In recent years, as economic inequality in America has grown, many researchers have tried to figure out just how much wealth and income those at the top have. These figures then come to stand for the magnitude of economic inequality in the U.S. One such finding is that of Emmanuel Saez, the Berkeley economist and Thomas Piketty-collaborator, who has found that America’s 1 percent’s share of total U.S. income is roughly 20 percent.

But a new paper from the Brookings Institution challenges this and other calculations of Saez and Piketty, arguing that while wealth and income inequality are indeed increasing (nobody disputes that), they haven’t done so at the magnitude that Piketty and his colleagues have found. The paper, authored by three Federal Reserve economists—Jesse Bricker, Alice Henriques, and John Sabelhaus—and Jacob Krimmel of Wharton, found that from 1992 to 2012, the top 1 percent’s share of wealth rose by 6 percentage points to 33 percent. This is substantially lower than estimates by Saez and his colleagues, which estimates that the share of wealth held by the 1 percent is 42 percent. The share of income estimate is more similar: The Brookings paper found that the share income earned by the 1 percent is 18 percent, while Piketty and Saez’s 2012 estimate is 23 percent.

So why the difference in these estimates? The Brookings paper uses a similar set of data and administrative tax records, but also incorporates the Federal Reserve’s Survey of Consumer Finance household survey as a macro data source. Additionally, the authors incorporated noncash government and employer transfers, such foodstamps, Medicaid, and health care, into their analysis. (Though they note that the current data sources don’t incorporate many employer-provided benefits and Social Security, which they believe would make inequality measures more accurate.) The researchers argue that methodology, data sources, and the mismatch between micro and macro data is the reason there’s a gap in these estimates.

“The macro data are key for understanding top wealth and income shares, because changes in the aggregate composition of income and balance sheets over time effect for whom the micro data are comprehensively capturing resources. Further, the two micro data sources measure income and wealth differently and rely on different income and wealth concepts. The biases that arise from different measurement and concepts can be quite large,” the researchers wrote.

But in conclusion, the researchers of the Brookings paper concede that though their estimates are lower, it’s still in line with the overall trend: “The estimates agree with the widely-held view that inequality, at least as reflected in top wealth and income shares, has been rising in recent decades. However, the levels and trends in our preferred top share estimates are more muted than in recent studies which are based directly on administrative income-tax data, but the levels and trends for top wealth shares are a bit larger than estimates based on estate-tax data.”