Bernie Sanders on inequality: what the evidence says · JRE #1330
“Over the last 30 years, the top 1% has seen a $21 trillion increase in their wealth. The bottom half of America has seen a $900 billion decline in their wealth.”
What the evidence says
Sanders has used versions of this "$21 trillion gain for the top 1%, $900 billion decline for the bottom half" statistic repeatedly since around 2019, typically attributing it to Federal Reserve wealth data. The Federal Reserve's Distributional Financial Accounts (DFA), compiled from the Survey of Consumer Finances, do show a real and well-documented trend: the top 1%'s share of total U.S. household wealth has risen substantially since the late 1980s/early 1990s, while the bottom 50%'s share has remained a small, sometimes-declining fraction of total wealth (roughly 1-3% of aggregate net worth in most years). However, the specific paired figures Sanders cites do not correspond cleanly to any single published Federal Reserve table for a fixed 30-year window; the bottom half's aggregate wealth in nominal dollars has generally risen over most 30-year spans (not declined by $900 billion), even though its wealth share has stagnated or fallen and its wealth is heavily concentrated in home equity and vulnerable to shocks like the 2008 crash. The $21 trillion figure for top-1% gains is roughly in the plausible range depending on the exact start/end dates chosen, since top-1% wealth has grown by tens of trillions of dollars in various multi-decade windows, but Sanders and other progressive politicians (including Elizabeth Warren, who cited similar figures) have used varying timeframes and slightly different numbers across speeches, making the claim difficult to verify as a single precise fact rather than an approximate, cherry-picked framing of real underlying Fed data. The overall direction, sharply rising wealth concentration at the top since the late 1980s, is well supported by Federal Reserve data; the specific dollar figures and the claim of an outright dollar decline for the bottom half are not clearly substantiated and are best characterized as an imprecise, rounded talking point built on real but loosely applied statistics.