Tuesday, June 11, 2013

Answer These Questions For A Better Cash Flow Forecast

The desk phone rings in your office at Joe's Agricultural Empire, Inc.

The sound of the CFO's voice can be heard even before you get the receiver to your ear.

"We need a cash flow forecast now! We had an investor meeting yesterday and the information we provided was completely unacceptable. Meet me in my office this afternoon."

As you replace the receiver, never having been given the chance to speak a word, your mind immediately generates a compelling list of questions: "What did we provide?" "What do investors want to see?", "Just what does a good cash flow forecast look like?" "What can be delivered by this afternoon?"

Why Do We Forecast?

Cash flow forecasting is a core organizational process for a number of reasons. On a daily basis we want to ensure that there is money in the bank for the payments we are making - we don't want our checks to bounce.

Longer term, there are financial aspects of our organization that we want to consider. Perhaps we have debt covenants that need to be maintained or financial ratios we target in order to preserve our credit ratings. Or we may want to have the capability to undertake "strategic transactions" such as buying out our competitor in an M&A transaction.

These objectives - liquidity management, budgeting, financial control - are important in a world where capital markets can dry up overnight in response to a crisis and where investors can lose faith in a firm rapidly.

There can be innumerable other reasons, but underlying all of the above is the fact that cash is the lifeblood of the business and you simply cannot afford to run out of it.

A Large Menu to Choose From

Factors that influence the construction of a particular cash flow forecast are:

    ::Purpose - as mentioned, the objectives for cash flow forecasting vary, such as managing near-term liquidity, establishing a financing plan as part of a budget process, maintenance of financial ratios, and support of strategic initiatives, among others.
    ::Period - cash flow forecasts can be performed for any given length of time - a day, a week, a month, a year, or a decade
    ::Process - there are a number of methods, templates, and statistical approaches we can use to develop our cash flow forecast, generally ranging along a continuum from 'top-down' tp 'bottom up'
    ::People and Resources - our cash flow forecast can involve minimal or intensive organizational involvement, ranging from creation within the Treasury or Finance function to complicated, involved and coordinated inputs from throughout the organization. The technology and software requirements may also range from simple spreadsheets to complex, interconnected software packages.

Given the number of factors and the choices available within each category there are thousands of possible combinations. Many of these, however, are simply not practical. For example, a day by day forecast for the year 2020 (it's now 2013) is just not going to be relied upon by anyone.

The approach we take to create a cash flow forecasting process depends on the answers to a series of questions related to the factors just outlined:

    ::Who Is The Customer of the Forecast?
    ::What Time Period Needs To Be Covered?
    ::What Level of Precision and Detail is Required?
    ::What Information is Available for Us to Use?
    ::How Much Money, Time and Effort are to be Spent?
    ::What Specific Methods and Techniques Will Produce the Most Reliable Results?
    ::What Information Needs to be Produced in Order to Support Decisions and Action?
Who Is The Customer?

We produce cash flow forecasts because the information is used in making decisions, so we need to understand who the decision maker is.

The decision maker may be an external party, such as investors, regulators, or analysts, internal areas of the organization, such as marketing or production, or someone within our finance organization.

Keeping the decision maker in mind - putting a face to the name so to speak - helps us answer questions as we go through setting up and executing the forecast, and emphasizes that our work is going to a real live person - a beneficial thing to keep in mind.

In thinking through the situation at hand, it can be a little unclear who the customer actually is. The firm's investors obviously want some additional cash flow information, so plainly they may be viewed as the customer. On the other hand, the CFO is the one presenting the information to them, and therefore perceptions as to the quality and responsiveness of the information, good or bad, will 'rub off' on them. Is the CFO the customer?

The Forecast Time Horizon and Precision
Figure A
Forecast Horizon and Utilization

If we create a 2x2 matrix, as shown in Figure A, with Time Horizon on one axis and Purpose on the other, in general we can focus on only two of the four boxes as shown by the blue line. Short-term horizon forecasts are generally for tactical purposes such as managing the upcoming day's cash and investment needs. The longer-term forecasts are for the more strategic level issues such as capital investment planning, financing activities and the like.

The purpose - tactical or strategic - also dictates in large part the level of, and type of, precision required. For tactical activities a higher level of 'fine-detail' precision is needed. If we are short by even a small amount in our bank account funding, we may find that all the payments clearing at that institution are not honored, resulting in a lot of unhappy vendors and a slew of competitors whispering dire warnings into our customer's ears about our survivability.

Strategic decisions require lower levels of 'fine-detail' precision. If we are "off" by a few thousand dollars this way or that, it will hardly matter if the overall activity we are contemplating is valued in the millions. This is not to say we can be 'sloppy' - decisions based on this information carry a much greater magnitude of impact across the organization than tactical level data, so we still need to be precise from a strategic point of view. This may mean casting a 'wider net' in terms of organizational input, for example, or evaluating a number of different scenarios involving the mix of business segments.

Taking the needs of investors into account, you believe a longer-term, strategic level cash flow forecast is the appropriate route to take.

What Information is Available and How Costly is it to Obtain It?

In an ideal world every relevant piece of information is available to us in an easy to use format.

Unfortunately, we live in the real world rather than the ideal one.

What this means is that acquiring and processing information involves a cost/benefit tradeoff. If we want more information than we will need to spend more time obtaining it. If we want to tap into all that "Big Data" people talk about we will need to invest in systems and interfaces, and then a) take the time to operate them and b) spend the money to maintain them.

In the 'big picture' of organizational life the forecast process is not a revenue generator. A good forecast may allow us to save a few costs here and there, but these savings will be crumbs compared to the revenue generation from the organization's core lines of business. For this reason, the cash flow forecaster generally faces a 'do more with less' dilemma.

Making matters worse, the value creation side of the cost/benefit tradeoff is not very amenable to quantitative analysis. We do not have the luxury of comparison to determine. What decision would be made without the information vs. the decision that was made with it? How do these translate into revenues and costs? The 'road not chosen' cannot yield any reliable answers.

We require both internal and external information for the cash flow forecast. Internal information comes from the procure-to-pay and order-to-cash processes, from groups within the company such as AR, AP, Supply Chain and the like. Areas responsible for large, "chunky" fund movements, such as tax or risk management (if we hedging using financial products) may also be sources of required information. Finally, major sources of revenue and costs might also be involved.

One primary external information source is the organization's bank data. Beyond that, the sky is really the limit. Is there a strong correlation between collections and new Twitter followers? Then perhaps we need to scrape that website everyday! Firm's that have a revenue or cost association with weather (e.g. agriculture, energy) might need to factor this type of data into their forecast processes. Longer term forecasts might require econometric inputs such as GDP, Money Supply, Unemployment rate, etc. into the mix if they are relevant cost or revenue drivers.

With your afternoon deadline looming large, you realize that whatever forecast is to be created must be done with readily available information, there simply is not the time or manpower available to research and produce additional items.

What Method Should We Use?

There are a variety of methods that might be utilized for a cash forecasting process, but all can be classified as falling into the 'top-down' or 'bottom-up' archetypes.

We use the term 'archetype' since, as a practical matter, all forecasting processes contain nuances which make them a particular blend of both. No two organizations do it exactly the same, just as Starbuck's coffee has a different 'flavor profile' from Dunkin' Donuts, and McDonald's tastes different than Burger King, even though both are called 'hamburgers'.

The top-down approach starts with higher order items and based on those derives the cash implications. Thus, the cash aspects of the process are derivatives, in the sense that they have been created from other data.

Figure B
Using Cash Conversion Cycle to Forecast Cash Flow

Shaded items represent inputs

For instance, if sales are going to be x, then we can apply our cash conversion cycle information to derive AR, AP and Inventory balances, and based on these work out the cash flow.

In Figure B, the cash conversion cycle metrics are inputs into the calculation. Ending Balance Sheet categories - AR, Inventory and AP - are calculated by using the cash conversion cycle equations and rearranging them to solve for Ending Balance. The month to month change in these balances represents change in cash. In Figure B, our maximum funding need over this period is 170, while our maximum amount of excess cash is 10.

A procedure similar to this is known as the "percentage of sales" method. In this technique we calculate each ending balance sheet category by applying a percentage of sales for the period.

For example, if ending AR historically ranges between 125% to 135% of period sales, then our ending AR balance for the forecast in Figure B would be between 125 to 135 for Month 1 (125% * 100 or 135% * 100) and 187.5 to 202.5 for Month 2 (125% * 150 or 135% * 150).

Once these ending balances are calculated, we would proceed with the process shown in Figure B - calculate the changes from period to period in each category and sum them together to arrive at net cash flow for the period.

The 'bottom-up' approach starts with individual items and/or categories and builds them layer by layer upward to the cash flow forecast.

One such approach is the Receipts and Disbursements Forecast. This method involves evaluating upcoming cash flows and places them within two broad categories - receipts and disbursements. It's financial statement equivalent is the Direct Method Cash Flow Statement, though normally this forecasting technique does not attempt to produce a full set of financial statements (i.e. Income Statement, Balance Sheet, Cash Flow Statement).

How each section is organized will vary from firm to firm. As I have often emphasized in this blog, there is no substitute for thinking as we go about our work. No magic bullets, one-size-fits-all templates, or standard methods.

Figure C shows an example of a Receipts and Disbursement Forecast for Joe's Agricultural Empire. The Receipts line items are broken down by revenue source - the farmstand that sells produce by the highway; the lockbox that processes invoice payments from customers such as restaurants, wholesalers, etc.; sales over the web; sales to biomass energy producers; and receipts of proceeds for any financing activities.

Figure C
Receipts and Disbursements Forecast

Note that the disbursement catagories are primarily by payment type as opposed to the revenue generating activities. One of the advantages of the Receipts and Disbursements approach is the flexibility it offers - we can have as many line items as we wish categorized in any way we decide is best. We might choose this disbursement approach due to how our historical data is organized or because the data is supplied by different identifiable areas (such as HR, Tax, etc.).

Given the long-term nature of the forecast you seek to develop, you decide that a top-down approach is more feasible given the time-frame. Historical percentage of sales figures for your financial statements can be done within an hour and evaluated. Your contacts in the sales and marketing areas have projections that will allow you to construct a reasonable span of time in the future, to which you can apply the percentage of sales method to construct a cash flow forecast.

What Can You Do?

Once we understand the elements and choices that go into the cash flow forecasting process, what are some of the things we should focus on in order for the process to 'click' within our firm?

    ::Give serious thought to each of the questions asked in the "Large Menu" section of this blog. The more up-front work you perform will allow you to a) develop forecasts that meet your needs more quickly, b) rapidly weed out unviable alternatives, and c) approach other areas within your organization with a stronger sense of purpose and clarity.
    ::Make sure your answers reflect the needs and capabilities of your organization from many different perspectives (e.g. production, sales, supply chain, etc.). A usable cash flow forecast requires 'buy-in' from other parts of the firm. Considering different points of view up-front demonstrates empathy with your organizational 'customers' and increases the chances that you will produce a viable forecast with a practical process.
    ::Experiment with multiple approaches and attempt to 'triangulate' a better collective result. The quality of the forecast is greatly enhanced if we can supply corroborating evidence.
    ::Store forecast data in a way that can be used for variance analysis or reproduced in required formats. This is required if we want to objectively evaluate performance through time.
    ::Engage others in all steps of the process. The more others are involved the more they will share in the ownership of the forecast. This is a critical feature if continuous improvement is a concern.
Key Takeaways

The cash flow forecast is a core operational element within the organization. There are a number of factors that will determine the best way to proceed with this process, which will depend in large part on the specific needs and circumstances of the firm.

You May Also Like
    ::What other 'top-down' forecast methods have you undertaken?
    ::What other 'bottom-up' forecast methods have you undertaken?
    ::Do you have any thoughts on the helpfulness of the story format used in this post?