*3.3. The Contribution of Mexico, the United States, and China to the Final Global Demand for Textiles and Apparel*

The previous section analyzed the evolution of the relative importance of Mexico, the United States, and China as the origin of the value added incorporated in their respective gross exports of textiles. However, for a better understanding of its implications, it is also interesting to study their importance in terms of their contribution to the final global demand for textile products.

This analysis shows the greater importance of China with a rising trend between 1995 and 2015 (from 6.2% to 38.3%), declining since then and contrasting notably with the declining share of the United States and Mexico, both with a share below 5% of global demand in 2018 (Figure 6).

**Figure 6.** Participation of China, Mexico, and the United States as origin of the global final demand value added in the textile and apparel industry (%). 1995–2018 Source: Authors based on TiVA (OECD 2021).

Moreover, the comparison between the share in final global demand and the share in world exports shows opposite trends. The Mexican and the U.S. value-added loses relative importance, while value-added originating in China is becoming increasingly relevant, both at the export level and from the final global demand for textile products.

These results are in line with the general hypothesis that there is a process of consolidation of China's hegemony in the global textile market to the detriment of the participation of the United States and Mexico. Furthermore, these findings show that this process is being mainly conducted through the flows of value added in GVCs.

#### *3.4. Econometric Results*

This section presents the econometric estimations, first, for the U.S., Mexico, and China, and second, for the global case. These analyses aim to explore the influence of variables such as FDI, textile tariffs, and textile wages on GVC participation. Concerning participation, as indicated in the methodological section, a distinction is made between total and backward participation in GVCs.

As discussed in Section 2, two econometric models have been estimated to capture some of the determinants of total participation (TPART) and backward participation (BPART). The independent variables selected are the tariff rates applied to the textile sector, where TARIF1 refers to textile raw materials (model I) and TARIF2 refers to textile products (model II). The ratio of inward FDI stock to GDP (FDI) and labor costs to value added in the textile industry (LABC). The main descriptive statistics for these variables are shown in Table 1.


**Table 1.** Main statistics of the variables included in the model.

Source: Author's elaboration based on information from OECD and UNCTAD.

The following subsections present the estimation results of the two models specified in Section 2. Section 3.4.1 presents the results for the group of countries formed by the United States, Mexico, and China, while Section 3.4.2 presents the results for the complete sample of 61 countries, adopting a global approach.

#### 3.4.1. The Case of Mexico, the U.S., and China

Table 2 presents the estimation results for the specific case of Mexico, the U.S., and China. For each of the two models, there are two dependent variables: total (TTPART) and backward (BPART) GVC participation. The results for tariff variables are, apparently, counterintuitive, revealing a positive impact on GVC participation rates. However, this apparent contradiction may be justified by the heterogeneous nature and different profiles of international insertion of these three countries: Mexico, with higher backward participation in the textile and apparel GVC, and higher tariffs affecting textile manufacturing than the U.S. and China. Meanwhile, lower labor costs in Mexican textiles may be counterbalancing their higher tariffs. Regarding FDI, the results are in line with the positive effect expected for this variable, which reveals a positive and significant effect on backward participation.

**Table 2.** Results of the econometric estimation (time fixed effects) for total (TPART) and backward (BPART) GVC participation. Period: 1995–2018. Countries: 3 (China, Mexico, and United States) Number of observations: 72.


Source: Author's elaboration based on information from OECD and UNCTAD Note: The standard error is indicated in brackets. "\*\*\*", "\*\*" and "\*" indicate significance at 0.01%, 0.05% and 0.1%, respectively.

#### 3.4.2. The Global Case

This subsection extends the analysis carried out in the previous point to a broader set of countries. In particular, a total of 61 countries are considered, grouped into developed and developing countries, thus, capturing different participation patterns, according to income level. Tables 3 and 4 present the main results of the estimation of the two models presented in expression (2) for three groups: full sample, developed economies, and developing economies (see Appendix B). It is of particular interest to consider both groups individually as the effects of FDI may vary on the performance of the textile industry, which is generally considered a central sector in the industrialization process of developing economies (Raei et al. 2019). As in Table 2, there are two cases for each model, considering total (Table 3) and backward (Table 4) GVC participation.

**Table 3.** Results of the econometric estimation (time fixed effects) for total GVC participation (TPART). Period: 1995–2018. Countries: 61. Number of observations: 1464.


Source: Author's elaboration based on information from OECD and UNCTAD Note: The standard error is indicated in brackets. "\*\*\*", and "\*\*" indicate significance at 0.01%, and 0.05%, respectively.



Source: Author's elaboration based on information from OECD and UNCTAD Note: The standard error is indicated in brackets. "\*\*\*" and "\*" indicate significance at 0.01% and 0.1%, respectively.

According to the results presented in Tables 3 and 4, the tariff protection level is revealed as a key variable in the participation degree in GVCs, regardless of whether total or backward participation is considered a dependent variable. In the full sample models, for developing and developed countries in the case of backward participation, a higher level of tariff protection would lead to lower GVC participation in textiles through a trade-reducing effect concerning the economic costs associated with export and import flows.

This result is particularly relevant considering that the textile industry, especially in developing countries, is sensitive to internal and external cost changes. However, in the case of the developed countries presented in Table 3, the results suggest that higher tariff protection on textile products (but not on raw materials) would increase total GVC participation. These results may make sense insofar as the participation of developed countries in the textile value chain is more linked to design and branding tasks and less to the manufacture of textile products. Cost sensitivity does not only stand out among developing countries. As Lawless and Morgenroth (2019) point out, Brexit is an excellent example of how a change in trade tariffs can affect cost-sensitive sectors, such as food and textiles.

Something similar happens with the other variables incorporated in the model. In the case of FDI, it is generally significant in all cases, while labor costs are significant in the case of developing countries. Regarding FDI, a greater inflow would imply higher participation in GVC in the case of developing countries; however, in developed economies, a greater inflow of FDI would have the opposite effect. In this regard, it should be noted that developed countries are, in many cases, the final markets for textile products. In this sense, an increase in FDI in these countries could well reduce the global textile insertion of developed economies and strengthen their domestic market. Finally, it should be noted that when analyzing the influence of FDI on the total GVC participation without discriminating by sector of activity, the literature supports a significant and positive relationship (Martínez-Galán and Fontoura 2019; Okah Efogo et al. 2022).

Labor costs are significant, especially in developing countries, because their insertion in the textile sector occurs mainly through a cost strategy. Thus, a higher labor cost would imply lower participation in GVCs (Fukase 2013; Javed and Atif 2021; McCaig 2011). However, regarding developed countries, labor costs are not a useful variable to explain the insertion of the textile sector in GVCs since, in this group of countries, the industry does not depend on its competitiveness in terms of costs, but rather on its innovative capacity, design, and brand prestige (Padilha and Gomes 2016; Vila and Kuster 2007).

#### **4. Conclusions**

This paper is based on the recognition of the important role played by GVC in production processes, because of international fragmentation of production, which particularly affects manufacturing sectors such as the textile and apparel industry. In this sense, different conclusions can be drawn.

From a comparative perspective, the analysis of the GVC participation shows interesting results. China, and especially, Mexico and the United States present high levels of total participation, with different trends. However, from this sectoral approach, it also can be observed an accentuation of general GVC insertion patterns. In this sense, the United States and China show clearer forward participation, while Mexico follows a different pattern, with a predominance of backward participation. This feature is common in countries specializing in manufacturing and assembly tasks.

Another relevant finding is China's growing importance in the origin of the valueadded incorporated in Mexican and U.S. textile exports. This result shows a steady increase in the backward linkages of Factory North America with the Asian giant. The contrary occurs in Mexico and especially the United States, which currently have residual importance as the origin of the value-added of China's textile exports and have even seen this importance decrease in the period of analysis (1995–2018).

These results are consistent with the characterization of the decline of Mexican and U.S. textile manufacturing, due to the presence of fragmented global trade flows not positively exploited. However, this contrasts with the innovative and competitive Asian textile and apparel industry, which adopts the full package strategy and FDI in a positive sense to increase its value-added with empowerment in productive fragmentation.

The analysis of the participation in the final global demand for textile products provides additional findings. The most remarkable result is that the Chinese share shows a clear and steady increase, approaching 40% of global demand. However, Mexico and the

United States show the opposite trend, becoming nearly residual. This contrasts with their respective shares in the textile GVC, where there are no significant differences in the levels of participation.

From an extended (61 countries) and explanatory perspective, the econometric analysis reveals tariff rates as a key factor for textile GVC participation, particularly in developing economies. However, the effect of other variables, such as FDI stock and labor costs, depends, to a large extent, on the group of countries considered. While among developed economies, FDI does not help to provide a simple understanding of their insertion in textile GVCs which, in developing economies, is a significant factor. This may be due to the type of tasks performed by each country. While the first group focuses on tasks such as branding and design, the developing countries mainly perform manufacturing tasks.

Although labor costs only reveal a significant and negative effect on backward participation for the full sample, this variable shows a negative impact mainly for developing countries, which is consistent with the expected influence of this factor. The counterintuitive result obtained for this variable for developed economies can be explained because the total GVC participation of these countries is driven by high labor costs (e.g., offshoring of manufacturing tasks to other countries), which causes a positive (and significant) effect on their total GVC participation. In contrast, the lower level of labor costs positively influences the GVC participation of developing economies.

Finally, this research highlights the profound changes the textile sector has undergone in a highly competitive and globalized context. In particular, it allows us to understand the changing role of economies in the international fragmentation of production, where countries follow different insertion patterns, with uneven results.

**Author Contributions:** Conceptualization, Ó.R.-M., H.C.-R. and A.L.G.P.; methodology, H.C.-R. and Ó.R.-M.; formal analysis, A.L.G.P. and H.C.-R.; investigation, A.L.G.P. and Ó.R.-M.; data curation, Ó.R.-M., A.L.G.P. and H.C.-R.; writing—original draft preparation, Ó.R.-M., A.L.G.P. and H.C.-R.; writing—review and editing, Ó.R.-M., A.L.G.P. and H.C.-R.; visualization, H.C.-R.; project administration, Ó.R.-M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research has been supported by the ICEDE research group, to which the authors belong, Galician Competitive Research Group financed by Xunta de Galicia (Ref. ED431C 2022/15).

**Data Availability Statement:** All data have been obtained from publicly available databases, such as TiVA (OCDE), UNCTAD, and WTO.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Appendix A**
