Compare study and work explicitly: three years of early earnings versus decades of annual income differences
Put both choices on paper. Starting work early gives you three years of earnings in jobs with low entry requirements, once. Continuing education may give you an earnings increment every year for the remaining 30–40 years of working life. Urban Chinese estimates associated an additional year of education with 4.0% higher income in 1988 and 10.2% in 2001. These old figures indicate an order of magnitude, not today's expected return.
What it takes
Nothing. Set aside about half an hour to calculate.
What you may gain
The study estimated returns to schooling in urban China from 1988 to 2001: the extra earnings associated with one more year of education. It reported a substantial rise from 4.0% per year of schooling in 1988 to 10.2% in 2001, mainly after 1992. This reflected a growing higher-education wage premium—the earnings difference associated with having attended university. The estimates offer a historical Chinese scale for converting extra schooling into an annual earnings difference. For the global average, see entry 7.
Context & considerations
First subtract the tuition waivers and grants in entry 4; also consider the mortality association in entry 2. These data end in 2001 and cover urban areas only. Neither they nor the global 9% average in entry 7 can be treated as a current personal forecast. Results will differ by person. A costly, lengthy program poorly matched to available jobs may genuinely fail the calculation, but that establishes that this particular program is poor value, not that education is useless. Some factors do not fit neatly into the formula: formal educational entry requirements and pathways, discussed in entry 5; resistance to replacement, in entry 10; and the ability to change industries. You are the beneficiary.
Research & references
Zhang J, Zhao Y, Park A, Song X (2005). Economic returns to schooling in urban China, 1988 to 2001. Journal of Comparative Economics, 33(4), 730–752. https://doi.org/10.1016/j.jce.2005.05.008