Shenzhen Didn't Conquer 90% of Electronics. The Real Story Is More Interesting.

Wendover Productions gets one important thing right: Shenzhen's manufacturing ecosystem is extraordinarily difficult to replicate. But its documentary, Why Chinese Manufacturing Wins, turns that advantage into a much larger story than its evidence can support. It presents an unsourced claim that 90% of the world's electronics are made at least partly in Shenzhen, treats low wages and favorable currency policies as central explanations, and predicts that automation will push manufacturing back towards America. Nearly a decade later, those arguments deserve a much harder examination.

The forecast of China's industrial decline went in the wrong direction

The strongest test of the documentary is what happened after it was published on August 8, 2017. According to UNIDO's 2025 International Yearbook of Industrial Statistics, China accounted for 32% of global manufacturing production in 2024, exceeding the combined shares of the United States, Japan, Germany and South Korea. China's share had been 2.8% in 1990. The predicted retreat from manufacturing dominance did not materialize. UNIDO's 2025 manufacturing statistics.

That does not mean the documentary was wrong about every relocation. Manufacturing jobs and production processes can move overseas while the original manufacturing economy grows. The important distinction is between losing individual factories and losing industrial dominance. The documentary blurred those two outcomes.

The famous 90% statistic has a surprisingly weak paper trail

The documentary's most memorable number is also its least defensible. Its statement that 90% of the world's electronics are made at least partly in Shenzhen is not accompanied by a source, a definition of electronics, or a calculation.

I traced the figure to a 2015 Inc. article by Will Yakowicz, which simply states that Shenzhen makes 90% of the world's electronics. It supplies no underlying study or statistical method. A November 2017 World Economic Forum article repeats the number and explicitly cites Inc. rather than independent research. The documentary repeats the same extraordinary figure without establishing it.

Shenzhen's official statistics tell a more measurable story. In 2023, its new-generation electronic information industry generated 571.7 billion yuan in value added, while the city's total GDP was 3.46 trillion yuan. These are substantial figures, but they do not establish a 90% share of global electronics production. Shenzhen Municipal Statistics Bureau.

A statistic repeated by several publications does not become independently verified merely because the publications repeat it. This is precisely the kind of number that should have been checked before becoming the documentary's headline evidence.

Automation strengthened China's position instead of sending factories home

The documentary treats automation as an escape route for Western manufacturers: if robots cost the same in America and China, American factories could return home and save on shipping. That argument ignores the expensive infrastructure surrounding a robot: specialist suppliers, tooling, technicians, component availability, production experience and established distribution networks.

China became the world's largest industrial-robot market rather than surrendering that advantage. The International Federation of Robotics reports that China installed 354,000 industrial robots in 2025, representing 59% of worldwide installations. Automation was being deployed on an extraordinary scale inside Chinese factories, not simply driving production out of the country. International Federation of Robotics.

The productivity figures make the distinction even clearer. UNIDO's data show that between 2015 and 2023, China's manufacturing value added increased approximately 50%, while manufacturing labour productivity rose approximately 45% and employment remained broadly flat. In Asia outside China, manufacturing value added increased approximately 27% and employment 13%. UNIDO's manufacturing employment and productivity analysis.

The documentary correctly anticipated diversification into cheaper Asian economies. What it failed to anticipate was that China could automate, increase productivity, retain enormous production capacity and expand its industrial capabilities simultaneously. The story was not China versus the rest of Asia; it was China becoming the centre of a more distributed Asian manufacturing network.

The tax and currency explanation is too convenient

China did manage its currency, but the documentary treats currency management as proof of an artificially depressed exchange rate without measuring the alleged advantage. The IMF's account is more complicated: China moved away from its fixed dollar peg in 2005, allowed substantial appreciation through 2008, stabilized the exchange rate during the financial crisis, and resumed greater flexibility in 2010. The renminbi remained managed, but the documentary leaves out the substantial appreciation that occurred over the intervening years. IMF, China's Evolving Exchange Rate Regime.

Its description of export taxation is more misleading. China did not simply exempt every export from tax. A 2014 World Trade Organization review documented export duties on approximately 4.2% of tariff lines, with interim duties on 320 product lines in 2013. The OECD's 2019 analysis also explains that Chinese exporters received different VAT rebates depending on the product, ranging from no refund to a full refund.

The American comparison is equally careless. The United States does not impose a universal VAT, but it does impose customs duties on imported goods according to product classification and applicable trade rules. Its own Customs and Border Protection agency confirms that imported merchandise is governed by the Harmonized Tariff Schedule. Some tablets may qualify for zero-duty treatment; that does not make American imports generally tax-free. U.S. Customs and Border Protection.

These distinctions matter. Export rebates, export duties, import tariffs and exchange-rate management are different policy instruments. Treating them as two countries simply choosing whether to tax trade conceals how the actual competitive advantages work.

Anker's access shaped the evidence, even if it did not control the conclusions

The documentary's central corporate case study was also the company that brought its creator to Shenzhen. At the end, the narrator explicitly thanks Anker for arranging the visit, says the company gave him complete creative control, describes his long-term use of its batteries, mentions receiving an upgraded product, and promotes its competition offering ten winners $2,000 each plus Anker products.

That relationship was disclosed. There is no evidence in the documentary that Anker dictated the editorial conclusions, and claiming otherwise would be unjustified. But editorial freedom and independent evidence are different things.

The problem is the sampling. Anker's CEO supplies the explanations for Shenzhen's development speed, PowerIQ's competitive advantage and the failure of AnkerBox in Seattle. The latter is particularly weak: the CEO attributes the unsuccessful trial to differences in American and Chinese bar culture, but the documentary provides no customer research, sales figures or examination of other possible causes.

A company's explanation of its own failed product is a useful interview, not an independently established explanation of an entire market. The film repeatedly crosses that line.

The opening statistics also need definitions

The documentary begins by claiming China produced 50% of global GDP around AD 1000. A 2020 Tsinghua-led reassessment of China's historical economic share instead estimates a peak of 34.6% around 1600. Historical GDP reconstruction is difficult and estimates differ, but a figure as dramatic as 50% requires a source and methodology, neither of which the film supplies. Tsinghua University research.

The claim that Shenzhen already had 18 million people also needs qualification. It was not necessarily invented: a Shenzhen government planning document estimated an actual managed population of approximately 18 million, while the official permanent-resident count was substantially lower. The documentary does not explain the distinction before comparing Shenzhen with New York and London. Shenzhen's population plan.

Cheap labour is not a complete manufacturing cost calculation

The narrator compares American factory wages of $10–$15 per hour with Shenzhen wages of $3–$4, then supports China's lower costs with personal observations about inexpensive meals and taxi rides.

But where is the comparison of equivalent jobs, working hours, employer contributions, productivity, production yields and total manufacturing costs? A $3 meal proves that lunch was inexpensive. It does not prove that a smartphone costs less to manufacture.

The documentary itself acknowledges that workers increasingly wanted permanent urban lives and that Shenzhen housing had become exceptionally expensive. Those observations undermine the simplistic idea that Chinese manufacturers could preserve their advantage indefinitely by paying workers less. Yet the film never calculates how those rising costs compared with productivity improvements or the benefits of remaining close to suppliers.

The important economic question was never simply how much an individual worker earned. It was how much it cost to manufacture a finished product, at the required quality and speed, in each competing location.

The industrial cluster was the real story

There is a much stronger explanation for Shenzhen's success than the documentary's wage comparisons: the concentration of the entire hardware-development process in one place.

A designer can source components, consult engineers, modify a prototype, arrange production and reach suppliers without coordinating across continents. The advantage compounds with every supplier, specialist and manufacturer joining the network.

A 2024 Financial Times investigation of Shenzhen's Huaqiangbei electronics district described prototype development cycles of just 24–48 hours. This independently reinforces the documentary's original observation about rapid prototyping. Financial Times, China tech — Shenzhen speed.

This is the competitive advantage the documentary should have investigated more deeply. A robot can be purchased in America, but the accumulated network of component suppliers, tooling specialists, engineers and experienced factories cannot simply be purchased and relocated.

Verdict: A compelling field trip is not an economic forecast

Wendover Productions documented something important about Shenzhen, but confused evidence of a manufacturing advantage with proof of its future trajectory.

The 90% electronics statistic has no established statistical foundation. The historical GDP figure is questionable. The wage and tax comparisons oversimplify international trade. The Anker interviews provide valuable access but cannot establish industry-wide conclusions. Most importantly, the prediction that automation would encourage manufacturing to return to Western countries underestimated China's ability to automate its own industrial base.

China did not need to preserve every low-wage factory to remain dominant. It could allow some labour-intensive production to spread across Asia while increasing productivity, investing in robotics, developing domestic technology and retaining the supplier networks that make rapid manufacturing possible.

That is the story the documentary nearly discovered but never properly told.

Shenzhen's achievement was not simply making things cheaply. It was building an industrial ecosystem so dense that moving a factory elsewhere could mean abandoning much of the advantage that made the factory successful.

The documentary explains why Shenzhen won yesterday. Its biggest mistake was assuming it understood why Shenzhen might lose tomorrow.

References

1. Why Chinese Manufacturing Wins — Wendover Productions, August 8, 2017. Original documentary and source of the claims, Anker interviews, predictions and commercial disclosures.

2. 5 Things to Know About Doing Business in Shenzhen — Will Yakowicz, Inc., February 24, 2015. The traceable source for the 90% electronics assertion; provides no supporting study or calculation.

3. The Astonishing Rise of Shenzhen, China's Gadget Capital — Briony Harris, World Economic Forum, November 9, 2017. Demonstrates how the unsupported Inc. statistic was subsequently repeated.

4. Shenzhen 2023 Statistical Communiqué — Shenzhen Municipal Statistics Bureau, April 28, 2024. Reports the city's electronics-industry value added and provides measurable local industrial statistics.

5. International Yearbook of Industrial Statistics 2025 — United Nations Industrial Development Organization, 2025. Documents China's 32% global manufacturing share and the relationship between output, productivity and employment.

6. Five Million Robots Now Operate in Factories Globally — International Federation of Robotics, September 24, 2026. Reports China's 354,000 industrial-robot installations in 2025 and its 59% share of global deployments.

7. China's Evolving Exchange Rate Regime — Sonali Das, International Monetary Fund, March 7, 2019. Explains the evolution of China's managed exchange-rate system and the limits of describing it as a simple fixed currency peg.

8. Trade Policy Review: China — World Trade Organization, May 27, 2014. Documents Chinese export duties, restricted products and the distinction between VAT treatment and export taxation.

9. Measuring Distortions in International Markets: The Aluminium Value Chain — Organisation for Economic Co-operation and Development, January 8, 2019. Explains how selective VAT rebates and export duties alter the competitive position of different Chinese export products.

10. Harmonized Tariff Schedule — General Information — U.S. Customs and Border Protection, February 13, 2026. Establishes that American imports are governed by product-specific customs tariff classifications.

11. The Evolution of China's Share in the Global Economy: 1000–2017 AD — Jin Xingye, Guan Hanhui, Li Daokui and Stephen Broadberry, Tsinghua University, August 13, 2020. Provides an alternative historical reconstruction that places China's peak global GDP share at 34.6% around 1600.

12. Shenzhen's Population and Social Development Plan — Shenzhen Municipal Government, November 9, 2016. Distinguishes the estimated 18-million managed population from the smaller permanent-resident population.

13. Transcript: China Tech — Shenzhen Speed — Financial Times, May 14, 2024. Provides independent reporting on rapid prototyping and the continuing importance of Shenzhen's electronics supply-chain ecosystem. 【turn454925news114】