But the findings are surprising and build an argument against raising grades.
Students who received soft grades were less likely to succeed in subsequent courses, post lower test scores later, were less likely to graduate from high school and enroll in college, and earned fewer years later.
The economic costs are not small. Denning estimates that when a teacher issues the highest grade (0.2 points or more on a 4-point scale, the difference between a B and a B-plus), a student in that class loses about $160,000 in lifetime earnings, measured in current dollars.
That is the result of one teacher, in one year. If the student meets several teachers who raise the level, the loss increases.
Evidence from two very different sources
Researchers tested students in two settings: Los Angeles and Maryland.
The Los Angeles Unified School District provided data on nearly a million high school students from 2004 to 2013, a time when the graduation rate was just over 50 percent. More than 70 percent of students were in Spanish, and failing grades were common.
The Maryland data followed about 250,000 high school students from 2013 to 2023. Graduation rates exceeded 90 percent, and the student population was very racially diverse. The Maryland data allowed researchers to track college enrollment, employment and earnings, while the Los Angeles data ended with high school.
Despite these differences, the pattern was similar.
Students taught by soft graders — defined as teachers who gave higher-than-expected grades based on standardized test scores and the student’s previous performance — did worse later in high school. In Maryland, where there was data on college and jobs, these students were also less likely to attend college or be employed, and earn less money.
Seeing the same pattern in two very different systems reinforces the case that this is not a one-state crisis or one-policy regime.
When letting go helps and when it doesn’t
Research makes an important difference. Teachers who still make A’s a challenge, but make it easier to pass — turn failing into low grades — have helped many students graduate from high school, especially those at risk of dropping out. That short-term benefit is real. For some students, passing Algebra I instead of failing it can keep them on track to graduate and possibly enroll in community college.
But the benefit ends there. Those students show no long-term gains in college graduation or earnings. Compassion helps them overcome an obstacle, but it doesn’t build the skills they need later.
In contrast, general grade inflation (teachers raising grades across the board, from C to B to A) does not reflect a price increase and hurts students’ chances of future success.
Why good intentions backfire
Research cannot explain exactly why higher grades lead to worse outcomes. But the machine is not hard to imagine. In a class with a low student, a bright student may quickly realize that he does not need to study hard or complete all the homework. If he earns a B in Algebra I without learning how to insert or solve quadratic equations, knowledge gaps follow him in geometry and beyond. He may come back again. Over time, deficits add up. Confidence fades. Learning is slow. In college or at work, the results are seen as lower skills and lower wages.
As Denning put it during the presentation, there appears to be a “chain of causes” for harm, even if he can’t directly measure how young students are learning or how far behind they are.
Don’t be too quick to blame the teachers
Increasing grades is not always a teacher’s decision. The 2025 survey documents the frustration of many grading teachers who say they feel pressured by administrators to comply with “equitable grading” policies that ban zeros, allow unlimited retakes and eliminate late work penalties.
Elementary students are not bad teachers. Research finds that they tend to be better at developing non-cognitive skills. Their students are better behaved, more cooperative, and less likely to be suspended. However, in this study, that does not translate into better health outcomes, as one would hope.
Students with strong grades tend to be better at raising student test scores in math, reading and other subjects. Despite that correlation, that doesn’t mean that all grade-schoolers are good teachers. Others are not.
This is preliminary research. More research is needed to understand whether there are similar workplace costs from college-level income. And there are questions about whether boys respond differently than girls to higher levels.
Teachers struggle to engage students in learning, which is fraught with obstacles, frustration and boring repetition. Perhaps low grades will not motivate students to do this hard work. But this preliminary evidence suggests that increased grades are not beneficial.
Contact a staff writer Jill Barshay at 212-678-3595, jillbarshay.35 at Signal, or barshay@hechingerreport.org.
