Mexico’s semiconductor ecosystem is focused on design and back-end manufacturing. This chapter provides an overview of the structure, regional composition and recent trends of the semiconductor and other electronics industry in Mexico and examines its integration into global value chains. Investment in the Mexican semiconductor industry and the availability of skills are also discussed at length.
Promoting the Development of the Semiconductor Ecosystem in Mexico
Abstract
The semiconductor industry has operated in Mexico since the 1960s. There are currently four main semiconductor firms in operation, one in the design segment (Intel) and three in back-end manufacturing (Infineon Technologies, Skyworks Solutions and Texas Instruments). No firm currently operates in the front-end segment. For statistical purposes, semiconductor manufacturing is aggregated under electronics manufacturing.
Although the broader “Semiconductor and other electronic component manufacturing” industry (hereafter, semiconductor and other electronics)1 presents one of the highest levels of gross production, net profit and value added among high-technology (high-tech) manufacturing activities in Mexico, there is room for improvement in labour productivity given the high prevalence of maquila activities.2
The geographic distribution and economic composition of the semiconductor and other electronics industry in Mexico exhibit significant heterogeneity. The north of the country (Northwest and Northeast regions) stands out with the highest number of firms, followed by the centre (Centre West and Centre regions). Notably, the Centre West region, particularly the state of Jalisco, has seen the highest influx of new firms in the past decade. In contrast, the number of firms in the Southeast is currently very small.
Semiconductor and other electronics firms in the North exhibit characteristics that distinguish them from other regions. Specifically, firms in the North (Northwest and Northeast) tend to be larger in terms of employment, older and rely more heavily on maquila activities for income. Nonetheless, they are also less productive compared to the smaller and younger firms found in the Centre (Centre West and Centre regions). In these regions, there is a lower reliance on income from maquila activities.
Mexico presents a trade deficit for chips and lacks specialisation in this segment or in other segments of the semiconductor supply chain such as raw materials or manufacturing equipment. Additionally, trade dependencies have increased for chips. Ensuring resilient supply chains for chips could mitigate the effect of supply disruptions for the semiconductor and semiconductor-using sectors in Mexico.
The need for acquisition of machinery and equipment remains significant. The relatively low ratio of machinery and equipment per worker may be attributed to suboptimal technology investments and could indicate a reliance on manual labour over automated processes within the semiconductor and other electronics industry. Furthermore, the industry exhibits a low ratio of machinery and equipment to total production, potentially stemming from financial limitations that restrict capital investment and, consequently, leads to fewer assets in relation to production output.
Mexico is the second largest economy in Latin America in gross domestic product (GDP) terms, after Brazil, and has large growth potential (OECD, 2024[1]). As of 2023, Mexico had the sixth highest GDP per capita among Latin American countries, behind Uruguay, Panama, Chile, Costa Rica and Argentina (World Bank, 2025[2]).
After exhibiting solid GDP growth in 2023, Mexico’s GDP grew by 1.5% in 2024 and is expected to have grown by 0.4% in 2025 and by 1.1% in 2026 (OECD, 2025[3]). Low unemployment and declining inflation are expected to support household consumption. However, public consumption and investment will remain modest due to ongoing fiscal consolidation. Exports are likely to be constrained by tariffs, uncertainty and slower global growth.3
Electronic manufacturing firms arrived in Mexico alongside the Maquiladora Export Industry (IME) during the 1960s. That decade also saw the arrival of the semiconductor industry within the electronics sector. Nowadays, Mexico has a presence in the design and back-end (assembly, testing and packaging or ATP, see Annex A) segments of the global semiconductor value chain, but currently does not have front-end (fabrication) semiconductor facilities.
With the current emphasis on diversification in global value chains (GVCs), in particular for semiconductors, Mexico could present a location of interest for the industry. In fact, the recent heightened geopolitical tensions between major economies has already triggered reorganisations of GVCs and the rise of net exports and product offerings from Mexican firms (Utar, Torres Ruiz and Zurita, 2023[4]). Furthermore, Mexico has become the largest source of United States (US) imports, replacing the People’s Republic of China (hereafter “China”) in 2023 (OECD, 2024[1]).
Mexico is home to clusters in high-tech manufacturing sectors, including the automotive and electronics sectors, which are important users of semiconductors. The border with one of the world’s leading semiconductor economies, the United States, and the availability of human capital enhance Mexico’s competitive edge.
2.1.1. Key players in Mexico’s semiconductor ecosystem
Table 2.1 presents an overview of the main firms active in the semiconductor ecosystem in Mexico. The country hosts four important players in the semiconductor GVC: Infineon Technologies, Intel, Skyworks Solutions and Texas Instruments.
The semiconductor ecosystem in Mexico dates back to 1969 when Skyworks Solutions, an integrated device manufacturer (IDM) started operating in Mexicali, Baja California (see Annex A for an overview of IDMs and other business models in the semiconductor industry). Since then, other foreign firms have joined, including Motorola, which was originally involved in front-end manufacturing and operated its discrete semiconductor division (now Onsemi) from 1969 until it relocated its front-end manufacturing to Asia in 1999. Currently, Motorola’s manufacturing activity in Mexico focuses on telecommunication equipment rather than semiconductors. Texas Instruments arrived in 1984, establishing semiconductor back-end manufacturing and packaging operations, followed by Intel in 1992, which launched the Guadalajara Design Center, and finally, Infineon Technologies arrived in 2001.
Notes: “N/A” indicates foreign ownership data are not available.
1. Location of firm within Mexico: all firms are foreign-owned and also operate in foreign economies.
2. The year when the firm started operating in Mexico.
All firms included in this table are private; there are no state-owned firms.
Source: Based on information provided by the Secretaría de Economía during the country consultation.
Beyond the four key players identified in Table 2.1, the semiconductor ecosystem is also composed of young firms, like QSM Semiconductores, a Mexican firm founded in 2022 focused on the design of integrated circuits and legacy chips production, among other services. In addition, Mexico has a cluster of electronic component manufacturing composed of firms like Jabil, Qualcomm and Vishay Intertechnology, listed in Annex B. Recent announcements suggest additional investment in the semiconductor ecosystem with Foxconn announcing a new plant for NVIDIA superchips and QSM intending to build a new plant in 2025 (Pro Mexico Industry, 2024[5]; Reuters, 2024[6]).
2.1.2. Evolution of the computer, electronic and optical equipment sector: Insights from macroeconomic data
Although total production from the electronics sector (International Standard Industrial Classification [ISIC] Division 26)4 has decreased over the past two decades – driven in part by the relocation of front-end manufacturing activities such as Motorola’s move to Asia – value added in the sector has increased in more recent years, from around MXN 200 billion (Mexican pesos) in 2013 to MXN 407 billion in 2020.5 In addition, the sector’s share of total value added in the economy has also been increasing since 2013, reaching 2% in 2022 (Figure 2.1, Panel A).
Note: Panel A shows the share of value added of ISIC Division 26 “Computer electronic and optical manufacturing” (electronics sector) and its share in total value added in constant MXN billion. Deflator: producer price indices (PPIs) for manufacturing, domestic sector (base is July 2019) without oil, from the Instituto Nacional de Estadística y Geografía (INEGI) (OECD, 2024[7]). Panel B shows the fixed-base Laspeyres physical volume index of production (2018=100) for the semiconductor and other electronics industry (North American Industry Classification System [NAICS] 3344), computer and electronic product manufacturing (NAICS 334) and total manufacturing (NAICS 31-33). Solid lines represent definitive values (until December 2023), dashed lines represent revised values (from January 2024 until July 2024) and dotted lines represent preliminary values (since August 2024).
Sources: Based on the OECD National Accounts Questionnaire in OECD (2025[8]), OECD Data Explorer - Annual value added and its components by economic activity, http://data-explorer.oecd.org/s/1zu (Panel A) and INEGI (n.d.[9]), Encuesta Mensual de la Industria Manufacturera 2025, Datos al mes de noviembre (2018=100), https://www.inegi.org.mx/rnm/index.php/catalog/1086 (Panel B, retrieved in June 2025). Please see Box 2.1 for more details.
Output growth (defined as volume of goods produced) in the electronics sector – particularly in the semiconductor and other electronics industry (NAICS 3344) and the broader computer and electronic product manufacturing sector – has outpaced the rest of manufacturing since the COVID-19 pandemic. Production in these two industries increased by approximately 56% and 61% from 2018 to 2025 respectively, compared to a 13% increase for total manufacturing (Figure 2.1, Panel B) in the same period. Both industries experienced a mild drop during the pandemic and recovered quickly, reflecting post‑pandemic shifts in global demand that favoured electronic products.
2.1.3. Evolution of the computer, electronic and optical equipment sector: Insights from sectoral data
Sectoral data provide a better understanding of the market structure and firm dynamics and shed light on important differences in performance across different categories of semiconductor firms, thereby helping policymakers better design and target policy actions. This report builds on aggregated data of the semiconductor and other electronics industry in Mexico to analyse potential bottlenecks and inform policies to facilitate its development (Box 2.1).6
Further disaggregated data would be important to develop the evidence base to help policymakers monitor changes in the dynamics of the semiconductor ecosystem and inform the development of relevant policies. According to the 2023 OECD Digital Government Index, Mexico scores below the OECD average in the “data-driven public sector” dimension, reflecting challenges in access and sharing of data across the public sector (OECD, 2024[10]). Mexico has rich disaggregated data available to different institutions and agencies, which could be leveraged to better understand the semiconductor ecosystem and other parts of the economy. Efforts to enhance access to and sharing of data across the public sector in line with best practices in other OECD countries would support well-informed policymaking for the semiconductor ecosystem, as well as other parts of the economy.
This publication employs data made available by the Secretaría de Economía for the purpose of this report. These sources provide insights into establishments and employees belonging to the semiconductor industry and include:
The Directorio Estadístico Nacional de Unidades Económicas (DENUE) is a public directory that offers identification information, geographical location, size category, economic activity and contact information of 5 million active establishments in Mexico. This is the main source of information on firms in the country. It does not cover agricultural firms covered by the rural census. Available to the OECD Secretariat on an annual basis at the firm level from 2016 to 2023.
The Censo Económico economic census containing information on all establishments producing goods, merchandise traders and service providers. These data make it possible to generate economic indicators for Mexico with a high level of geographical, sectoral and thematic detail. Available to the OECD Secretariat at the sector level for 2019 and 2024.The Encuesta Anual de la Industria Manufacturera is an annual survey providing statistical information on economic outcomes, including wages, employment, investment and total production, covering the manufacturing sector. Available to the OECD Secretariat on an annual basis from 2018 to 2023.
The Encuesta Mensual de la Industria Manufacturera is a monthly survey providing statistical information on labour, wages and production, covering the manufacturing sector. Available to the OECD Secretariat on a monthly basis from 2018 to 2025.
The Encuesta Nacional de Ocupación y Empleo (ENOE) is a labour force survey providing statistical information on workers, including their occupation, education level or economic activity, facilitating the analysis of employment trends and sociodemographic changes. Available to the OECD Secretariat on a quarterly basis from 2012 to 2023.
Given that only four large firms are active in the semiconductor industry in Mexico (see Filippo et al. (2022[11]) and Table 2.1), the analyses for the semiconductor industry were conducted on the broader NAICS 3344 “Semiconductor and other electronic component manufacturing” industry (hereafter semiconductor and other electronics industry). This aggregation covers the manufacturing of the following activities: i) bare printed circuit boards; ii) semiconductor and related devices; iii) capacitors, resistors, coils, transformers and other inductors; iv) electronic connectors; v) printed circuit assembly (electronic assembly); and vi) other electronic components.
Notes: The OECD worked with the Secretaría de Economía to obtain aggregate statistics from the INEGI. The number of establishments in the NAICS 3344 “Semiconductor and other electronic component manufacturing” category differs by source: for example, the economic census listed 444 establishments in the sector in 2019 (the census was carried out in 2018), whereas DENUE listed 482 in in 2018. In this document, due to data availability in ENOE, the NAICS code used to identify the semiconductor sector is 334, corresponding to the computer and electronic product manufacturing sector. This is a higher-level aggregation including NAICS 3344.
Source: Filippo, A. et al. (2022[11]), “México y la cadena de valor de los semiconductores: oportunidades de cara al nuevo escenario global, México y la cadena de valor de los semiconductores: oportunidades de cara al nuevo escenario global”, https://doi.org/10.18235/0004276.
Establishment composition by age and size
The Mexican semiconductor and other electronics industry saw a steady increase in the number of establishments operating until the COVID-19 pandemic. The number of establishments increased from 450 to 609 between 2016 and 2019 (Figure 2.2). With the onset of the pandemic, there was a significant drop in firm creation and a gross decrease of 100 establishments. Afterwards, the sector began a slight recovery and reached 515 establishments by 2023.
Prior to the COVID-19 pandemic, 61% of establishments within the sector were categorised as young, defined as being less than five years old. However, the pandemic had a profound impact on this segment of establishments, leading to a notable decrease in their share. As of 2023, the share of young establishments within the sector has declined to 38% (Figure 2.2).
Large establishments accounted for 35% of the industry’s total number of establishments in 2015 (Figure 2.3). This share decreased to 30% by 2023, despite an increase in the total number of establishments. This change can be attributed to the rise in the number of micro establishments, which increased from 33% to 38% during this period. Although micro establishments were the most impacted segment during the pandemic, they managed to increase their presence in the sector. Small and medium‑sized establishments hold a similar share, and both these segments account for approximately one-sixth of the firms.
Note: The figure shows the distribution of establishments by age category (based on the first year of incorporation available from DENUE): the first wave is from 2010; there was no unified directory for all establishments [economic units] in Mexico before 2010).
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx. Please see Box 2.1 for more details.
Note: Size groups follow those set by the Secretaría de Comercio y Fomento Industrial (Government of Mexico, 1999[13]). Micro establishments in the manufacturing sector employ less than 30 workers, small establishments between 31 and 100, medium-sized establishments between 101 and 500 and large establishments more than 500 workers. However, considering DENUE’s available size categories, this figure considers the medium-sized segment as establishments with 101-250 workers and large establishments with 251 or more workers.
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx. Please see Box 2.1 for more details.
Geographic distribution
The Mexican states with the largest presence of semiconductor and other electronics establishments are located along the Mexico-United States border, followed by the Bajío states (Centre West region, from Jalisco to Querétaro) and the Metropolitan area of the Valley of Mexico (in the Centre region), including the state of Mexico and the capital Mexico City (Figure 2.4, Panel A).
Note: In Panel A, the intensity of the shade indicates the number of establishments, with 120 being the maximum and 0 the minimum. In Panel B, the intensity of the shade indicates the increase or decrease in the number of semiconductor and other electronics establishments, from ‑2 (negative, minimum) to 38 (maximum).
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx. Please see Box 2.1 for more details.
The Northwest region is primarily known as an industrial hub for manufacturing electronic components used in the telecommunications, aerospace and automotive industries. The Northeast, on the other hand, is characterised by manufacturing the electronic components used in computers, household appliances and other consumer electronics (i.e. televisions, smartphones and audio systems). Meanwhile, the Centre is known for high-tech manufacturing and electronic subassembly. The Centre West region is widely known as the location of Mexico’s “Silicon Valley”, with a focus on the manufacturing of electronic products with world-class firms such as Bosch, Intel and Oracle, among others (Filippo et al., 2022[11]). The list of states that belong to each region are presented in Table 2.2.
Notes: The term region reflects the “mesoregion” definition used by the Secretaría de Economía.
Source: Regionalisation component from Secretaría de Economía (2022[14]), “Rumbo a una Política Industrial”, https://www.gob.mx/cms/uploads/attachment/file/761984/Rumbo_a_una_Pol_tica_Industrial.pdf.
Business dynamics
Firm and regional heterogeneity are the key features of the landscape, and the creation of new establishments further underscores these differences. This subsection examines age and size dynamics among establishments in the semiconductor and other electronics industry, using data from DENUE between 2016 and 2023.
Age
Mexico’s firm dynamics show a weakening in business creation despite positive net entry. Nationwide, Mexico experienced a net entry of 65 establishments between 2016 and 2023, reflecting the exit of 35 young establishments and the entry of 100 new establishments (Figure 2.5). Despite this positive net entry, the rate of new firm creation has declined since the mid-2010s. The slowdown in new entries contributes to an ageing establishment base, raising concerns about the sustainability of this trend.
In 2023, the Northwest region had the highest number of mature establishments in the semiconductor and other electronics industry, with an additional 53 establishments that were more than 5 years old between 2016 and 2023. This is in contrast to the decline in the number of young establishments (54 fewer young establishments) in this region during the same period, likely reflecting an ageing of the existing establishment and lack of market dynamism. Overall, the Northwest region saw a net loss of one establishment during the period.7
The other regions saw a net increase in the number of establishments. For instance, the Centre West presented the highest growth from 2016 to 2023, with 42 new establishments, 17 new establishments in the Northeast and less than 5 new establishments in the Centre and Southeast. For more details on the establishment composition by state see Annex D.
Note: For Panels A to F, young (light blue) establishments are defined as establishments aged 1-5 years and mature (dark blue) as establishments aged 6 or more years, based on the first year of incorporation available from DENUE: the first wave is from 2010; there was no unified directory for all establishments (economic units) in Mexico before 2010. The two columns on the left of each chart represent the number of young and mature establishments in selected years (2016 and 2023). The diamond marker represents the net difference between 2016 and 2023 for all types of establishments.
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx. Please see Box 2.1 for more details.
Size
Most of the large semiconductor and other electronics establishments are located in the North (Figure 2.6). The North accounts for 78% of the large establishments in Mexico, with a nearly equal split between the Northwest and Northeast in 2023. Additionally, these regions account for 40% and 34% of the medium‑sized establishments respectively in the Northwest and Northeast.
Note: Size groups follow those set by the Secretaría de Comercio y Fomento Industrial (Government of Mexico, 1999[13]). Micro establishments in the manufacturing sector employ fewer than 30 workers, small establishments between 31 and 100, medium-sized establishments between 101 and 500 and large establishments more than 500 workers. However, considering DENUE’s available size categories, this figure considers the medium-sized segment as establishments with 101-250 workers and large establishments with 251 or more workers. The columns represent the number of establishments by size. The diamond marker represents the net entry between 2016 and 2023.
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx. Please see Box 2.1 for more details.
In terms of small establishments, the Northwest leads with 39%, followed by the Centre West with 29%. When it comes to micro establishments, the distribution is relatively equal among the main four regions, each accounting for approximately one‑quarter of the total, with the Centre West experiencing significant growth in this category. Overall, the Centre and Centre West regions are predominantly composed of micro establishments.
The Northwest region (defined in Table 2.2) stands out as the primary hub for semiconductors and other electronics in the country, with 178 establishments, accounting for 35% of all active establishments in this industry in 2023 (Figure 2.7). The Northeast region comes second with 142 establishments (28% of the industry’s total). Interestingly, in 2015, both the Centre West and Centre regions had a similar percentage of establishments in this industry; however, the Centre West experienced a twofold increase from 56 to 118 establishments between 2015 and 2023 and now accounts for 23% of all active establishments in this industry, while the Centre’s share remained at 13% in 2023. The Southeast region lags behind with a mere 2%.
Note: In Panel B, the columns represent the number of establishments in each region. Regions and their codes are defined in Table 2.2.
Source: OECD calculations based on DENUE in INEGI (2026[12]), Directorio Estadístico Nacional de Unidades Económicas, https://www.inegi.org.mx/app/mapa/denue/default.aspx, for Panel A and the 2024 economic census in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/, for Panel B. DENUE is constructed on the basis of the Economic Census and is subsequently updated on a continuous basis. As a result, while both DENUE and the Economic Census rely on the same establishment-level concept (economic units), differences in periodicity lead to differences in establishment counts for the periods presented. Please see Box 2.1 for more details.
Based on the economic census, Baja California leads with 105 establishments (Figure 2.7, Panel B), primarily located in, Mexicali, Tecate and Tijuana. This strategic positioning, adjacent to the US border, provides a significant advantage for these cities in terms of access to the US market and GVCs. In fact, Tijuana alone was home to 64 establishments in 2019, accounting for nearly 15% of the country’s total. Sonora and Chihuahua (driven by Ciudad Juarez), both states within the same region, were the second and fourth states with the most establishments, with 61 and 50 respectively.
The Northeast is the second region in the country in terms of establishments. This region, which also shares a border with the United States, is led by Nuevo León (sixth state in the country) followed by Tamaulipas and Coahuila (seventh and tenth respectively).
The presence of the semiconductor and other electronics industry in the Centre West is growing. Jalisco, a prominent state within this region, takes the third spot nationwide with 51 establishments. Notably, 4 municipalities within the Guadalajara metropolitan area were among the top 20 nationwide in 2019, with Zapopán ranking third with 30 establishments. Querétaro, another state within the Centre West region, hosts 37 establishments, a number that has almost doubled since 2019.
The Centre region of Mexico is home to a total of 51 establishments, with a significant presence around Mexico City and the state of Mexico. The Southeast region of the country has the lowest number of establishments, with six in total (Figure 2.7, Panel A).
2.1.4. Performance indicators
This sub-section describes the main performance indicators of the industry using information from the Censo Económico (hereafter economic census) on value added, profitability and labour productivity, including information on income from maquila activities. Furthermore, it offers a comparative analysis between the industry and other high-tech manufacturing activities in the country.8 These indicators are crucial for assessing the overall state of the industry and shedding light on the industry’s standing within the Mexican high-tech manufacturing landscape.
According to data from the economic census, the semiconductor and other electronics industry represents a sizeable share of the country’s high-tech manufacturing activity, 1.3% of total manufacturing production in 2024 (up from 0.8% in 2019, see complete list of economic activities in Annex C) and ranks second among the very high-tech industries (or Level I, dark blue industries in Figure 2.8, Panel A),9 just behind pharmaceuticals.
The semiconductor and other electronics industry ranks first in profitability with almost 2.3% of total manufacturing profits (an increase from 1.6% in 2019), followed by petroleum and coal (2.1%) and basic chemicals (1.9%) (Figure 2.8, Panel B). Within Level I high-tech industries – the highest technological-intensity category – the semiconductor and other electronics industry overtook pharmaceuticals as the most profitable industry in 2024.
The industry’s value added (gross production minus intermediate consumption) grew by 36% between 2019 and 2023, followed closely by the medical devices and pharmaceutical industries (Figure 2.9, Panel A). In the same period, value-added growth in the semiconductor and other electronics industry was slower than the broader computer and electronics sector (NAICS 334) that showed a 40% increase in value added, but faster than overall manufacturing (13%) (Figure 2.9, Panel B).
Note: Profit is measured as the difference between revenues and expenditures from the supply on goods and services (profit proxy). Taxes, labour costs and capital costs are not considered. High-tech manufacturing is classified according to Hecker (2005[16]). The dark blue columns correspond to Level I of high-tech manufacturing (Annex C), whereas Levels II and III are represented in medium blue.
Source: OECD calculations based on the economic census (2019 and 2024) in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
Productivity followed a similar upward pattern, with the semiconductor and other electronics industry and the computer and electronics sector clearly outpacing overall manufacturing. The semiconductor and other electronics industry experienced consistent growth, reaching a 20% productivity increase in the 2019-2023 period (Figure 2.10, Panel A). Nonetheless, the semiconductor and other electronics industry recorded the least marked increase among industries within the broader computer and electronics sector. Other industries, such as computer and peripheral equipment manufacturing (NAICS 3341) and navigational, measuring, electromedical and control instruments (NAICS 3345) saw more pronounced productivity improvements and were key drivers of the overall productivity growth in the sector. These trends point to stronger productivity gains across electronics industries over the broader manufacturing sector since 2019 (Figure 2.10, Panel B).
Note: The figure shows the evolution of the indexed gross value added (gross production minus intermediate consumption) for selected manufacturing industries and sectors in Mexico. Gross value added is expressed in real terms and is obtained by deflating nominal values using industry-specific PPIs, with 2019 as the base year (2019=100).
Source: OECD calculations based on INEGI (n.d.[17]), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, https://www.inegi.org.mx/rnm/index.php/catalog/1061. Please see Box 2.1 for more details.
Note: The figure shows the relative labour productivity for selected high-tech industries and sectors in Mexico. Labour productivity is defined as the ratio of gross value added divided by total employed personnel, whether directly employed or contracted through another entity provided they worked at least one-third of the regular working hours. Gross value added is expressed in real terms and is obtained by deflating nominal values using industry-specific PPIs, with 2019 as the base year (2019=100).
Source: OECD calculations based on INEGI (n.d.[17]), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, https://www.inegi.org.mx/rnm/index.php/catalog/1061. Please see Box 2.1 for more details.
According to the economic census (Figure 2.11, Panel A), labour productivity varies markedly across regions, reflecting their specialisation in different segments of the value chain. The Centre West has the highest labour productivity at MXN 0.66 million per worker, supported by its broader range of activities, including design. The North regions (Northwest and Northeast) host the highest number of firms, yet their labour productivity lags behind that of the Centre West and only above the Centre and Southeast regions, the latter having the smallest industry presence nationwide.
Note: In Panel A, the columns represent the value added per worker. The number of establishments in the industry differs between the economic census (Figure 2.7, Panel A) and DENUE across regions as follows: Northwest (169), Northeast (133), Centre West (119), Centre (51) and Southeast (6). In Panel B, the dark blue columns represent Level I of high-tech manufacturing whereas Levels II and III are represented in medium blue. The statistical aggregate “Semiconductor and other electronics” includes several activities beyond semiconductors, such as electronics assembly and printed circuit boards, based on data reported by the economic census at the NAICS 4-digit industry group level. More granular data focused solely on semiconductor manufacturing are not available.
Source: OECD calculations based on the economic census (2019 and 2024) in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
Labour productivity also varies significantly across industries. Available statistics aggregate semiconductor activities together with other electronics manufacturing, which as a whole show relatively low labour productivity levels (Figure 2.11, Panel B). This reflects the nature of the activities captured by this industry classification, which largely correspond to maquila processes in electronics assembly and in printed circuit board operations, rather than semiconductor manufacturing (see the following subsection on maquila income). Labour productivity in this industry has been increasing, recording the third-largest rise between 2019 and 2024, just behind the basic chemical industry and the electrical equipment industry.
Maquila income
In Mexico, the manufacturing activities with the highest technological intensity earn the most from maquila operations (Figure 2.12). Four out of the top five economic activities in high-tech manufacturing fall into this group. In 2024, maquila operations accounted for a considerable share of the income of the semiconductor and other electronics industry (78%). Other high-tech manufacturing activities such as communication equipment, aerospace products and parts and computer equipment also rely heavily on maquila income, with 75%, 58%, and 51% respectively.
Note: The dark blue columns represent Level I of high-tech manufacturing whereas Levels II and III are represented in medium blue.
Source: OECD calculations based on the economic census (2019 and 2024) in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
2.1.5. Position in the global value chain
This subsection describes the GVC of products needed in the semiconductor ecosystem. The primary data source used in the analysis is the BACI database10 (for years 2012-2023), which enables the monitoring of global trade statistics by product and trading partner at the six-digit level. For more detailed information on the methodology, refer to Box A F.1.
OECD data on trade in value added (TiVA) provide additional insights on the integration of Mexico into GVCs and global production networks. TiVA data consist of a selection of principal indicators that track the origins of value added in exports, imports and final demand for years 1995-2022
Mexico has a significantly higher share of re-exported intermediate imports and a lower domestic value-added share in gross exports compared to other Latin American countries. With a large proportion of electronics inputs sourced from abroad, Mexico could benefit from enhancing the ecosystem of local suppliers through geographic cluster development programmes, innovation hubs and by facilitating networking events to connect local suppliers with existing firms within the semiconductor ecosystem.
To guide this assessment, the analysis begins by identifying the main products for each segment of the semiconductor supply chain, building on previous research by the OECD (2025[18]). Differentiating by segment, the subsequent step involves analysing the evolution of Mexico’s exports and imports and examining the degree of specialisation, concentration and number of trade dependencies.
As illustrated in Figure 2.13, exports of chips have grown significantly in recent years, reaching more than USD 5 billion in 2023. Nonetheless, imports of chips have also increased notably in recent years, reaching approximately USD 23 billion in the same year. This surge in imports has caused a worsening trade balance, resulting in a deficit of around USD 18 billion in 2023.
In 2023, Mexico’s chip imports were led by electronic integrated circuits; processors and controllers (Harmonized System [HS] code 854231), which accounted for 36% of imports. This category was followed by other electronic integrated circuits not elsewhere classified (HS code 854239) with 19%, and flash memory cards (HS code 852341) with 12% of imports.
Note: The figure includes products classified as chips under the six-digit HS classification, as described in Annex E. Trade values are expressed in real terms and are obtained by deflating nominal exports and imports using the PPI for Sector 3340, with July 2019 as the base period.
Source: OECD calculations based on the BACI database in CEPII (2026[19]), The CEPII‑BACI dataset, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html, for trade data and INEGI (n.d.[20]), Índice Nacional de Precios Productor (INPP). Año base julio 2019, https://www.inegi.org.mx/programas/inpp/2019/ for the deflator.
Figure 2.14 shows that the United States is the most important supplier of chips, representing 32.8% of Mexican chip imports in 2023, followed by Viet Nam (11.2%) and Korea (11.1%). Chip imports from Malaysia increased significantly from USD 2.3 billion in 2012 to USD 6.2 billion in 2022, reverting to USD 2.3 billion in 2023 (9.8% of total imports). Imports from China amounted to USD 2.2 billion in 2023 (9.5% of total imports), remaining below 2022 (USD 4.5 billion) and 2021 (USD 3.6 billion) figures. It remains to be seen if the decline in imports from these two economies reflects a temporary fluctuation or a more persistent change in chip sourcing patterns.
As illustrated in Figure 2.15, trade dynamics differed in other segments of the supply chain. Exports of manufacturing equipment have steadily increased since 2017, while imports have also risen, but at a slower pace. This resulted in a trade surplus of around USD 1.5 billion in 2023. In 2023, exports were mainly driven by machinery and apparatus for filtering or purifying gases (HS code 842139), accounting for 59% of the exports, and air or vacuum pumps (HS code 841459), contributing to 9% of the exports. At the same time, imports were driven by the same products, with 41% and 18% of imports respectively. The prevalence of the same products in both imports and exports suggests significant re-exports within Mexico’s manufacturing sector. This trend likely reflects the influence of the maquiladora industry, where imported equipment is used for processing or modification before being re-exported.11
Note: The figure includes products classified as chips under the six-digit HS classification, as described in Annex E. Trade values are expressed in real terms and are obtained by deflating nominal exports and imports using the PPI for Sector 3340, with July 2019 as the base period.
Source: OECD calculations based on the BACI database in CEPII (2026[19]), The CEPII‑BACI dataset, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html, for trade data and INEGI (n.d.[20]), Índice Nacional de Precios Productor (INPP). Año base julio 2019, https://www.inegi.org.mx/programas/inpp/2019/ for the deflator.
Activity in the raw materials segment remains limited. In 2023, exports and imports remained close to USD 0.1 billion (Figure 2.15). Nonetheless, establishing a raw materials supply chain is a key priority in the Master Plan for the Development of the Semiconductor Industry in Mexico, published by Mexico’s technology trade association CANIETI (see Section 3.1). Specifically, the country could focus on developing the production of materials such as graphite, zinc, barite, manganese and tungsten, among others (CANIETI, 2024[21]).
To complement the analysis, the Revealed Comparative Advantage (RCA) can assess Mexico’s degree of specialisation within the different segments of the supply chain. The RCA indicator compares a country’s share of exports in a particular segment to the global share of those exports, thereby providing evidence of countries’ specialisation in certain segments and products. An RCA value above 1 would indicate comparative advantage (see Annex E for further details).
Figure 2.16 shows that Mexico does not exhibit a high level of specialisation in any of the segments under analysis. However, compared to other Latin American countries, Mexico’s RCA is the highest for semiconductor manufacturing equipment. Among the selected countries, only Costa Rica shows a high specialisation in the export of chips.
The existence of few countries supplying certain goods can induce trade dependencies that may ultimately result in supply bottlenecks, shortages and higher prices. To identify dependencies, the Herfindahl-Hirschman Index can be computed for each commodity imported to Mexico. This analysis determines whether Mexico’s imports of a specific product are concentrated in a limited number of economies. Further to the analysis of concentration in the market, additional criteria are used to determine if Mexico is trade dependent on a specific product (see Annex E for more details).
Note: The figure includes products classified as chips under the six-digit HS classification, as described in Annex E. Trade values are expressed in real terms and are obtained by deflating nominal exports and imports using the PPI for Sector 3340, with July 2019 as the base period.
Source: OECD calculations based on the BACI database in CEPII (2026[19]), The CEPII‑BACI dataset, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html, for trade data and INEGI (n.d.[20]), Índice Nacional de Precios Productor (INPP). Año base julio 2019, https://www.inegi.org.mx/programas/inpp/2019/ for the deflator.
Note: The values are presented using a logarithmic scale.
Source: OECD calculations based on the BACI database from CEPII (2026[19]), The CEPII‑BACI dataset, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html.
Following this approach, in the latest period of available data (2021-2023), seven dependencies are identified for Mexico, of which six correspond to trade flows with the United States and one corresponds to trade flows with China (Figure 2.17). This represents a decline from the previous period (2012-2014), when 12 dependencies were recorded.
Note: Each arrow denotes one dependency at the economy-product level. Node sizes represent degree centrality, indicating the number of direct connections each node has.
Source: OECD calculations based on the BACI database from CEPII (2026[19]), The CEPII‑BACI dataset, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html.
Data on the number of trade dependencies in different segments of the semiconductor value chain show that the number of dependencies increased for chips (from one in the earlier period to two in the most recent period) and were concentrated in parts of electrical resistors (HS code 853390). Trade dependencies also increased for raw materials, from two to three over the same period. In contrast, trade dependencies declined across other segments, particularly in manufacturing equipment, where they decreased from four in the earlier period to just one in the latest period.
OECD data on TiVA provide additional insights into the integration of Mexico into GVCs (OECD, 2024[22]). Figure 2.18 shows that domestic value added in electronics exports peaked at about 55% in 1998 and has since fluctuated within a lower range of 40%-50%. This downward trajectory, followed by the decline in domestic content of Mexican exports of electronic products, is likely a reflection of the dependence on the Maquiladora Export Industry (IME). Figure 2.18 also shows that the contribution of domestic value added is much lower in Mexico in comparison with other Latin American economies, such as Chile and Costa Rica.
Figure 2.19 shows the share of re-exported intermediate imports (REII) by the electronics sector as a share of intermediate imports, that is how much of the imports are exported in the sector. This indicator grew from 57% in 1997 to 73% in 2008 – a much higher level than in Chile and Costa Rica – and has increased steadily since then. With a high share of electronics inputs coming from abroad, Mexico could consider increasing its role as a key hub in GVCs, facilitating the movement of semiconductor-related goods between trading partners.
Figure 2.20 shows how the Mexican electronics sector’s value added is distributed across foreign economies. This measure reflects how this value added is connected to the final demand of consumers in other economies, through exports of final goods or services and, indirectly, via exports of intermediates that reach foreign final consumers (households, government, business investment) through other economies (OECD, 2024[22]). The Mexican electronics sector is most closely connected to the United States, as expected due to their geographical proximity and trade agreements, followed by Canada and China, which have increased their final demand in recent decades.
Figure 2.21 shows the distribution of the electronics sector’s foreign value added by economy of origin. This indicates that the dependence of the Mexican electronics sector partly shifted from Japan and the United States to China. Other economies that increased their connectedness with Mexico to a lesser extent are Korea, Chinese Taipei, Malaysia and Singapore. In 1995, 69% of foreign value added to electronics embodied in Mexican final demand came from the United States. This share fell to less than 18% in 2022, while China’s share increased from less than 1% to 27% over the same period.
Note: This figure shows the contribution domestic value added (VA) as a percentage of gross exports (EXGR_DVASH TiVA series). This figure uses ISIC Rev.4 Division 26 (electronics sector), as TiVA sectoral classifications follow ISIC; national sources in this report rely on NAICS 3344 (semiconductor and other electronics manufacturing).
Source: Based on TiVA data in OECD (2024[22]), Trade in value added (indicator), https://doi.org/10.1787/data-00648-en.
Note: This indicator reflects the share of intermediate inputs imported by the electronics sector that is used domestically by Mexico in producing goods and services for export (both indirectly and directly), as a percentage of total intermediate inputs (from the electronics sector) (OECD, 2024[22]). It provides a measure of the importance of intermediate imports to produce goods and services for export and their role as a source of international competitiveness (IMGRINT_REII TiVA series). This figure uses ISIC Rev.4 Division 26 (electronics sector), as TiVA sectoral classifications follow ISIC; national sources in this report rely on NAICS 3344 (semiconductor and other electronics manufacturing).
Source: Based on TiVA data in OECD (2024[22]), Trade in value added (indicator), https://doi.org/10.1787/data-00648-en.
Note: The figure represents how Mexican electronics sector value added is distributed across economies, with each column representing the share of each economy (summing to 100%) for 1995 (diamond markers) and 2022 (columns). Domestic value from the Mexican electronics sector embodied in foreign final demand by partner country (FFD_DVAPSH TiVA series) provides a value-added perspective of domestic industries’ relative connectedness with other economies, independent of whether domestic (upstream) industries are direct exporters (OECD, 2024[22]). This figure uses ISIC Rev.4 Division 26 (electronics sector), as TiVA sectoral classifications follow ISIC; national sources in this report rely on NAICS 3344 (semiconductor and other electronics manufacturing).
Source: Based on TiVA data in OECD (2024[22]), Trade in value added (indicator), https://doi.org/10.1787/data-00648-en.
Note: The figure represents how foreign value added of the electronics sector in Mexican final demand is distributed by economy of origin, with each column representing the share of each economy (summing to 100%) for 1995 (diamond markers) and 2022 (columns) (DFD_FVAPSH TiVA series), providing a value-added perspective of a domestic economy’s relative connectedness to production in other economies and regions (OECD, 2024[22]). This figure uses ISIC Rev.4 Division 26 (electronics sector), as TiVA sectoral classifications follow ISIC; national sources in this report rely on NAICS 3344 (semiconductor and other electronics manufacturing).
Source: Based on TiVA data in OECD (2024[22]), Trade in value added (indicator), https://doi.org/10.1787/data-00648-en.
Attracting investment is a crucial component for the development of the semiconductor ecosystem. Increased investment enables the semiconductor industry to boost R&D efforts, expand manufacturing capacity, increase competitiveness and ultimately strengthen supply chains.
The bulk of foreign direct investment (FDI) in the semiconductor industry comes from the United States, mainly from the firms Jabil and Intel, with a sizeable amount of investment going towards the Guadalajara metropolitan area. Almost one‑third of total investment is allocated to R&D activities and two‑thirds to manufacturing activities.
Mergers and acquisitions (M&A) activity involving manufacturers of computer, electronic and optical products grew significantly in recent years and is largely driven by firms within the same industry, although the involvement of external firms has grown over time.
Cross-border investments are primarily driving M&A activity in Mexico. Throughout the period 2009-2022, 92% of total deal value involved a Mexican firm being acquired by a foreign firm, with Asian firms primarily driving this result.
Innovation is another crucial component for the development of the semiconductor ecosystem and for the expansion to high value-added activities within the supply chain. At the global level, patent filings in the semiconductor industry have experienced an increasing trend since the early 1980s. However, in Mexico, only 12 semiconductor-related Patent Cooperation Treaty (PCT) patents were filed between 2001 and 2010 while 17 PCT patents were filed between 2011 and 2022. Additionally, R&D investment has declined in the last decade and remains substantially below the OECD average.
2.2.1. Foreign direct investment - greenfield
Between January 2003 and August 2025, according to fDi Markets data, foreign firms invested USD 1.6 billion in the Mexican semiconductor and other electronics industry through 29 investments. Figure 2.22 highlights that investments stem predominantly from the United States, with more than USD 1.2 billion. The biggest investments recorded were made by Jabil and Intel: the former invested USD 400 million and USD 150 million in 2022 (two separate investments) to expand its factories in Guadalajara, while the latter invested USD 189 million in 2010 for its plant in the same city. Jabil is responsible for investing USD 638 million in the country and creating 11 000 new jobs over the whole period. Overall, Jabil accounts for 40% of total investment and 55% of new job creation, whereas Intel accounts for 12% of total investment.
Apart from those two firms, the most relevant investments included those made by Advanced Semiconductor Engineering, Freescale Semiconductor, NVIDIA, Snowbush Microelectronics and Solidigm. As of 2025, the most recent investments were announced by Micron Technology and Vishay Intertechnology. Almost two‑thirds of the investments support manufacturing activities and around one‑third support R&D activities, with minimal investment in sales, marketing and support activities. Figure 2.23 shows the evolution of FDI and gross new job creation in the semiconductor industry between 2003 and 2025. The spike in 2022 illustrates the recent interest by Jabil in the country.
Note: There are 29 investment projects recorded in this period, no missing values either on capital investment or jobs created. The data include announced deals.
Source: OECD calculations based on fDi Markets (2024[23]), “Homepage”, https://www.fdimarkets.com/ (accessed on 14 January 2025).
Source: OECD calculations based on fDi Markets (2024[23]), “Homepage”, https://www.fdimarkets.com/ (accessed on 14 January 2025).
2.2.2. Domestic investment
Regarding the investment made by establishments (gross fixed capital formation plus change in inventories), the semiconductor and other electronics sector holds the second spot in the high-tech manufacturing segment (Figure 2.24, Panel A), behind basic chemicals, receiving an investment of MXN 3.5 billion in 2024, or 2% of total investment in the manufacturing sector. The petroleum and coal industry received over 50% of total investment in high-tech manufacturing in 2019 but was the industry with the lowest investment in 2024.
Based on the geographical distribution of investments, the Centre West region accounts for a significant 41% of total investment in the semiconductor and other electronics industry (Figure 2.24, Panel B). Notably, the states of Jalisco and Querétaro are experiencing a substantial influx in gross capital formation, suggesting a promising path for future growth in these states.
Note: Total investment by establishments is defined as the increase in assets, inputs and outputs experienced by economic units during the reference year. It is obtained by adding the variation in inventories to gross fixed capital formation, as defined by INEGI (2014[24]). The dark blue columns in Panel A represent Level I of high-tech manufacturing, whereas Levels II and III are represented in medium blue.
Source: OECD calculations based on the economic census (2019 and 2024) by INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
Gross fixed investment in machinery and equipment in the semiconductor and other electronics industry has declined sharply since 2019, falling by nearly 47% by 2023 (Figure 2.25). The drop was more pronounced than in the computer and electronics sector, which saw a 21% decrease over the same period. Investment in overall manufacturing also saw a downward trend of 42% in the 2019-2023 period, which may reflect a contraction in productive capacity and underscores vulnerabilities in high-tech manufacturing production in Mexico.
Note: The figure shows the change in indexed gross fixed investment (GFI) across asset types between 2018 and 2023. According to INEGI, GFI is defined as the expenditures made for the purchase of machinery and production equipment. The indicator includes expenditures on the purchase of fixed assets, own-account production of assets and significant improvements, upgrades and renovations made by third parties, minus the value of asset sales during the year. Investment values are expressed in real terms by deflating the nominal series using the NAICS 2013 PPI for final goods and services, by destination, with July 2019 as the base period (100).
Source: OECD calculations based on INEGI (n.d.[17]), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, https://www.inegi.org.mx/rnm/index.php/catalog/1061. Please see Box 2.1 for more details.
2.2.3. Mergers and acquisitions
This subsection describes M&A activity in the semiconductor ecosystem. It shows that M&A activity involving manufacturers of computer, electronic and optical products has grown significantly over recent years and is largely driven by firms within the same sector. Deals involving at least one firm outside the sector (either on the target or the acquiror side) played a more limited role in this growth, suggesting a contained, although growing, cross-sectoral spillover (see Figure 2.26).
Notes: This figure covers the deals in the following categories: genuine acquisition, further acquisition, minority stakes and joint venture. The manufacture of computer, electronic and optical products sector corresponds to Division 26 of the ISIC classification. A deal is considered internal when both firms belong to this sector and external when only one of them belongs to the sector.
Source: OECD calculations based on Zephyr (v.2022), currently Orbis M&A (2026[25]), “Welcome to Orbis M&A”, https://login.bvdinfo.com/R1/OrbisMA.
Cross-border M&A can bring benefits in terms of firm performance and profits for home and host countries if successful industrial restructuring brings greater efficiency and does not imply undue market concentration (Kang and Johansson, 2000[26]). The results show a notable interest in Mexican firms by international counterparts. Throughout the period 2009-2022, 92% of the total deal value involved one Mexican firm being acquired by a foreign firm, with Asian firms primarily driving this result. Acquisitions from Hong Kong (China), Singapore and Chinese Taipei accounted together for more than 60% of the total value of the deals.
2.2.4. R&D and patents in the semiconductor field
Patent statistics provide insightful information on the general state-of-art in the concerned technology area and can help identify the maturity of certain technologies or technological trends (EPO, n.d.[27]; WIPO, 2023[28]). While patent data may not provide the full picture of the innovative nature of an industry in a certain country, they are often used as a proxy for measuring technological innovation. In evaluating technological progress, this information may be complemented with other indicators such as investments in R&D and digital technologies.
At the global level, patent filings in the semiconductor industry have experienced an increasing trend since the early 1980s (see Box 2.2 for details on the use of patents as a measure of innovation activity). As shown in Panel A of Figure 2.27, the number of patents filed for semiconductor inventions saw a significant increase between 1995 and 2005, with an annual growth rate of 19.1% in the number of patent applications filed under the PCT, a proxy for high-potential patents. Since then, PCT applications have slowed down: from 2015 to 2020, PCT applications grew at an annualised growth rate of 4.7%. Additionally, as illustrated in Panel B of Figure 2.27, the share of semiconductor patents in total PCT applications has remained broadly stable since 2002.
Notes: Data refer to patent applications filed under the PCT, by earliest filing date. Patent application follows fractional counting. The latest observations are for 2022.
Source: OECD (n.d.[29]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats.
Patents provide a detailed source of information on the inventive activity of a country and are frequently used as a proxy for innovation measures, following a long tradition in the literature (see, for example, Griliches (1990[30])). Patents are used by firms, research institutions or individuals to protect inventions in a given market. Patent indicators convey information on the output and on the processes of inventive activities, as described in the OECD Patent Statistics Manual (2009[31]).
The patent data derive from the intellectual property (IP) data of the OECD Science, Technology and Innovation (STI) Micro-data Lab, that mostly relies on the European Patent Office (EPO) Worldwide Patent Statistical Database, also known as PATSTAT Global, in its Spring 2025 edition. The PATSTAT database includes information on patents filed worldwide, including their administrative steps, the origin of the invention (with the name and location of inventors or applicants), the technologies that are covered by the invention (e.g. the patent classes, the title, abstract and claims describing the invention), the countries in which the inventions are protected, the citations made to or by other patents, as well as references to the scientific literature.
Semiconductor patents are identified as those allocated to International Patent Classification (IPC) codes H01N or H10 and complemented with patents featuring the keywords related to semiconductors in their abstract that are not classified in the IPC codes. Keywords used to identify trends in patents related to semiconductors are as follows: “semiconductor”, “transistor”, “integrated circuit”, “silicon wafer”, “logic chip” and “memory chip”.
The latest trends in semiconductor patents worldwide are shown using patent applications filed under the PCT: because of administrative delays observed during the patent filing process at different IP offices, IP5 patent families are less timely. Notably, PCT patent applications aim to protect patents across multiple jurisdictions and therefore typically include patents of higher economic value than patents protected in a single jurisdiction.
IP5 patent families are defined as sets of patent applications protecting the same invention filed in at least two IP offices – with at least one application filed in one of the five largest IP offices worldwide: the EPO, the Japan Patent Office, the Korean Intellectual Property Office, the State Intellectual Property Office of the People’s Republic of China or the United States Patent and Trademark Office (USPTO).
Patents are assigned to the country where the applicant filing the patent is located, using fractional counts (if two firms located in two countries jointly file a patent, each country gets assigned one‑half of a patent). Owing to the scarcity of patent citation records in PATSTAT, citation-based measures only rely on EPO or USPTO patents.
Sources: Griliches, Z. (1990[30]), “Patent Statistics as Economic Indicators: A Survey”, https://www.jstor.org/stable/2727442; OECD (2009[31]), OECD Patent Statistics Manual, https://doi.org/10.1787/9789264056442-en.
Patenting activity amongst Mexican inventors experienced a significant upward trend from the early 2000s through 2015. However, this patenting activity slowed down after 2015. Despite these broader patterns, patenting activity in the semiconductor industry has remained relatively limited. Based on the most recent patent data, only 12 semiconductor-related PCT patents were filed between 2001 and 2010 by Mexican inventors while 17 PCT patents were filed between 2011 and 2023 (see Figure 2.28). These results underscore the sector’s limited contribution to overall innovation and IP development.
Several stakeholders expressed the need to foster innovation and research activities and avoid policies that hinder the development of patents. Some universities such as the Monterrey Institute of Technology and Higher Education (Tecnológico de Monterrey) have put initiatives in place for staff to engage in knowledge transfer activities (OECD/IDB, 2022[32]).
Note: Data refer to patent applications filed under the PCT by earliest filing date. Patent application follows fractional counting. The latest observations are for 2023.
Source: OECD (n.d.[29]), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats.
R&D expenditure, an often-used input measure of innovation, can shed further light on the Mexican patenting performance. Figure 2.29 shows that in 2022, Mexico allocated less than 0.3% of its GDP to R&D annually, below the 2013 figure (0.4%) and significantly below the OECD average. In 2022, OECD Member countries, on average, dedicated 2.7% of GDP to R&D, with approximately 74% of this investment coming from the business sector (OECD, 2024[33]). Furthermore, Chile, Costa Rica and Uruguay surpass Mexico in R&D investment, while expenditure on R&D in Panama remains below 0.2% of GDP.
Note: Non-government expenditure includes R&D expenditure performed by business enterprises (public and private), higher education and private non-profit organisations. Total R&D expenditure is the sum of non-government and government R&D expenditures. The latest observations for Chile are for 2021.
Source: OECD calculations based on RICYT (n.d.[34]), “Homepage”, https://www.ricyt.org/ (accessed on 15 November 2024).
The decline in Mexico’s R&D spending reflects both a drop in government funding and a stagnation in private sector investment. Government contribution to R&D investment fell from 0.17% of GDP in 2013 to 0.07% in 2022. Meanwhile, private sector investment also declined, from 0.24% of GDP in 2013 to 0.19% in 2022, leading to the overall decline in R&D expenditure. Enhanced efforts to increase R&D investment in Mexico would be essential to foster innovation, boost productivity and maintain global competitiveness, especially when compared to other OECD Member countries where both public and private sectors contribute significantly more to R&D.
2.2.5. Machinery and equipment
Acquisition of machinery and equipment is one of the main bottlenecks for the development of the semiconductor industry in Mexico. As Figure 2.30 shows, the share of machinery and equipment over total net assets of the semiconductor and other electronics industry in Mexico is lower than the average in the high-tech segment (38% compared to 56%). Figure 2.30 also shows that the share of machinery in total assets is about the same as that of buildings, possibly suggesting a relatively limited capital allocation towards productivity-enhancing assets.
Source: OECD calculations based on INEGI, (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
The semiconductor and other electronics industry presents the second lowest value of machinery and equipment (at replacement cost) per worker, when compared to other high-tech manufacturing industries in Mexico (Figure 2.31, Panel A). The low intensity of machinery and equipment capital suggests that this industry in Mexico might rely more heavily on manual labour than on automated processes. In addition, one of the biggest challenges identified in the industry is machinery and equipment acquisition. Financial and logistical constraints restrict capital investment, resulting in a lower share of machinery and equipment over total production, as shown in Panel B of Figure 2.31.
Note: In Panel A, average value of machinery and equipment owned by the economic unit (at replacement costs) is directly linked to production processes, with respect to the number of workers employed. In Panel B, dark blue columns represent Level I of high-tech manufacturing, while medium-blue columns correspond to Levels II and III.
Source: OECD calculations based on the economic census (2019 and 2024) in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
Stakeholders across countries cite insufficient skills as a critical barrier in the development of semiconductor ecosystems globally. Developing an appropriate set of skills is crucial for countries aiming to expand their role within the semiconductor GVC. Recent studies on semiconductor workforce development (Younkin, 2024[35]) highlight several core components, namely fostering the growth of required talent, refining curricula, aligning knowledge, skills and abilities with critical job functions, ensuring access to state-of-the-art educational and training facilities, facilitating experiential learning and providing recruitment support. Addressing these challenges will help narrow the gap between skill demand and supply to support successful ecosystem development.
Mexico has expanded its talent base for the semiconductor industry, but foundational gaps persist. Learning outcomes in mathematics and science from the OECD Programme for International Student Assessment (PISA) test remain below OECD levels, and upper secondary completion is low despite long-term declines in dropout rates supported by public programmes. Tertiary education produces a comparatively high share of engineering graduates, including women, though regional imbalances and shortages in specialised fields persist.
Labour-market data show rising demand concentrated among a few large firms, with strong needs for transversal and specialised manufacturing skills that overlap with other industries, particularly motor vehicle parts manufacturing. Employment in the semiconductor and electronics sector continues to grow, especially in northern regions, but spending on scientific and technical roles remains limited. At the same time, sustained negative net migration and limited attractiveness for highly skilled workers underscore the need for stronger long-term retention and deeper engagement with Mexico’s diaspora.
2.3.1. Supply
Secondary education
Technical aptitude (i.e. understanding and applying technical concepts) is particularly relevant to the engineering profiles required for the semiconductor industry. The efficacy of education systems in preparing students for these kinds of tertiary education tracks can be evaluated, for instance, through OECD PISA, which assesses the knowledge and skills of 15-year-old students in mathematics, reading and science and serves as a benchmark for international comparison.
Mexico witnessed a decline in students’ performance in mathematics and science in 2022 compared to 2018, remaining below the OECD average. Although Mexico’s PISA results in mathematics and science surpass Panama’s, it remains below other Latin America countries such as Chile or Uruguay, as illustrated in Figure 2.32. In mathematics, only 34% of students achieved at least level 2 proficiency, significantly below the OECD average of 69%. Similarly, in science, only 49% of students attained level 2 or higher, compared with the OECD average of 76% (OECD, 2023[36]). Level 2 is the baseline level of proficiency that students need to participate fully in society.
According to stakeholders consulted for this report, one of Mexico’s priorities is enhancing education at the secondary level and reducing dropout rates. Currently, 43% of young adults in Mexico have not obtained an upper secondary qualification, a figure significantly surpassing the OECD average of 14% (OECD, 2023[37]). However, dropout rates have considerably decreased in Mexico across the main educational segments over the past two decades. As illustrated in Figure 2.33, dropout rates at the federal level in secondary education decreased in the 2022/2023 school year compared to rates in 2010/2011. However, some regions, such as the state of Oaxaca, witnessed an increase; Oaxaca is also the region with the highest dropout rate (8%).
Among the reasons for not enrolling in the 2020/2021 school year, 4.8% of those aged 13 to 15 years and 6.8% for the 16-to-18-year-old group did not enrol due to COVID-19 pandemic‑related reasons.12 While 1.1% of the population aged 13 to 15 years cited lack of resources as the reason for not enrolling, this reason increased to 5.6% for the 16-to-18-year-old group. In the 2019/2020 school year, upper secondary education had the highest non-completion rate at 3.6% (INEGI, 2021[38]). Although the COVID-19 pandemic caused a spike in the higher education dropout rate, recovery followed in the aftermath (Figure 2.34). The upper secondary segment still has the highest dropout rate, but it also saw the largest reduction, with an 8.8 percentage point (p.p.) decline. Public programmes aimed at reducing dropout rates, such as the Benito Juárez Scholarship (see Section 3.2.1), may have played a crucial role in these improvements.
Note: Scores are standardised to follow a normal distribution, with a mean of 500 points in OECD Member countries, and a standard deviation of 100 points. These scores summarise the mean performance of 15-year-old students, averaged across each country’s nationally representative sample. PISA data support rigorous cross-country comparisons of student achievement, highlighting differences in educational outcomes. Mexico’s scores meet PISA’s technical standards, ensuring their validity and comparability within this global frame (OECD, 2024[39]).
Source: Based on OECD (2023[40]), OECD PISA 2022 Database, https://www.oecd.org/pisa/data/2022database/.
Note: The figure shows the percentage of students who leave school activities during the school year and at the end of it, relative to the total number of students enrolled in the school year. The percentage may be negative, because during the studied school year there were more students who enrolled as "new entries" than those who left as "dropouts".
Source: OECD calculations based on INEGI (n.d.[41]), Tasa de abandono escolar por entidad federativa según nivel educativo, ciclos escolares seleccionados de 2000/2001 a 2023/2024, https://www.inegi.org.mx/app/tabulados/interactivos/?pxq=Educacion_Educacion_11_c6aa7c65-4d89-4eaf-972e-431727fc686d.
Note: Time is represented in five-year intervals, but annual reporting began with the COVID-19 pandemic. Preliminary data for academic year 2022/2023 are included to provide the latest available observation.
Source: OECD calculations based on INEGI (n.d.[41]), Tasa de abandono escolar por entidad federativa según nivel educativo, ciclos escolares seleccionados de 2000/2001 a 2023/2024, https://www.inegi.org.mx/app/tabulados/interactivos/?pxq=Educacion_Educacion_11_c6aa7c65-4d89-4eaf-972e-431727fc686d.
Tertiary education
While secondary education lays a crucial foundation, industry stakeholders emphasise the importance of tertiary education in preparing future engineering and science, technology, engineering and mathematics (STEM)-related professionals. As shown in Figure 2.35, in 2021, 17% of total students in tertiary education graduated from engineering and related subjects. This share not only exceeds that of other countries of the region like Chile, Costa Rica and the United States, but also surpasses the OECD average.
Moreover, this trend is particularly noteworthy for women, with 10% of female students graduating from engineering and related subjects compared to 5% for Chile and Costa Rica, 3% for the United States and 6% for the OECD average (Figure 2.35). Nonetheless, closing the gap between male and female student enrolment should remain a priority and the effective implementation of programmes, such as NiñaSTEM, can help in this purpose.13
By regions, the highest share of semiconductor-related graduates is in the Northeast, with around 16% of the total graduates specialising in the field.14This figure is notably lower for the other regions, especially the Southeast, which could reflect the presence of top-ranked universities in the Northeast region, such as Tecnológico de Monterrey, which is amongst the top 500 universities worldwide according to the QS ranking.15 Nonetheless, labour income is the highest in the Northwest region, likely due to the influence of large semiconductor firms (Figure 2.36).
At an aggregated level, Mexico exhibits a surplus of engineers (Filippo et al., 2022[11]). However, after a closer examination of the disaggregated regions, Filippo et al. (2022[11]) find a shortage of electronic and automation, geological and information-and-communication-technology-specialised engineers in the Northwest region. Conversely, the Northeast region faces a deficit of mechanical engineers. Ensuring cross-region mobility could help alleviate the shortage of specialised talent in some regions.
Note: Data represent graduates from engineering, manufacturing and construction.
Source: OECD (2023[37]), Education at a Glance 2023: OECD Indicators, https://doi.org/10.1787/e13bef63-en.
Note: Data are restricted to people answering yes to the question “Did you finish the studies or subjects of this degree?”, corresponding to Question 16 of the sociodemographic module of ENOE. The bars represent the share of graduates specialising in the relevant fields by region. The graduates are classified by region following the federal entity reported in the survey. Income refers to monthly net labour income. Data correspond to the fourth quarter of 2023.
Source: OECD calculations based on ENOE in INEGI (n.d.[42]), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, https://www.inegi.org.mx/programas/enoe/15ymas/. Please see Box 2.1 for more details.
2.3.2. Demand
Based on the analysis of online job postings in the semiconductor and other electronics industry, it is evident that the US firm Jabil emerges as the primary provider of employment opportunities for this industry in Mexico, offering 10 000 vacancies between 2020 and 2022 (Figure 2.37). This estimate accounts for nearly 40% of total job postings in the industry. IBM and TE Connectivity follow in second and third place, with 3 500 and 2 400 job postings respectively.
In terms of semiconductor firms (identified in Table 2.1), a total of 2 500 job postings were advertised online between 2020 and 2022, constituting 10% of the overall observations. Among them, Intel leads with 1 300 job postings, followed by Skyworks Solutions with 859 and Infineon with 217. Texas Instruments occupies the 21st position with under 80 job postings.
The remaining job postings are predominantly advertised online by the other electronic component manufacturing segment, where other significant players are Plexus, Sanmina and Vishay Intertechnology. As Figure 2.37 shows, the semiconductor vacancies are a small fraction of the broader electronics sector in Mexico.
Note: The dark blue bars represent the firms identified as key players in the Mexican semiconductor ecosystem in Table 2.1, and the medium blue represent firms belonging to firms in other electronics manufacturing. Lightcast data capture flows of advertised positions and are not a measure of the stock of workers in the semiconductor and other electronics industry. Coverage reflects online vacancy postings, not the full labour market; some postings lack complete identifiers, and not all vacancies have an employer name attached.
Source: OECD calculations based on Lightcast (2023[43]), “Lightcast Data”, https://lightcast.io/products/data/overview.
Most common skills in the semiconductor and other electronics industry
Analysing Lightcast online job postings provides valuable insights into the evolving demand of the labour market and identifies core competencies required by firms in the semiconductor and other electronics industry. The job postings analysed in the context of this report are from Indonesia, Mexico and Viet Nam, countries that mainly focus on ATP activities: out of the 28 999 total job postings for semiconductors, 1 124 were from Indonesia, 26 491 from Mexico and 1 384 from Viet Nam.
The semiconductor and other electronics industry requires a blend of transversal and technical skills. Transversal skills, such as communications, management and operational competencies, are in highest demand (Figure 2.38, Panel A). These broader skills complement the technical proficiencies required in manufacturing processes (such as molding, dicing and swaging) (Figure 2.38, Panel B), which form the technical toolkit for working in the semiconductor and other electronics industry. The complete list of generic skills is listed in Annex G and specific skills are listed in Annex H.
Note: Panel A shows generic skills and Panel B semiconductor-specific skills, where the y-axis corresponds to the number of online job postings by skill in the semiconductor and other electronics industry (NAICS code 3344). The complete list of generic skills is listed in Annex G and the semiconductor-specific skills are listed in Annex H.
Source: OECD calculations based on Lightcast (2023[43]), “Lightcast Data”, https://lightcast.io/products/data/overview.
Synergies in skill demand across some industries
Semiconductor-specific skills are generally in high demand in manufacturing but are even more prominently required within the semiconductor and other electronics industry. Particularly, Figure 2.39 compares the most sought-after skills in the industry to those in other manufacturing activities. This comparison highlights that communication skills are found in 31% of job postings within the industry, compared to 25% in other manufacturing industries. Additionally, security policies emerge as the most relatively demanded skill, with a 13 p.p. higher prevalence in job postings within the industry. This is followed by operating systems (10 p.p.), data collection (10), arithmetic (10) and memos (9).
In addition, synergies between industries can be explored. By computing the distances among industries considering the most common 1 000 skills associated with the semiconductor and other electronics industry, the most closely aligned industries in terms of skills demand can be identified (using the Mahalanobis distance, see the note relating to Figure 2.40 below). These distances serve as a metric for evaluating the similarity between industries, specifically focusing on the distance to NAICS 3344 (semiconductor and other electronics industry). Industries closest in this skill space are therefore most likely to compete for talent and could be targeted simultaneously through similar policy initiatives aimed at fostering human capital development.
Motor vehicle parts manufacturing is by far the closest to the semiconductor and other electronics industry in terms of skill demand Figure 2.40. This industry involves the production of components vital for automotive assembly, which tend to require precision engineering and adherence to stringent safety regulations. This skill profile is similar to the manufacturing of semiconductors, especially considering that Mexico harbours mainly ATP activities.
Note: Panel A shows the frequency share of the 100 most sought-after skills in the semiconductor and other electronics industry in the same sector (x-axis) and all the other manufacturing industries (y-axis). Panel B omits the top five skills.
Source: OECD calculations based on Lightcast (2023[43]), “Lightcast Data”, https://lightcast.io/products/data/overview.
Note: The Mahalanobis distance serves as a metric for evaluating the similarity between economic industries using the manufacturing activities from NAICS at the four-digit level. It provides a distance between industries in terms of skills, weighting the different skills by the number of job postings in which they appear. The last three industries of the graph correspond to the furthest distances to the semiconductor and other electronics industry. Distances were standardised, with a maximum distance of 1 (between semiconductors and lime and gypsum industries) and the remaining distances ranging between 0 and 1 relative to the furthest distance.
Source: OECD calculations based on Lightcast (2023[43]), “Lightcast Data”, https://lightcast.io/products/data/overview.
Other similar industries, albeit to a much lesser extent than vehicle parts manufacturing, include the manufacturing of beverages and pharmaceuticals, which require competencies related to ingredient sourcing, production processes and packaging, along with cutting-edge R&D. Manufacturing of medical equipment and supplies, as well as navigational, measuring and electromedical manufacturing also share similarities with semiconductors and require proficiency in software, precision machinery and electronics assembly.
2.3.3. Labour-market developments
The semiconductor and other electronics industry employs the largest number of workers among all high‑tech manufacturing industries, according to the economic census (Figure 2.41, Panel A). Employment in the semiconductor and other electronics industry is highly concentrated around the Northwest, Northeast and Centre West regions (Figure 2.41, Panel B), reflecting the clustering of workers around established hubs such as the Guadalajara metropolitan area, Mexicali and Tijuana.
In 2024 alone, the industry represented 247 000 jobs (up from 174 000 in 2019), surpassing the combined employment placements offered by the second, third and fourth high-tech industries, namely electrical equipment, pharmaceutical and other general machinery manufacturing. Semiconductors account for 25% of the overall labour market within the high-tech segment. Most workers are located in the North, with 41% in the Northeast, 32% in the Northwest, and 26% in the Centre West. In addition, the semiconductor and other electronics industry exhibits a balanced gender composition, with women representing 53% of the workforce (Figure 2.42).
Note: In Panel A, dark blue columns represent Level I of high-tech manufacturing, and medium blue ones represent Levels II and III. In Panel B, the spider graph represents the number of workers by region.
Source: OECD calculations based on the economic census in INEGI (n.d.[15]), Censos Económicos (CE) 2024, https://www.inegi.org.mx/programas/ce/2024/. Please see Box 2.1 for more details.
Note: The dark blue bars represent the women’s share and the light blue the men’s share. The list of regions is available in Table 2.2.
Source: OECD calculations based on INEGI (n.d.[17]), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, https://www.inegi.org.mx/rnm/index.php/catalog/1061. Please see Box 2.1 for more details.
Despite having one of the highest shares of maquila income, manufacturing of semiconductors and other electronics ranks seventh in terms of expenditure on scientific, technical and professional workers within high-tech manufacturing, accounting for 2.04% of the total expenditure (Figure 2.43). This position underscores the industry’s demand for specialised knowledge and skills in the production process.
Note: Dark blue columns represent Level I of high-tech manufacturing and medium blue Levels II and III. The share corresponding to audio and video equipment manufacturing in 2024 is 22.6%.
Source: OECD calculations based on the economic census (2019 and 2024) in INEGI (n.d.[17]), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, https://www.inegi.org.mx/rnm/index.php/catalog/1061. Please see Box 2.1 for more details.
Occupational composition
In terms of occupations, assembly-related roles dominate, representing around 50% of the positions held, especially assemblers and installers of electrical and electronic parts. This is closely followed by assembly supervisors. Within the computer and electronics sector, employers also require operational occupations for maintenance, such as support workers. Additionally, there is a growing need for software and multimedia developers who can facilitate automation and digitalisation processes. For this occupation, the share has increased from approximately 1% in the fourth quarter of 2012 to 2.8% in the fourth quarter of 2023 (see Figure 2.44).
Notes: The computer and electronics sector corresponds to Code 334 in the NAICS classification. The occupations selected are the five most common occupations in 2023 Q4. Data correspond to the fourth quarter of 2023.
Sources: OECD calculations based on ENOE in INEGI (n.d.[42]), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, https://www.inegi.org.mx/programas/enoe/15ymas/. Please see Box 2.1 for more details.
As shown in Figure 2.45, transportation equipment manufacturing employs around 50% of the workers with assembly-related roles as described above, therefore sharing similarities with the computer and electronics sector in terms of occupational structure and skill needs. This similarity could be beneficial for cross-sectoral spillovers and knowledge transfer, fostering collaboration and innovation across industries.
Mexico is the fourth-largest exporter of motor vehicles globally. The robust growth of the automotive industry has fostered the development of an ecosystem that supplies not only Mexico but also a significant portion of Latin America. Consequently, this industry generates substantial demand for semiconductors, leading to synergies between the two industries (Filippo et al., 2022[11]).
Electrical equipment, appliance and component manufacturing is another sector requiring assembly-related occupations, as shown in Figure 2.45. Together with computer and electronics, these three sectors concentrate more than 90% of the assemblers of electronic and electronic parts and around 80% of the supervisors in the assembly process.
By examining the median income of relevant occupations in the computer and electronics sector compared to other non-computer and electronics sectors, it is possible to determine whether the semiconductor industry provides a sector premium for workers. Nonetheless, several reasons may drive this wage premium, including age or productivity composition.
Notes: The chart shows the three sectors with the highest share of workers for the first two occupations described in Figure 2.44.
Source: OECD calculations based on ENOE in INEGI (n.d.[42]), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, https://www.inegi.org.mx/programas/enoe/15ymas/. Please see Box 2.1 for more details.
The median labour income, shown in Figure 2.46, is higher for assemblers in the computer and electronics sector than in other sectors. Conversely, median labour income for assembly supervisors tends to be slightly higher in other sectors, albeit with a greater dispersion observed across the lower and upper ends of the distribution.
Note: The lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentiles). The upper whisker extends from the hinge to the largest value no further than 1.5 * IQR from the hinge (where IQR is the inter-quartile range, or distance between the first and third quartiles). The lower whisker extends from the hinge to the smallest value at most 1.5 * IQR of the hinge. Outliers are excluded from the graph. Data show monthly net income. The sample is restricted to workers who worked more than 34 hours per week.
Source: OECD calculations based on ENOE from INEGI (n.d.[42]), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, https://www.inegi.org.mx/programas/enoe/15ymas/. Please see Box 2.1 for more details.
As illustrated in Figure 2.47, the computer and electronics sector, along with related sectors, saw a significant increase in demand for assemblers and assembly supervisors between the fourth quarter of 2013 and the fourth quarter of 2023. The growth in employment for assemblers was particularly strong in the transport equipment manufacturing sector, increasing by 61% compared to 31% in the computer and electronics sector. Similarly, for assembly supervisors, employment increased by 80% for the transport equipment manufacturing sector versus a 46% increase for the computer and electronics sector.
For assemblers in the computer and electronics sector, median labour income growth was in line with that in other sectors like transport equipment manufacturing. For supervisors, however, wage growth in the computer and electronics sector surpassed that of transport equipment manufacturing and electrical equipment manufacturing. Assembler and supervisor employment growth was more modest in the computer and electronics sector, when compared to the other two sectors shown in Figure 2.47. These results might reflect increased demand (and competition amongst firms) for supervisors in this sector, whereas the comparable wage growth for assemblers across sectors indicates that labour-market conditions for these workers are similar. Wage growth has also helped avoid the impact of inflation and recover from the loss of purchasing value.
Note: For income, the sample is restricted to workers who worked more than 34 hours per week.
Source: OECD calculations based on ENOE in INEGI (n.d.[42]), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, https://www.inegi.org.mx/programas/enoe/15ymas/. Please see Box 2.1 for more details.
Net migration
Human capital flight (or “brain drain”) presents significant challenges for countries experiencing it, as the outflow of skilled professionals in search of better opportunities abroad depletes the local talent pool.
Net migration provides a quantitative measure of workforce loss and brain drain, since emigration also includes highly skilled individuals. Latin America and the Caribbean are characterised by sustained negative net migration, and Mexico shares a similar pattern. In fact, as shown in Figure 2.48, Mexico experienced only two years of positive net migration between 1960 and 2022.
According to the latest OECD Indicators of Talent Attractiveness (Andersson, 2025[44]), which assess countries’ strengths and weaknesses in attracting and retaining different types of talented migrants, Mexico is relatively unattractive to highly educated workers. Among OECD Member countries, Mexico ranked in the bottom half for quality of opportunities and future prospects, and second to last for the skills environment in 2023 (OECD, 2023[45]).
While Mexico has addressed short-term brain drain by requiring recipients of some scholarships, such as those funded by the Secretariat of Science, Humanities, Technology and Innovation (SECIHTI, previously the National Council for Humanities, Science and Technology, CONAHCYT) to reside in Mexico for at least six months after their studies, the lack of long-term opportunities requires further efforts to foster meaningful brain circulation and increase talent retention in the country.
Both the Global Network Mx (RGMX) and the Consortium of Higher Education Institutions for the Educational Development of Mexicans Overseas (Consorcio de Instituciones de Educación Superior para el Desarrollo de las personas Mexicanas en el Exterior, CIESDEMEX) promote brain circulation by establishing institutional channels for collaboration between Mexicans abroad and academic, scientific and productive sectors in Mexico. The Global Network Mx mobilises high-skilled professionals in the diaspora through regional chapters to support research, technology transfer and entrepreneurship aligned with development priorities (Red de Talentos, n.d.[46]). CIESDEMEX brings together Mexican higher education institutions in educational programmes for communities abroad, while enabling their academic reintegration or remote collaboration with institutions in Mexico (IME, 2024[47]).
Number of years with positive and negative net migration
Note: LAC: Latin America and the Caribbean.
Source: OECD calculations based on UN (2024[48]), World Population Prospects: 2022 Revision, https://population.un.org/wpp/.
References
[44] Andersson, L. (2025), “Measuring and assessing talent attractiveness in OECD countries, second edition: Methodology paper”, OECD Social, Employment and Migration Working Papers, No. 318, OECD Publishing, Paris, https://doi.org/10.1787/133b6085-en.
[21] CANIETI (2024), Master Plan for the Development of the Semiconductors’ Industry in Mexico 2024-2030, National Chamber of Electronics, Telecommunications, and Information Technologies Industry.
[19] CEPII (2026), The CEPII-BACI dataset, Centre d’études prospectives et d’informations internationales, https://www.cepii.fr/DATA_DOWNLOAD/baci/doc/baci_webpage.html (accessed on 16 January 2026).
[27] EPO (n.d.), “Why are patent statistics important?”, European Patent Office, https://www.epo.org/en/service-support/faq/searching-patents/patent-statistics/why-are-patent-statistics-important.
[23] fDi Markets (2024), “Homepage”, https://www.fdimarkets.com/ (accessed on 14 January 2025).
[11] Filippo, A. et al. (2022), “México y la cadena de valor de los semiconductores: oportunidades de cara al nuevo escenario global”, Inter-American Development Bank, Washington, DC, https://doi.org/10.18235/0004276.
[49] Gaulier, G. and S. Zignago (2010), BACI: International Trade Database at the Product-level The 1994-2007 Version, http://www.cepii.fr/anglaisgraph/bdd/baci.htm (accessed on 3 February 2026).
[13] Government of Mexico (1999), Acuerdo de estratificación de empresas micro, pequeñas y medianas, Diario Oficial de la Federación, https://www.dof.gob.mx/nota_detalle_popup.php?codigo=4946386 (accessed on 3 December 2024).
[30] Griliches, Z. (1990), “Patent Statistics as Economic Indicators: A Survey”, Journal of Economic Literature, Vol. 28/4, pp. 1661-1707, https://www.jstor.org/stable/2727442.
[16] Hecker, D. (2005), “High-technology employment: A NAICS-based update”, Monthly Labor Review, https://www.bls.gov/opub/mlr/2005/07/art6full.pdf (accessed on 29 April 2024).
[47] IME (2024), “CIESDEMEX”, Instituto de los Mexicanos en el Exterior, https://ime.gob.mx/educacion/programa/ciesdemex (accessed on 5 June 2025).
[12] INEGI (2026), Directorio Estadístico Nacional de Unidades Económicas, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/app/mapa/denue/default.aspx (accessed on 16 January 2026).
[38] INEGI (2021), Encuesta para la Medición del Impacto COVID-19 en la Educación (ECOVID-ED) 2020, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/investigacion/ecovided/2020/ (accessed on 13 June 2024).
[24] INEGI (2014), Sistema Automatizado de Información Censal (SAIC), Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/contenidos/app/saic/saic_metodo.pdf (accessed on 31 May 2024).
[50] INEGI (2001), “El ABC de la Estadística de la Industria Maquiladora de Exportación”, https://www.inegi.org.mx/contenidos/productos/prod_serv/contenidos/espanol/bvinegi/productos/historicos/2104/702825175429/702825175429_1.pdf (accessed on 24 May 2024).
[15] INEGI (n.d.), Censos Económicos (CE) 2024, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/programas/ce/2024/ (accessed on 16 January 2026).
[17] INEGI (n.d.), Encuesta Anual de la Industria Manufacturera 2024. Serie 2018, Información 2023, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/rnm/index.php/catalog/1061 (accessed on 16 January 2026).
[9] INEGI (n.d.), Encuesta Mensual de la Industria Manufacturera 2025, Datos al mes de noviembre (2018=100), Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/rnm/index.php/catalog/1086 (accessed on 16 January 2026).
[42] INEGI (n.d.), Encuesta Nacional de Ocupación y Empleo (ENOE), población de 15 años y más de edad, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/programas/enoe/15ymas/ (accessed on 16 January 2026).
[20] INEGI (n.d.), Índice Nacional de Precios Productor (INPP). Año base julio 2019, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/programas/inpp/2019/ (accessed on 16 January 2026).
[41] INEGI (n.d.), Tasa de abandono escolar por entidad federativa según nivel educativo, ciclos escolares seleccionados de 2000/2001 a 2023/2024, Instituto Nacional de Estadística y Geografía, https://www.inegi.org.mx/app/tabulados/interactivos/?pxq=Educacion_Educacion_11_c6aa7c65-4d89-4eaf-972e-431727fc686d (accessed on 16 January 2026).
[26] Kang, N. and S. Johansson (2000), “Cross-Border Mergers and Acquisitions: Their Role in Industrial Globalisation”, OECD Science, Technology and Industry Working Papers, No. 2000/01, OECD Publishing, Paris, https://doi.org/10.1787/137157251088.
[43] Lightcast (2023), “Lightcast Data”, https://lightcast.io/products/data/overview (accessed on 14 January 2025).
[18] OECD (2025), “Mapping the semiconductor value chain: Working towards identifying dependencies and vulnerabilities”, OECD Science, Technology and Industry Policy Papers, No. 182, OECD Publishing, Paris, https://doi.org/10.1787/4154cdbf-en.
[8] OECD (2025), OECD Data Explorer - Annual value added and its components by economic activity, http://data-explorer.oecd.org/s/1zu.
[3] OECD (2025), OECD Economic Outlook, Volume 2025 Issue 1: Tackling Uncertainty, Reviving Growth, OECD Publishing, Paris, https://doi.org/10.1787/83363382-en.
[10] OECD (2024), “2023 OECD Digital Government Index: Results and key findings”, OECD Public Governance Policy Papers, No. 44, OECD Publishing, Paris, https://doi.org/10.1787/1a89ed5e-en.
[1] OECD (2024), OECD Economic Surveys: Mexico 2024, OECD Publishing, Paris, https://doi.org/10.1787/b8d974db-en.
[33] OECD (2024), “OECD Main Science and Technology Indicators. R&D and related highlights in the March 2024 Publication”, Directorate for Science, Technology and Innovation, OECD, Paris, https://www.oecd.org/sti/msti2024march.pdf.
[39] OECD (2024), PISA 2022 Technical Report, PISA, OECD Publishing, Paris, https://doi.org/10.1787/01820d6d-en.
[7] OECD (2024), “Producer price indices (PPI) (indicator)”, https://doi.org/10.1787/a24f6fa9-en (accessed on 22 May 2024).
[22] OECD (2024), Trade in value added (indicator), OECD Statistics on Trade in Value Added, OECD, Paris, https://doi.org/10.1787/data-00648-en (accessed on 22 May 2024).
[37] OECD (2023), Education at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/f5fb5b52-en.
[40] OECD (2023), OECD PISA 2022 Database, OECD, Paris, https://www.oecd.org/pisa/data/2022database/ (accessed on 16 April 2024).
[36] OECD (2023), PISA 2022 Results: Factsheets - Mexico, OECD Publishing, Paris, https://www.oecd.org/publication/pisa-2022-results/country-notes/mexico-519eaf88/.
[45] OECD (2023), “What is the best country for global talents in the OECD”, OECD Migration Policy Debates, No. 29, OECD, Paris, https://web-archive.oecd.org/2023-03-09/652850-What-is-the-best-country-for-global-talents-in-the-OECD-Migration-Policy-Debates-March-2023.pdf (accessed on 16 June 2024).
[31] OECD (2009), OECD Patent Statistics Manual, OECD Publishing, Paris, https://doi.org/10.1787/9789264056442-en.
[29] OECD (n.d.), STI Micro-data Lab: Intellectual Property Database, OECD, Paris, http://oe.cd/ipstats (accessed on 15 December 2025).
[32] OECD/IDB (2022), “Innovative and Entrepreneurial Universities in Latin America”, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/ca45d22a-en.
[25] Orbis M&A (2026), “Welcome to Orbis M&A”, https://login.bvdinfo.com/R1/OrbisMA (accessed on 16 January 2026).
[5] Pro Mexico Industry (2024), “QSM Semiconductors Advances Chip Factory Integration”, https://www.promexicoindustry.com/en/article/qsm-semiconductors-advances-chip-factory-integration (accessed on 23 May 2025).
[46] Red de Talentos (n.d.), “Red Global MX”, https://www.redtalentos.nl/en/red-global-mx (accessed on 5 June 2025).
[6] Reuters (2024), “Foxconn building Nvidia superchip facility in Mexico, executives say”, https://www.reuters.com/technology/foxconn-says-it-is-building-worlds-largest-manufacturing-facility-for-nvidias-2024-10-08/ (accessed on 23 May 2025).
[34] RICYT (n.d.), “Homepage”, Red Iberoamericana de Indicadores de Ciencia y Tecnología, https://www.ricyt.org/ (accessed on 15 November 2024).
[14] Secretaría de Economía (2022), Rumbo a una política industrial, https://www.gob.mx/cms/uploads/attachment/file/761984/Rumbo_a_una_Pol_tica_Industrial.pdf (accessed on 6 June 2025).
[48] UN (2024), World Population Prospects 2024, United Nations, https://population.un.org/wpp/ (accessed on 14 January 2025).
[4] Utar, H., L. Torres Ruiz and A. Zurita (2023), “The US-China Trade War and the Relocation of Global Value Chains to Mexico”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.4568757.
[28] WIPO (2023), “How Patent Information Can Help Spot Technology Gaps and Catalyze Economic Development”, World Intellectual Property Organization, https://www.wipo.int/wipo_magazine_digital/en/2023/article_0034.html.
[2] World Bank (2025), GDP per capita (current US$), World Bank, Washington, DC, https://data.worldbank.org/indicator/NY.GDP.PCAP.CD (accessed on 22 May 2025).
[35] Younkin, T. (2024), “SRC’s Microelectronics and Advanced Packaging Technologies (MAPT) roadmap: an era of innovation driven by energy, workforce, and environmental sustainability”, Proceedings Volume PC12953, Optical and EUV Nanolithography XXXVII, https://doi.org/10.1117/12.3027041.
← 1. Firms manufacturing semiconductors, printed circuit boards (PCB) and related components are grouped under one industry: Semiconductor and other electronic component manufacturing (hereafter “semiconductor and other electronics industry”, NAICS 3344). This industry is part of the broader industry group Computer and electronic product manufacturing (hereafter “computer and electronics sector”, NAICS 334).
← 2. Maquiladora establishments (or maquilas) develop the industrial processes or services intended for the transformation, elaboration or repair of merchandise of foreign origin, temporarily imported for subsequent export (INEGI, 2001[50]).
← 3. See the following page for OECD’s latest macroeconomic information on Mexico: https://www.oecd.org/economy/mexico-economic-snapshot.
← 4. OECD statistics used in this document focus on International Standard Industrial Classification of All Economic Activities (ISIC) Division 26 (manufacture of computer, electronic and optical products or electronics sector). Statistics from Mexican authorities refer to NAICS 3344 industry group (semiconductor and other electronic component manufacturing or semiconductor and other electronics industry).
← 5. Manufacture of computer, electronic and optical products, using ISIC rev. 4. This document uses ISIC for international data sources, instead of the North American Industry Classification System (NAICS) used by Mexican statistical authorities.
← 6. Sectoral data originate from establishment-level aggregations from the sources described in Box 2.1.
← 7. The interpretation of a declining number of young firms or establishments requires caution, as entries and exits are observed only at the aggregate level, without accounting for firm age. For instance, a unit that entered in 2017 and survived until 2023 would be classified as mature. Therefore, entry category for mature firms may include previously young firms that survived throughout the period of analysis.
← 8. Each economic activity of the high-tech manufacturing sector can be classified into three groups considering their R&D intensity, share of scientific, professional and technician occupations and product and process innovation. Level I is the group with the highest technological requirements (Hecker, 2005[16]). See Annex C for the list of high-tech manufacturing activities using NAICS.
← 9. High-tech industries are associated with new products and processes, which support productivity, international competitiveness and economic growth (Hecker, 2005[16]).
← 10. French acronym of Base pour l’Analyse du Commerce International, Database for International Trade Analysis. The BACI database can offer several advantages over the raw Comtrade database, which is the most comprehensive database on world trade, building on data on bilateral trade flows reported by countries to the United Nations Statistical Division (for more information, see https://comtradeplus.un.org/). However, while countries report both their imports and their exports, mirror flows (which should be identical for the reporter and partner country) between partners may differ in the Comtrade database. BACI employs a procedure to reconcile flows reported by importers and exporters to provide consistent trade flow estimates (Gaulier and Zignago, 2010[49]).
← 11. See the subsection on maquila income for information on the performance of the maquiladora industry and the subsection on investment incentives for details on programmes specific to this industry (e.g. IMMEX).
← 12. Additionally, non-enrolment due to lack of money or resources stands out, indicating economic reasons that may be intensified due to the financial hardship caused by the pandemic, even if not directly associated with the COVID-19 phenomenon.
← 13. For more information, please see https://www.oecd.org/latin-america/regional-programme/ninastem/.
← 14. The relevant fields are mechanical, electrical, electronic and chemical engineering, and related professions as defined in group 071 of the Mexican Classification of Study Plans by Academic Fields (CMPE) of 2016.
← 15. The QS score is a weighted average of several indicators in the areas of academic reputation, employer reputation, faculty student ratio, citations per faculty, international faculty ratio, international students ratio, international research network, employment outcomes and sustainability. The scores are calculated on the subset of all analysed institutions, as opposed to the subset of finally published (ranked) institutions. For more information, see https://www.topuniversities.com/university-rankings/world-university-rankings/2022.