The empirical motivation is striking. Based on tests taken in 2022, the top five countries in the aggregated Programme for International Student Assessment score, combining mathematics, science and reading, were Singapore, China, Japan and Taiwan tied, and South Korea, while the United States ranked eighteenth. Time-use research reinforces the gap: one cited study found South Korean teenagers spend on average 49.43 hours per week studying, against an OECD average of 33.92 hours, and earlier work found adolescents in East Asia spend 5.5 to 7.5 hours per day studying compared with 3.0 to 4.5 hours in North America. The causal effect of studying on grade performance, the authors note, has been well documented in prior research.

Attribution differences are equally well established. Studies cited in the paper found that Chinese and Japanese mothers stress effort as the primary determinant of academic performance to a greater degree than American mothers, who give greater emphasis to ability, and similar East-West contrasts have been reported across multiple studies from the 1980s onward. Psychology research also links these beliefs to behavior: the incremental view, the belief that intelligence and academic skills can be increased through practice, is associated with greater academic effort, whereas the entity view, the belief that they cannot be changed, is linked to learned helplessness. A natural temptation is to conclude that East Asian students simply hold more motivating beliefs. The authors resist that move, arguing that beliefs inconsistent with observed reality are unlikely to be sustained over time.

Their alternative is a Tullock contest model with heterogeneous ability. Two players, a high-ability type and a low-ability type, each choose an effort level while competing for a prize that represents the reward for educational success relative to alternative opportunities. The probability that a player wins depends on a ratio-form contest success function in which ability and effort jointly produce effective effort through a Cobb-Douglas production function, so that ability and effort are complementary. Each player weighs the marginal benefit of effort, which depends on how it shifts the winning probability, against a marginal cost that is increasing and weakly convex in effort.

A key structural result is that in the Nash equilibrium both types of players exert exactly the same level of effort, even though ability and effort are complementary in producing performance. This happens because the marginal benefit function, which combines the complementarity effect with the diminishing marginal probability of winning, turns out to be identical for both players. The equilibrium effort level is then pinned down by a single equation involving the prize, the sensitivity parameters of the contest technology, and the two ability levels. Success attribution is modeled through the ratio of the high-ability player’s winning probability to the low-ability player’s winning probability: the larger this ratio, the more success stories are realized by high-ability students, and the more observers attribute success to ability.

The comparative statics deliver the paper’s central predictions. When the dispersion of the ability distribution is smaller, meaning the gap between high and low ability narrows, equilibrium effort rises and the ratio of success probabilities falls, so success is attributed less to ability. When the mean ability level is larger, effort again rises and the attribution ratio again falls. The intuition is that with a larger ability gap, the high-ability player can win more easily and therefore exerts less effort, while the low-ability player’s effort becomes less productive, so both scale back; meanwhile, high-ability students collect a disproportionate share of the wins, reinforcing the belief that ability matters. A higher average ability raises effort because ability and effort complement each other, and it shrinks the relative gap that drives ability attribution.

The third factor, the size of the prize for educational success, behaves differently. A larger prize increases the marginal benefit of effort for both players and therefore raises equilibrium effort, but because both players increase their effort in lockstep, their winning probabilities remain unchanged and the attribution ratio is unaffected. In other words, a society that places a high relative value on education motivates harder studying without changing what its members believe about the respective roles of ability and effort. The paper also analyzes the contest technology parameters, finding that effort is lower and ability attribution higher when ability contributes more to effective effort, and that effort is higher, with attribution unchanged, when the contest outcome is less random.

Mapping these results onto the observed cross-regional differences, the authors argue that East Asian countries combine a smaller dispersion and a larger mean in the distribution of cognitive ability with a larger relative reward for educational success than North America. Cited international IQ comparisons report higher average IQ in East Asian countries than in the United States, and the paper suggests that the United States, as a diverse country of immigrants with many geographic origins, has a more dispersed IQ distribution than relatively homogeneous East Asian countries. The United States’ large and diverse economy also rewards effort across many fields, arguably reducing dependence on formal education as the exclusive route to productive skills, whereas East Asian countries, for cultural reasons rooted in Confucianism or non-cultural reasons such as limited opportunities for people with limited formal education, place a higher value on education. All three differences point in the same direction: more effort and less attribution of success to ability in East Asia.

The authors are explicit that they are not treating attribution beliefs as a given cultural factor used to explain effort differences. Instead, both differences are derived from more fundamental conditions, and of those conditions the mean and dispersion of the ability distribution are non-cultural, while the value of education has a cultural component. They also note supporting evidence that Asian-American students’ superior academic performance over other ethnic groups in the United States is mainly attributable to greater academic effort rather than advantages in cognitive abilities or socio-demographics.

The model carries a pointed policy implication. Because success attribution beliefs are equilibrium outcomes determined by the statistical characteristics of the ability distribution, which are only marginally changeable in the short run, attempts to instill more incremental attribution beliefs in students in order to boost effort are unlikely to succeed. The paper is a theoretical exercise, not an empirical test, and its predictions rest on assumptions about contest technology, risk neutrality and the formation of beliefs from realized winning probabilities. Still, it offers a unified account of why the same regions that produce the most studying also produce the strongest belief in effort, and it suggests that changing what students believe about ability may require changing the deeper conditions that generate those beliefs in the first place.

Subject of Research: A theoretical contest model explaining cross-country differences in student effort and success attribution

Article Title: Student Effort and Success Attribution across Countries

Article References: Gronberg, T., Jansen, D. W., & Liu, L. (2026). Student Effort and Success Attribution across Countries. Atlantic Economic Journal. https://doi.org/10.1007/s11293-026-09867-1

Image Credits: AI Generated

DOI: 10.1007/s11293-026-09867-1

Keywords: student effort, success attribution, East Asia, North America, Tullock contest model, ability distribution, educational economics, PISA, study hours, theoretical model, comparative statics, education policy