What this page is
A reference list of the figures that actually govern the money side of a major life event, each one linked to the body that published it, with a note on what it measures. It is not original research. PivotReset runs no survey, holds no household dataset, and publishes no statistic of its own. Where a number is widely quoted but has no authoritative source, this page says so rather than estimating one — there is a list of those at the end, and it is longer than you might expect.
Three distinctions do most of the work here, and getting them wrong is how good figures turn into bad advice. A median is not a mean, and for job searches and legal fees the two are far apart. A survey is not a census: a self-selected reader survey describes who answered it, not the country. And a market rate is not a paid price: what providers charge is not what families net of subsidy actually pay.
Job loss and unemployment
| Figure | Value | What it measures |
|---|
| Duration of unemployment | Median 11.4 weeks, mean 26.3 weeks; 27.0% unemployed 27 weeks or longer | The gap between median and mean is the long tail. Budget against the mean. BLS Table A-12, August 2026. |
| State unemployment maximums | $235 to $1,152 a week | Mississippi lowest, Washington highest. Most states pay up to 26 weeks; fourteen pay fewer. DOL, January 2026. |
| COBRA premium | About $793/month single, $2,294/month family (2025 averages) | 102% of the full average employer premium, which is what you pay once the employer stops contributing. KFF, 2025. |
| Election windows | 60 days each, counted separately | 60 days to elect COBRA, and a 60-day marketplace special enrollment period on loss of coverage. DOL. |
| Severance review period, age 40+ | 21 days, or 45 in a group programme, plus 7 to revoke | A statutory minimum for a valid age-discrimination waiver, not a courtesy. 29 U.S.C. § 626(f). |
| Median job tenure | 3.9 years | Lowest since 2002. A biennial series; January 2024 is still the newest published. BLS. |
Divorce
Divorce cost is the single worst-sourced subject in consumer finance. No federal agency collects it. The only substantial dataset is a reader survey, and the figures below are far lower than the ranges usually published.
| Figure | Value | What it measures |
|---|
| Divorces granted | 672,502 in 2023, a rate of 2.4 per 1,000 | 45 states and DC report; California, Hawaii, Indiana, Minnesota and New Mexico do not. No agency counts filings. CDC NVSS. |
| Attorney fees | Average $11,300, median $7,000; 42% paid $5,000 or less | Full-scope representation. Martindale-Nolo reader survey — self-selected respondents, not a population sample. |
| Other routes | DIY average $925 (median $300); online services $150–$750; private mediation $3,000–$10,000 total | Same survey, same caveat. Mediation is usually split between spouses. |
| Duration and age | Median 12 years (women) / 13 (men) at first divorce; median age 41.9 / 43.3 | Both at historic highs. NCFMR. |
| Income change after 50 | Women −41%, men −23%; assets down about 40% for both | Applies only to divorce or separation after age 50. Widely misquoted as a general divorce outcome. GAO-12-699. |
| Gray divorce | Doubled between 1990 and 2010 | Date-bound. The open-ended "since 1990" implies a continuing trend the study does not establish. Brown and Lin, 2012. |
| Credit impact | Not recorded at all | Marital status never appears on a credit report; damage is downstream of joint accounts. CFPB. |
A new baby
| Figure | Value | What it measures |
|---|
| Out-of-pocket childbirth | Average $2,743 out of $20,416 total | Large-employer coverage, 2021–2023 claims. Excludes out-of-network balance billing. Peterson-KFF, September 2025. |
| Childcare price | $6,552–$15,600 a year, or 8.9%–16% of median family income | County-level medians compiled from state market rate surveys, 2022 data. DOL publishes no national average on purpose. This is what providers charge. DOL Women's Bureau. |
| Paid family leave access | 27% of private-industry workers | March 2023, and there is no newer figure. BLS is redefining the measure for March 2026 to include state-mandated plans, so the next number will not be comparable. BLS. |
| FMLA | 12 weeks, unpaid | Requires 12 months of service, 1,250 hours, and 50 employees within 75 miles. DOL Fact Sheet #28. |
| Child Tax Credit | $2,200 for 2026, up to $1,700 refundable | Raised by the 2025 reconciliation law and indexed. Rev. Proc. 2025-32. |
Career change
| Figure | Value | What it measures |
|---|
| How many careers people have | Not measured | BLS "never has attempted to estimate the number of times people change careers" because no definition is agreed. The "five to seven careers" line corrupts a jobs statistic. BLS. |
| Pay change when switching jobs | 48% earn less; median change +2.6% | Direct job-to-job moves in Census SIPP data. Measures changing jobs, not fields, so treat it as the optimistic case. St. Louis Fed. |
| Switchers against stayers | 5.0% vs 3.6% median wage growth | Twelve-month growth, current reading. Atlanta Fed, August 2026. |
| Employer-paid retraining | $5,250 a year, tax-free | Section 127 exclusion, unchanged for 2026; indexing begins 2027. Student-loan repayment is now a permanent qualifying use. IRS FS-2026-10. |
| Job-search expense deduction | None, permanently | The 2025 reconciliation law terminated miscellaneous itemized deductions for tax years beginning after 2025. Several government pages still show the old sunset date. JCT. |
| GI Bill private-school cap | $30,908.34 for AY2026–27 | Public schools are covered at net in-state cost with no dollar cap. VA. |
Health coverage and medical bills
| Figure | Value | What it measures |
|---|
| Medical debt on credit reports | Still reportable. Paid medical collections and balances under $500 are excluded | The bureaus removed those voluntarily in 2022–23. The CFPB rule that would have barred medical debt entirely was vacated on 11 July 2025 in Cornerstone Credit Union League v. CFPB. CFPB. |
| Nonprofit hospital assistance | Required, facility by facility | Section 501(r) requires a written financial assistance policy, capped charges for eligible patients, and no extraordinary collection before checking eligibility. IRS. |
| Surprise billing | Banned in three situations | Emergency care, out-of-network providers at in-network facilities, and air ambulance. Ground ambulance is not covered — the most common misstatement about this law. CMS. |
| ACA cost-sharing cap | $10,600 individual / $21,200 family for 2026 | The ceiling on in-network essential-benefit cost sharing. Revised upward mid-cycle from $10,150/$20,300, which much published material still carries. 90 FR 27074. |
| HSA-qualified plan caps | Deductible $1,700/$3,400; out-of-pocket $8,500/$17,000 | A different and lower ceiling than the ACA cap above. Both apply to an HSA plan; the HDHP one binds. Rev. Proc. 2025-19. |
| The $400 question | 63% would cover it with cash or equivalent | Fed SHED 2025, published May 2026. It asks how someone would pay a hypothetical expense, not whether they have $400 — the usual rendering misstates it. Federal Reserve. |
The numbers that reset every January
| Item | 2026 | Source |
|---|
| 401(k), 403(b), 457(b) deferral | $24,500 | IRS Notice 2025-67 |
| Catch-up, age 50+ / ages 60–63 | $8,000 / $11,250 (totals $32,500 / $35,750) | Same |
| IRA limit and catch-up | $7,500 / $1,100 — the catch-up is indexed for the first time | Same |
| HSA contribution | $4,400 self-only / $8,750 family | Rev. Proc. 2025-19 |
| Estate and gift exclusion | $15,000,000 per person; the scheduled sunset was repealed and the amount is now indexed | Rev. Proc. 2025-32 |
| Annual gift exclusion | $19,000 — unchanged, despite three consecutive prior increases | Same |
| Social Security wage base | $184,500 | SSA |
| Federal poverty guideline | $15,960 for one person; $27,320 for three | 91 FR 1797 |
Federal student loans
| Figure | Status | What it means |
|---|
| SAVE | Terminated by court judgment on 10 March 2026 | Not "blocked" or "paused". Borrowers still described as enrolled are in a forbearance. |
| SAVE forbearance interest | Accruing since 1 August 2025 | Balances have grown for over a year while earning no credit toward forgiveness. |
| RAP | $10 monthly floor, no income exemption | Every earlier plan could produce a $0 payment; this one cannot. |
| Credit between plans | One-way | Legacy months count toward RAP. RAP months count toward nothing else. |
| Tiered Standard | Earns zero PSLF credit in every tier | Including its ten-year tier — the sharpest trap in the current rules. |
| Involuntary collections | Paused since 16 January 2026 | Discretionary, no end date, and FFEL guaranty agencies are not covered and can still garnish. |
| IDR forgiveness tax | Federally taxable again | The ARPA exclusion lapsed on 31 December 2025. PSLF remains tax-free under a separate permanent provision. |
Figures we could not verify, and no longer publish
Each of these circulates widely in this subject area. We looked for a primary source for every one and could not find it. Several were previously published on this site; all have been removed.
- "The average first-year cost of a baby is $X." No authoritative figure exists. The USDA report usually invoked last used 2015 data and was published in January 2017.
- "Median out-of-pocket cost of a major medical event." No agency defines "a major medical event", so no median for one exists.
- "Americans change careers five to seven times." BLS has never estimated it and says so plainly.
- "X% of career changers recover their income within five years." No dataset tracks this.
- "The average credit score drop from divorce is N points." Divorce is not a scoring input at all.
- "Divorce costs $15,000 to $30,000 per spouse." No survey supports it; see the figures above.
- "X% of Americans cannot afford a $400 emergency." Misstates what the Federal Reserve asked.
- Bootcamp placement rates and salary lifts for 2026. Independent audited reporting has collapsed to three schools, ending 2023–24. Any current figure is the school's own.
- "Medical debt has been removed from credit reports." The rule that would have done it was vacated in July 2025.
A figure without a source is not a conservative estimate. It is a number someone made up, and on this subject it usually travels in the direction that sells something.
Sources
Every figure above links to its source inline. The bodies relied on are: the Bureau of Labor Statistics, the Census Bureau, the Centers for Disease Control and Prevention, the Department of Labor, the Department of Veterans Affairs, the Internal Revenue Service, the Social Security Administration, the Government Accountability Office, the Consumer Financial Protection Bureau, the Centers for Medicare and Medicaid Services, the Federal Reserve Board and the Federal Reserve Banks of St. Louis and Atlanta, the Joint Committee on Taxation, the National Center for Education Statistics, the National Center for Family and Marriage Research, KFF and the Peterson-KFF Health System Tracker, and Martindale-Nolo. Where a source is a survey rather than a count, the table says so.