Can American Adults Still Do Math?

Scores from the latest round of the Program for the International Assessment of Adult Competencies (PIAAC) were released on December 10 by the Organization for Economic Cooperation and Development (OECD). PIAAC is a large-scale assessment of adults aged 16 to 65 that measures competencies in numeracy, literacy, and problem solving. The United States plus 30 other countries participated in this latest round of testing. Two earlier rounds of testing in the US occurred in 2012/2014 and again in 2017.  

AEI’s Nat Malkus, based on US students’ results from the 2023 Trends in Mathematics and Science Study, recently argued that the “bottom is falling out for US test scores.” It’s not just students; the PIAAC survey results make clear that American adults are struggling as well. 

I have often joked that with my relatively high level of education and my commitment to reading, writing and ‘rithmetic, I was a priest of a dying religion. These newest PIAAC results have turned that joke into a sad reality. 

Here, I concentrate on numeracy. Turning first to scale scores, American adults landed 10 points below the OECD average on the numeracy scale—a score that fell by over seven points across the last decade. 

OECD also reports levels of performance, ranging from the lowest, Level 1, to the highest, Levels 4 and 5. Over one in four US adults, 28 percent, fell at or below Level 1. Even more disturbing, 12 percent of American adults are fully below Level 1, meaning they might be unable to perform basic addition and subtraction. The percentage of American adults below Level 1 is higher than the OECD’s average of nine percent and exceeds fellow OECD members like Japan (where three percent are below Level 1), the United Kingdom (nine percent), and Germany (eight percent). 

At the other end of the skills distribution are Levels 4 and 5, at which adults can understand rates and ratios as well as interpret complex graphs and statistics. These levels are reported together because of the small proportion of respondents who reach Level 5. At this higher end, American adults don’t measure up either. Just 12 percent of American adults scored at Level 4/5, lagging the OECD average of 14 percent. In Germany, 18 percent of adults reached Level 4/5. That number was 15 percent in the UK and 25 percent in Japan.  

Low numeracy doesn’t just matter in international comparisons—it drives down economic success. Numeracy affects labor force participation rates (62 percent at Level 1 versus 93 percent at Level 4/5), unemployment rates (two percent versus seven percent), and median hourly wages ($19 versus $42). I’ve focused on numeracy here, but patterns for literacy are just as bad. Indeed, the Economist, in its review of PIAAC’s new findings, headlined its story “Are adults forgetting how to read?” 

Americans today are reexamining the value of higher education based on the belief that many degree programs provide graduates with low, if any, returns on their large investment of time and money. They have reason to doubt—PIAAC also documents the mismatch between the qualifications and skills working adults have versus those that are needed in their jobs.  

According to the survey, over one-quarter of employed adults aged 25 to 65 say they are overqualified for the jobs they have, and almost 40 percent say they are over-skilled. Both these numbers are higher than the OECD averages.  

In any analysis of today’s American labor market, a common theme is the need for “upskilling” the workforce. But the already-large mismatch between skills and qualifications of workers and their jobs should give us pause.  

It could be that many American college graduates who studied gender or area studies, or who immersed themselves in Foucault rather than mathematics, feel that their skills are not being tapped in their jobs. It’s also worth asking if the students who graduate from our K–12 system have the skills they need for the job market. 

The large number of American adults that score so low on numeracy and literacy seems cause for concern, but that might be a failure of imagination. Maybe we simply won’t need the same level of literacy and numeracy in the future that we think adults should have today. For instance, platforms like YouTube have expanded beyond entertainment to become a valuable information source where viewers can learn many skills that are transmitted visually, without the need for high numeracy or literacy. 

Of course, the biggest wildcard is how the growth of artificial intelligence will affect the skills workers need in the future. AI is already rewriting the skillset needed for an ever-expanding set of jobs—and maybe it will render our current conception of the need for numeracy and literacy outdated.