Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
Fact check: How many registered voters were in the United States in 2020 according to the Census?
Executive Summary
The U.S. Census Bureau’s Current Population Survey (CPS) estimates that 168.3 million U.S. citizens of voting age were registered to vote for the 2020 election, and about 155 million people actually cast ballots, producing a reported turnout rate of 66.8% of citizens 18 and older [1] [2]. Claims circulating that only 133 million people were registered in 2020 conflict with the Census CPS figures and with subsequent fact-checking that cites the Census estimates [3] [2]. These differences arise from mixing alternative measures and administrative lists rather than a single Census-based metric [4].
1. What people claimed and why it matters — a close look at the competing numbers
Multiple public claims presented a low figure — commonly 133 million registered voters for 2020 — as proof that official turnout or counts were inflated. The Census CPS data provide a different measure: 168.3 million registered citizens of voting age in 2020, which is higher than the 133 million claim and therefore undermines that specific argument [1] [3]. Separately, the CPS reports 155 million actual voters, a figure that explains discrepancies between registered totals and ballots cast and highlights why comparing different numbers without context can mislead. The debate has real implications for how people interpret turnout, registration completeness, and claims about election integrity; the CPS measure is a nationally consistent household survey estimate, not an administrative tally of state voter lists, and that distinction matters when parties and advocates push simplified comparisons [1] [4].
2. What the Census CPS actually reports — the numbers and the methodology
The official Census reporting tied to the 2020 election and the CPS indicates 168.3 million citizens of voting age reported being registered to vote in 2020, and 155 million were reported as having voted, yielding a citizen turnout rate of 66.8% [1]. The CPS is a household survey administered by the Census Bureau with questions on registration and voting; it produces estimated counts that are widely used for demographic breakdowns and trend analysis [4]. Because the CPS is survey-based, it captures self-reported registration status among citizens and is designed to provide comparability across demographic groups and time, which differs from state-maintained voter registration databases that can include noncitizens, duplicates, or inactive records depending on local practice [1] [4].
3. Why different sources and social-media claims diverge — three common confusions
Three recurring reasons explain divergent totals: first, the difference between administrative state voter rolls and the CPS household survey; second, varying definitions of the universe being counted (citizens only versus all adults, registered versus eligible); third, misreading turnout versus registration statistics [4] [1]. Social-media claims that assert 133 million registered voters commonly conflate partial administrative extracts, selective subsets, or outdated roll-cleaning snapshots with the Census survey estimate, producing a lower figure that does not match the CPS methodology or the Census’ published tables [3] [1]. Fact-checkers and analysts who compared the 133-million claim to the CPS and other official sources concluded the 133-million number is inconsistent with the Census’s 168.3 million estimate and therefore inaccurate as a Census-based claim [3] [2].
4. What the numbers imply for turnout, registration efforts, and oversight debates
The CPS figures—168.3 million registered, 155 million voting, 66.8% turnout of citizens—indicate a high level of registration and participation relative to recent history and frame policy debates about registration accuracy, voter-list maintenance, and access to registration [1]. Analysts focused on election administration point out that survey-based registration estimates and administrative roll totals serve different analytic purposes: CPS is better for demographic trends and turnout rates, while state lists are operational tools for election management; conflating them can create misleading narratives about fraud or undercounting [4] [1]. Fact-checking published in 2023 reiterated that simplified social-media claims omitting these distinctions lead to false conclusions about the size of the registered electorate [3].
5. Bottom line, sources, and where to go next for verification
The Census Bureau’s CPS-based reporting for the November 2020 election shows 168.3 million registered voters and 155 million actual voters, with a 66.8% turnout among citizens 18 and older; claims of 133 million registered voters contradict those CPS figures and were judged inaccurate by subsequent fact-checks [1] [3]. For verification, consult the Census “2020 Presidential Election Voting & Registration Tables” and the CPS voting and registration data files for exact table values and methodology notes [1] [4]. Differences between administrative voter rolls and CPS estimates explain most of the public confusion; authoritative comparisons must state which measure is being used before drawing conclusions about election results or registration levels [4] [1].