US Divorce Statistics: Rates, Counts, and Trends

Two document folders, each with a house key, beneath a dotted US map, calendar pads, and rows of people icons

The latest federal divorce statistics show 662,950 divorces and annulments in 2024, a rate of 2.3 per 1,000 people. Those figures come from the CDC's National Center for Health Statistics. They cover 45 states and the District of Columbia, not the whole country, and the rate is not the share of marriages that end in divorce.

That second point clears up the biggest confusion around divorce statistics. A count of events, a rate per 1,000 residents, the share of adults who are divorced, and a lifetime projection all measure different things. This page keeps them apart.

Every figure below was matched to the CDC's published tables on November 10, 2026.

Key Takeaways

US Divorce Counts and Rates, 2015 to 2024

The CDC publishes a provisional national table for 2000 through 2024. These are the last ten years.

YearDivorces and AnnulmentsReporting-Area PopulationRate per 1,000
2015800,909258,518,2653.1
2016776,288257,904,5483.0
2017787,251270,423,4932.9
2018782,038271,791,4132.9
2019746,971272,842,7482.7
2020630,505274,190,4252.3
2021695,509276,585,1042.5
2022673,989278,154,4542.4
2023672,502279,800,0792.4
2024662,950284,385,8972.3

Source: the CDC's national marriage and divorce rate table for 2000 to 2024.

The rate in 2024 is lower than in 2015. The table does not explain why. Changes in marriage patterns, the age of the population, reporting coverage, the economy, and court operations can all move the series.

The 25-Year Trend

The longer view shows a steady slide. In 2000, the reporting area recorded 944,000 divorces and a rate of 4.0 per 1,000 people. By 2010 the rate was 3.6. It reached 2.3 in 2024.

YearDivorces and AnnulmentsRate per 1,000
2000944,0004.0
2005847,0003.6
2010872,0003.6
2015800,9093.1
2020630,5052.3
2024662,9502.3

The rate is the better trend line. Raw counts move up and down partly because the set of reporting states changes. In 2000, for example, California, Indiana, Louisiana, and Oklahoma were excluded. In 2024, the excluded states were California, Hawaii, Indiana, Minnesota, and New Mexico.

What a Rate of 2.3 Means

A rate of 2.3 means the reporting area recorded 2.3 divorces and annulments for every 1,000 people living there.

The denominator is the whole population. It includes children, never-married adults, and people who are already divorced or widowed. None of them can divorce that year. That is why demographers call this a crude rate.

A crude rate is good for tracking one series over time. It is weak for judging the chance that any one marriage will end.

Which States the National Count Leaves Out

The CDC builds its national table from records that states send in. Not every state supplies qualifying divorce data every year.

The CDC's FastStats summary states the coverage plainly: its divorce figures come from 45 reporting states and the District of Columbia. The 2024 table footnote names the five excluded states: California, Hawaii, Indiana, Minnesota, and New Mexico.

The national system also collects less detail than many readers assume. The CDC's marriage and divorce program page says the collection of detailed data was suspended beginning in January 1996. Today, national rates are calculated from provisional counts that state health departments send in. That is why the CDC table gives totals and rates, but not the age, duration, or reasons behind each divorce.

You can see the gap in the population column. The 2024 divorce table covers 284.4 million people. The marriage table for the same year covers 340.1 million. By our calculation, the divorce figures cover about 84% of the population (284,385,897 divided by 340,110,988).

So 662,950 is not the exact number of all US divorces in 2024. It is the number recorded in the reporting area. California alone is the most populous state, so the true national total is higher.

Marriages and Divorces Side by Side

The same CDC release reports marriages. These figures cover a wider area than the divorce figures.

YearMarriagesMarriage Rate per 1,000Divorce Rate per 1,000
20222,065,9056.22.4
20232,041,9266.12.4
20242,128,4466.32.3

The marriage rate rose in 2024 while the divorce rate dipped. Both moves are small. The marriage rate also fell sharply to 5.1 in 2020 before recovering, which mirrors the dip in divorces that year.

Do not divide the divorce count by the marriage count. The two tables cover different states, and the people divorcing in 2024 married in many different years.

Four Measures That Should Not Be Mixed

Good divorce statistics name their numerator and denominator. Here are the four most common measures.

MeasureNumeratorDenominatorQuestion It Answers
Divorce countDivorces and annulmentsNoneHow many events were recorded?
Crude divorce rateDivorces and annulmentsTotal populationHow common were events relative to population?
Refined divorce rateDivorcesMarried populationHow common was divorce among married people?
Divorced sharePeople currently divorcedSelected populationWhat share has divorced status now?

A fifth measure, the lifetime chance of divorce, needs a model and assumptions about future behavior. It should never be presented as if it were the annual CDC rate.

What Survey-Based Divorce Statistics Add

Because the CDC stopped collecting detail, researchers turn to surveys for the rest of the picture. A Pew Research Center analysis of federal data, 8 facts about divorce in the United States, uses the American Community Survey and other sources. Its divorce statistics answer questions the CDC table cannot.

Survey-Based MeasureFigureYear
Refined divorce rate, per 1,000 married women14.42023
Refined rate at its peak, around 198022.6About 1980
Divorce rate for adults 50 and older, per 1,000 married women10.32023
Ever-married adults whose first marriage ended in divorceOne-third2023
Ever-divorced adults who have remarried66%2023

The refined rate uses married women age 15 and older as its denominator, so it tracks risk among people who are married and able to divorce. It tells the same story as the crude rate: divorce has become less common since its peak.

Pew is careful about one point. The finding that one-third of ever-married adults saw a first marriage end in divorce does not mean 33% of marriages end in divorce. It describes people alive in 2023, whose marriages began in many different decades.

Pew also reports that the median length of marriages ending in divorce rose from 10 years in 2008 to 12 years in 2023. Keep survey figures and CDC records in separate columns. They use different methods and can never be added together.

What Happened in 2020

The count fell to 630,505 in 2020, and the rate dropped to 2.3. The count then rose to 695,509 in 2021.

Court closures and reporting delays may have shifted some cases from one year to the next. The CDC table does not assign causes, though. The safe conclusion is narrow: the reporting-area count was lower in 2020 and higher in 2021.

Explaining that swing would need court-level data, which the national table does not include.

The Half of Marriages Myth

The claim that half of all marriages end in divorce is repeated often, usually without a year, population, or method. It does not come from the annual CDC table.

You cannot get it by dividing one year's divorces by one year's marriages. People divorcing in 2024 may have married decades earlier. People marrying in 2024 will reach their outcomes over many future years.

A credible lifetime estimate follows groups of couples who married in the same period, or models how their behavior will change. Results vary by age at marriage, first marriage versus remarriage, education, and the years studied. This page does not turn the annual rate into a lifetime probability.

How Age and Marriage History Affect the Numbers

National totals blend first marriages and remarriages, short marriages and long ones, and every age group.

An older population can show a lower crude rate even when the risk inside each married age group stays the same. That is one more reason not to treat the crude rate as a personal forecast.

Survey data from other agencies can describe current marital status, age at marriage, and recent marital events. Those measures add context, but they use survey definitions and margins of error that differ from CDC event records.

What the Trend Supports

The CDC series supports three conclusions you can state with confidence:

  1. The crude divorce rate in the reporting area is far lower than in 2000.
  2. The rate moved from 3.1 in 2015 to 2.3 in 2024.
  3. Counts and rates can move differently because population and state coverage change.

It cannot show that marriages became more successful for every group. It also cannot describe the legal or financial outcome of any single divorce.

Where Divorce Statistics Fit in Legal Data

Divorce counts are administrative records, like many legal statistics. They capture filings that reached a recorded outcome, not every separation. The same caution applies to medical malpractice statistics, where payment reports are not the same as proven negligence.

Family law is also one of the largest areas of legal work. If you are weighing a legal career, our breakdown of law school acceptance rates shows how competitive admissions became in 2025. For a wider view of the profession, see the legal industry statistics by firm size.

How to Read Divorce Statistics You Find Online

Most divorce statistics you will see online fall into one of the measures above. Before you quote one, run four quick checks.

  1. Find the year. A figure without a data year may be decades old.
  2. Find the denominator. Per 1,000 people, per 1,000 married women, and percent of marriages are not interchangeable.
  3. Find the coverage. CDC counts leave out five states. Survey figures carry margins of error.
  4. Find the source. A law firm roundup that cites another roundup is not a primary source. Trace the number back to the CDC, the Census Bureau, or a named research team.

A number that fails any of these checks may still be accurate, but you cannot tell. Leave it out until you find the original table.

How We Checked the Data

All counts, populations, and rates come straight from the CDC's 2000 to 2024 provisional table. The reporting-area note comes from CDC FastStats and the table's footnotes. We did not estimate any missing state.

We made one calculation: reporting-area coverage, which divides the 2024 divorce-table population by the 2024 marriage-table population. We use "divorce" as shorthand, while the CDC counts divorces and annulments together.

Frequently Asked Questions

What Is the Current US Divorce Rate?

The latest CDC figure is 2.3 divorces and annulments per 1,000 people in the 2024 reporting area.

How Many Divorces Were There in 2024?

The CDC reporting area recorded 662,950 divorces and annulments in 2024, excluding five states.

Is the Divorce Rate 50%?

No annual government table says half of marriages ended in 2024. The CDC rate was 2.3 events per 1,000 people.

Has the Divorce Rate Gone Down?

Yes. Within the CDC series, the rate fell from 4.0 in 2000 to 2.3 in 2024.

Why Are Some States Missing?

California, Hawaii, Indiana, Minnesota, and New Mexico do not supply qualifying divorce data to the national table.

What Is a Refined Divorce Rate?

A refined rate divides divorces by the married population rather than the total population, so it better reflects risk among married people.

Resources

These CDC sources hold the national tables and the coverage notes.

Verdict Point provides general statistical information, not legal advice.