by Robert van der Zee, Treasurer & Deputy Director of Finance, UN World Food Programme
Improving cash and treasury management processes is not simply a tactical means of reducing costs or demonstrating best practices in an academic sense. Instead, as Robert van der Zee outlines in this article, for an organisation like the World Food Programme, optimising cash and treasury management is having a direct impact on its ability to deliver food assistance and reduce hunger in crisis-struck communities around the globe.
Key facts: WFP Treasury
- 9 treasury professionals
- Nearly 400 bank accounts across 88 banks in 93 countries
- USD 2.1bn in investments (2015)
- USD 0.6bn in FX transactions each year, from hard currencies to 32 local currencies
WFP Treasury Organisation
WFP has a small, centralised treasury organisation of eight people including myself based at our headquarters in Rome, Italy. I act as both treasurer and deputy director of finance. As a small team, with enormous responsibilities to beneficiaries and donors, it is essential that we optimise our use of head office resources, and minimise the amount of time spent by field offices on administration. Consequently, over the past four years we have embarked on a process of review and optimisation of our treasury processes and infrastructure to ensure the flow of transactions and information is as efficient, transparent and cost effective as possible.
Rationalising bank relationships and connectivity
A key element of our treasury optimisation strategy was to centralise cash and harmonise cash management processes wherever possible. As a first step, we wanted to rationalise our banking partners to reduce fragmentation of cash, and manage our counterparty risk more effectively. In 2012, we launched a request for proposal (RFP) for global banking services, and ultimately appointed four banks: two global banks, and two regional African banks, given the large number of programmes we operate in Africa. Although we still need to maintain relationships with local banks for access to branch networks, selecting global banking services was a major step forward in rationalising and simplifying our cash management arrangements.
The next step was to communicate securely and reliably with our banks, both at a group treasury level and in our field locations. Although we had implemented a single instance of SAP as our ERP system across the organisation, we had very little connectivity between SAP and our banking systems, so payments had to be input into each system individually, resulting in duplication of effort and a potential loss of control and accuracy. We considered implementing a specialist treasury management system, but the business case was not sufficiently compelling, as our treasury activities are relatively straightforward: for example, we outsource most of our investment management, and we do not make use of complex derivatives for FX hedging. Furthermore, we operate as a single entity, with a number of branches, which further simplifies our treasury needs.
Therefore, we decided to create a more integrated payments infrastructure based on SAP, and designed our process roadmap with PWC. The new design created a system for setting up payment instructions and approving payments either centrally or remotely (using SAP’s Bank Communication Manager). Files are then formatted as appropriate (using SAP’s Process Integrator) and transmitted to our banks via SWIFT. We use MT101 messages for treasury payments and FileAct protocol for all other payment types.
SWIFT in Africa
The decision to use SWIFT rather than banks’ proprietary electronic banking solutions was a significant one, and leveraging a bank-neutral, multi-bank platform rather than individual electronic banking systems has been crucial in providing greater security, flexibility and efficiency in bank communications. The decision to adopt SWIFT is relatively uncommon among organisations with significant emerging market activity. We recognised, however, that although it would be more difficult to implement SWIFT in some countries compared with others, we estimated that we would be able to connect 60 percent of our offices to partner banks (although this is proving to be slightly higher).This also equates to around 90 percent of our total volumes. As we did not have the resources to manage the SWIFT infrastructure directly, we appointed SunGard (now FIS) as our service bureau.[[[PAGE]]]
We started our SWIFT roll-out with our two international banks, Citi and Standard Chartered, including our headquarters and a selection of pilot countries. We have now rolled out the solution to around 30 operations, and while there are challenges with file formatting in some countries that have a less-developed payment infrastructure, it has been worth spending the time to resolve these issues rather than reverting to proprietary electronic banking systems. In such countries as Sudan and Ethiopia, it is not feasible to implement SWIFT in the foreseeable future due to lack of international banking partners with a presence, but we will continue to review the possibilities.
Managing cultural and operational change
While it is important to measure the operational benefits – which have been considerable - of implementing a new banking and liquidity framework, it is just as important to evaluate the response from users, both at headquarters level and in the field. In some countries, such as Uganda, implementing automated processes and electronic payments represented a major cultural and operational change, as around 80% of payments had previously been made via cheque. Consequently, there was significant effort required to educate both WFP users and our suppliers on the benefits, and to collect the relevant supplier data in order to make electronic payments. Inevitably, there were some ‘teething troubles’ when converting from manual to electronic payments, with some payment rejections to start with, but we have worked with field offices to resolve master data issues quickly.
Now that we are seeing the benefits of efficient payment processes and SWIFT connectivity, there is greater demand among field offices, thus creating a new challenge to prioritise countries as part of the roll-out plan. For example, while Turkey hasn’t been a key market for WFP in the past, WFP is now providing significant humanitarian assistance in the country, so it has become a much higher priority to make sure we can maximise the impact of the work we do, and support cash-based transfers.
Centralising cash, managing liquidity and risk
In addition to building improved processes and infrastructure to optimise financial and information flows, cash pooling was an important aspect of our project. We already had euro and USD cash pools in place, and subsequently converted funds to local currencies required to fund each programme. However, with major programmes in regions such as west and central Africa, particularly Niger, Chad, Cameroon, Central African Republic and Mali, we decided to work with Citi to build local currency cash pools. We looked to two groups of countries in the region - the Central African Economic and Monetary Union (CAEMU), and West African Economic and Monetary Union (WAEMU) – and within each is a common currency (XAF and XOF respectively), a common central bank and clearing network. Electronic payments within both CAEMU and WAEMU are treated as domestic rather than cross-border payments through bank accounts located in any country in the region. Cash pooling is permitted, and transaction pricing is consistent within each group of countries.
In 2014, we implemented a XOF cash pool in seven WAEMU countries: Benin; Burkina Faso; Cote d’Ivoire; Guinea-Bissau; Mali; Niger, and Senegal. We located the master account in Dakar, Senegal (although some organisations that have implemented XOF cash pools have set up their header account in Cote d’Ivoire) and set up a sub account for each country office with Citi. We also implemented a similar structure in CAEMU, with a XAF-denominated cash pool covering Cameroon, Central African Republic, Chad and Congo Brazzaville, with a header account in Yaoundé, Cameroon. Local offices conduct payments from their sub-account, reconcile their bank account movements etc., in the usual way, but balances are zero balanced at the end of each day.
These cash pools have helped us to manage our XOF and XAF liquidity and counterparty credit risk more effectively by centralising balances in one place rather than holding balances in multiple banks and accounts. In addition, we exchange messages using XML ISO 20022 standards and receive bank statements in MT940 formats, allowing us to standardise and streamline processes. While we still maintain local bank providers for certain payment types, we are able to direct high-value payments through Senegal or Cameroon respectively via SWIFT.
Minimizing costs in FX risk management
Each year we conduct around USD 600m in FX transactions from hard currencies into 32 local currencies, and XOF and XAF are particularly important currencies that account for around one-third of this total value. As the margins can be quite high on exotic currency conversion, we need to seek competitive bids on each of our transactions, although in reality there is usually a limited number of banks that are able to offer the best price for each currency pair. We therefore implemented independent trading portal 360T to allow us to invite bids from all twelve of our FX banks for each transaction. This means that we have a transparent and auditable process, allowing us to reduce our FX costs whilst providing feedback on banks’ success rate in winning bids and demonstrating what they would need to do to win the business.
Driving project success
There is often a perception that Africa is a ‘difficult’ continent from a cash and treasury management perspective and while there are undoubtedly challenges (as indeed there are in every region) these are not insurmountable. For example, we have been able to implement SWIFT in order to streamline and automate payment processes, retrieve bank statements in a standardised format, and put in place efficient cash pooling mechanisms. We continue to maintain around 80 bank relationships for operational reasons, but we now hold three-quarters of our bank balances globally with our four core banking partners. While it is not always more economical to work with international rather than local banks in terms of bank charges, we have increased visibility and control over liquidity and risk, while reducing costs by implementing more efficient business processes.[[[PAGE]]]
There have been a variety of factors that have contributed to the success of our cash and treasury management initiatives so far. Both WFP and our banks needed to dedicate project management and IT resources, and we adopted a disciplined approach to aligning our test and production environments. In addition to the infrastructure, transaction automation depends on the completeness and accuracy of master data, so it was important to focus on this and ensure field offices received appropriate training. Given that Africa is such a diverse continent, we needed clear visibility of data and format requirements, central bank reporting, currency conventions, etc., in each country. Our banks have played an important role in helping us to comply with requirements in individual countries.
Looking ahead
We have now channelled around 54% of our disbursement flows through SWIFT, and will continue with our rollout project in all countries in which it is feasible. We are also finding that cash-based transfers are becoming an increasingly important means of providing food assistance to beneficiaries, a change prompted by the Syrian crisis in particular.
In most of the countries in which we operate, we have historically delivered traditional food aid. However, since the Syrian crisis began, we have been increasingly assisting refugees in other ways in countries such as Jordan, Lebanon and Turkey where the need for aid is acute, yet functioning supply chains are in place and fairly sophisticated financial services are available. In these situations, we can often deliver aid more quickly through more innovative programmes such as prepaid debit cards. These allow beneficiaries to purchase food and essential supplies from selected retailers while offering the added advantage of bringing income in to the local economy.
The delivery mechanism for cash-based transfers varies across countries. For example, we are now piloting a mobile money solution in north-eastern Nigeria to meet the need there for food assistance, which has in part resulted from the effects of the Boko Haram violence and the collapse in world oil prices.
It is clear that the changes brought by technology have been dramatic. In 2012, we delivered around 7 percent of food assistance through cash-based transfers; in 2016, we expect that to reach 29 percent. This is likely to increase even more over time as technology infrastructure and digital, financial and physical supply chains develop in emerging continents. We therefore have wanted to implement a mechanism that would enable us to rapidly deliver aid globally via cash-based transfers, leveraging the payments and telecoms infrastructure within the country. To this end, we launched an invitation to tender, inviting a variety of different organisations including banks, card providers, mobile network operators etc., to participate in a global programme that would allow us to respond to disasters – whether natural or manmade - quickly, according to the needs of each country and programme.
We are now coming to the end of this tender process, and once we have implemented the programme, we will be in a better position to quickly operationalise cash-based transfers to alleviate poverty and hunger and increase food security wherever environmental or humanitarian disasters strike.