Effective Cash Flow Forecasting Techniques

by Bennett Quillen

There are four steps the cash management
professional needs to apply in developing and
refining periodic cash flow forecasts. This article
outlines the key steps.
Bennett Quillen,
Too often business managers focus on
the means rather than on the ends of
cash flow forecasting. Tomes have been
written on sophisticated cash flow forecasting
techniques and computer models.
Sometimes a treasury professional gets
the impression that by merely applying
these tools, she or he can perform better
cash flow forecasts.
That is not to say that these methods
are unnecessary, but they need to be judiciously
applied in the context of an organization’s
business objectives. The cash
management professional responsible for
developing cash flow forecasts needs to
ask: what is the nature of my business
and what is the business trying to accomplish
with its cash flow?
This article outlines four steps the cash
management professional needs to apply
in developing and refining periodic cash
flow forecasts. The steps are:
1. Define end-user cash requirements.
2. Defino and segment the factors that
influence demand for cash roquiraments
and sources.
3. Apply forecasting tools to the cash requirements
and demand factors.
4. Refine the forecast as appropriate.
Define End-User Cash Requirements
To define end-user cash requirements,
functional and timing objectives need to be
identified. Functional objectives can usually
come straight from the general ledger.
Unfortunately, the typical general ledger is
a poor vehicle from which to make forecasts.
It is simply a tool to identify appropriate
cash flow categories such as:
• Accounts receivable
• Investments
• Trade payables
• Payroll
• Capital expenditures: plant and
• Tax payments.
Timing will typically fall into weekly,
monthly, and quarterly needs.
To understand the end -ueer cash requirements,
the cash manager should discuss
each of the functional categories (and
their subledger accounts) with the individuals
that make decisions for these accounts.
These individuals will include representatives
from the sales, treasury, purchasing
(including vendors) and accounting
Cash management professionals should
x x
Figure 2: Industry/Cash Flow Factors
Industry Examples
Cash Flow
Dependent Factors Utility
Processor Publisher
..,2 years
* 2 months
Customer Base
product Line
x x
Segmenting Industry Cash Flow-
Dependent Factors
Consequently, each industry or business
has a unique set of factors that determine
its cash flow. Figure 2 illustrates an example
of industries and the factors that determine
their sources of cash.
then deducts outlays that do not affect
profits (e.g., capital acquisitions and dividend
Balance Sheet Projections
Both the Balance Sheet Projection and
Working Capital Extrapolation methods
start with the balance sheet.
The Balance Sheet Projection looks at
changes in the balance sheet to determine
cash inflows and outflows. This method is
more applicable to seasonal trends.
Working Capital Extrapolations
Beginning with the balance sheet, this approach
uses anticipated changes to working
capital by assessing each revenue and
expense category as to whether it will be a
source or use of cash. The method is best
applied to long-term forecasting.
Accrual Addback (1)
The Accrual Addback method is somewhat
unique in that it can be prepared relatively
easily with a limited number of data
sources. The approach is similar to the
Adjusted Earnings method, but it finetunes
historical data and the distribution
of non-cash accruals. It is extremely accurate
in the 3- to IS-month timeframe.
Forecasting Tools
After end-user cash requirements and dependent
factors have been specifically defined
for a certain business, the cash manager
may apply one of five prevalent and
well-documented cash flow forecasting
methods. These are:
• Cash receipts and disbursements
• Adjusted earnings
• Balance sheet projections
• Working capital extrapolation
• Accrual addback
Some of the principal features of each of
these methods are outlined as follows:
Cash Receipts and Disbursements
This very straightforward method is frequently
used by smaller corporations. It
does, however, require a very high level of
accuracy on anticipated disbursements
and receipts. Consequently, it may not be
applicable to many companies requiring a
forecast in excess of 60 or 90 days.
Adjusted Earnings
Adjusted Earnings is highly dependent
upon the accuracy of profit projections.
The approach begins with net after-tax
profit and adds expenses that are not actna
1 N~Rh OlJtl:WR (p..P” .. rlenreciatirm t Tt.
Applying Cash Forecasting Methods
All of these methods have common sources
of information: historical and budgeted
balance sheet and/or profit and 1088 data.
Their differences stem largely from fonn
. and time horizon. The key 113 to apply the
selected method consistently over an extended
ensure that the detailed functional categories
account for more than 80 percent of
the company’s cash flow. Once this is determined,
a cash flow requirement decision
matrix should be completed, based on
detailed functional sources and uses of
cash and timing needs.
Figure 1 illustrates an example of a
functional/timing matrix. Estimated dollar
amounts would be inserted into the appropriate
matrix celL Of course, the categories
and timing vary by industry and
Factors Influencing Cash Demands
For the individual responsible for cash
flow forecasting, the identification and
definition of the factors influencing cash
demand (sources and uses) constitute the
most important step. This step involves a
thorough knowledge of both the company
and the industry in which it operates. Understanding
the factors of demand is crucial
to the success of any forecast.
Cash flow requirements are affected
both by macroeconomic factors and by industry-
specific developments. Consequently,
interviews with cash managers representing
utility, grocery distributors, health
claims processors and publishing companies
indicate that there is no unique set of
factors and methods for predicting cash
flow requirements.
Examples of Industry-Dependent
• Utility
For an electric utility, the two most influential
factors are seasonality and
geography. One might assume the mix
of business and residential users in
the market served by the utility would
have the most significant effect on utility
demand and, thus, cash flow.With
a given mix of commercial and residential
users, however, cash sources and
applications remain relatively stable.
Utilities have found that residential
customers nearly always pay five to
seven days in advance of the due date.
The typical utility’s cash forecasting
horizon is 60 to 90 days.
Figure 1: Functional/Timing Matrix
1—- Timing
Functional Categories Weekly Monthly Quarterly
Cosh Sources
Investments X X
Cosh Uses
Payroll X X
Capital Expenditures X
Tax Payments X
• Grocery Distributor
The grocery distribution business is an
example of the combination of empirical
data and seasonality. Forecasting
and managing cash flow for grocery
distributors require a cash manager to
constantly monitor vendor payments
in order to capitalize on payment
terms. The day of the week is also a
predictive element in forecasting; on
Monday, for example, the company
would experience a heavy cash inflow
while the inflow on Tuesday is usually
very low.
In this case, the forecasting horizon
is more critical in the short term;
so, the cash manager requires weekly,
monthly and quarterly forecasts.
• Health Claims Processor
Cash flow for health claims processing
is almost entirely dependent upon the
recipient base and its mandated coverage
Once a health claims processor
knows the number of its recipients and
the extent of coverage options that are
allowed by the state, the company has
sufficient data to estimate its cash
flow on a monthly and quarterly basis.
The number of claims per recipient
within a state provides the processor
with quantified results on the expected
number and dollar amounts of claims.
Monthly dollar amounts vary, depending
upon the number of Fridays and
holidays within the month.
The accuracy of cash flow forecasting
can be improved by using some basic projection
techniques. These include moving
averages, least squares, or exponential
Prelection Techniques
• Moving Averages
This technique simply smooths the data.
It produces consecutive chronological
projected data for a selected period.
• Least Squares
This technique is one of the most widely
used. It produces an equation and a
line that bears a definite and easily
understandable relationship to the actual
data points. Least Squares may
be used to fit a straight line or a
curved line. Generally, the most complicated
curved line is a quadratic or
second degree equation.
• Exponential Equations
This technique is useful in projecting
geometric progressions. The progressions
vary according to a multiplier factor
determined from observation and
experience with the historical data.
These techniques may be applied to one
or more of the factors affecting the cash
flow forecast. The accuracy of a particular
technique can be closely monitored by
evaluating its relative error for an actual
versus forecast.
Refine the Forecast
Circumstances change over time. Even if
the underlying factors of a business do not
alter, actual cash flow experience will
For example, a company may change its
lockbox, controlled disbursement operations
or billing procedures. These conditions
need to be factored into the analysis
of a cash flow experience. They may suggest
one forecasting technique iristead of
another or, more likely, a modification in
the mathematical projection technique
To improve the accuracy of cash flow
forecasts, cash managers will need to periodically
review their results and margins
of error. Acceptable margins of error depend
upon the magnitude of the dollars
involved and the time horizon of the forecast.
Of course, uncertainties will always
remain. If cash managers base their cash
forecasts on the underlying end-user requirements
and factors that influence demand
for cash, however, they stand a better
chance for accuracy. •
(1) The technique was developed by Alan Cunningham
of Cash Forecasting Associates in Berthoud,