H.P. Bunaes, Director of Banking Practice at DataRobot
I don’t know that I’ve come across a problem better suited to machine learning than Anti Money Laundering (AML) in banking. For many other predictive applications, banks find that the availability of data for machine learning is an issue. The data may not exist, or if it does it may be of dubious quality. Not so in AML.
Banks have been collecting client information as part of their Customer Identification Programs (CIP) and Know Your Customer (KYC) programs for years.
Transactional details and suspicious activity alerts are accessible and in place. The findings and outcomes from investigations are available – specifically, whether a suspicious activity report (SAR) was filed or not. This constitutes ideal training data for machine learning which can be used to solve many of the common challenges throughout the typical AML program.
The most immediate opportunity is to screen out false positives from suspicious activity alerts. Many banks use rules-based systems for detecting suspicious activity and generating alerts. Since the downside risk of not detecting truly suspicious activity is severe, these rules tend to be conservative — better to get too many alerts than too few. The result can be an overwhelming number of alerts, the bulk of which turn out to be innocuous activity. But, it’s too risky to restrict alerts by narrowing detection rules.
Fortunately, banks can leverage their own data, and machine learning, to filter out the vast majority of false positive alerts with little or no downside risk. By definition, banks know the outcome of their investigations — whether a SAR was filed or not. Using machine learning, banks can use this historical data to train a model to screen out false positives (or at the very least, prioritise them lower) using the known outcomes. The model may learn, for example, to eliminate an alert for a particular combination of product, transaction size, KYC risk score, and location that has never resulted in a SAR.
To ease concerns about truly suspicious activity getting missed, models can be tuned so that there are zero false negatives. Even when tuned to “zero false negatives,” we have found that typically more than half of the false alarms can still be detected and eliminated. And the false positive rate on the remaining population also falls significantly.
In addition to screening false positives, a trained model can point out data features – client or transactional data for example – that are strong indicators of money laundering based on SARs generated when those patterns are evident. Using this information banks can create “smart” rules to detect suspicious activity that are unique to their client set, the product set, and the investigation outcomes. These detection rules are also far more dynamic: they will self-adjust for changes in products and client behavior as the model is periodically retrained.
Banks may also want to use these insights to validate their KYC process. Features of the data that are strong AML predictors can be built into the KYC procedure. Clients that fit these criteria, or whose product use is likely to “trip” an alert, would be subject to enhanced due diligence (a higher than normal degree of scrutiny upfront). KYC questions that are not good predictors of money laundering risk could be eliminated entirely.
Either way, banks have an ironclad defense of their detection rules and KYC process: their own data with rigorous analytics applied. In my experience, a data-based analysis is always easier to justify than an expert opinion. Expert opinions vary.
Finally, there are unsupervised machine learning algorithms that can be applied for anomaly detection. This can be a “last line of defence” to detect previously unseen activity for a client, or to identify a client acting completely unlike similar clients – unusual activity that might otherwise go undetected.
Money launderers are smart, and AML programs need to be smarter to stay one step ahead. Machine learning is a great way to do just that.
About the Author:
H.P. Bunaes leads the banking practice at DataRobot, helping banks leverage AI and machine learning for predictive analytics and data mining. H.P. has 35 years’ experience in banking, with broad banking domain knowledge and deep expertise in data and analytics. Prior to joining DataRobot, H.P. held a variety of leadership positions at SunTrust and FleetBoston. H.P. is a graduate of the Massachusetts Institute of Technology where he earned a Masters Degree in Management Information Systems and of Trinity College where he achieved a Bachelor of Science degree in Computer Science and Mechanical Engineering.
These 5 Payments Trends Once Seemed Revolutionary. In 2021, They’ll Continue to Become the Norm
By Warren Hayashi, President, Asia-Pacific, Adyen.
The pandemic forced brands to transform their businesses in ways that are here to stay
After a year of such great uncertainty, attempting to predict the future may seem risky. But even as brands and retailers faced unprecedented upheaval in 2020, one constant has held true. The pandemic has accelerated trends toward digitisation—and that’s as true in payments as in so many other areas of business and society. The stark reality of needing to avoid close contact with others has driven transformations for retailers and brands in a matter of months that in the past might have taken years. In the process, behaviours and expectations have changed for good.
As 2021 begins, much uncertainty remains but we feel confident that the digital transformation of payments will only get faster. Even after the pandemic has receded and consumers have the option to go back to their old behaviours, many won’t. The rapid increase in e-commerce seen under COVID-19 will persist, especially among previously digital-hesitant consumers. Merchants can no longer assume that their digital customers are limited to younger, tech-savvy shoppers. As brands have shown flexibility during the pandemic, consumers have also come to expect the flexible arrangements to continue. On that note, these are the key trends in payments that should be top-of-mind for brands and retailers in Singapore and Asia Pacific in 2021:
- Contactless will extend its reach into every corner of retail
From the start, the pandemic forced merchants to find ways to minimize the amount of physical contact necessary to complete a transaction. Customers and workers alike sought to avoid handing over credit and debit cards, touching keypads, and handling cash. According to our 2020 Agility Report 58% of APAC respondents preferred to use contactless payment methods because of hygiene concerns.
Our data also showed that the use of services such as Apple Pay and Google Pay has significantly increased over the last year too. Research from Kantar reiterates this, revealing that the frequency of e-Wallets transactions in Southeast Asia rose from an average of 18% pre-COVID-19 to 25% post-COVID-19, indicating a shift from one payment method to another.
In the post-pandemic world, the transition to contactless will only become more widespread now that the bar has been raised among consumers for what checking out can be, from one-click payments to same-day delivery options. Not to mention, the value of QR codes has also been made apparent in anchoring a seamless experience, not just at point-of-sale but at multiple points along the customer journey too, such as viewing menus and placing orders. The pandemic may have driven the change in behavior, but the superior user experience will cement contactless as the new normal.
- The distinction between offline and online will fade into irrelevance
As countries went into different forms of lockdown, many shoppers were unable to enter brick-and-mortar stores throughout 2020. Unifying offline and online became an issue of survival for retailers, who quickly pivoted to make app-powered deliveries and self-pick up options a reality.
Even while most physical stores in Singapore have opened their doors to consumers again, the digital infrastructure will remain in place. Many shoppers continue to prefer the convenience of deliveries and expect the options to continue, and retailers will find they’re able to forge better customer relationships thanks to the rich data generated by digital transactions.
One of the biggest learnings for the industry is the need to rethink the traditional split between offline and online stores. With lines increasingly blurred, retailers will benefit from adopting a unified commerce approach where brand interactions on and across all channels are important.
- The membership model will reign in retail and also in food and beverage
The membership model is another emerging trend for 2021. Amazon Prime is a great example of this, where customers pay an annual fee that in effect encourages them to buy more from Amazon in an effort to ensure they’re getting their money’s worth from their Prime memberships. Quick-serve restaurants especially are seeking to seize some of that flywheel effect. In addition to improved incremental spend, membership programs enable QSRs to get to know their customers in ways that were never possible when they were just anonymous faces standing in line.
Meanwhile, subscription passes encourage loyalty and more frequent use. Our 2020 Agility Report found that 38% of Singapore respondents (compared to 27% in APAC and 22% in Europe) signaled their interest in using these for products, including food passes, to reduce the amount of times they need to shop. Expect to see more retailers offering memberships in 2021 as brands seek to own the customer relationship and the data that goes along with it.
- Installments will become an everyday way to pay
The twin forces of increased convenience and tightened household budgets have brought pay-by-installment options mainstream, a trend that will only grow in 2021. Machine learning algorithms have become more adept than ever at assessing risk instantaneously, making it easy to offer “buy now, pay later” options right at checkout. For small and mid-ticket items, shoppers know that, say, instead of paying $100 now, they’ll pay $25 per month for four months. That kind of transparency makes it easier for shoppers on the fence to commit, which appeals to merchants hoping to avoid the dreaded abandoned shopping cart.
In 2021, providers of “buy now, pay later” options themselves will start to diverge, as some focus on higher-end, multi-year agreements, while others seek to offer installment plans for shopping baskets as small as $50. For households increasingly accustomed to paying by the month for everything from streaming services to food delivery premium memberships, installment plans start to look like subscriptions that just happen to have a fixed end date.
- The checkout-less experience will draw shoppers back to brick-and-mortar
In 2020, the appeal of an in-store experience offering limited human contact took on a new dimension, accelerating interest in doing away with the checkout counter altogether. For instance, in Singapore, BHG is looking to expand its endless aisle offering. By using interactive screens in-store, customers are able to check on inventories across all of BHG’s stores and e-commerce platform and can opt to have items to be delivered directly to their homes. Post-pandemic, shoppers will still find appeal in the human touch. The physical store continues to be relevant, especially in Asia Pacific and eliminating checkout counters frees staff to interact with shoppers in a more personal way, while also making lines a thing of the past.
In 2021, more stores will find various ways to make checkout a less prominent part of how people shop in-store. Multiple providers are creating their own versions of checkout-less experiences, where instead of going to the counter, customers will scan their items with their phones’ cameras, pay via app, and head out the door—a combination of increased trust and decreased friction that helps cultivate customer loyalty. In the case of Love, Bonito in Singapore, if customers are unable to find a particular item in store, they can go to an iPad within the premises, buy it online and have it shipped to their homes.
Across the five trends, this paradigm shift in the retail sector is underpinned by the under-tapped potential of technology to elevate the customer experience. Looking ahead in the new year, we expect retailers to increasingly harness digital solutions. Not only does this streamline operations, it also gives retailers the flexibility to pivot in line with changing preferences, and provide a seamless consumer journey across multiple channels.
Bitcoin heads for worst weekly loss in months
By Tom Westbrook
SINGAPORE (Reuters) – Bitcoin wavered on Friday and was heading toward its sharpest weekly drop since September, as worries over regulation and its frothy rally drove a pullback from recent record highs.
The world’s most popular cryptocurrency fell more than 5% to an almost three-week low of $28,800 early in the Asia session, before steadying near $32,000. It has lost 11% so far this week, the biggest drop since a 12% fall in September.
Traders said a report posted to Twitter by BitMEX Research https://twitter.com/BitMEXResearch/status/1351855414103715842 suggesting that part of a bitcoin may have been spent twice was enough to trigger selling, even if concerns were later resolved.
“You wouldn’t want to rationalise too much into a market that’s as inefficient and immature as bitcoin, but certainly there’s a reversal in momentum,” said Kyle Rodda, an analyst at IG Markets in Melbourne, in the wake of the BitMEX report.
“The herd has probably looked at this and thought it sounded scary and shocking and it’s now the time to sell.”
Bitcoin was trading more than 20% below the record high of $42,000 hit two weeks ago, losing ground amid growing concerns that it is one of a number of price bubbles and as cryptocurrencies catch regulators’ attention.
During a U.S. Senate hearing on Tuesday, Janet Yellen, President Joe Biden’s pick to head the U.S. Treasury, expressed concerns that cryptocurrencies could be used to finance illegal activities.
That followed a call last week from European Central Bank President Christine Lagarde for global regulation of bitcoin.
Still, some said the pullback comes with the territory for an asset that is some 700% above the 2020 low of $3,850 hit in March.
“It’s a highly volatile piece,” said Michael McCarthy, strategist at brokerage CMC Markets in Sydney. “It made extraordinary gains and it’s doing what bitcoin does and swinging around.”
Second-biggest cryptocurrency ethereum intially slipped to a one-week low on Friday before rising 6% late in the Asia session to $1,177.
(Reporting by Tom Westbrook; editing by Leslie Adler & Simon Cameron-Moore)
Oil prices fall as China’s surging COVID-19 cases trigger clampdowns
By Sonali Paul and Koustav Samanta
MELBOURNE/SINGAPORE (Reuters) – Oil prices dropped on Friday, retreating further from 11-month highs hit last week, weighed down by worries that new pandemic restrictions in China will curb fuel demand in the world’s biggest oil importer.
U.S. West Texas Intermediate (WTI) crude futures dropped 53 cents, or 1%, to $52.60 a barrel at 0445 GMT, after slipping 18 cents on Thursday.
Brent crude futures fell 45 cents, or 0.8%, to $55.65 a barrel, erasing a 2 cent gain on Thursday.
Recovering fuel demand in China underpinned market gains late last year while the United States and Europe lagged, but that source of support is fading as a fresh wave of COVID-19 cases has sparked new restrictions to contain the spread.
“Indeed, investors are struggling to see through short-term pain for long-term gain heading into the weekend as COVID case counts in China are the most significant demand concern for traders,” Axi chief market strategist Stephen Innes said in a note.
The commercial hub of Shanghai reported its first locally transmitted cases in two months on Thursday, and Beijing is urging people not to travel during the upcoming Lunar New Year holiday, when tens of millions of urban workers typically head back to their villages.
A seasonal boost to China’s gasoline demand that is typically seen during the New Year holidays will be moderated by the tightened restrictions this year, consultancy FGE said in a note.
“We now have some data on vaccine rollouts, which show that acceptability is a bit on the low side, so pace of implementation may be slow… There may well be a bearish momentum developing (in oil markets),” said Sukrit Vijayakar, director of energy consultancy Trifecta.
The market is awaiting official oil inventory data from the U.S. Energy Information Administration on Friday, after industry data on Wednesday showed a surprise 2.6 million barrel increase in U.S. crude inventories last week compared with analysts’ forecasts for a 1.2 million barrel draw. [API/S]
(Reporting by Sonali Paul in Melbourne and Koustav Samanta in Singapore; Editing by Ana Nicolaci da Costa & Simon Cameron-Moore)
CEO Hisham Itani and Resource Group Recognized in the 2020 Global Banking & Finance Awards®
Global Banking & Finance Review has awarded Hisham Itani the Chairman and CEO of Resource Group, Technology CEO of the...
Euro zone business activity shrank in January as lockdowns hit services
By Jonathan Cable LONDON (Reuters) – Economic activity in the euro zone shrank markedly in January as lockdown restrictions to...
Volkswagen’s profit halves, but deliveries recovering
BERLIN (Reuters) – Volkswagen reported a nearly 50% drop in its 2020 adjusted operating profit on Friday but said car...
Global chip shortage hits China’s bitcoin mining sector
By Samuel Shen and Alun John SHANGHAI/HONG KONG (Reuters) – A global chip shortage is choking the production of machines...
Iran’s oil exports rise ‘significantly’ despite sanctions, minister says
DUBAI/LONDON (Reuters) – Iran’s oil exports have climbed in recent months and its sales of petroleum products to foreign buyers...
Nissan to source more UK batteries as part of Brexit deal ‘opportunity’
By Costas Pitas LONDON (Reuters) – Nissan will source more batteries from Britain to avoid tariffs on electric cars after...
Muted recovery for UK retailers in December ends worst year on record
By David Milliken and Andy Bruce LONDON (Reuters) – British retailers struggled to recover in December from a partial coronavirus...
Chinese phone maker Honor partners with key chip suppliers after Huawei split
By David Kirton SHENZHEN, China (Reuters) – Chinese budget phone maker Honor said on Friday it had signed partnerships with...
Oil down $1 as China COVID-19 cases trigger clampdowns
By Noah Browning LONDON (Reuters) – Oil prices fell on Friday, retreating further from 11-month highs hit last week, weighed...
Stocks ease as gloomy data takes shine off U.S. stimulus hopes
By Tom Arnold and Swati Pandey LONDON/SYDNEY (Reuters) – Global shares slipped off record highs on Friday as gloomy data...