Crypto handovers in Japanese fraud rose 1,446% in a year

Why this case matters

The headline is that Japanese special fraud losses rose 162% in half a year. The useful information is underneath it, in how the money physically leaves.

The channel shift

Bank transfer is winning, and internet banking is why. Transfers carried 8,213 first-half cases, up 105.4%, and 62.2% of everything. Within that, internet banking accounted for 38.6% of the cases and 59.5% of the money — a victim moved to their own banking app sends much larger amounts than one sent to a counter.

Cryptocurrency has arrived. Direct crypto handovers went from 24 cases to 371 — an increase of 1,445.8% — with losses up 496.7%. Add transfers into exchange accounts and cryptocurrency is 4.4% of cases and 15.0% of losses.

The old channels are collapsing. Electronic money fell 10.7 percentage points to 6.6%. Cash card handover and theft fell 8.4 points. Cash in hand is down to 13.2%.

Every part of that movement runs from methods that require physical presence towards methods that do not. A cash pickup needs somebody to send to a doorstep, and that person gets arrested. A transfer instruction needs nobody in the country at all.

What has not changed

The telephone. 79.1% of all approaches still begin with a phone call, and the share went up 2.9 points. For ore-ore-type fraud and refund fraud it is above 90%.

This is worth stating plainly because it contradicts the shape of most scam advice, which is overwhelmingly about links, messages and websites. In the country with the most detailed fraud statistics in the world, four out of five approaches are a voice on a phone.

The exception proves the rule: for fictitious billing fraud the mix is different — 39.2% by message, 27.4% by pop-up, only 26.9% by phone. Different scheme, different door.

The 165,292 calls before the calls

The NPA counts something no other agency on this site publishes: 予兆電話, precursor calls. Calls that probe for a name, an address, what assets somebody has, which bank they use — without asking for anything.

There were 165,292 of them in six months, up 131.3%. An average of 27,549 a month.

And their geography differs from the fraud’s. Cases concentrate in eight metropolitan prefectures (67.6% of cases, against 51.4% of the population). Precursor calls are more evenly spread — 55.2% in those same prefectures. The NPA’s own reading is that the reconnaissance is reaching regional Japan even where the losses have not yet followed.

That is the closest thing to a leading indicator anywhere in this site’s data.

The per-victim figure

Average loss per completed case rose to ¥4.646 million, up 78.0%.

Split by pretext, it separates cleanly: ¥8.287 million for fake police fraud, ¥2.549 million for everything else. The NPA identifies that single category as the main reason the overall average rose.

Daily losses across the country reached ¥330 million.