People talk about earthquakes like buses.
“The big one is overdue.”
It sounds intuitive, and it isn’t entirely wrong. But it’s misleading in a way that matters for anyone deciding whether to retrofit a house or buy earthquake insurance.
What “overdue” gets wrong — and right
California absolutely has substantial seismic hazard. The state’s official long-term forecast (known as UCERF3) gives these odds of a major earthquake in the next 30 years:
| Region | M6.7 or larger | M7 or larger | M7.5 or larger |
|---|---|---|---|
| Los Angeles area | 60% | 46% | 31% |
| San Francisco Bay Area | 74%* | 51% | 20% |
*Updated from 72% by the USGS in September 2026, partly because more time has passed. Other figures from the 2015 UCERF3 release.
That footnote is the honest complication. The forecast is partly time-dependent: on faults where scientists know when the last big rupture happened, the modeled odds do rise slowly as quiet years pass. Research on the southern San Andreas describes its history as “quasi-periodic” — more regular than pure chance, far less regular than a timetable.
The U.S. Geological Survey put it well: since some fault sections have exceeded their average interval, many people use the term “overdue,” “but it’s more complicated than that.”
Here’s why. On one stretch of the San Andreas in the Bay Area, there were three large earthquakes in about 70 years (1838–1906) — and none in the 500 years before that. Averages hide enormous variation. “Overdue” implies a due date and imminence. What the science actually offers is elevated odds that change gradually, not a countdown clock.
The fault you haven’t heard of
There’s a second problem with the countdown framing: it focuses attention on one famous fault. The 1994 Northridge earthquake (magnitude 6.7) struck on a blind thrust fault that doesn’t reach the surface and wasn’t fully understood beforehand. It killed 33 people, injured more than 7,000 and caused roughly $40 billion in property damage. The California Geological Survey’s summary is blunt: damaging shaking is possible anywhere in the state.
Start with the map
That’s why property owners should look at actual hazard information for their address rather than countdown language. California offers two free tools:
- EQ Zapp (California Geological Survey): enter an address to see mapped zones for fault rupture, liquefaction and earthquake-induced landslides. It doesn’t show tsunami zones, and it may not show every fault — no zone doesn’t mean no hazard.
- Earthquake shaking potential map (Map Sheet 48): statewide expected shaking intensity.
Those lead to much more useful questions than “is it overdue?”:
- Is the property near a mapped active fault?
- Is the soil susceptible to liquefaction?
- Is it on terrain susceptible to earthquake-triggered landslides?
- How was the building constructed, and when?
- Has it been retrofitted? (California’s Earthquake Brace + Bolt program offers grants of up to $3,000, and more for income-eligible owners, to retrofit older raised-foundation wood homes.)
Why earthquake insurance costs what it does
Earthquake insurance is expensive partly because earthquake losses are unusual. Most years produce no catastrophic insured loss. Then one event can cause enormous, correlated losses across thousands of properties at once. That is a very different risk from a dripping pipe in one kitchen.
Northridge proved the point: insurers paid out more in claims for that one quake than they had collected in earthquake premiums over the previous 30 years. Many then pulled back, and in 1996 California created the California Earthquake Authority, which now writes about two-thirds of the state’s residential earthquake policies.
Few Californians buy it. In 2024, only about 12.5% of residential policies carried earthquake coverage, according to the Department of Insurance. The average CEA premium was about $876, with deductibles of 5% to 25% of the dwelling’s coverage.
Which raises the useful insurance question:
If homeowners decide earthquake coverage is prohibitively expensive, is that telling us the insurance is overpriced? Or that the underlying risk is genuinely expensive to transfer?
Those aren’t the same thing. Low take-up is consistent with both — and with a third possibility, that many households are quietly choosing to self-insure. Before arguing about the price, understand the risk.
Start with the map. For wildfire, the equivalent tool is covered in Is your California house actually in a fire zone?
The Zesht box
Follow the incentive
- What behavior does this reward?
- Retrofitting (grants and lower deductible minimums), avoiding mapped rupture, liquefaction and landslide sites, and learning a property's specific hazard.
- What does it discourage?
- Building on mapped hazard zones without investigation, and leaving older raised-foundation homes unbolted.
- What becomes more expensive?
- Coverage for older unretrofitted or high-value homes, which face a 15% minimum deductible, and development on hazard-zoned sites, which requires geotechnical study.
- What becomes scarcer?
- Low-deductible earthquake coverage, and private insurers' appetite for concentrated, correlated earthquake exposure.
- Who receives the benefit?
- Retrofitted and insured owners, the lenders holding their mortgages, and taxpayers if fewer households need aid after a disaster.
- Who ultimately pays?
- Policyholders through premiums; uninsured owners through their own losses; taxpayers through disaster assistance after a large event.
- What happens when people have another option?
- Many households choose to self-insure by not buying coverage; some retrofit instead or rely on expected government aid. Low take-up is feedback that price, deductibles and perceived risk are misaligned.
Goal → mechanism → result
The scorecard
- 1The goal
- Public hazard maps inform buyers, builders and local governments; the California Earthquake Authority aims to keep coverage available after private insurers pulled back following Northridge.
- 2The mechanism
- Official zone maps trigger site investigations and disclosure; insurers must offer earthquake coverage; the CEA supplies a standardized policy; retrofit grants lower vulnerability.
- 3The incentive
- Owners are nudged to learn their site hazard, retrofit vulnerable homes, and decide whether to buy coverage at risk-based prices with high deductibles.
- 4The observed result
- Take-up is low: 12.48% of residential policies in 2024. Hazard information is widely available and retrofit grants operate in more than 1,100 ZIP codes.
- 5The unintended consequences
- Low take-up means a large quake would leave most losses uninsured, pushing pressure onto disaster aid, lenders and household balance sheets; “overdue” rhetoric swings between panic and complacency.
- 6What supporters say
- Risk-based pricing with high deductibles keeps the CEA solvent against correlated losses; maps and grants help owners reduce risk rather than just transfer it.
- 7What critics say
- Premiums and 15–25% deductibles make coverage poor value for many households, and relying on voluntary purchase leaves a large protection gap.
- 8What the evidence actually shows
- Southern California faces high 30-year odds; recurrence is irregular but not purely random; losses are highly correlated; Northridge showed pre-event premiums can badly undershoot losses.
- 9What remains uncertain
- The timing of any particular fault's next rupture, how multi-fault ruptures change damage scenarios, and whether low take-up reflects mispricing, misperceived risk or rational self-insurance.
Questions people ask
Is Southern California overdue for a big earthquake?
Parts of the southern San Andreas fault have gone longer than their average interval, and the USGS notes people call that overdue — but says it is more complicated than that. Intervals vary widely, so elapsed time raises the odds gradually rather than setting a date. The official forecast puts the chance of a magnitude 6.7+ quake in the Los Angeles area at 60% in 30 years.
How do I check earthquake hazards for my address in California?
Use the California Geological Survey's EQ Zapp, which shows whether a property is in a mapped fault-rupture, liquefaction or earthquake-induced landslide zone. Not being in a zone does not mean no hazard; the Survey says damaging shaking is possible anywhere in California.
How many Californians have earthquake insurance?
About 12.5% of residential policies (14.9% of homeowners policies) carried earthquake coverage in 2024, according to the California Department of Insurance. The average California Earthquake Authority premium was about $876.
Why is earthquake insurance so expensive?
Earthquake losses are rare but hit many properties at once. After the 1994 Northridge earthquake, insurers paid more in claims than they had collected in earthquake premiums over the previous 30 years. Coverage also carries high deductibles — the CEA offers 5% to 25%, with a 15% minimum for some older, unretrofitted or high-value homes.
Sources
- Back to the Future on the San Andreas Fault
- Quasi-periodic recurrence of large earthquakes on the southern San Andreas fault
- What is the probability that an earthquake will occur in the Los Angeles area? In the San Francisco Bay area?
- New numbers show odds spike for major Bay Area earthquake in next 30 years
- Third Uniform California Earthquake Rupture Forecast (UCERF3)
- EQ Zapp: California Earthquake Hazards Zone Application
- Earthquake Shaking Potential for California (Map Sheet 48)
- M6.7 January 17, 1994 Northridge, California Earthquake
- Earthquake Brace + Bolt Retrofit
- Earthquake Insurance (consumer guide)
- Background on: Earthquake insurance and risk
- History of the California Earthquake Authority
- Earthquake Premium and Policy Count Data Call: Summary of 2024 Residential Totals
Cite or link to this article
See something wrong? Zesht changes the theory when the evidence changes. Send a correction · Our standards