CorporateTreasurer

J.P. Morgan’s View: Best Practices in Cash Flow Forecasting

By J.P. Morgan | Feb 1, 2018

In this new monthly column, J.P. Morgan responds to pressing questions around cash management expressed by treasurers and finance practitioners in Asia-Pacific. This instalment will provide you with simple tips on improving cash forecasting practices.

Q. What are the best practices in cash flow forecasting?

Cash flow forecasting is an essential tool for all companies and provides treasurers with numerous benefits. Good forecasting can help treasurers optimise their cash buffers as well as serve as an early warning system for potential cash shortfalls.

Treasurers should consider three areas to ensure high standards in cash flow forecasting:

Focus on both short-term and medium-term forecasting. Cash flows should be forecasted on a daily, weekly and monthly basis. Short-term forecasting covers periods of up to 30 days and includes daily and weekly views. Medium-term forecasting provides monthly projections of up to one year.

While short-term cash forecasting helps the treasurer effectively execute daily investments and funding actions, a medium-term forecast can help optimise the duration of investments, hedge maturities and minimise funding mismatches.

Automate everything as much as possible. Almost 50% of companies cite a lack of automated tools as a key challenge to forecasting, according to a 2017 survey by PwC [1]. Having automated processes that gather and consolidate different inputs can significantly improve forecast accuracy and reliability. They also provide greater efficiency to cash managers, allowing them to focus on the more qualitative aspects of forecasting.

Review cash flow forecasting variances regularly. Forecast variance – or actual versus projected cash flows – should be analysed on a daily basis. Large variances should be investigated and promptly escalated to improve the quality of forecasts. There are technology tools that can automate and improve this process.

Q. How can I improve the accuracy of the forecasts?

Forecasting is as much an art as it is a science. Accuracy is largely dependent on the experience of cash managers who supplement the forecast with adjustments based on historical trends and patterns.

Cash flow forecasting is generally compiled from treasury and business flows. Treasury flows include liquidity balances and cash movements expected as a result of investments, funding and foreign exchange transactions, while business flows include accounts payable and collections projections received from the finance and sales teams. Finer adjustments to the forecast are then carried out by the treasury considering historical trends, seasonal patterns and end-of-period adjustments.

Gourang Shah
Gourang Shah, Head of APAC
Solutions and Outbound Sales
Treasury Services, J.P. Morgan

Even companies with sophisticated TMS applications that have automation functions sometimes face issues with non-standard data formats, multiple information sources and system integration problems. A lack of integration between ERP and TMS results in a process that remains largely manual, often requiring treasury teams to complete the forecast.

Is data analytics the answer?

With technology becoming increasingly sophisticated, many companies are looking to data analytics to enhance the accuracy of forecasting. Analysing historical payments and collection flows can provide valuable insights and visibility to working capital trends, seasonality and anomalies.

But the process of creating a data lake – a storage repository that holds a vast amount of raw data – ­to run analytics tends to be costly and onerous for many organisations. A more immediate step may be to leverage the analytics services offered by banks on transactions already flowing through their networks. This not only provides meaningful business insights to treasury, but also helps companies visualize and plan for longer-term data strategy.

Over the past year at J.P. Morgan, we have seen an increasing number of clients leveraging our interactive transaction analytics dashboards for gaining working capital insights. Via the dashboard – which is offered as a complementary value-added service to our key clients in the region – users are able to view their payment and collection corridors, volume trends, transaction frequency as well as the analytics around supplier and customer concentration.

Q. Tell me more about automation in cash forecasting. What tools do I need?

Treasurers should consider enhancing their existing platforms like TMS/ERPs to automate the collection and consolidation of forecasting inputs.  However, the level of customisation required in this approach may translate to significant cost, integration efforts and implementation timelines for some companies.

Alternatively, treasurers can take a modular approach and use platforms offered by third parties or banks that are cost-effective, fast to implement, and flexible with integrating data and setting up the necessary forecasting templates.

Varoon Mandhana, Senior
Advisor, APAC Solutions,
Treasury Services, J.P. Morgan

At J.P. Morgan, we have successfully addressed this need for a large number of clients using our proprietary J.P. Morgan ACCESS® InsightSM, a wizard that uses a Microsoft Excel add-in to auto-populate balances on client accounts held with us as well as third-party banks. The platform, which is offered as a free value-added tool for clients, also allows users to create customised Excel templates, import transaction information required for forecasting, define rules and trigger payments in a secured manner.

Q. What role can new technology play?

Traditionally, cash flow forecasting tools have used statistical regression models that rely on repeating trends in data to forecast an outcome. However, as business cash flows are a function of large number of variables, including industry trends, seasonality, business sales cycles, collection patterns, cost of sales, statistical models may not fully capture and process the complex patterns and relationships between these variables. As a result, the forecasting process has largely remained dependent on the competence of finance practitioners.

Thanks to advances in artificial intelligence, intuitive cash forecasting methods involving neural network technologies that closely replicate the way people think are changing the game. Unlike statistical methods, these neural networks identify complex patterns in cash usage and automatically adapt and adjust the forecasting model.

At J.P. Morgan, we recently piloted machine learning to help clients predict operating cash inflows and outflows, as well as determine the optimal level of operating liquidity required to support fluctuations in working capital. These sophisticated tools and techniques will help improve the overall quality of forecast.

Q. How do I forecast cash flow for the long-term?

Long-term forecasting is critical to corporate decision-making in areas like capital planning, budgeting, strategic investments and long-term funding decisions.

The forecasting is usually done by corporate finance and planning teams to capture the accounting projection of revenue, expenses and changes in balance sheet over three-to-five years. This normally involves a budgeting and planning exercise which looks ahead at sales and expense projections, the ratio of working capital items to sales and the projection of other balance sheet items that capture cash flows in financing and investment activities.

Long-term forecasting should be subject to sensitivity analysis to make allowances for factors like, but not limited, to currency fluctuations, interest rate movements, inflation impact, economic influences and other industry and market changes. Companies using sensitivity analysis produce cash forecast under multiple scenarios for corporate decision-making.

To learn more, please contact:

Gourang Shah
Head of APAC Solutions and Outbound Sales
Treasury Services, J.P. Morgan
[email protected]

Varoon Mandhana
Senior Advisor, APAC Solutions,
Treasury Services, J.P. Morgan
[email protected]

 


[1] Source: The ‘virtual reality’ of treasury – Global Treasury Benchmark Survey 2017, PwC

 

© Haymarket Media Limited. All rights reserved.