand of the architectural design market. Based on the specific manifestations of designer mismatch, policy measures should target the development trends of local economies, particularly in sectors like construction, real estate, and civil engineering. When designer supply is insufficient, priority should be given to the possibility of an overheated construction economy; conversely, incentives for the development of the construction industry should be strengthened.
Structural mismatch refers to the supply–demand imbalance in designers’ skills and specialties caused by industrial restructuring in the construction industry, representing inevitable growing pains during industrial upgrading [
54]. For instance, with the advent of the intelligent construction era, digital and smart design has become a new trend in the development of the construction industry. A plethora of new tools and technologies, such as IBM and prefabricated building design, continue to emerge [
55,
56]. Many traditional designers, constrained by factors like age and education, are unfamiliar with these new tools and technologies, and their mastered design skills have become obsolete [
57]. As the construction industry transitions toward green and intelligent construction, traditional designers face unemployment. Influenced by developmental stages, regional cultures, industrial policies, and other factors, the growth of prefabricated buildings, green buildings, and intelligent construction in China remains uneven [
58]. The root of structural mismatch lies in the disconnection between the supply of designers’ skills and the demand for architectural design positions, leading to “employment difficulties” in traditional design fields and “labor shortages” in emerging design sectors [
59]. The coexistence of “employment difficulties” and “labor shortages” may occur within the same province or across different provinces, depending on the developmental divergence between traditional and emerging design markets. Jilin, Heilongjiang, Hainan, Xizang, Gansu, Qinghai, Ningxia, and other declining or underdeveloped regions in the northwest, northeast, and southwest border areas have long been in a state of Positive Matching, indicating that due to economic conditions, geographical factors, and quality of life, it is difficult to attract designers from emerging fields to settle down, likely exacerbating structural mismatches in local design markets. In contrast, Tianjin exhibits Negative Matching. As a major first-tier city in close proximity to Beijing, Tianjin exhibits strong appeal to emerging designers in terms of its economic environment, geographical location, and quality of life. This has led to a concentration of highly skilled designers in Tianjin, thereby avoiding the intense internal competition among designers in Beijing. Overall, the structural designer mismatch originates from the microeconomic structure of the construction industry, with a greater emphasis on the alignment between supply and demand for designers in both traditional and emerging sectors. Addressing the structural designer mismatch primarily hinges on the transformation of talent cultivation, guided by the evolving needs of the construction industry’s transition. Schools and enterprises must collaborate to establish a new designer training system.
Frictional mismatch represents a temporary state arising from designers’ job transitions. Owing to information asymmetry in the designer supply–demand market and mobility costs, job-seeking designers struggle to promptly access matching position information, while companies also require time to screen suitable candidates [
60,
61]. Meanwhile, time and economic costs associated with geographic relocation, skill certification, and contract termination slow down the matching process between architectural designers and design firms [
62]. For instance, regions like Hebei, Shanxi, Jiangxi, and Fujian exhibit designer mismatch fluctuations between Positive Matching and Negative Matching, with no state persisting long-term, indicating they are in a benign frictional designer mismatch state. Their architectural design talent market can self-adjust and correct mismatch conditions without human intervention. It is worth noting that Jiangsu transitioned from Positive Mismatch to Negative Mismatch during fluctuations, Hubei shifted from Positive Matching to Negative Matching, and Guizhou changed from Negative Matching to Positive Matching. Their frictional designer mismatch conditions may be non-benign. The architectural design talent market is unable to correct the mismatch of designers through self-regulatory mechanisms, making human intervention measures urgently necessary. Overall, frictional designer mismatch focuses more on whether the mismatch is temporary or long-term, which can be caused by macro-environmental fluctuations or micro-level industry changes. The key to resolving frictional designer mismatch lies in reducing transaction costs in the architectural design market, such as improving employment and entrepreneurship information platforms for designers, providing re-employment and entrepreneurship education for designers [
63], and correcting human capital mismatches caused by incomplete market information in the design industry.
It should be pointed out that both cyclical and structural designer mismatches have significant negative effects on the high-quality development of the construction economy. In contrast, frictional designer mismatch exhibits duality. Similar to cyclical and structural designer mismatches, excessive frictional designer mismatch may lead to designer wastage or shortages, thereby threatening the development of the construction economy. The difference is that moderate frictional designer mismatch reflects the vitality of the designer market, indicating that designers are pursuing better matches with design firms and careers. For example, when frictional mismatch manifests as a slight surplus of designers, it compels designers to proactively transform or engage in self-directed learning, thereby enhancing their design capabilities, and it will drive business upgrades for design firms, further leveraging the transformation of the local architectural design industry and market. Another example is when frictional mismatch appears as a slight shortage of designers. In such cases, design firms incentivize designers to expand their skill sets and enhance the diversity and complexity of their skills through rewards, compensation, professional titles, and promotions [
64]. These measures similarly drive business upgrades for design firms and catalyze the expansion of the design industry and market. Academia holds a negative stance towards human capital mismatch, believing that designer mismatch hinders the development of the construction economy. In contrast, this paper argues for embracing frictional designer mismatch to facilitate a smooth transition during the transformation and upgrading of the construction industry and design sector.
5.3. Robustness Check and Limitation Analysis
Threshold selection and outlier handling often exert a significant impact on analytical results. Given the presence of an outlier in Xizang in 2015, we incorporate robustness analysis here. This study’s robustness analysis focuses on two aspects: First, verification of the authenticity of the outlier data, i.e., whether the 2015 outlier in Xizang constitutes a data error, with a focus on analyzing whether its existence is consistent with the objective reality of Xizang’s construction industry and economic and social development. Second, an impact check of the outlier on the analytical results in both temporal and spatial dimensions, i.e., examining whether this extreme value significantly affected the data analysis results of adjacent years of 2015 or other provinces and cities in the same year, covering four aspects: the geographical pattern, change trends, mismatch types, and mismatch contribution degree of architectural designer supply. The robustness analysis method adopted in this study is to exclude the outlier of “the number of architectural designers in Xizang in 2015”, recalculate the data analysis results of the four core statistical indicators, and compare the differences between the two sets of results (“with outlier” and “without outlier”). The judgment criterion is if the clustering structure of the two sets of results remains unchanged and the rank changes are minor (generally no more than 3), it indicates that the outlier has little impact on the conclusions, and the conclusions are robust. Conversely, it is necessary to supplement the explanation of “the particularity of Xizang’s 2015 data” and its impact on partial conclusions.
From the perspective of temporal effects, the outlier appeared in 2015, and its potential impact might involve the analytical results of 2013–2014 and 2016–2017. This is because China’s architectural design cycle is generally within 1 year, and the time from design to construction start usually does not exceed 2 years. From the perspective of temporal changes in Xizang’s architectural designers, the changes were continuous in 2013–2014, with an extreme value appearing in 2015, which is highly consistent with the data of 2016–2017. The data showed long-term continuity both before and after the abrupt change, indicating that the 2015 outlier in Xizang is most likely not a data error, requiring further verification by consulting more data on Xizang’s construction industry, social, and economic development. Through comparative analysis, this study concludes that the massive influx of engineering survey and design personnel into Xizang in 2015 was mainly driven by factors such as increased demand for local construction projects and infrastructure development, the implementation of talent introduction policies, and reform and development in the engineering survey and architectural design industry. First, in 2015, Xizang launched numerous water conservancy projects and resettlement programs, such as the Laluo Water Conservancy Project and its supporting irrigation district project, as well as resettlement projects. As the largest investment water conservancy project in Xizang’s water conservancy development history, these initiatives significantly boosted the demand for engineering survey and architectural design talents in Xizang. Second, Xizang introduced its first high-level talent introduction policy in 2015, which greatly enhanced the attractiveness to engineering survey and architectural design talents. In that year alone, Xizang attracted over 6000 graduates from other provinces in fields related to engineering technology, design, and construction. Third, Xizang initiated reforms in the bidding system for engineering survey and architectural design, strengthening the competitiveness of the engineering survey and architectural design market, thereby promoting the mobility of designers, particularly their inflow into Xizang. In summary, combined with the characteristics of data changes and the analysis of the underlying social and economic environment, the 2015 outlier in Xizang is a normal phenomenon rather than a data error.
Table 1 shows the ranking of China’s designer supply: Xizang ranked last (31st) nationwide from 2013 to 2016, with Qinghai ranking 30th. In 2017, the rankings of Xizang and Qinghai swapped (Xizang 30th and Qinghai 31st), with a rank change of 1 (less than 3), which did not affect the spatial clustering results. Therefore, it can be clearly judged that the 2015 extreme value in Xizang did not affect the geographical pattern characteristics of China’s designer supply, and its impact on the analytical results can be excluded.
Table 2 shows the ranking of the change rate of China’s designers: Xizang’s national rankings from 2013 to 2017 were 31st, 13th, 1st, 7th, and 9th, respectively (with rank changes much greater than 3), indicating that the outlier significantly affected the spatial clustering results of China’s designers’ change trends. After excluding Xizang’s 2015 data, the analysis results of China’s designer mismatch types remained unchanged, demonstrating the robustness and validity of the analysis. After excluding Xizang’s 2015 data, the values of China’s designer mismatch contribution degree changed significantly (all decreased), but their rankings were slightly affected, with only Guizhou and Qinghai swapping positions (their national rankings were 27th and 28th, with a rank change of 1 < 3), which had no impact on the spatial clustering structure (
Table 8). In summary, the 2015 outlier in Xizang had almost no impact on the geographical pattern and mismatch types of China’s designers. It significantly affected the values of mismatch contribution degree but had little impact on their rankings. Notably, the 2015 outlier exerted a significant impact on the spatial structure of designers’ change trends, affecting the robustness of data analysis results. Fortunately, the analysis results of designers’ change trends are relatively independent of those of mismatch types and mismatch contribution degree, without interfering with each other, thus not affecting the core analysis results of this study.
Therefore, it is ultimately determined that the 2015 outlier in Xizang is not a data error, as it is consistent with Xizang’s construction industry and economic and social development environment at that time. Meanwhile, this outlier had almost no impact on the analysis results of the geographical pattern, mismatch types, and mismatch contribution degree of China’s designer supply, ensuring the robustness of the analysis conclusions. It is worth noting that the outlier significantly affected the analysis results of designers’ change trends, failing to maintain robustness, but it did not impact the analysis results of other core statistical indicators in this paper, thus still ensuring the validity of most key research conclusions. In future studies, great attention should be paid to data outliers and their impact on threshold selection and the robustness of analysis results. The mean is highly affected by outliers and may not be the optimal threshold; considering the median or other criteria based on data characteristics may be more appropriate.
The analysis of the spatial mismatch of architectural designers falls within the research scope of talent geography. Currently, international talent geography is exhibiting a core trend of shifting from broad-spectrum talent research to in-depth exploration of specific talent types [
82,
83]. The spatial allocation and mobility patterns of talents in specific fields—such as technology-oriented [
84], highly educated [
85], high-level [
86], medical [
87], PhD [
88,
89], high -skilled [
90], and overseas [
91] talents—have become research hotspots. Against this backdrop, conducting research on the spatial mismatch of architectural designers provides new cases for understanding the formation mechanisms of spatial mismatch among “industry-specific talents” and offers Chinese empirical support for the “industry-talent adaptation” theory in international talent geography, thus holding significant innovative value. On the one hand, architectural design talents possess both technical attributes and creative characteristics, and are deeply tied to the development of regional construction industries. Their spatial mismatch has not yet been systematically analyzed. This study incorporates architectural design talents into the analytical framework of talent geography, expanding the scope of its research on segmented talents [
92]. On the other hand, international talent geography research is increasingly emphasizing the spatial adaptability between “talent supply and industrial demand.” By focusing on the mismatch between designer supply and construction industry demand, this study accurately responds to the cutting-edge topic of “talent-industry” spatial interaction, addressing the gap in existing research regarding insufficient attention to the engineering design field [
93]. The paper’s most notable innovation resides in developing a technical framework that integrates “designer mismatch characteristics–influencing factors–design talent cultivation strategies”. By systematically applying the spatial mismatch model and Geodetector, this study quantitatively assesses the relationships and magnitudes of designer mismatch in China, empirically examines its influencing factors, and further proposes forward-looking design talent cultivation strategies. These findings offer practical references for designer management and talent development in China and comparable countries globally.
It is noted that this study also has limitations. The rationality of designer allocation requires time for verification. While dynamic mismatch models are more accurate than static ones, constrained by the researcher’s capabilities and limitations in data and methodology, this paper adopts a static mismatch model. Additionally, the influencing factors of designer mismatch are exceptionally complex and significantly influenced by government policy interventions. However, constrained by data availability, this analysis did not incorporate policy factors as influencing variables, warranting further in-depth exploration in future research. Furthermore, empirical research on the microdata of firm-level employment or vacancy designers can further enhance the depth of analysis. However, due to the absence of relevant official data, our study failed to extend the analysis scale to the micro level. For instance, under China’s provincial administrative divisions, there are cities and counties. Empirical analyses that directly use provincial aggregate data will inevitably obscure spatial heterogeneity at the city and county scales. Therefore, from the perspective of enhancing the precision of analytical results, empirical analyses at finer scales (i.e., cities and counties) or empirical studies based on disaggregated provincial data are both desirable. In future research, we look forward to collaborating with interested scholars to obtain firm-level employment or vacancy data of designers through corporate surveys or internet big data scraping, and analyze the mismatch mechanism at a more refined level.