The simple version

On Wednesday the Bureau of Economic Analysis releases personal income and outlays for July. The report is watched mainly for the inflation gauge inside it, which we cover separately, and it also contains the personal saving rate.

That rate is not a survey result. No household is asked how much it saved. In the agency's own description, income left over after people spend money and pay taxes is personal saving, and the rate is that leftover expressed as a share of after-tax income.

One clarification before the rest, because the words collide. This is not the interest rate on a savings account. It is a national statistic about the gap between what households take in and what they spend.

The numbers

• The Bureau of Economic Analysis defines the measure plainly: the personal saving rate is personal saving as a percentage of disposable personal income (Bureau of Economic Analysis)
• The agency describes personal saving as income left over after people spend money and pay taxes, which is a description of a remainder rather than of something collected (Bureau of Economic Analysis)
• In the most recent published month, June 2026, personal saving was $646.1 billion and the personal saving rate was 2.7% (Bureau of Economic Analysis)
• That same month shows the mechanism directly. Disposable personal income rose $48.3 billion while personal outlays rose $70.0 billion, so the gap between them narrowed by $21.7 billion (Bureau of Economic Analysis)
• Personal outlays are not only shopping. The measure is the sum of personal consumption expenditures, personal interest payments, and personal current transfer payments (Bureau of Economic Analysis)
• The figures behind the rate get revised. The June release revised its April and May estimates to reflect updated Bureau of Labor Statistics data (Bureau of Economic Analysis)
• Capital gains are excluded from income and from saving in these accounts. Selling an appreciated asset at a profit does not raise the national saving rate (Bureau of Economic Analysis)
• July personal income and outlays releases Wednesday, August 26, at 8:30 a.m. Eastern. This article makes no claim about what it will show (Bureau of Economic Analysis)
• The core personal consumption expenditures price index publishes in the same release, which is why the saving rate usually appears below the inflation headline (Bureau of Economic Analysis)

Why a leftover behaves differently from a measurement

When a number is measured directly, an error in it comes from the measurement. When a number is the difference between two other numbers, it inherits every error in both, and it inherits them amplified.

Personal income runs to trillions of dollars a year and so does personal spending. Saving is the gap between two very large estimates, and the gap is small relative to either. A revision of a fraction of a percent to spending moves the saving rate by a visibly larger fraction, because the same dollar change is a much bigger share of the smaller number.

June is a clean illustration. After-tax income rose and spending rose more, so saving fell by about $21.7 billion. Nothing in that sentence describes a decision anyone made.

That is also why the causal reading is often backwards. A headline saying the saving rate fell invites the conclusion that people chose to save less. What the arithmetic shows is that outlays rose relative to income, which is the same fact stated from the other side and carries no information about intent.

What the number does not count

There is a second reason the rate resists the readings people give it. These accounts exclude capital gains from both income and saving, so an asset rising in value contributes nothing to the measure.

The agency gives its reasoning openly. A capital gain is less available to fund spending, because the asset has to be sold first, and gains are volatile in a way ordinary income is not. It cites 2002, when capital losses on stocks and mutual funds cut personal net worth by more than 20% of disposable personal income.

The consequence for reading the number is specific. A household whose retirement balance grew substantially in a year, and which felt wealthier as a result, contributed nothing to the saving rate through that growth. The measure is tracking flows of ordinary income, not changes in what anyone owns.

The Real Cost lens on the arithmetic

Run the identity on a stated household so the mechanism is visible. Every figure below is an illustration rather than a measurement of anyone.

• A stated household with $6,000 of monthly after-tax income spending $5,700 saves $300, which is a saving rate of 5.0%
• If a later estimate revises that household's outlays up by $120 with income unchanged, saving becomes $180 and the rate becomes 3.0%
• Nothing about the household changed between those two lines. One estimate did, and the rate moved by two full percentage points
• That is the national figure in miniature. The rate is sensitive to revisions in a way a directly measured number would not be

None of this makes the saving rate useless. It is one of the few regular indicators of the relationship between what households take in and what they spend, and its trend across years is informative. It is a weak tool for reading any single month, and a weaker basis for conclusions about how carefully anyone is managing money.

What this means

When Wednesday's number lands, the question worth asking is which side of the subtraction moved. The same report publishes income and outlays, so the answer sits in the release rather than requiring inference.

The broader idea carries to any statistic described as a rate or a gap. If a number is the difference between two larger numbers, it is more volatile than either of them and more sensitive to revision, and it usually should not carry a story about intent.

What this is NOT

This is not a prediction of Wednesday's release or of any figure in it. This is not advice about saving, spending, budgeting, or any financial decision, and nothing here suggests what saving rate is appropriate for any household. This is not advice about any security, fund, or asset. This is not a claim that American households are saving too little or too much, or that they are under financial strain, which are judgments this article does not make. This is not a criticism of the Bureau of Economic Analysis, which publishes its definitions, its methodology, and its revisions openly; this article describes how the figure is built. The personal saving rate is a national statistic and is unrelated to the interest rate paid on a savings account. The household figures are stated illustrations rather than measured amounts. This is not investment or financial advice of any kind.

Sources

• U.S. Bureau of Economic Analysis, Personal Saving Rate: https://www.bea.gov/data/income-saving/personal-saving-rate
• U.S. Bureau of Economic Analysis, Personal Income and Outlays, June 2026: https://www.bea.gov/sites/default/files/2026-07/pi0626.pdf
• U.S. Bureau of Economic Analysis, Personal Income: https://www.bea.gov/data/income-saving/personal-income
• U.S. Bureau of Economic Analysis, Why do the NIPAs exclude capital gains from income and saving: https://www.bea.gov/help/faq/67
• U.S. Bureau of Economic Analysis, news release schedule: https://www.bea.gov/news/schedule