A 93-Country Study About Cats Has Been Everywhere This Week. In Shelters, a Microchip Took a Cat's Odds of Going Home From 1.8% to 38.5%.

Cats have been having a data moment. A study published in PLOS ONE on September 30 found a correlation between how many cats a country has and how happy its people report being, across 93 countries [1]. Euronews reported on October 3 that Russia leads the world in cat ownership — somewhere around 59% of households — while carrying an estimated 2.8 million stray cats, up to 98% of which were once somebody's pet [2]. And the American Pet Products Association reported in September that US cat-owning households had grown 5% year over year to 53 million [3].

The happiness study is less than it looks, and it says so itself. It's an ecological correlational study — the authors put that phrase in their own title. It compares countries, not people; nobody was asked about their own cat. National GDP correlated with happiness more strongly (r = 0.800) than cat density did (r = 0.620), and the authors write that all the associations "should be interpreted as statistical relationships rather than evidence for causal pathways" [1]. More cats did not make anyone happier as far as this paper can tell. Richer countries are both happier and more cat-dense, and cat density surviving as a predictor once wealth is in the model is the mildly interesting part.

The more actionable cat statistics are elsewhere, and they're not flattering.

19% and 3%

Shelter Animals Count's 2025 annual report covers national intake and outcomes. Cats now out-number dogs arriving at US shelters — 3.0 million cats against 2.8 million dogs. Of all the dogs taken in, 19% were returned to an owner. Of all the cats, 3% [4].

Narrow it to animals that arrived as strays, which is the fair comparison, and it's 34% of dogs against 6% of cats. Either way it's roughly a six-fold gap, and it's been documented in the peer-reviewed literature for longer than that: a study across 53 shelters found 21.9% of stray dogs returned to owners against 1.8% of stray cats [5].

A cat that ends up in a shelter is, statistically, a cat that doesn't go home.

56% of lost cats were carrying nothing

The reason is mostly identification, and the numbers are stark.

A national telephone survey of 1,015 US households asked owners whose pet had gone missing what the animal had been wearing. The gap between species runs through every single category [6]:

	Dogs	Cats
Wearing a collar	90%	43%
Personalized tag with a phone number	61%	25%
Microchipped	24%	15%
No identification whatsoever	11%	56%

More than half of the cats that went missing had nothing on them that could connect them to an address. An earlier single-county study of 138 lost cats found only 19% had an ID tag, rabies tag, or microchip at the time they were lost — against 48% of the dogs in its companion study [7][8].

And this is the mechanism behind the shelter gap. An unidentified cat in a shelter is indistinguishable from the enormous population of unowned free-roaming cats. There is nothing to process it as a lost pet rather than a stray.

The microchip number

Across those 53 shelters, researchers tracked what happened to animals that arrived with a microchip. Microchipped cats were returned to their owners 38.5% of the time, against 1.8% for stray cats generally [5].

For dogs the same study found 52.2% against 21.9% — a lift of roughly 2.4 times. For cats the lift is more than twenty-fold. Not because chips work better in cats, but because cats have so little else going for them: only 2% of recovered cats in the national survey were found via a tag or chip, and under half wear a collar at all [6].

That's the one place it's worth saying plainly what this site is for. The chip is only half of it, and the study is unusually clear about which half fails. Among cases where a shelter had a chip and still couldn't reunite the animal [5]:

• 35.4% — the phone number on file was wrong or disconnected
• 24.3% — the owner never responded
• 17.2% — the chip was registered in a database that didn't match the manufacturer's records
• 9.8% — the chip had never been registered at all

So the most common failure isn't an unregistered chip. It's a chip registered to a phone number the owner has since changed. Roughly 27% had some registration problem and about a third had a dead phone number — which means the single highest-value thing most people reading this can do takes about ninety seconds, and it's checking that the number on the registration is the number in your pocket.

"She won't wear a collar" is usually wrong

The common objection to cat ID is that cats won't tolerate it. That's been tested.

A study enrolled 538 cats from 338 households and asked owners to keep a collar on for six months. 391 of them — 72.7% — wore it successfully for the full six months. 56.3% of owners reported their cat tolerated the collar better than they had expected. Collars caught on something — a mouth, a forelimb, an object — in 3.3% of cases [9].

The finding that matters: cats were significantly more likely to fail to wear a collar when their owners hadn't expected them to accept it. Owner expectation predicted the outcome better than the cat did. Microchip migration from between the shoulder blades, incidentally, occurred in 3 of 478 cats scanned, about 0.6%.

If a cat does go missing, the search area is smaller than you think

The largest study of missing cats followed 1,210 cases. 61% were found within a year; 34% were recovered alive by their owner within seven days; the median time to recovery for cats who were found was six days [10].

The distances are the useful part:

• Median distance from home for all found cats: 50 metres
• 75% of found cats: within 500 metres
• Indoor-only cats: median 39 metres (interquartile range 9 to 137 metres)
• Indoor-outdoor cats: median 300 metres

An indoor cat that gets out is, typically, still on the property or next door — 39 metres is about 128 feet. 83% of found cats were located outside; 11% turned up inside someone else's house, and cats their owners rated as highly curious were far more likely to be the ones in a neighbour's garage (17% against 6%) [10].

Two caveats worth stating, because the study states them. A physical search was conducted for 96% of these cats, so there's effectively no un-searched comparison group — this tells you where to look, not that searching beats waiting. And cats were less likely to be found alive when owners advertised or contacted facilities, which the authors attribute to timing rather than ineffectiveness: those methods tend to get deployed later, after the best window has closed. It isn't evidence against putting up posters.

The behaviour gap

There's one more finding, with small numbers but a hard edge. In the national survey, among owners who never recovered their pet, 6 of the 8 dog owners had visited a shelter to look. Among cat owners, 4 of 18 had [6].

Those are tiny denominators and shouldn't be dressed up as a clean percentage. But they point at something the recovery rates echo: cats are returned home on their own far more often than dogs are — 59% of recovered cats came back by themselves, against 20% of dogs [6] — and that may teach owners that a missing cat is a cat that will reappear. Most of the time it is. The 25% that don't come back are the ones where somebody needed to be out walking the street with a torch on night one, and checking the shelter on day two.

The short version

• Put a collar and a tag on the cat. The odds it tolerates one are better than three in four, and if you assume it won't, that assumption is itself the risk factor [9].
• Microchip, then register, then check the phone number is current. A dead phone number is the most common reason a chipped animal doesn't get home [5].
• If an indoor cat escapes, search your own property and the two nearest ones first, thoroughly, on the first night. Median distance is 39 metres [10].
• Check the shelter in person, and keep checking. A cat with no ID will not be filed as a lost pet, because there's nothing to file it as.

The happiness paper may or may not survive replication. The 3% return rate is not in dispute.

---

Sources

1. Borráz-León JI, Rantala MJ, Krams I, Maestripieri D, "Wealth, health, cats and Toxoplasma gondii: An ecological correlational study of subjective well-being across 93 countries," PLOS ONE, September 30, 2026 — DOI 10.1371/journal.pone.0357911
2. Aleksandrova I, "Russia leads world in cat ownership — yet has 2.8 million strays," Euronews, October 3, 2026 — euronews.com. Figures as reported by Euronews, attributed to a 2023 Ingosstrakh/Financial University survey and a 2025 study by the Centre for Animal Nutrition and Well-being Studies with the Nika charity and Ipsos.
3. American Pet Products Association, 2026 Cat Report, September 14, 2026 — americanpetproducts.org. Note that the AVMA's 2025 Pet Ownership and Demographics Sourcebook puts US cat-owning households lower, at 43.1 million (32.6% of households) and 76.3 million cats — avma.org
4. Shelter Animals Count, 2025 Annual Data Report — shelteranimalscount.org
5. Lord LK, Ingwersen W, Gray JL, Wintz DJ, "Characterization of animals with microchips entering animal shelters," JAVMA 235(2):160–167, 2009 — summary via Ohio State
6. Weiss E, Slater M, Lord L, "Frequency of Lost Dogs and Cats in the United States and the Methods Used to Locate Them," Animals 2(2):301–315, 2012 — DOI 10.3390/ani2020301
7. Lord LK, Wittum TE, Ferketich AK, Funk JA, Rajala-Schultz PJ, "Search and identification methods that owners use to find a lost cat," JAVMA 230(2):217–221, 2007 — avmajournals.avma.org
8. Lord LK, Wittum TE, Ferketich AK, Funk JA, Rajala-Schultz PJ, "Search and identification methods that owners use to find a lost dog," JAVMA 230(2):211–216, 2007 — avmajournals.avma.org
9. Lord LK, Griffin B, Slater MR, Levy JK, "Evaluation of collars and microchips for visual and permanent identification of pet cats," JAVMA 237(4):387–394, 2010 — PubMed
10. Huang L, Coradini M, Rand J, Morton J, Albrecht K, Wasson B, Robertson D, "Search Methods Used to Locate Missing Cats and Locations Where Missing Cats Are Found," Animals 8(1):5, 2018 — DOI 10.3390/ani8010005