Skip to content
FinObservatory

Research note: crisis event study

The bottom half held 1.7% of US net worth at both the 2007:Q4 and 2019:Q4 peaks, and the two paths afterward could hardly differ more

The Federal Reserve's Distributional Financial Accounts put the bottom 50% of US households at the same 1.7% of total household net worth at each of the last two business-cycle peaks. After the 2007:Q4 peak that share fell to 0.4% by 2010:Q4, a drop of -1.3 percentage points, and did not regain its 1.7% starting level until 2019:Q3, 47 quarters later. After the 2019:Q4 peak it did not fall at all: it was 1.8% one quarter on and 2.5% by 2026:Q1. This note reports that timing and nothing more. It does not say why the paths differ, and it makes no claim about what comes next.

1.7%
bottom-50 net-worth share at both peaks (2007:Q4 and 2019:Q4)
-1.3 pp
bottom-50 share change to the 2010:Q4 trough after 2007:Q4
47 quarters
from the 2007:Q4 peak until the share regained 1.7%, at 2019:Q3
+0.0 pp
bottom-50 share change after 2019:Q4: no post-peak decline

What we anchor to

Event time here is measured in quarters from the National Bureau of Economic Research's own quarterly business-cycle peaks, so both episodes start at the officially dated turn rather than at a hand-picked quarter. The 2007-2009 recession peaked in 2007:Q4 (December 2007) and reached its trough in 2009:Q2 (June 2009), 6 quarters of contraction. The 2020 recession peaked in 2019:Q4 (February 2020) and reached its trough in 2020:Q2 (April 2020), the shortest recession in the NBER's record. Those quarterly peaks are the t equals zero of each path below.

Source: Federal Reserve Distributional Financial Accounts (Z.1 household net worth distributed with Survey of Consumer Finances weights), bottom-50% share by quarter. Event time is quarters from the NBER Business Cycle Dating Committee’s quarterly peaks (2007:Q4 and 2019:Q4). Public-domain US government data, displayed with attribution. The NBER reference dates are the Business Cycle Dating Committee's published monthly and quarterly turning points, verified against the committee's table at write time.

What we measure, and how it is counted

  • One series, shared with the wealth module. The line is the bottom-50% share of household net worth from the Distributional Financial Accounts, the same series the site's wealth page plots, read here through the same query rather than re-fetched. The DFA distributes the Z.1 "Households; Net Worth" aggregate across percentile groups with Survey of Consumer Finances weights, quarterly since 1989:Q3, 147 quarters through 2026:Q1.
  • A share of a total, not a dollar figure. Every number here is the bottom half's share of aggregate net worth, in percent. The DFA has no medians, it is a distribution of a total rather than a survey of families, so this note never reports a typical household's wealth, only the group's slice of the whole.
  • Event offsets are calendar quarters, never row steps. Each quarter is placed relative to the peak by its calendar position (year and quarter), not by counting a fixed number of rows back or forward. The DFA series is gapless, but a row-offset step over any quarterly series is a defect this site has hit before and now anchors every window to the calendar to avoid.
  • Depths and recoveries use the displayed numbers. Shares are rounded to a tenth of a point, as shown, and the trough depth and the quarters-to-recover are computed from those rounded values, so the arithmetic on the page matches the arithmetic behind it. The recovery quarter is the first one after the trough at which the rounded share is back to its peak-quarter level.

The two paths, on one clock

Both lines start from the same 1.7% at their peak. After 2007:Q4 the 2008 line falls for three years to 0.4% at 2010:Q4, sits near that low for 6 quarters, and then climbs back so slowly that it does not reach 1.7% again until 2019:Q3, one quarter before the next recession began. After 2019:Q4 the 2020 line does the opposite: it never dips below its peak level, rising to 2.2% within a year and to 2.5% by 2026:Q1.

NBER peak0.0%0.5%1.0%1.5%2.0%2.5%3.0%3.5%-4048121620242832364044quarters from the NBER business-cycle peak2008, t = -4: bottom 50% held 2.3% of net worth2008, t = -3: bottom 50% held 2.2% of net worth2008, t = -2: bottom 50% held 2.1% of net worth2008, t = -1: bottom 50% held 2.0% of net worth2008, t = 0: bottom 50% held 1.7% of net worth2008, t = 1: bottom 50% held 1.5% of net worth2008, t = 2: bottom 50% held 1.5% of net worth2008, t = 3: bottom 50% held 1.2% of net worth2008, t = 4: bottom 50% held 1.0% of net worth2008, t = 5: bottom 50% held 0.7% of net worth2008, t = 6: bottom 50% held 0.7% of net worth2008, t = 7: bottom 50% held 0.7% of net worth2008, t = 8: bottom 50% held 0.6% of net worth2008, t = 9: bottom 50% held 0.5% of net worth2008, t = 10: bottom 50% held 0.5% of net worth2008, t = 11: bottom 50% held 0.5% of net worth2008, t = 12: bottom 50% held 0.4% of net worth2008, t = 13: bottom 50% held 0.4% of net worth2008, t = 14: bottom 50% held 0.4% of net worth2008, t = 15: bottom 50% held 0.4% of net worth2008, t = 16: bottom 50% held 0.4% of net worth2008, t = 17: bottom 50% held 0.4% of net worth2008, t = 18: bottom 50% held 0.6% of net worth2008, t = 19: bottom 50% held 0.6% of net worth2008, t = 20: bottom 50% held 0.7% of net worth2008, t = 21: bottom 50% held 0.7% of net worth2008, t = 22: bottom 50% held 0.8% of net worth2008, t = 23: bottom 50% held 0.9% of net worth2008, t = 24: bottom 50% held 0.9% of net worth2008, t = 25: bottom 50% held 0.9% of net worth2008, t = 26: bottom 50% held 1.0% of net worth2008, t = 27: bottom 50% held 1.0% of net worth2008, t = 28: bottom 50% held 1.0% of net worth2008, t = 29: bottom 50% held 1.0% of net worth2008, t = 30: bottom 50% held 1.1% of net worth2008, t = 31: bottom 50% held 1.1% of net worth2008, t = 32: bottom 50% held 1.2% of net worth2008, t = 33: bottom 50% held 1.2% of net worth2008, t = 34: bottom 50% held 1.2% of net worth2008, t = 35: bottom 50% held 1.3% of net worth2008, t = 36: bottom 50% held 1.3% of net worth2008, t = 37: bottom 50% held 1.3% of net worth2008, t = 38: bottom 50% held 1.4% of net worth2008, t = 39: bottom 50% held 1.4% of net worth2008, t = 40: bottom 50% held 1.4% of net worth2008, t = 41: bottom 50% held 1.5% of net worth2008, t = 42: bottom 50% held 1.6% of net worth2008, t = 43: bottom 50% held 1.5% of net worth2008, t = 44: bottom 50% held 1.6% of net worth2008, t = 45: bottom 50% held 1.6% of net worth2008, t = 46: bottom 50% held 1.6% of net worth2008, t = 47: bottom 50% held 1.7% of net worth20082020, t = -4: bottom 50% held 1.6% of net worth2020, t = -3: bottom 50% held 1.6% of net worth2020, t = -2: bottom 50% held 1.6% of net worth2020, t = -1: bottom 50% held 1.7% of net worth2020, t = 0: bottom 50% held 1.7% of net worth2020, t = 1: bottom 50% held 1.8% of net worth2020, t = 2: bottom 50% held 2.0% of net worth2020, t = 3: bottom 50% held 2.1% of net worth2020, t = 4: bottom 50% held 2.2% of net worth2020, t = 5: bottom 50% held 2.3% of net worth2020, t = 6: bottom 50% held 2.3% of net worth2020, t = 7: bottom 50% held 2.4% of net worth2020, t = 8: bottom 50% held 2.4% of net worth2020, t = 9: bottom 50% held 2.5% of net worth2020, t = 10: bottom 50% held 2.7% of net worth2020, t = 11: bottom 50% held 2.7% of net worth2020, t = 12: bottom 50% held 2.6% of net worth2020, t = 13: bottom 50% held 2.6% of net worth2020, t = 14: bottom 50% held 2.6% of net worth2020, t = 15: bottom 50% held 2.5% of net worth2020, t = 16: bottom 50% held 2.5% of net worth2020, t = 17: bottom 50% held 2.5% of net worth2020, t = 18: bottom 50% held 2.5% of net worth2020, t = 19: bottom 50% held 2.4% of net worth2020, t = 20: bottom 50% held 2.5% of net worth2020, t = 21: bottom 50% held 2.5% of net worth2020, t = 22: bottom 50% held 2.5% of net worth2020, t = 23: bottom 50% held 2.5% of net worth2020, t = 24: bottom 50% held 2.5% of net worth2020, t = 25: bottom 50% held 2.5% of net worth2020

Source: Federal Reserve Distributional Financial Accounts (Z.1 household net worth distributed with Survey of Consumer Finances weights), bottom-50% share by quarter. Event time is quarters from the NBER Business Cycle Dating Committee’s quarterly peaks (2007:Q4 and 2019:Q4). Public-domain US government data, displayed with attribution. The dashed rule is the NBER peak (t equals zero). The 2008 path runs to 2019:Q3, one quarter before the 2019:Q4 peak; the 2020 path runs to 2026:Q1.

The two episodes, side by side

MeasureAfter 2007:Q4 (2008)After 2019:Q4 (2020)
Bottom-50 share at the peak1.7%1.7%
Lowest share after the peak0.4% (2010:Q4)1.7% (no decline)
Change from peak to that low-1.3 pp+0.0 pp
Quarters from peak to the low120 (none)
Quarters to regain the peak share47 (2019:Q3)1 (2020:Q1, never fell)
Share at the end of the window1.7% (2019:Q3)2.5% (2026:Q1)

Source: Federal Reserve Distributional Financial Accounts (Z.1 household net worth distributed with Survey of Consumer Finances weights), bottom-50% share by quarter. Event time is quarters from the NBER Business Cycle Dating Committee’s quarterly peaks (2007:Q4 and 2019:Q4). Public-domain US government data, displayed with attribution. Each cell is computed from the rounded quarterly bottom-50 share; the change and the quarter counts follow from those displayed values.

What this cannot tell you

  • This is timing, not cause. The note records when the bottom-50 share fell and rose around each peak. It does not identify why the 2020 path differs from the 2008 path. The 2020 episode coincided with the shortest recession on record, with large federal transfers, and with a fast rebound in equity and house prices, all widely documented context, but this note attributes the movement of the share to none of them and tests no such mechanism.
  • A share can rise while a group gains or loses ground. The bottom-50 share is a slice of a total, so it can move with the group's own assets, with the assets of everyone else, or with both. A larger slice is not by itself evidence that the bottom half grew richer in dollars, and this note reports no levels, only shares.
  • The decline began before the dated peak. In the 2008 window the bottom-50 share was already easing in the quarters before 2007:Q4, from 2.3% four quarters earlier. Anchoring to the NBER peak is a convention, not a claim that the turn in the share happened exactly then.
  • The DFA is revised, and it is a distribution of an aggregate. The DFA is built by distributing the Z.1 household totals with SCF weights and is revised as those inputs are revised; intra-wave movements are interpolations between triennial surveys. The quarter-to-quarter path should be read as the published series, not as a survey measurement of the same households over time.

The finding, in one line

From the same 1.7% at both business-cycle peaks, the bottom half's share of US net worth fell -1.3 points to 0.4% and took 47 quarters to come back after 2007:Q4, and it never fell after 2019:Q4, standing at 2.5% by 2026:Q1. That is a difference in timing, computed from the Distributional Financial Accounts; it is not a statement of cause.

Data as of 2026:Q1 (Federal Reserve Distributional Financial Accounts)