Beneficial ownership explained
Beneficial ownership may sound complicated, but once you understand what’s involved, it really isn’t. We recently sat down with Norman Hodne, Program Director, Microsoft ACTS (Advanced Cloud Transparency Services), who works closely with partners around the globe to implement advanced technology solutions that support government transparency and build citizen trust. In our conversation, he helps break down beneficial ownership as a concept, why it’s an important challenge to address, and what ACTS is doing to help governments prevent it.
So, what exactly is beneficial ownership?
Beneficial ownership indirectly asks, “Who gains the most from doing something illegal?” That illegal action could be a bribe, favors, a job for a family member, and so on. The gain is that the individual or company gets more contracts, or makes more profit, perhaps by double or triple charging for products. An individual can also do things that are beneficial to them, by the way, and not necessarily to the company.
Walk me through an example of beneficial ownership.
My government coworker is going to all these ballgames and getting private suites. How can he afford that? In an investigation, it’s discovered that he's been approving high-priced contracts from a particular supplier, even though that supplier made higher bids than other qualified suppliers. In return, the government worker is getting these season tickets that the supplier writes off as a business expense.
In such scenarios, is there always purposeful intent, meaning do these individuals or companies go in with a plan to commit corrupt acts?
Not necessarily. Maybe when they started, they charged a reasonable price and got a contract. The supplier is super happy with the government person they worked with. In time, they build a business friendship, and then one day, they see each other on the street in front of a stadium. The supplier says, “Hey, I'm going to the ballgame. Would you like to go with me? I’ve got a suite.” And the government person thinks, “Sure,” because it's not necessarily related to that transaction. It's just friends who know each other. That’s how it can start.
I assume competition often plays a role in all of this too?
Yes. These large corporations that you read about getting caught for bribing government officials to gain huge multi-year, hundred-million-dollar construction contracts. I think that oftentimes, when they originally go in, they're trying to legitimately compete for a contract, but ultimately they're set up with a situation where either they have to pay some fees to a government official, or they know that the other competing company is going to give bribes to gain favor. That puts them at a competitive disadvantage. If you’re someone who thinks your job’s on the line or knows your whole company might be on the line with this deal, then although you weren't planning on doing anything, when push comes to shove, you might.
How does corruption within government procurement typically get discovered?
Historically, corruption has been found out when someone reports it. There's a whistleblower that says, “Hey, I think something’s happening here.” It could be as simple as a government worker with season tickets always seeing a peer seated in the suites. So, then someone asks, “Who owns the tickets for the suite?” That’s where the investigation starts. You track down the information. Oh, it's owned by this company, which does business as a supplier. Through investigation, you figure out who gains or is the one pulling the strings on the operation. This is how we define corruption—it’s relationship based. You then can further investigate that person and the relationships they have with other things.
How do the bad actors in these scenarios try to cover their tracks?
They mask their data. I can be Norman Hodne, owner of this company. I could also be Mr. Hodne and own another company or Norm or use my middle name. Masking my name isn’t truly lying. Maybe I’m doing misspellings. In these examples, I’d be the beneficial owner of each company. Or maybe someone has one company in one country and another company in another country. They move contracts and payments back and forth between companies and across jurisdictions, which makes it harder to track because then you need data from two different countries and compare and contrast and do all types of processing. Or someone says that their son and daughter each own 20 percent of their company. The truth is, they're both 10 years old. So, that person’s really the controller of those stocks. When you try to manually look at it all, it can be fairly hard and time-consuming to do.
How complicated does it get?
It can be really hard to spot who's moving the chess pieces on the board and the person that's benefiting from all of this. If there’s a big company that owns all kinds of small companies, you can start following the chain of ownership of those small companies and see they're owned by other small companies, which are owned by some medium-sized companies, which are owned by other medium-sized companies. Eventually, you get to the big kingpin. We saw a demo from a commercial vendor that showed us how beneficial ownership went from a tiny company in one country all the way up to an oligarch in another region. It took like 15 steps to get there.
What motivated ACTS to get involved in beneficial ownership?
Microsoft launched ACTS believing that technology can make a difference in turning back the tide on corruption. Beneficial ownership is a great example of a situation like that, one where we believe modern technology and services can lend governments a hand by helping them monitor potential illegal activities with government transactions.
What is ACTS doing to help prevent this type of corruption?
We apply new technology and techniques with machine learning to look at all available data about the basic transactions of a contract and see if we can spot any irregularities with the people involved in the contract.
So it’s more a preventative measure as opposed to a typical reactive one?
Yes. We’re highlighting potential problems, not necessarily actual problems. We’re helping governments to prevent illegal activity from happening by spotting trends or relatedness from a set-up network. We try to get as many data points as possible to see how all these pieces fit together into the puzzle.
Where does the data you use come from?
It comes from many different places. There’s open data, like telephone numbers, names, addresses of all past transactions, future transactions that are currently being processed, etc. The governments themselves could add proprietary information related to corporations, individuals, and transactions that perhaps aren't in an open data environment. And there can be corporate registries, which include the company’s name, location, phone number and email, websites, and a list of all the major owners—I think it's usually owners of 25 percent or more. Some places have something called wealth reporting that’s required by government officials, where you must report the cars you own, the real estate you own, the stock you own, and so on, on a regular basis.
What exactly does machine learning do?
Machine learning takes all of this combined data, millions—perhaps billions—of records, and gives a more holistic and networked view of everything related to a contract. Based on different relatedness calculations, you can see if a transaction looks OK and is ready to approve. If there are any “red flags,” it signals potential risks that the government can dive more deeply into for more information.
What do you think are some of the greatest benefits of this technology?
It helps organizations be more effective and efficient because they're not manually culling through a bunch of information to determine what they should investigate. The technology helps expose “red flags” that need investigation or further review. They can also set system parameters to try and see where the highest risk may be. A lower-risk item could still have a “red flag,” but as a lower risk, it means less exposure for the government.
To learn more, visit our Beneficial ownership page.