Causal Analysis
Introduction
[Note – follows immediately after trend and compositional analysis]
[conduct the session interactively by asking the participants what factors are likely
causes of the observed variables – especially observed trends, notable changes or
patterns ]
The objectives of Causal Analysis are to identify
1. the main factors that cause the changes in the numbers reported in the
financial statements, or performance measures and
2. the causal factors that are, at least to some extent, under the firm’s control
Thus causal analysis can be regarded as an extension of trend analysis, although
causal analysis is useful even when no trends are apparent – eg data that appears to
fluctuate regularly – eg exhibiting seasonal fluctuations
The data that is of greatest interest, in terms of factors for which the causes are
sought, are those that relate most directly to the organization’s objectives.
For example commercial firms usually have earnings maximization (or some
variant of it such as return on investment) as the primary objective [note ROE
is better than earnings alone. Compare 2 equal earnings outcomes – if one is
associated with a higher level of investment it is indicative of a lower level of
performance ie ROE is lower but earnings are the same].
SOEs will usually have the fulfillment of some public policy objective(s) as
the primary goal, but probably along with a financial objective such as
breaking even or earning some specified return on investment
We will call the variables that reflect the firm’s main objectives “performance
measures (PMs)”
PMs are affected by variables that are to some extent under the control of
management (call these “control variables”) and other variables that reflect conditions
in the business environment that are beyond management control.
Symbollically we can write PM=f(C1,…,CN;X1,…,XM) where the Ci’s are
the control variables and the Xi’s are variables that affect or “cause” PM.
Hence the task of management can be regarded as manipulating the control variables
to move the PMs to the desired levels or as close as possible to target levels.
Example
o Suppose the firm’s objective is to maximize net income
o According to the income statement, by definition:
NI = Revenue – Costs (by definition)
and
Revenue = price of good sold x quantity sold ( by definition)
The price is a control variable (although if the market is highly
competitive it may be difficult to charge a price different from the
market price without having a large impact on quantities sold)
the quantity sold is partly controlled and partly not – it must be broken
down further into its causal components
Quantities sold are affected or caused by:
controlled variables:
prices of goods sold,
quality of goods sold
convenience of location
availability and cost of credit (if provided by the firm itself)
uncontrolled variables
prices of competing goods,
quality of competing goods,
prices and quality of complementary goods,
incomes of customers,
credit availability,
interest rates,
unemployment levels or change in unemployment
o Costs = input quantities and qualities, prices of inputs (including labour),
interest expense,depreciation, taxes
Note – revenues and costs can be affected by acquisitions, divestitures,
litigation, strikes, natural disasters, major capital projects and other
unusual or one- off events
o Volatility – High volatility may suggest high risk and it is important to
understand the sources of volatility [more on this below]
Example - Port Facility
o To make the example more concrete consider a government port facility that
caters to privately owned cruise liners. The port charges fees to ships that use
its pier based on the number of passengers per ship.
o Revenues = [per passenger fee] x [# of passengers]
o We observe that revenues and net income of the port are falling gradually
over time. What are some of the possible reasons?
o This can also be written in symbols: R= PxQ and Q=a-bP+cX so that R=
Px(a-bP+cX) Hence by controlling P the firm can partly control R and
drive to its desired level by manipulating P. [draw graph of R(P) and note
that it is maximized at an intermediate point]
o Note that reducing prices generally increases the quantity purchased – a
relationship called the demand curve: Q=a-bP, which characterizes
customer behaviour. The effect on total revenue depends on the shape of the
demand curve ie customer behaviour.
o Thus to choose the price that maximizes revenue the firm must know or be able
to estimate the behaviour of its customers as reflected by the demand curve.
o There are many other variables that affect total revenue, not all off which can be
identified. Thus the explanatory factors are never perfect there is always
substantial error. Other factors that can affect the firm’s revenues include for
example
actions by competitors, such as lowering their prices, can reduce the
firm’s revenue
Income and wealth of the firm’s customers
Interest rates and Credit availability (if customers frequently borrow
to purchase the firm’s products
Unemployment or fear of it
Other factors that are specific to the industry or the firm
o For the tourist industry – exchange rates, economic conditions
in home countries, local crime rates
Some factors that affect the target variable do not change significantly or
continuously – rather they may change slowly and have little effect on the target – but
over a longer time period can have a major influence on the target
Some factors change rarely but when they do they may have a significant impact
o Causal factors that do not change over a given period cannot affect the target
variables during that same period.
o But changes in causal factors do not necessarily have to change within the
same period as the resulting change in the target – eg the response may be
delayed - but a change in a target cannot precede a change in causal factor
(clairvoyance is not allowed)
Specifying the objectives (target variables). Usually firms do not set revenue
maximization as their primary objective – rather some version of profit maximization
(eg ROE) is the ultimate goal. But profit= Revenue – costs. The firm therefore will
want to know what variables affect the firms costs as well as its revenues, and which
of those variables it can control
There are many different costs
o Some vary with the quantity produced over a specified period (these are called
variable costs) others are independent of the quantity produced (these are
called fixed costs)
o A key concept is the average cost per unit produced (or unit cost in short).
Some firms use the unit cost to set prices – that is they calculate or estimate
the unit cost and then add a reasonable amount for profit and hope customers
will agree to pay this price and purchase roughly the quantity that was used to
calculate the unit cost. Only by accident will this approach result in the profit
maximizing price.
o The problem is shown by graphing the unit cost curve and the demand curve
together
o Capital investment is usually considered a fixed cost – and capital investment
must earn profit over the asset’s lifetime to earn any positive target rate of
return thus to justify making the investment the firm must forecast revenues
for a period at least as long as its longest lived substantial asset (and no
longer) – firms without substantial fixed costs do not need to make forecasts
far into the future (unless there are strategic reasons that require it)
What about comparative analysis – comparison with other similar firms