Driving Major Improvements in Treasury Processes

Toyota Financial Services (TFS) is the captive finance company that provides financing, leasing, insurance, and protection plans to both consumers and auto dealers in support of the North American divisions of Toyota and Lexus. The largest captive finance company in the world, TFS has more than $115 billion in managed assets and originates about $3 billion in new contracts each month.

When Covid-19 hit, TFS discovered—like many other businesses around the world—that processes which were efficient in an environment where everyone worked in the same office space became less effective when most staff began working from home. The treasury team launched a series of cross-functional initiatives to digitalize, automate, and build real-time data and analytics into the firm’s business processes, with a goal of enhancing productivity throughout the treasury function.

“Years ago, we started building a datamart to support our smaller SFCs [sales finance companies], whose staff may not have expertise in every specific area of risk management,” explains Sylvia Baharet, national manager of balance sheet strategy for TFS. “They needed someone to provide feedback on their risk metrics and help guide hedging and similar decisions.”

The goal of the datamart initially was to consolidate risk information from several countries in the Americas. The U.S. division of TFS also provides treasury services for Mexico and Canada; a custom-built tool pulls information from that group’s treasury management system into the centralized datamart. External information from Bloomberg, credit rating agencies, and other third parties provides additional context to the internal treasury feed.

“Information on our cash positions, debt, commercial paper, derivatives, and investments is refreshed every five minutes or so,” says Irvin Lee, Toyota Financial Services’ national manager of treasury data and automation. “We pull in market valuation information. We even do some Web scraping, pulling certain rates off the Fed site, for example. It’s pretty extensive.”

Because they have access to so much information in one place, TFS treasury and risk functions can react more quickly to changes in the business or external environment. “Our treasury risk team uses the datamart to figure out our liquidity metrics,” Lee says. “The front-office team uses it to look at maturity schedules and monitor investments based off ratings, concentration limits, and similar factors. And finance and management teams can answer questions independently. If the data indicates a looming problem with a counterparty, everyone can access the data rather than calling around to try to get a picture of all our exposures.”

Baharet adds that the datamart is particularly helpful to those managing TFS’s commercial paper issuance. “We can see immediately what yield we’re issuing, what our derivative position is, whether we’re hedging the notional or counterparty exposure, and what currency it’s in,” she says. “The real-time availability of this information is very useful.” In fact, the regional datamart has been so effective that TFS management has improved and extended it.

“After our parent company saw what the tool had done on a regional level, they asked us to make it global,” Baharet explains. “They wanted to ensure we were all speaking the same language and calculating our numbers in the same way. Expanding the tool to more than 35 affiliates across the globe, which have treasury data in multiple systems and locations, was a journey of interpreting data, cleaning up data, and dealing with nuances around how different liability instruments work in different places.”

Baharet’s team upgraded the tool that pulls information into the datamart, leveraging bank application programming interfaces (APIs) to retrieve real-time account balances and transaction data from banks around the world. They also developed a system of tests to validate data that the tool retrieves. As an example, Baharet points to double-checking of intercompany loans.

“Let’s say we have an intercompany loan from Canada to the U.K.,” she says. “We would compare the positions in the general ledger between the two entities. One would need to have an asset and the other a liability, and if they didn’t match, that would indicate a hole in the data that corporate treasury would need to look into.

“Coming up with new checks is an ongoing process,” she adds. “We build them into the code, and if they reveal that something isn’t consistent, then we work with the business units to figure that out. We find things all the time, but we know that little by little we’re getting better.”

At the same time its automated data retrieval processes were improving, TFS treasury embarked on an initiative designed to streamline collateral management. When the pandemic struck, the company was using a decade-old platform to handle daily collateral servicing for the derivatives with which it mitigates interest rate and currency risk. The legacy system required several different internal groups to perform various manual tasks.

“We post collateral and receive collateral on a daily basis from our counterparties,” Lee says. “Folks were spending hours every day pulling valuation statements and searching through emails, looking for correspondence from counterparties. That was really time-consuming before Covid, but when we all started working from home, it became even more difficult because there was less open communication. We wanted to make the process much more automated.”

Lee’s team developed an application in-house that both generates and retrieves margin call emails to and from counterparties; retrieves and parses valuation statements from more than 40 disparate counterparties each day; and generates and delivers cash receipts, payments, and forecasts. The application pulls information from TFS’s valuation system, pulls mark-to-market statements from the firm’s counterparties, and compares the two.

See also:

• Improving Investment Management via Better Portfolio Visibility
• The on-demand webcast celebrating our winning projects: Innovative Strategies to Improve Treasury Operations Through Technology

“It will use the agreement logic to determine the appropriate movement amount,” Lee says. “This saves a huge amount of time because collateral management involves so much data collection. The application also monitors our emails for counterparty correspondence and pulls data from them automatically. So, if a counterparty sends a margin call or responds to a margin call, the logic embedded in the application enables it to understand the sender’s intention. Are they fully agreeing to our margin-call request, or partially agreeing? In computer science speak, the application uses regular expressions to decipher the text of those emails.”

Lee’s group also leveraged bank APIs to further automate processes. “After our operations team has gone through the process of reviewing and approving collateral for a counterparty, the application will use the bank APIs to validate the amount of funds to move,” he says. “It’s really an end-to-end process that requires human intervention only for the review and approval phase.”

The end result is that the collateral servicing application has freed people up to work on more analytical work, while also removing the risk of human error because individuals aren’t typing in so much data. And, Lee says, “it enables our accounting team to get a better read on our collateral balances, accrued interest off of those balances, and things like that.”

Lee and Baharet give their teams kudos for the outside-the-box thinking needed to so dramatically revamp treasury processes. “Before, we were stuck in our ways,” Lee says. “People were essentially just checking items off the list every day—like, step one, step two, step three to get the job done. But when we started redesigning our processes, we took on a new way of thinking. We looked at ‘maybe we don’t necessarily need step number three.’ Or ‘maybe we can do step one and step four at the same time.’ It boils down to looking at everything with a fresh set of eyes, seeing how processes can be rearranged and what technologies can make that possible.”