What is different about manufacturing in China in 2020?

The answer to this question is, it depends.

In the 1990's and 2000's, the de-facto way to get almost any product manufactured is to find a contract manufacturer in China and get it made there. In those days, R&D is typically done Stateside for US based companies, and manufacturing is dominated by Chinese factories. This is because labor costs were much lower, and component manufacturing took hold quickly in China. This means that final assembly of finished goods is also more easily done in China where the supply chain is fully developed to support mass production of all types of products (but especially low-value, high volume products including all sorts of consumer electronics devices).

Fast forward to 2020. Times have changed. China has gone from a production economy to a consumer economy. Wages and land prices are going up exponentially. The Pearl River Delta (also called the Greater Bay Area) in South China used to be a hotbed of manufacturing . Now Shenzhen is a major tech metropolis, and manufacturers can no longer afford to stay in Shenzhen.

Prices are going up everywhere within China. Young people have much better choices than work on the manufacturing floor putting cell phones together. It has been increasingly difficult for Chinese manufacturers to hire direct labor - the same trend that US and other developed economies experienced 10 or 20 years ago.

As a result, the low value, high volume manufacturing that characterized China's manufacturing sector is increasingly moving to lower cost countries like Indonesia and Vietnam. Wages are coming up quickly in those countries too, and eventually, off shoring will no longer make sense. For startups experimenting with their products, in many cases it is easier and more cost effective to find a manufacturer in North America to build their product and then investigate off shore locations as their business scales.

These days the right way to think about it is not "off-shoring" but "right-shoring".  A hardware startup would do well to consider its specific needs and make informed decisions about where to design and develop their product, where to make prototypes and iterate quickly, where to build their first production run, and where to scale their manufacturing. It may not be one country at the end of the day.