Beyond digital access: how internet use patterns shape labour income returns—evidence from China
Using Chinese panel data, this study demonstrates that while the mere digital divide does not significantly boost earnings due to selection bias, the specific composition of internet use—particularly productive and mixed activities—drives substantial income gains that are disproportionately higher for women and older workers, suggesting that policy should prioritize converting access into productive engagement rather than simply expanding connectivity.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the modern world, the internet is often treated as a single, uniform tool, like a road that everyone can drive on. Economists have long debated whether simply having access to this road makes people richer. The prevailing theory suggests that when workers can go online, they find better jobs, learn new skills, and connect with more opportunities, leading to higher wages. This idea has driven massive public investments in connectivity across the globe, including in China, where hundreds of millions of people are now online. However, a critical question remains: does the internet itself create wealth, or do people who are already skilled and successful simply happen to be the ones who go online? To answer this, researchers must look beyond the simple fact of connection and examine how people actually use the technology. They must distinguish between using the web to learn and work, versus using it for entertainment, and determine if these different habits lead to different financial outcomes.
A team of researchers from Thammasat University set out to solve this puzzle using data from China, a country with more than 1.1 billion internet users. They analyzed a massive, long-running survey of Chinese families that tracks the same people over several years, allowing them to see how individual lives change over time. Their goal was to separate the effect of simply being online from the effect of what people do while they are there. They broke down internet activity into three distinct categories: productive use, such as taking online courses to learn new skills; mixed use, like shopping online which involves comparing prices and managing transactions; and consumptive use, such as playing video games for entertainment. By comparing the income of people who used the internet for these different purposes, they sought to find the true source of any financial benefit.
The study first confirmed what many previous reports had suggested: in a simple snapshot of the population, people who use the internet earn significantly more than those who do not. The data showed a premium of roughly eleven percent, meaning internet users appeared to earn about that much more than non-users. However, when the researchers looked deeper, using the long-term data to track the same individuals over time, this advantage vanished. They found that the higher earnings were not caused by the internet itself. Instead, the people who earned more were already more capable and better educated before they ever went online. Once the researchers accounted for these pre-existing differences, the idea that simply connecting to the internet boosts income collapsed to essentially zero. The initial gap was not a reward for going online; it was a reflection of who chose to go online in the first place.
The real story emerged when the researchers looked at how people used the internet once they were already connected. They discovered that the type of activity mattered immensely. People who used the internet for productive tasks, like online learning, or for mixed tasks, like shopping, saw a clear financial boost. These groups earned approximately eight to nine percent more than their peers who were online but did not engage in these specific activities. In stark contrast, those who used the internet primarily for gaming saw no financial benefit at all. The data showed a precise zero effect for gaming, suggesting that using the web for pure entertainment does not translate into higher wages. This finding indicates that the value of the internet lies not in the connection itself, but in the quality of the engagement. It is the act of processing information, learning, or managing economic transactions that drives income, not the passive consumption of content.
Perhaps the most surprising discovery was who benefited the most from these productive uses. Conventional wisdom often suggests that technology favors the young, the urban, and the highly educated. Yet, this study found a complete reversal when it came to online learning. Women, older workers between the ages of forty-six and sixty, and people living in rural areas gained the most from using the internet to learn. These groups saw returns that were two to three times larger than those enjoyed by men, younger workers, or city dwellers. This suggests that for people who have historically faced barriers to education and job information, the internet acts as a powerful equalizer. While the initial access to the internet might still favor the privileged, the ability to use it productively offers a unique opportunity for those who have been left behind to catch up.
The researchers also explored why these benefits occur. The evidence points to two main pathways. First, the internet provides access to information that was previously hard to find, helping people understand job markets and wage standards. Second, and more strongly, productive internet use appears to help workers secure formal employment contracts, which come with better pay and protections. This shift from informal to formal work seems to be a key mechanism for the income gains. The study concludes that policies focused solely on expanding internet access are insufficient. To truly improve livelihoods, especially for disadvantaged groups, the focus must shift to digital literacy and encouraging productive use. Simply handing someone a connection is not enough; the real dividend comes from teaching them how to use that connection to learn, work, and build a better economic future.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.