A recent Gallup poll indicates that 89% of U.S. adults believe government corruption is widespread, marking the highest level of concern in 20 years. This figure represents a 10 percentage point increase from the previous year and is significantly higher than the 72% to 79% range reported between 2010 and 2025.
The perception of corruption has increased most notably among Democrats, with 91% stating that corruption is widespread in 2026, up from 57% in 2024. In comparison, 90% of political independents share this view, while 83% of Republicans do. The Republican perspective has remained relatively stable over the past two years, with around 80% consistently indicating that corruption is widespread.
The differing perceptions among party lines suggest that Democrats' views may be influenced by the party in power, while Republicans maintain a consistent view of government corruption regardless of political leadership.
In an international context, the U.S. has diverged from other advanced economies regarding perceptions of government corruption. According to Gallup data, the median perception of corruption among OECD countries has decreased from 69% in 2009 to below 60% since 2022. In 2025, the median perception of corruption among OECD countries was reported at 59%.
The U.S. now ranks highest among OECD countries in perceived government corruption, with a 20-point gap compared to the OECD median. Only four countries have reported higher levels of perceived corruption than the U.S. in recent years.
Additionally, perceptions of business corruption in the U.S. have also risen, reaching 71%, but remain significantly lower than perceptions of government corruption. The gap between perceptions of government and business corruption is the largest recorded in U.S. history.
The findings reflect a rare moment of bipartisan agreement among Americans regarding concerns about government corruption, with the difference in concern between Democrats and Republicans being marginal.
The data is based on telephone interviews conducted from May 1 to June 10, 2026, with a random sample of 1,000 U.S. adults, and has a margin of sampling error of ±4.1 percentage points at the 95% confidence level.