First time user of WU sending more than $500 to Nigeria or Ghana seems pretty reasonable to me. Let alone thousands or tens of thousands.
At the end of the day though, scammers will be scammers and victims will be victims. You can't prevent everything but you can still try harder than WU.
Some people have family in Nigeria or Ghana, and everyone is a first time use once. The whole point of Western Union is sending money to places that don't have a solid banking system.
Asking if they are sure it's not fraud is not necessarily preventing them from using the service. People with family in those areas will presumably know based on the wording that it doesn't apply to them. A notice that the person has to acknowledge seems a fairly innocuous defense mechanism.
I think it is valid to think in terms of expected value here, where value is negative (a loss), since there is a large volume of transactions so the law of large numbers will apply. Set a threshold of expected loss below which you will not flag transactions. Then either very large payments or relatively smaller payments to countries with a higher known percentage of fraudulent transactions will flag a transaction. You can further refine this if for example there is a relationship between observed fraud rate and transaction amount.
If a company profits off of fraudulent transactions and is aware of the fraud, and the attorney general or equivalent stops viewing this as negligence, there is an incentive for the company to encourage this fraud to increase profits.
It is as reasonable as proactively detaining everyone using tatoos or sports clothing.
Correlation does not imply causation.
Ethically, I do not think it is correct to outright flag an entire country. It can be used as one factor, but that in itself is not enough.
Japan for example, a country that is not usually associated with crime, has one of the most profitable crime organizations in the world, the Yakuza. The UK might be where most transactions tied to money laundering are conducted, and so on so forth. In the grand scale of things, Nigerians are in a lower order of magnitude when it comes to dirty money.
Fraud detection, at scale, takes not one or two but many factors in consideration. That would produce a level of confidence that will translate into a suggested action. If the confidence is high, a fraud detection system will take an action without human intervention. If the confidence is low, a human needs to intervene and take action.
You are confusing legal restrictions specifically set up to limit the power of the state with tools every company will use to limit fraud.
For example, you speak of "presumption of innocence" and "burden of proof" as "core concepts of our justice system". These are only core concepts of our criminal justice system; our civil justice system has very different rules for burden of proof, for example.
Some countries have much, much higher percentage of fraudulent transactions than other countries, and it's perfectly reasonable for Western Union to use country of origin as a factor in raising the fraud red flag. Western Union doesn't have the power to arrest anybody, and they have no reason to "presume innocence" for any one of their customers. If anything, when it comes to monetary transfers, I think it's safer to "presume guilt", that is assume all transactions have reason to be fraudulent, and then only let transactions through if you have strong reason to believe they are not fraudulent.
The burden of proof is still on the party forwarding an accusation in civil cases. Presumption of innocence still holds in civil cases.
Presumption of innocence is even in the UN universal declaration of human rights.
What is considered burden of proof would be different, but it is still responsibility of the person forwarding the accusation. What you describe would be an accusation through "probable cause", described as reasonable grounds to conduct a search or investigation.
That being said I think country of origin are not reasonable grounds. Could be an individual sending money home, could be a merchant, could be anything.
> Presumption of innocence still holds in civil cases.
This is just flat out, 100% wrong. Mainly because "innocence" and "guilt" aren't even concepts that apply in civil cases.
All of your responses still seem to assume that businesses somehow have the same level of responsibilities and restrictions of a government, and that is just not true.
It matters less what the ratio of bad actors to population is, and more what the ratio of bad actors to legitimate orders is. In the examples I mentioned above, they were seeing hundreds, perhaps thousands of bad orders without seeing any genuine ones. From countries with known issues of trafficking stolen goods.
The truly genuine customers can contact the company directly and deal with them directly, rather than buying through a site.
When your per-capita GDP is a few dollars a day, the odds that the order for a multi-thousand-dollar consumer electronics item is genuine gets awfully close to zero.
You call it discrimination against an entire country, I call it a sensible policy for a seller to enact. If you'd like to volunteer to police these fraudulent purchases, I'm sure they'd appreciate the help, but when you have an entire department that spends most of its time dealing with orders from two countries, orders which almost always turn out to be fraud, the most sensible way forward becomes nearly self-evident.
As others have mentioned, you seem to draw no distinction between constraints on governments, justice systems, and private policy determined by individuals and companies. I think until you stop confusing the two concepts (and the vocabulary thereof), you'll have a difficult time persuading anyone to come around to your thinking. (To say nothing of forming a coherent ethical argument for why you think this is some great injustice and what possible solutions might be.)
What are the attributes of the source object that we need to determine if it's suspicious? That's the hard part.