Singapore recorded fewer scam cases and lower scam losses in the first half of 2026. Yet the way most victims lost money became even more pronounced: they made or authorised the transactions themselves after being deceived.
The Singapore Police Force Mid-Year Scam and Cybercrime Brief 2026 reported 16,821 scam cases and about S$410.6 million in losses from January to June. Self-effected transfers were involved in 80.8% of reported scams, up from 78.8% in the same period of 2025.
Those figures need to be read separately. S$410.6 million is the total lost across all scam cases. The report does not say that self-effected transfers accounted for 80.8% of that amount. What it does show is that deception and social engineering remain central to the scam problem, even as the headline totals improve.
The main statistics below separate case volume, financial loss and transfer method. That distinction matters because each measure describes a different part of the risk.
| 1H 2026 Measure | Result | Change From 1H 2025 |
|---|---|---|
| Reported Scam Cases | 16,821 | Down 14.4% |
| Total Scam Losses | About S$410.6 Million | Down 17.9% |
| Median Loss Per Case | S$1,350 | Down 19.8% |
| Cases Involving Self-Effected Transfers | 80.8% | Up From 78.8% |
| Cases With Losses Below S$5,000 | 69.5% | Not Stated |
| Cases With Losses Of At Least S$100,000 | 4.7% | Down 24.5% |
Table of Contents
- What The S$410.6 Million Figure Actually Means
- Why Self-Effected Transfers Remain So Common
- Which Scam Types Caused The Largest Losses
- Why Lower Overall Losses Do Not Mean Lower Risk Everywhere
- Why Banking Controls Cannot Catch Every Authorised Payment
- A Practical Check Before You Transfer Money
- What To Do If You Have Already Transferred Money
- What The 1H 2026 Numbers Should Change
- Frequently Asked Questions
What The S$410.6 Million Figure Actually Means
The first-half results were an improvement. Scam cases fell from 19,644 to 16,821, while losses declined from about S$500.2 million to S$410.6 million. The median loss also fell from S$1,682 to S$1,350.
However, a lower total does not make the remaining loss modest. It still works out to an average of more than S$2.2 million lost each day over the six-month period. This daily figure is a simple calculation from the rounded total, not a separate statistic published by the Police.
The distribution is also uneven. More than two-thirds of cases involved losses below S$5,000, but 4.7% involved at least S$100,000. A relatively small number of severe cases can therefore have a large effect on the national total.
“Self-effected” describes the transaction method, not informed consent to the fraud. The victim may have logged in, entered the amount and approved the payment, but did so because the scammer misrepresented the identity of the recipient, the purpose of the transfer or the consequences of refusing.
That distinction avoids two mistakes. The first is assuming that most losses came from hacked bank accounts. The second is blaming victims as though they knowingly chose to give money to criminals. In reality, the transfer may be technically authorised while the decision behind it has been engineered through deception.
Why Self-Effected Transfers Remain So Common

The Scam Targets The Decision Before The Transaction
Many security controls test whether the person making a payment has the correct password, device or one-time password. A self-effected scam can pass those checks because the genuine account holder is completing the transaction.
The scammer’s work happens earlier. They create a convincing reason to act: protect an account, assist an investigation, secure an investment opportunity, complete a job task, pay a supplier or help someone facing an emergency. By the time the banking app opens, the victim may believe the transfer is necessary.
This is why an apparently normal payment can still be the endpoint of a scam. Authentication confirms who approved the transaction. It does not confirm whether the story behind the payment is true.
Credibility Is Built In Layers
The strongest scams rarely depend on one claim. They combine familiar names, convincing documents, copied branding, professional-looking platforms and several people playing different roles.
The 2026 Police brief describes investment scammers using advertisements that impersonated organisations such as the National University of Singapore and Moomoo. Victims were then invited into WhatsApp groups where supposed mentors offered tips and other fake members praised the advice. Each layer made the next request appear more credible.
A similar chain appears in government official impersonation scams. The approach may begin with someone claiming to represent a bank, telco or insurer. The call is then transferred to a supposed government officer who alleges that the victim is involved in money laundering and must move assets to a “safety account”.
Loan scams use a more direct pressure point: an urgent need for cash. Requests for upfront fees, Singpass credentials or unusual transfers should be checked against the common loan scam and account-security warning signs before any payment or disclosure.
Also Read: How To Protect Yourself From Personal Loan Scams In Singapore
One Payment Can Lead To The Next
After an initial transfer, a scammer may say that a tax, clearance fee, deposit or verification payment is needed to release the money. An investment platform may display fabricated profits but block withdrawals until another sum is paid.
The victim is no longer deciding only whether to trust a new opportunity. They may also be trying to recover money already sent. That makes stopping psychologically harder and gives the scammer another reason to demand urgency or secrecy.
The safest response to an unexpected request for more money is not to negotiate inside the same conversation. Stop, preserve the messages and verify the claim through a channel that the other party did not provide.
Which Scam Types Caused The Largest Losses
Case counts alone do not show where the financial damage is concentrated. Based on the Police report’s rounded figures, the five scam types with the highest losses accounted for about S$362.4 million, or roughly 88% of all scam losses in 1H 2026.
| Scam Type | Reported Cases | Amount Lost | Average Loss Per Case Reported By SPF |
|---|---|---|---|
| Investment Scams | 2,256 | S$169.8 Million | S$75,267 |
| Government Officials Impersonation Scams | 1,363 | S$90.8 Million | S$66,643 |
| Business Email Compromise Scams | 262 | S$57.3 Million | Not Shown In The Case-Volume Table |
| Job Scams | 2,247 | S$34.9 Million | S$15,534 |
| Phishing Scams | 3,104 | S$9.6 Million | S$3,108 |
| E-Commerce Scams | 3,865 | S$8.3 Million | S$2,160 |

Investment scams alone contributed about 41% of total losses, based on the rounded figures. Their 2,256 cases caused S$169.8 million in losses, making them the most damaging scam type by amount.
Government official impersonation scams ranked second by loss at S$90.8 million. Business email compromise came third at S$57.3 million despite only 262 reported cases, showing how a small number of diverted business payments can produce substantial losses.
E-commerce scams present the opposite pattern. They were the most common scam type, with 3,865 cases, but their aggregate loss was much lower than that of investment or impersonation scams. Prevention therefore needs two lenses: how frequently a scam occurs and how severe the loss can be.
Why Lower Overall Losses Do Not Mean Lower Risk Everywhere
The national totals moved in the right direction, but several categories worsened.
E-commerce scam cases rose 19.3% to 3,865, while losses increased 18.9% to S$8.3 million. Pokémon trading card cases more than doubled to 605, driven largely by pre-orders in which buyers paid deposits or the full price long before the promised delivery date.
Social media impersonation scam cases rose 47.9% to 587, and losses increased 71.3% to S$3.9 million. The Police noted that one cryptocurrency case accounted for about S$2.1 million of that amount. This is another reason to look beyond percentage changes: one extreme case can materially shift a smaller category.
Business email compromise recorded an even sharper movement. Cases increased 67.9% to 262, while losses rose 193.1% to S$57.3 million. These scams can redirect a genuine invoice or payment instruction to a fraudulent account by spoofing or compromising an email account.
The risk is not confined to one age group. People below 50 made up 64.4% of scam victims. Victims aged 65 and above accounted for 14.4%, but had the highest average loss at S$42,347. Different groups encounter different scripts; access to digital services does not by itself protect someone from a persuasive false story.
Scammers also need accounts, mobile lines and digital identities through which to receive money and maintain credible contact. Crawfort’s coverage of a Police probe involving suspected scammers and money mules explains why lending out a bank account, SIM card or Singpass access can support crimes even when the account holder never approaches a victim.
Learn More: How Scam And Illegal Moneylending Networks Use Ordinary Accounts
Why Banking Controls Cannot Catch Every Authorised Payment
Banks can flag unusual activity, warn customers, delay selected transactions and freeze accounts linked to known scams. These controls matter, but self-effected transfers expose a difficult boundary: the customer appears to be giving a valid instruction while the bank may not know that the reason for it is false.
A blanket block on every new payee, large transfer or investment payment would also interrupt legitimate transactions. Effective controls therefore depend on combining transaction signals with timely intelligence, customer warnings and a way to intervene when a person is clearly caught in an ongoing scam.
Singapore’s Protection from Scams Act allows the Police to direct banks to restrict the banking and credit facilities of people assessed to be at risk of transferring money to scammers. The Police had issued 18 restriction orders as at 30 June 2026 to victims who remained deeply entrenched despite engagement.
Restriction orders are a targeted last-resort measure, not the main defence for every transaction. The small number issued, compared with 16,821 scam cases, also illustrates why intervention cannot depend on formal restrictions alone. Most risky decisions must be interrupted earlier—before the payment is approved.
A Practical Check Before You Transfer Money

A warning such as “be careful” is easy to agree with and hard to apply under pressure. A more useful rule is to test four parts of the request before paying.
1. Verify The Person
End the call or pause the chat. Contact the organisation through its official website, banking app or a number you already trust. If a friend or relative asks for emergency money, call them on a known number and ask something only they would know.
Do not use the callback number, link or account-verification contact supplied in the suspicious message. That keeps you inside the scammer’s controlled environment.
2. Verify The Reason
Ask why money must move at all. Government officers do not require transfers to “safety accounts”. A legitimate investigation does not depend on secrecy from your family or bank. An investment return should not require repeated fees before withdrawal.
Treat urgency, secrecy and threats as reasons to stop. They are not substitutes for evidence.
3. Verify The Destination
Check whether the beneficiary name and account belong to the party you independently verified. For business payments, confirm any change of bank details using a known telephone number and a second authorised colleague.
For loan offers, do not transfer an advance fee to release a promised loan. A claimed lender’s name or licence number is not enough; scammers can copy legitimate details. A remote-only process should be checked against how a licensed moneylender application can legally proceed online.
Also Read: How To Recognise A Fake Crawfort Website
4. Add A Deliberate Pause
Create a personal or family rule for high-risk transfers. For example, a new payee combined with urgency, secrecy or a request to borrow money triggers a pause and a second-person check.
Businesses can apply the same idea through dual approval and independent confirmation of changed supplier details. The point is not to predict every scam script. It is to prevent one persuasive conversation from controlling the entire decision.
If you are unsure, use the official ScamShield checking tools or call the 24-hour 1799 helpline before acting.
What To Do If You Have Already Transferred Money
Speed matters, but recovery is not guaranteed. The official MoneySense response guide for suspected scam victims recommends securing the affected accounts and reporting the incident promptly.
- Stop communicating with the scammer and do not send another payment to recover the first.
- Contact your bank immediately through its official fraud hotline. Ask about freezing the account, cards or e-wallets and using any available emergency “kill switch”.
- Make a Police report while the details are fresh.
- Preserve messages, transaction records, phone numbers, email addresses, usernames and website links.
- Change compromised passwords from a trusted device and review affected accounts for unfamiliar activity.
- Call ScamShield at 1799 if you need scam-related guidance or help checking a suspicious approach.
The Police reported that the Anti-Scam Centre recovered more than S$97.7 million in the first half of 2026. In the report, “recovered” refers to funds frozen in bank accounts and cryptocurrency wallets associated with scammers or mules. It does not mean that all of that money had already been returned to victims.
What The 1H 2026 Numbers Should Change
The most useful lesson from the 2026 figures is that account security and decision security are different. Strong passwords, transaction alerts and authentication protect access. They cannot independently verify a story told by a convincing caller, chat group, supplier or supposed official.
Before a consequential transfer, move the decision outside the conversation that created the urgency. Use a known contact channel, involve another person and verify both the reason and the destination. That pause directly targets the mechanism behind most reported scam cases.
The falling case and loss totals are encouraging. The rise in self-effected transfers is a reminder that the next stage of prevention must help people recognise manipulation before a genuine banking customer authorises an irreversible payment.
Review Note: This article was checked against the Singapore Police Force’s Mid-Year Scam and Cybercrime Brief 2026 and government guidance available on 3 September 2026.
Important Note: This article is for general information only and is not financial or legal advice. If you suspect a scam, contact your bank and the relevant authorities promptly.
Frequently Asked Questions
What Is A Self-Effected Transfer?
It is a monetary transaction made or authorised by the victim after being deceived by a scammer. The scammer may not have direct control of the victim’s account, but uses social engineering to influence the payment decision.
Does 80.8% Mean Self-Effected Transfers Caused 80.8% Of Scam Losses?
No. The Police report states that 80.8% of reported scam cases involved self-effected transfers. It does not publish that percentage as a share of the S$410.6 million lost.
Which Scam Type Caused The Most Loss In 1H 2026?
Investment scams caused the highest aggregate loss at S$169.8 million across 2,256 reported cases. Government official impersonation scams ranked second by loss at S$90.8 million.
Which Scam Type Had The Most Cases?
E-commerce scams had the highest number, with 3,865 reported cases and S$8.3 million in losses. Phishing scams were next with 3,104 cases.
Can A Bank Reverse A Scam Transfer?
Recovery may be possible in some cases, especially when the victim reports the transfer quickly and the funds can be frozen before they move further. It is not guaranteed. Contact the bank immediately, secure affected accounts and make a Police report.


