As China’s economy advances, investments into public education expand

Historically speaking, China’s public education system is very new. China has a centuries-long tradition of using its imperial examination system to select its corps of civil servants, but for much of this time education was primarily accessible to social elites only. Universal education first began in earnest with the founding of the People’s Republic of China in 1949, when the Chinese Communist Party saw education and literacy as important elements of peasant empowerment and a means to promote rapid industrialization. Swept up in political turmoil, public schools nationwide shut down for several years during the Cultural Revolution in the late 1960s; they slowly reopened over the following years, with merit-based university examinations reinstated in 1973.

Deng Xiaoping, who eventually became China’s paramount leader shortly after Chairman Mao Zedong’s death in 1976, was a pragmatist who emphasized public education—especially in science and technology—to promote economic development. It was during Deng’s leadership that compulsory education (through the equivalent of 8th grade in the U.S.) was formally adopted into law in 1986.

This same period also marked the beginning of a 4% education spending target, which eventually became a key benchmark in China. In 1982, the government commissioned a group of preeminent Chinese scholars to conduct a cost and benefit analysis to determine the appropriate expenditure on public education as a percentage of the nation’s GDP. Four years later, the project issued its report comparing foreign countries’ expenditures on education relative to GDP and recommended a target of 4%. China’s spending on education at the time was far below this level, but the commission encouraged incremental increases over time to meet the goal by the year 2000.

Though both compulsory education and the 4% spending targets were settled on separately in 1986, they became integrated over time. The 4% spending goal was later adopted into policy documents and five-year plans, with frequent references to it as a policy yardstick. The 1995 Education Law not only formally adopted the spending guideline but also promoted “three increases” guiding how the funding was to be spent; these guidelines explicitly called for teachers’ salaries and average per-pupil spending to steadily increase over time. In other words, education spending that expanded merely in proportion with student enrollment was insufficient, meaningful increases in spending directed toward students and teachers were a priority.

China failed to meet the 4% goal by its original 2000 deadline but persisted and met it in 2012. Since then, China’s education spending has consistently exceeded the 4% mark. The 4% benchmark maintains its significance even today: The Ministry of Education speaks of the last decade as entering a new era with new priorities in addressing inequities in school quality while maintaining spending above 4% of GDP, and the Ministries of Education and Finance jointly founded the 4% Office upon completion of the goal to ensure progress on education spending does not roll back.

Intrigued by China’s education spending approach relative to GDP, we conducted our own cross-country analysis of trends in both China and the U.S. We collected China data from the National Bureau of Statistics and U.S. education spending data from the Digest of Education Statistics at the National Center for Education Statistics. We converted all spending numbers to 2017 U.S. dollar values using the Consumer Price Index. To keep the numbers in both countries as comparable as possible, public spending on all levels of education (K-12 and higher education) were included from all government sources (including federal, state, and local sources). GDP numbers come from China’s National Bureau of Statistics and the U.S’s Bureau of Economic Analysis.

Figure 1 below shows how total public education spending, as a percentage of GDP, has shifted in both countries from 1998 through 2020. As Figure 1 shows, China’s education spending as a share of GDP saw (mostly) steady increases over the 2000s, going from 2.55% of GDP in 1998, surpassing 4% in 2012, and then fluctuating in the low 4% range since.

Education spending in the U.S., meanwhile, has been comparably stable over this period, only modestly fluctuating from slightly less than 5% in the late 1990s to highs near 5.5% through much of the 2000s and dropping slightly since. Readers familiar with the context of American educational policy will recognize the surges in education spending during the 2000s as coinciding with the No Child Left Behind Act, and spending slowly tapered following the Great Recession and the exhaustion of federal stimulus funds.

We also broke these spending numbers down to the state level in the U.S. and to provinces (and other provincial-level jurisdictions) in China, to understand whether there are underlying spending patterns related to economic development across different jurisdictions. Figure 2 presents a scatterplot of these spending figures for 2020, where the blue dots on the left of the figure represent Chinese provincial jurisdictions and the orange dots represent U.S. states (Washington, D.C. was omitted from the analysis as a high outlier).

Interestingly, the range of education spending as a percentage of GDP (on the y-axis) across jurisdictions is roughly the same across both the U.S. and China, where the mass of observations in both countries have values between about 4% and 7%. China does show a higher spread across provinces, which ranges from lows below 3% in Fujian Province (2.66%) and Jiangsu Province (2.67%) to a high outlier of over 16% in the Tibet (Xizang) Autonomous Region; Qinghai Province is the next highest at 9.03%. Compare this against the smaller range observed in the U.S., which had a low of 3.85% in Nevada to a high of 9.07% in New Mexico.

And in both countries, we see a consistent relationship where the most economically developed jurisdictions (along the x-axis) spend relatively less on education as a GDP percentage (the strength of this relationship is notably weaker in the U.S. compared to China). For example, Beijing (China’s richest jurisdiction) spent 3.65% of its 2020 GDP on public education; compare this against 4.91% spent in New York (the state with the highest per-capita GDP). Both values are below the mean spending levels in their respective countries.

But we need some caution in interpreting here. These figures represent spending as a share of GDP, this does not necessarily imply actual spending in nominal values was larger in poorer jurisdictions; they’re often not. Even if school spending were perfectly equal across settings, a larger economy in a jurisdiction mechanically lowers education spending as a share of its GDP. Yet, at the same time, through the lens of purchasing power, it costs more money to maintain school buildings and pay teachers in New York than it does in Nevada. The number of students in school may also vary across jurisdictions, too. Consequently, this GDP-benchmarked measure alone cannot say which jurisdiction is over- or under-spending.

Beyond offering a starting point for our foray into this series investigating China’s public education system, this discussion of spending offers several key insights. We offer three for consideration here, discussed in the context of American research on these issues.

First, we should pay attention to GDP share as a useful benchmark of education spending. Those scholars initially tasked with developing a percentage recommendation for education spending in China felt the framing was itself a novel way of approaching the issue. As we learned more about this policy, we agreed. While this metric is common in international comparisons now, it feels novel as a guiding strategy for spending allocations in the school system. Digging deeper, we found that American spending on K-12 education is quite consistent with international patterns (given the level of economic development), and U.S. higher education spending is much higher. This is consistent with the U.S.’s status as a net exporter of higher education. This measure should be considered a benchmark only, as its limitations were also apparent in trying to make sense of subnational spending measures; but it offers a useful perspective that spending numbers alone miss.

Second, runaway education spending is surprisingly stable through this GDP lens. Reports on public education spending in the U.S. over time point out, correctly, that even after adjusting for inflation, education spending continues to climb—for example, more than doubling since 1977. These narratives have spawned fears about the sustainability of public education spending and the diminishing marginal production of schools. Some analysts have even warned that as costs climb (particularly in a labor-intensive sector like education), they may potentially become a crippling disease. Yet, disease symptoms have largely not materialized in practice. Our analysis corroborates this conclusion, though from a different approach: education spending in the U.S. as a share of GDP has demonstrated remarkable stability in our analysis, despite the growth in real spending amounts. The disconnect between these two trends is explained by economic growth itself, where GDP has roughly tripled since 1977. So long as the growth in education spending does not exceed real GDP growth, education spending is sustainably contained within the economic budget, so to speak. Many educators are concerned that the incoming Trump administration may be eyeing public schools for major spending cuts, motivated in part by these increases in school spending over time; but through the lens of GDP, these increases are hardly noteworthy and new cuts will represent a significant setback.

Third, a GDP benchmark could help smooth education allocations over time. Arguably, the most common allocation approach in the U.S. is to minimize spending necessary to meet the task and then adjust annually as needs arise (at least this is the case for K-12 funding at the state level). Consequently, spending amounts stay mostly flat within states (adjusted for inflation) but reset in episodic shocks responding to either major policy shifts or legal challenges around resource allocations. Conversely, China’s 4% approach prioritizes investments in public education that is more reminiscent of, say, setting aside a certain share of one’s personal income for retirement, smoothing that investment value over time. By indexing against GDP, school budgets automatically adjust as a sort of “dividend” from economic growth. When measured against this GDP benchmark, we find public education spending in the U.S. has been generally stable over time, despite the underlying volatility of each state (note that aggregation dampens individual volatility). Thus, from our perspective, it seems the general spending pattern is well established in the U.S., but the current model requires the input of teacher unions, advocates, and lawyers to trigger the episodic ratcheting up of resource dollars instead of automatically allocating them as the economy grows. States could analyze their own historical spending patterns and establish GDP benchmarks that could serve as guideposts for future education budgets, offering a potential avenue to mitigate these resource fights.  

Finally, both countries recognize that spending alone is insufficient to achieve their goals. In its pursuit of the 4% goal, China augmented its GDP benchmark by articulating related spending goals, such as increasing teachers’ salaries and per-student expenditures (i.e., the “three increases” principles). Even after entering the post-4% era, China continues to restate that average per-student education expenses should increase yearly and spending gaps across regions should decrease.

In the U.S., we have had our own decades-long debate over whether money matters for students. A key lesson from this debate is how schools spend new dollars can matter a lot (even if spending values alone may not predict educational gains).

We will have more to say on these priorities in future installments, including discussion of teachers’ responsibilities and their salaries—topics of active policy discussion in both countries. Stay tuned.